From 0394100be1cebea5a66238abdb95e79a94f8efdc Mon Sep 17 00:00:00 2001 From: Daniel Ari Friedman <6911384+docxology@users.noreply.github.com> Date: Thu, 12 Nov 2026 10:51:28 -0800 Subject: [PATCH 1/3] Refresh the roadmap to the observed release state and current gates The roadmap still described a repository with no published refs, a 4.6.0 SkillTree pin that was current, and a coverage aggregate that had not been re-measured. All three are stale against the repository. - asf-public-release-metadata: blocked -> done. `git ls-remote` now returns 27 refs (3 heads, annotated tags v1.0.0 = 5f32086 and v1.0.1 = c4e34e7, 19 pull-request refs) and Zenodo record 21865644 returns DOI 10.5281/zenodo.21865644, version 1.0.0 published 2026-08-10, whose deposited PDF is byte-identical to the tracked copy. - asf-public-archive-provenance-release: drop the zero-refs premise. The real remaining work is the version drift (every metadata file claims 1.0.1; Zenodo holds only 1.0.0), the stale release-readiness inventory, and the generator line that still renders "remains blocked" unconditionally. - asf-skilltree-460-live-reverify: report upstream skills-service 5.0.5 (2026-11-12) so the pin is not read as current, and name the maintainer's destructive run as the only closer. - asf-review-coverage-floor-illusion: record the 2026-11-12 measurement, 989 passed at 90.92% aggregate, the 23 modules that sit under 90%, and the blocked-status branches in src/roadmap.py and src/figures/plots/roadmap.py that only the real roadmap data covered. - add asf-gnn-upstream-5-1-adoption: this repository pins GNN v5.0.3 and was not part of the v5.1.0 cutover. Regenerated: TODO.md, docs/todo/, roadmap_risk.png, manuscript_variables.json, the hydrated 04_conclusion.md, and the combined PDF with its provenance receipt. --- TODO.md | 23 ++- docs/todo/asf-gnn-upstream-5-1-adoption.md | 56 +++++++ .../asf-public-archive-provenance-release.md | 7 +- docs/todo/asf-public-release-metadata.md | 64 -------- .../asf-review-coverage-floor-illusion.md | 2 +- docs/todo/asf-skilltree-460-live-reverify.md | 5 +- output/data/manuscript_artifacts.json | 4 +- output/data/manuscript_variables.json | 4 +- output/figures/roadmap_risk.png | Bin 111302 -> 103266 bytes output/manuscript/04_conclusion.md | 2 +- output/pdf/Active_Skillference_combined.pdf | Bin 5425948 -> 5419540 bytes .../Active_Skillference_combined.receipt.json | 6 +- tasks.yaml | 139 ++++++++++++++---- 13 files changed, 194 insertions(+), 118 deletions(-) create mode 100644 docs/todo/asf-gnn-upstream-5-1-adoption.md delete mode 100644 docs/todo/asf-public-release-metadata.md diff --git a/TODO.md b/TODO.md index 690c302..fd8bed8 100644 --- a/TODO.md +++ b/TODO.md @@ -9,29 +9,26 @@ ## Active Roadmap -### External Blockers - -- [!] **[asf-public-release-metadata](docs/todo/asf-public-release-metadata.md)** (metadata): Publication DOI and repository target decision - - Scope: The public repository target is owner-confirmed as ActiveInferenceInstitute/Active_Skillference and is set in manuscript/config.yaml. The owner also confirmed MIT for authored code/content and generated artifacts, Daniel Ari Friedman as the copyright holder, and the need for a counsel-review checklist before formal release. The private docxology/Active_Skillference mirror is current and local offline validators are green. Probed on 2026-08-09, the public target now exists and is public, but git ls-remote returns zero refs and the GitHub API reports size 0, so the target is reachable and still unpublished. Existence is not publication, and no commit, no tag, no archive deposit, and no DOI exist for it yet. This task stays blocked rather than todo because the remaining step is an owner-executed push, release, and Zenodo deposit performed outside this repository, not repository work that can be implemented here. DOI, release-date, and repository-URL fields stay blank rather than carrying fabricated metadata. - - Evidence: `uv run python scripts/validate_live_counts.py`; `uv run python scripts/validate_manuscript_variables.py`; `uv run python scripts/z_generate_manuscript_variables.py`; `uv run python scripts/validate_rendered_freshness.py`; `git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git HEAD` - - Scholarship lanes: delivery_and_reporting, applications_and_scope - ### Minor Improvements -- [ ] **[asf-skilltree-460-live-reverify](docs/todo/asf-skilltree-460-live-reverify.md)** (skilltree): Live re-verify local SkillTree stack against skills-service 4.6.0 - - Scope: Opened 2026-08-01 after bumping the pinned runtime from skills-service 4.5.2 to 4.6.0 (upstream release 2026-07-28). docker-compose.yml postgres:15 digest and the skills-service tag+digest were refreshed via the documented Docker Hub procedure (skills-service 4.6.0 = sha256:071220ce..., single-arch; postgres:15 = sha256:74e110c4..., multi-arch incl. arm64), and AuthMode was re-confirmed as FORM/PKI/SAML2 at the 4.6.0 tag (no PKI_OR_FORM). The digest pin and docs were updated, but no live full-app boot against 4.6.0 was recorded because Docker was not running on the reconciliation host. Per the provenance-over-assertion principle, the 4.5.2 live records (asf-skilltree-452-live-reverify) do NOT prove anything about 4.6.0; run the full-app proof (reset_local_skilltree.py --confirm-local-reset --timeout 1200) against 4.6.0 and record proof.ok here before reading this pin as live-verified. 2026-08-09 partial progress on a Docker-capable host, recorded without claiming the phase exit: validate_docker_digests.py passes against the running stack, and the strict full-parity proof (validate_full_app_stack.py --skip-regenerate, see asf-live-stack-ci-rehearsal) is green against skills-service 4.6.0 with 1199/1199 live dependencies, 630 live admin skills, raw FORM login, and the browser console gate. What is still NOT proven is the fresh-volume path this task actually asks for: whether the documented default root credentials bootstrap correctly on a newly created 4.6.0 volume. That requires the destructive reset (docker compose down -v), which would also discard the 1199/1199 parity that took roughly an hour of per-edge writes to reach on this Apple/ARM host, so it is deliberately left for an explicit maintainer run rather than folded in here. +- [ ] **[asf-skilltree-460-live-reverify](docs/todo/asf-skilltree-460-live-reverify.md)** (skilltree): Live re-verify the pinned SkillTree stack (4.6.0; upstream is 5.0.5) + - Scope: Opened 2026-08-01 after bumping the pinned runtime from skills-service 4.5.2 to 4.6.0 (upstream release 2026-07-28). docker-compose.yml postgres:15 digest and the skills-service tag+digest were refreshed via the documented Docker Hub procedure (skills-service 4.6.0 = sha256:071220ce..., single-arch; postgres:15 = sha256:74e110c4..., multi-arch incl. arm64), and AuthMode was re-confirmed as FORM/PKI/SAML2 at the 4.6.0 tag (no PKI_OR_FORM). The digest pin and docs were updated, but no live full-app boot against 4.6.0 was recorded because Docker was not running on the reconciliation host. Per the provenance-over-assertion principle, the 4.5.2 live records (asf-skilltree-452-live-reverify) do NOT prove anything about 4.6.0; run the full-app proof (reset_local_skilltree.py --confirm-local-reset --timeout 1200) against 4.6.0 and record proof.ok here before reading this pin as live-verified. 2026-08-09 partial progress on a Docker-capable host, recorded without claiming the phase exit: validate_docker_digests.py passes against the running stack, and the strict full-parity proof (validate_full_app_stack.py --skip-regenerate, see asf-live-stack-ci-rehearsal) is green against skills-service 4.6.0 with 1199/1199 live dependencies, 630 live admin skills, raw FORM login, and the browser console gate. What is still NOT proven is the fresh-volume path this task actually asks for: whether the documented default root credentials bootstrap correctly on a newly created 4.6.0 volume. That requires the destructive reset (docker compose down -v), which would also discard the 1199/1199 parity that took roughly an hour of per-edge writes to reach on this Apple/ARM host, so it is deliberately left for an explicit maintainer run rather than folded in here. 2026-11-12 currency check, recorded without claiming the phase exit: upstream's latest release is now skills-service 5.0.5 (published 2026-11-12), so the pinned 4.6.0 is a major version behind and validate_docker_digests.py can keep passing on a pin that is itself stale. Settle the order before spending a live run: prove 4.6.0 as pinned and bump in a separate change, or retarget the pin first (docker-compose.yml tag+digest, docs/gates.md, vendor/README.md) and prove that. Either path needs a Docker-capable host and a maintainer's explicit destructive run (docker compose down -v) to reach the fresh-volume bootstrap this task actually asks about, so the maintainer is the only party who can close it. - Depends on: `asf-skilltree-452-live-reverify` - - Evidence: `uv run python scripts/reset_local_skilltree.py --confirm-local-reset --timeout 1200`; `uv run python scripts/validate_docker_digests.py` + - Evidence: `gh api repos/NationalSecurityAgency/skills-service/releases/latest --jq .tag_name`; `uv run python scripts/reset_local_skilltree.py --confirm-local-reset --timeout 1200`; `uv run python scripts/validate_docker_digests.py` + - Scholarship lanes: delivery_and_reporting +- [ ] **[asf-gnn-upstream-5-1-adoption](docs/todo/asf-gnn-upstream-5-1-adoption.md)** (validation): Adopt the v5.1.0 GNN upstream pin (currently v5.0.3) + - Scope: gnn/upstream.json pins Generalized_Notation_Notation at v5.0.3 (8f293d9ba810e6eda5c09842d9c7ed9ffa5f92c1), while the estate cutover moved the importer repositories to v5.1.0 (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf); this repository was not part of that cutover, so its pin is accurate but behind. The tag is duplicated in five places that must move together: gnn/upstream.json, gnn/README.md, docs/gates.md, .github/GNN.md, and the pinned-record constant plus the scratch-checkout tag in tests/test_gnn_conformance.py, with README.md linking the v5.0.3 tag at three places. scripts/validate_gnn_spec.py extracts the pinned commit with git archive and parses the specification with that checkout's parser, so a metadata-only edit would be drift rather than proof: the bump counts only once ./gates.sh gnn passes against a v5.1.0 checkout. Blocker: the arm fails, never skips, without a prepared v5.1.0 checkout and interpreter (GNN_ROOT/GNN_PYTHON), and the hosted `gnn conformance (pinned)` job cannot stand in for it while Actions is billing-locked org-wide. A local Generalized_Notation_Notation checkout at v5.1.0 already exists on the working host (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf), so the maintainer who runs it can close this. + - Evidence: `git ls-remote --tags https://github.com/ActiveInferenceInstitute/Generalized_Notation_Notation.git (v5.1.0 = 75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf)`; `GNN_ROOT= GNN_PYTHON= ./gates.sh gnn`; `uv run pytest tests/test_gnn_conformance.py -q` - Scholarship lanes: delivery_and_reporting ### Major Improvements - [ ] **[asf-public-archive-provenance-release](docs/todo/asf-public-archive-provenance-release.md)** (metadata): Public archive, DOI, and provenance release package - - Scope: The public target exists and is public but currently holds zero refs, so reachability is no longer the gate; published content is. After the target carries a real published release, complete the publication metadata chain: release tag, Zenodo DOI, CFF/CodeMeta/manuscript URL fields, artifact hashes, third-party notices, counsel-review checklist, and release inventory refresh. This task must not fabricate DOI or repository metadata before the upstream blocker is cleared. + - Scope: Upstream dependency cleared on 2026-11-12: the public target carries main, 19 pull-request refs, and the annotated tags v1.0.0 (5f32086) and v1.0.1 (c4e34e7), and Zenodo record 21865644 holds version 1.0.0 (published 2026-08-10; concept DOI 10.5281/zenodo.21865643). The release chain can therefore be read from observed records; what remains is making it agree with itself. (1) Version drift: the only deposit is 1.0.0, while .zenodo.json (version 1.0.1, publication_date 2026-08-09), CITATION.cff (version 1.0.1, date-released 2026-08-09), codemeta.json (version 1.0.1, datePublished 2026-08-09) and manuscript/config.yaml (paper.version and book.edition 1.0.1) all claim a release Zenodo does not hold, the v1.0.1 tag has no deposit behind it, and manuscript/config.yaml keeps publication.version_doi and version_record blank. Either deposit v1.0.1 and record the version DOI it returns, or downgrade the claim to the observed 1.0.0. (2) docs/release_readiness_inventory.md is stale at HEAD: validate_release_readiness_inventory.py fails on the tracked-diff summary, on the listed manuscript/README.md modification, and on the recorded figure hashes, and src/release_readiness.py renders "Public DOI/repository remains blocked until git ls-remote succeeds" unconditionally, so a --write refresh would publish a sentence the current probe no longer supports. Fix the generator before refreshing the snapshot. (3) The third-party notice sweep and the counsel-review checklist are still outstanding. Every value must come from the observed records; nothing here may be fabricated. - Depends on: `asf-public-release-metadata` - - Evidence: `git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git HEAD`; `uv run python scripts/validate_release_readiness_inventory.py --write`; `uv run python scripts/validate_release_readiness_inventory.py`; `uv run python scripts/validate_manuscript_variables.py`; `uv run python scripts/validate_rendered_freshness.py` + - Evidence: `git ls-remote --tags https://github.com/ActiveInferenceInstitute/Active_Skillference.git (v1.0.0 = 5f32086, v1.0.1 = c4e34e7)`; `curl -s https://zenodo.org/api/records/21865644/versions (a single version — 1.0.0)`; `uv run python scripts/validate_release_readiness_inventory.py --write`; `uv run python scripts/validate_release_readiness_inventory.py`; `uv run python scripts/validate_manuscript_variables.py`; `uv run python scripts/validate_rendered_freshness.py` - Scholarship lanes: delivery_and_reporting, applications_and_scope - [ ] **[asf-review-coverage-floor-illusion](docs/todo/asf-review-coverage-floor-illusion.md)** (validation): FIX-ME major: >=90% coverage aggregate masks 32% security-module coverage - - Scope: Progress through 2026-08-01: aggregate src coverage is 92.03% (up from 90.30%) and the previously-thin logic/validation modules are now genuinely covered: summary.py 85% -> 99%, curriculum/alignment.py 66% -> 95%, pilot/contract.py 83% -> 95%, form_login_validation guard branches 32% -> 50%, github_sources retry paths (5xx/network/exhaustion), and recommendations ordering/cycle/remediation. The aggregate floor no longer greenwashes these core modules. Remaining sub-90 modules are all live/network/browser gated and need a Docker/browser-capable host (they are folded into the live-stack tasks): full_app_stack.py ~61% (Docker), form_login_validation.py ~50% (browser subprocess), source_health_network.py ~72% (live HTTP), dashboard/payload.py ~68% (live-auth block, httpserver). Add live/browser-capable coverage for those before claiming the floor is uniformly green. 2026-08-09 update: aggregate is 92.08% over 923 tests. The four named modules are still the sub-90 set (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), and this pass confirmed empirically why the exception matters rather than assuming it: the dashboard CSP bug that blanked the official SkillTree iframe was invisible to all 900+ offline tests and was caught only by the live browser console gate, and the reconciler's abort-on-slow-write bug was caught only by a real 1199-edge run. Offline unit coverage of these modules would not have found either. The regression tests added here are the mechanically checkable part (browser-session restart, post-timeout confirmation, retry bounds, CSP directives, deferred parity, gzip), and the live gates remain the proof for the rest. + - Scope: Progress through 2026-08-01: aggregate src coverage is 92.03% (up from 90.30%) and the previously-thin logic/validation modules are now genuinely covered: summary.py 85% -> 99%, curriculum/alignment.py 66% -> 95%, pilot/contract.py 83% -> 95%, form_login_validation guard branches 32% -> 50%, github_sources retry paths (5xx/network/exhaustion), and recommendations ordering/cycle/remediation. The aggregate floor no longer greenwashes these core modules. Remaining sub-90 modules are all live/network/browser gated and need a Docker/browser-capable host (they are folded into the live-stack tasks): full_app_stack.py ~61% (Docker), form_login_validation.py ~50% (browser subprocess), source_health_network.py ~72% (live HTTP), dashboard/payload.py ~68% (live-auth block, httpserver). Add live/browser-capable coverage for those before claiming the floor is uniformly green. 2026-08-09 update: aggregate is 92.08% over 923 tests. The four named modules are still the sub-90 set (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), and this pass confirmed empirically why the exception matters rather than assuming it: the dashboard CSP bug that blanked the official SkillTree iframe was invisible to all 900+ offline tests and was caught only by the live browser console gate, and the reconciler's abort-on-slow-write bug was caught only by a real 1199-edge run. Offline unit coverage of these modules would not have found either. The regression tests added here are the mechanically checkable part (browser-session restart, post-timeout confirmation, retry bounds, CSP directives, deferred parity, gzip), and the live gates remain the proof for the rest. 2026-11-12 re-measurement of the current tree (`python -m pytest --cov=src`, no floor override): 989 passed, aggregate src coverage 90.92% (10348 statements, 738 missed, 3228 branches, 355 partial), i.e. 0.92 points above the 90% gate. The four modules this task names are unchanged (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), but they are not the whole sub-90 set: 23 src modules sit under 90%, and the extremes are src/publication/zenodo.py 0.00% (98 statements that no test touches, although scripts/publish_zenodo_deposit.py calls it), form_login_validation.py 49.76%, src/screenshots/capture.py 47.79%, full_app_stack.py 63.77%, and src/curriculum/materials/* between 75.48% and 84.38%. The same measurement found a second illusion: dropping the last blocked task from tasks.yaml moved the blocked-status branches of src/roadmap.py (94.23% -> 92.99%) and src/figures/plots/roadmap.py (92.62% -> 89.34%) out of coverage, because those paths were exercised only by the real roadmap data and have no fixture, so the next blocked task silently restores the appearance of coverage. A roadmap-state fixture is therefore part of closing this. Blocker for the live-gated share is a Docker/browser-capable host, and the maintainer who can provide it is the owner; the per-module coverage exception record this task keeps pointing at still does not exist in the repository. - Evidence: `uv run pytest --cov=src --cov-fail-under=90` - Scholarship lanes: delivery_and_reporting, applications_and_scope diff --git a/docs/todo/asf-gnn-upstream-5-1-adoption.md b/docs/todo/asf-gnn-upstream-5-1-adoption.md new file mode 100644 index 0000000..69568d7 --- /dev/null +++ b/docs/todo/asf-gnn-upstream-5-1-adoption.md @@ -0,0 +1,56 @@ +# Adopt the v5.1.0 GNN upstream pin (currently v5.0.3) + + + +## Roadmap Link + +- Task id: `asf-gnn-upstream-5-1-adoption` +- Status: `todo` +- Scale: `minor` +- Surface: `validation` +- Owner: `gnn/upstream.json` +- Top-level TODO: [`TODO.md`](../../TODO.md) + +## Scope + +gnn/upstream.json pins Generalized_Notation_Notation at v5.0.3 (8f293d9ba810e6eda5c09842d9c7ed9ffa5f92c1), while the estate cutover moved the importer repositories to v5.1.0 (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf); this repository was not part of that cutover, so its pin is accurate but behind. The tag is duplicated in five places that must move together: gnn/upstream.json, gnn/README.md, docs/gates.md, .github/GNN.md, and the pinned-record constant plus the scratch-checkout tag in tests/test_gnn_conformance.py, with README.md linking the v5.0.3 tag at three places. scripts/validate_gnn_spec.py extracts the pinned commit with git archive and parses the specification with that checkout's parser, so a metadata-only edit would be drift rather than proof: the bump counts only once ./gates.sh gnn passes against a v5.1.0 checkout. Blocker: the arm fails, never skips, without a prepared v5.1.0 checkout and interpreter (GNN_ROOT/GNN_PYTHON), and the hosted `gnn conformance (pinned)` job cannot stand in for it while Actions is billing-locked org-wide. A local Generalized_Notation_Notation checkout at v5.1.0 already exists on the working host (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf), so the maintainer who runs it can close this. + +## Deliverables + +- Update the `validation` surface owned by `gnn/upstream.json` only as needed for this task. +- Refresh generated artifacts or documentation surfaces named by the evidence commands. +- Keep `tasks.yaml` canonical and regenerate `TODO.md` rather than hand-editing it. + +## Out of Scope + +- Do not stage, commit, push, publish, or mint a DOI as part of this task unless a later request explicitly asks for it. +- Do not weaken validators, coverage thresholds, provenance checks, or source-role requirements to make evidence pass. + +## Dependencies and Blockers + +- Depends on: none declared in `tasks.yaml`. +- Current blocker: none; this item is open for implementation. + +## Evidence Commands + +- `git ls-remote --tags https://github.com/ActiveInferenceInstitute/Generalized_Notation_Notation.git (v5.1.0 = 75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf)` +- `GNN_ROOT= GNN_PYTHON= ./gates.sh gnn` +- `uv run pytest tests/test_gnn_conformance.py -q` + +## Acceptance Criteria + +- All evidence commands listed above pass, or the task remains blocked with exact blocker evidence. +- Generated roadmap surfaces validate with `uv run python scripts/validate_roadmap.py`. +- Live counts remain current with `uv run python scripts/validate_live_counts.py`. + +## Scholarship and Evidence Plan + +- Lanes: delivery_and_reporting +- Existing source keys: none +- Planned source targets: +- none +- Claim-support types: scope_limit +- Boundaries: +- A pin bump is specification-conformance maintenance, not scientific or learner-outcome evidence. +- Phase exit: +- gnn/upstream.json records tag v5.1.0 and commit 75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf, the five duplicated pin references agree, and `./gates.sh gnn` reports clean against that checkout with no skips. diff --git a/docs/todo/asf-public-archive-provenance-release.md b/docs/todo/asf-public-archive-provenance-release.md index 53bce44..85a6264 100644 --- a/docs/todo/asf-public-archive-provenance-release.md +++ b/docs/todo/asf-public-archive-provenance-release.md @@ -13,7 +13,7 @@ ## Scope -The public target exists and is public but currently holds zero refs, so reachability is no longer the gate; published content is. After the target carries a real published release, complete the publication metadata chain: release tag, Zenodo DOI, CFF/CodeMeta/manuscript URL fields, artifact hashes, third-party notices, counsel-review checklist, and release inventory refresh. This task must not fabricate DOI or repository metadata before the upstream blocker is cleared. +Upstream dependency cleared on 2026-11-12: the public target carries main, 19 pull-request refs, and the annotated tags v1.0.0 (5f32086) and v1.0.1 (c4e34e7), and Zenodo record 21865644 holds version 1.0.0 (published 2026-08-10; concept DOI 10.5281/zenodo.21865643). The release chain can therefore be read from observed records; what remains is making it agree with itself. (1) Version drift: the only deposit is 1.0.0, while .zenodo.json (version 1.0.1, publication_date 2026-08-09), CITATION.cff (version 1.0.1, date-released 2026-08-09), codemeta.json (version 1.0.1, datePublished 2026-08-09) and manuscript/config.yaml (paper.version and book.edition 1.0.1) all claim a release Zenodo does not hold, the v1.0.1 tag has no deposit behind it, and manuscript/config.yaml keeps publication.version_doi and version_record blank. Either deposit v1.0.1 and record the version DOI it returns, or downgrade the claim to the observed 1.0.0. (2) docs/release_readiness_inventory.md is stale at HEAD: validate_release_readiness_inventory.py fails on the tracked-diff summary, on the listed manuscript/README.md modification, and on the recorded figure hashes, and src/release_readiness.py renders "Public DOI/repository remains blocked until git ls-remote succeeds" unconditionally, so a --write refresh would publish a sentence the current probe no longer supports. Fix the generator before refreshing the snapshot. (3) The third-party notice sweep and the counsel-review checklist are still outstanding. Every value must come from the observed records; nothing here may be fabricated. ## Deliverables @@ -36,7 +36,8 @@ The public target exists and is public but currently holds zero refs, so reachab ## Evidence Commands -- `git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git HEAD` +- `git ls-remote --tags https://github.com/ActiveInferenceInstitute/Active_Skillference.git (v1.0.0 = 5f32086, v1.0.1 = c4e34e7)` +- `curl -s https://zenodo.org/api/records/21865644/versions (a single version — 1.0.0)` - `uv run python scripts/validate_release_readiness_inventory.py --write` - `uv run python scripts/validate_release_readiness_inventory.py` - `uv run python scripts/validate_manuscript_variables.py` @@ -56,7 +57,7 @@ The public target exists and is public but currently holds zero refs, so reachab - none - Claim-support types: delivery, scope_limit, literature - Boundaries: -- DOI, public URLs, and release timestamps are blank until the repository and archive events occur. +- Deposited version identifiers and dates must match the Zenodo record that actually exists (1.0.0, published 2026-08-10), not the locally claimed 1.0.1. - Metadata standards support release reproducibility, not scientific validity, legal advice, or learner outcomes. - Phase exit: - A real public tag, archive deposit, DOI, CFF/CodeMeta metadata, hashes, notices, and inventory agree. diff --git a/docs/todo/asf-public-release-metadata.md b/docs/todo/asf-public-release-metadata.md deleted file mode 100644 index 0ed991c..0000000 --- a/docs/todo/asf-public-release-metadata.md +++ /dev/null @@ -1,64 +0,0 @@ -# Publication DOI and repository target decision - - - -## Roadmap Link - -- Task id: `asf-public-release-metadata` -- Status: `blocked` -- Scale: `external` -- Surface: `metadata` -- Owner: `manuscript/config.yaml` -- Top-level TODO: [`TODO.md`](../../TODO.md) - -## Scope - -The public repository target is owner-confirmed as ActiveInferenceInstitute/Active_Skillference and is set in manuscript/config.yaml. The owner also confirmed MIT for authored code/content and generated artifacts, Daniel Ari Friedman as the copyright holder, and the need for a counsel-review checklist before formal release. The private docxology/Active_Skillference mirror is current and local offline validators are green. Probed on 2026-08-09, the public target now exists and is public, but git ls-remote returns zero refs and the GitHub API reports size 0, so the target is reachable and still unpublished. Existence is not publication, and no commit, no tag, no archive deposit, and no DOI exist for it yet. This task stays blocked rather than todo because the remaining step is an owner-executed push, release, and Zenodo deposit performed outside this repository, not repository work that can be implemented here. DOI, release-date, and repository-URL fields stay blank rather than carrying fabricated metadata. - -## Deliverables - -- Confirm the public repository target is reachable before adding public DOI or repository metadata. -- Keep blocked DOI, CFF, CodeMeta, and manuscript URL fields blank until release proof exists. -- Update the `metadata` surface owned by `manuscript/config.yaml` only as needed for this task. -- Refresh generated artifacts or documentation surfaces named by the evidence commands. -- Keep `tasks.yaml` canonical and regenerate `TODO.md` rather than hand-editing it. - -## Out of Scope - -- Do not invent a public repository, DOI, release URL, or publication timestamp. -- Do not stage, commit, push, publish, or mint a DOI as part of this task unless a later request explicitly asks for it. -- Do not weaken validators, coverage thresholds, provenance checks, or source-role requirements to make evidence pass. - -## Dependencies and Blockers - -- Depends on: none declared in `tasks.yaml`. -- Current blocker: blocked by the external condition described in the Scope section. - -## Evidence Commands - -- `uv run python scripts/validate_live_counts.py` -- `uv run python scripts/validate_manuscript_variables.py` -- `uv run python scripts/z_generate_manuscript_variables.py` -- `uv run python scripts/validate_rendered_freshness.py` -- `git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git HEAD` - -## Acceptance Criteria - -- The blocker is cleared and rerun evidence replaces the current blocked status. -- All evidence commands listed above pass, or the task remains blocked with exact blocker evidence. -- Generated roadmap surfaces validate with `uv run python scripts/validate_roadmap.py`. -- Live counts remain current with `uv run python scripts/validate_live_counts.py`. - -## Scholarship and Evidence Plan - -- Lanes: delivery_and_reporting, applications_and_scope -- Existing source keys: `smith2016softwarecitation`, `citationfileformat2026`, `codemeta2026schema`, `github2026citingcontent`, `zenodo2026github`, `wilkinson2016fair`, `nosek2015top` -- Planned source targets: -- none -- Claim-support types: delivery, scope_limit, literature -- Boundaries: -- The empty public repository remains an external blocker, not a failed scholarly claim. -- DOI and repository metadata remain blank until real GitHub and Zenodo events are observed. -- Phase exit: -- The public target carries published content and a real release, and the publication environment can complete the actual release path. -- Any release metadata is copied from observed repository and archive records. diff --git a/docs/todo/asf-review-coverage-floor-illusion.md b/docs/todo/asf-review-coverage-floor-illusion.md index 60db9e6..3d79fd0 100644 --- a/docs/todo/asf-review-coverage-floor-illusion.md +++ b/docs/todo/asf-review-coverage-floor-illusion.md @@ -13,7 +13,7 @@ ## Scope -Progress through 2026-08-01: aggregate src coverage is 92.03% (up from 90.30%) and the previously-thin logic/validation modules are now genuinely covered: summary.py 85% -> 99%, curriculum/alignment.py 66% -> 95%, pilot/contract.py 83% -> 95%, form_login_validation guard branches 32% -> 50%, github_sources retry paths (5xx/network/exhaustion), and recommendations ordering/cycle/remediation. The aggregate floor no longer greenwashes these core modules. Remaining sub-90 modules are all live/network/browser gated and need a Docker/browser-capable host (they are folded into the live-stack tasks): full_app_stack.py ~61% (Docker), form_login_validation.py ~50% (browser subprocess), source_health_network.py ~72% (live HTTP), dashboard/payload.py ~68% (live-auth block, httpserver). Add live/browser-capable coverage for those before claiming the floor is uniformly green. 2026-08-09 update: aggregate is 92.08% over 923 tests. The four named modules are still the sub-90 set (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), and this pass confirmed empirically why the exception matters rather than assuming it: the dashboard CSP bug that blanked the official SkillTree iframe was invisible to all 900+ offline tests and was caught only by the live browser console gate, and the reconciler's abort-on-slow-write bug was caught only by a real 1199-edge run. Offline unit coverage of these modules would not have found either. The regression tests added here are the mechanically checkable part (browser-session restart, post-timeout confirmation, retry bounds, CSP directives, deferred parity, gzip), and the live gates remain the proof for the rest. +Progress through 2026-08-01: aggregate src coverage is 92.03% (up from 90.30%) and the previously-thin logic/validation modules are now genuinely covered: summary.py 85% -> 99%, curriculum/alignment.py 66% -> 95%, pilot/contract.py 83% -> 95%, form_login_validation guard branches 32% -> 50%, github_sources retry paths (5xx/network/exhaustion), and recommendations ordering/cycle/remediation. The aggregate floor no longer greenwashes these core modules. Remaining sub-90 modules are all live/network/browser gated and need a Docker/browser-capable host (they are folded into the live-stack tasks): full_app_stack.py ~61% (Docker), form_login_validation.py ~50% (browser subprocess), source_health_network.py ~72% (live HTTP), dashboard/payload.py ~68% (live-auth block, httpserver). Add live/browser-capable coverage for those before claiming the floor is uniformly green. 2026-08-09 update: aggregate is 92.08% over 923 tests. The four named modules are still the sub-90 set (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), and this pass confirmed empirically why the exception matters rather than assuming it: the dashboard CSP bug that blanked the official SkillTree iframe was invisible to all 900+ offline tests and was caught only by the live browser console gate, and the reconciler's abort-on-slow-write bug was caught only by a real 1199-edge run. Offline unit coverage of these modules would not have found either. The regression tests added here are the mechanically checkable part (browser-session restart, post-timeout confirmation, retry bounds, CSP directives, deferred parity, gzip), and the live gates remain the proof for the rest. 2026-11-12 re-measurement of the current tree (`python -m pytest --cov=src`, no floor override): 989 passed, aggregate src coverage 90.92% (10348 statements, 738 missed, 3228 branches, 355 partial), i.e. 0.92 points above the 90% gate. The four modules this task names are unchanged (full_app_stack.py 63.77%, form_login_validation.py 49.76%, source_health_network.py 72.39%, dashboard/payload.py 70.37%), but they are not the whole sub-90 set: 23 src modules sit under 90%, and the extremes are src/publication/zenodo.py 0.00% (98 statements that no test touches, although scripts/publish_zenodo_deposit.py calls it), form_login_validation.py 49.76%, src/screenshots/capture.py 47.79%, full_app_stack.py 63.77%, and src/curriculum/materials/* between 75.48% and 84.38%. The same measurement found a second illusion: dropping the last blocked task from tasks.yaml moved the blocked-status branches of src/roadmap.py (94.23% -> 92.99%) and src/figures/plots/roadmap.py (92.62% -> 89.34%) out of coverage, because those paths were exercised only by the real roadmap data and have no fixture, so the next blocked task silently restores the appearance of coverage. A roadmap-state fixture is therefore part of closing this. Blocker for the live-gated share is a Docker/browser-capable host, and the maintainer who can provide it is the owner; the per-module coverage exception record this task keeps pointing at still does not exist in the repository. ## Deliverables diff --git a/docs/todo/asf-skilltree-460-live-reverify.md b/docs/todo/asf-skilltree-460-live-reverify.md index a260bfd..ac7053c 100644 --- a/docs/todo/asf-skilltree-460-live-reverify.md +++ b/docs/todo/asf-skilltree-460-live-reverify.md @@ -1,4 +1,4 @@ -# Live re-verify local SkillTree stack against skills-service 4.6.0 +# Live re-verify the pinned SkillTree stack (4.6.0; upstream is 5.0.5) @@ -13,7 +13,7 @@ ## Scope -Opened 2026-08-01 after bumping the pinned runtime from skills-service 4.5.2 to 4.6.0 (upstream release 2026-07-28). docker-compose.yml postgres:15 digest and the skills-service tag+digest were refreshed via the documented Docker Hub procedure (skills-service 4.6.0 = sha256:071220ce..., single-arch; postgres:15 = sha256:74e110c4..., multi-arch incl. arm64), and AuthMode was re-confirmed as FORM/PKI/SAML2 at the 4.6.0 tag (no PKI_OR_FORM). The digest pin and docs were updated, but no live full-app boot against 4.6.0 was recorded because Docker was not running on the reconciliation host. Per the provenance-over-assertion principle, the 4.5.2 live records (asf-skilltree-452-live-reverify) do NOT prove anything about 4.6.0; run the full-app proof (reset_local_skilltree.py --confirm-local-reset --timeout 1200) against 4.6.0 and record proof.ok here before reading this pin as live-verified. 2026-08-09 partial progress on a Docker-capable host, recorded without claiming the phase exit: validate_docker_digests.py passes against the running stack, and the strict full-parity proof (validate_full_app_stack.py --skip-regenerate, see asf-live-stack-ci-rehearsal) is green against skills-service 4.6.0 with 1199/1199 live dependencies, 630 live admin skills, raw FORM login, and the browser console gate. What is still NOT proven is the fresh-volume path this task actually asks for: whether the documented default root credentials bootstrap correctly on a newly created 4.6.0 volume. That requires the destructive reset (docker compose down -v), which would also discard the 1199/1199 parity that took roughly an hour of per-edge writes to reach on this Apple/ARM host, so it is deliberately left for an explicit maintainer run rather than folded in here. +Opened 2026-08-01 after bumping the pinned runtime from skills-service 4.5.2 to 4.6.0 (upstream release 2026-07-28). docker-compose.yml postgres:15 digest and the skills-service tag+digest were refreshed via the documented Docker Hub procedure (skills-service 4.6.0 = sha256:071220ce..., single-arch; postgres:15 = sha256:74e110c4..., multi-arch incl. arm64), and AuthMode was re-confirmed as FORM/PKI/SAML2 at the 4.6.0 tag (no PKI_OR_FORM). The digest pin and docs were updated, but no live full-app boot against 4.6.0 was recorded because Docker was not running on the reconciliation host. Per the provenance-over-assertion principle, the 4.5.2 live records (asf-skilltree-452-live-reverify) do NOT prove anything about 4.6.0; run the full-app proof (reset_local_skilltree.py --confirm-local-reset --timeout 1200) against 4.6.0 and record proof.ok here before reading this pin as live-verified. 2026-08-09 partial progress on a Docker-capable host, recorded without claiming the phase exit: validate_docker_digests.py passes against the running stack, and the strict full-parity proof (validate_full_app_stack.py --skip-regenerate, see asf-live-stack-ci-rehearsal) is green against skills-service 4.6.0 with 1199/1199 live dependencies, 630 live admin skills, raw FORM login, and the browser console gate. What is still NOT proven is the fresh-volume path this task actually asks for: whether the documented default root credentials bootstrap correctly on a newly created 4.6.0 volume. That requires the destructive reset (docker compose down -v), which would also discard the 1199/1199 parity that took roughly an hour of per-edge writes to reach on this Apple/ARM host, so it is deliberately left for an explicit maintainer run rather than folded in here. 2026-11-12 currency check, recorded without claiming the phase exit: upstream's latest release is now skills-service 5.0.5 (published 2026-11-12), so the pinned 4.6.0 is a major version behind and validate_docker_digests.py can keep passing on a pin that is itself stale. Settle the order before spending a live run: prove 4.6.0 as pinned and bump in a separate change, or retarget the pin first (docker-compose.yml tag+digest, docs/gates.md, vendor/README.md) and prove that. Either path needs a Docker-capable host and a maintainer's explicit destructive run (docker compose down -v) to reach the fresh-volume bootstrap this task actually asks about, so the maintainer is the only party who can close it. ## Deliverables @@ -33,6 +33,7 @@ Opened 2026-08-01 after bumping the pinned runtime from skills-service 4.5.2 to ## Evidence Commands +- `gh api repos/NationalSecurityAgency/skills-service/releases/latest --jq .tag_name` - `uv run python scripts/reset_local_skilltree.py --confirm-local-reset --timeout 1200` - `uv run python scripts/validate_docker_digests.py` diff --git a/output/data/manuscript_artifacts.json b/output/data/manuscript_artifacts.json index d289333..ef7b1d7 100644 --- a/output/data/manuscript_artifacts.json +++ b/output/data/manuscript_artifacts.json @@ -257,13 +257,13 @@ "width": 1875 }, { - "bytes": 111302, + "bytes": 103266, "channels": 4, "filename": "roadmap_risk.png", "height": 960, "maxValue": 1.0, "minValue": 0.0, - "stddev": 0.22593, + "stddev": 0.220876, "width": 2010 }, { diff --git a/output/data/manuscript_variables.json b/output/data/manuscript_variables.json index e4bd6d4..430556c 100644 --- a/output/data/manuscript_variables.json +++ b/output/data/manuscript_variables.json @@ -83,10 +83,10 @@ "posterior_mean_precision_shift": "2.64706", "precision_weight_on_observation": "0.058824", "quiz_count": "630", - "roadmap_completed_count": "74", + "roadmap_completed_count": "75", "roadmap_next_surfaces": "metadata, skilltree, validation", "roadmap_open_count": "4", - "roadmap_task_count": "78", + "roadmap_task_count": "79", "roadmap_updated": "2026-07-31", "scholarship_lane_count": "6", "scholarship_source_count": "188", diff --git a/output/figures/roadmap_risk.png b/output/figures/roadmap_risk.png index 7d9466f9c3f17128511f0e629fab7d8126dcd61d..f6347e52a60c08e88a7b117b9a53c54628d17964 100644 GIT binary patch literal 103266 zcmeFZhgXx^*Di_;wguRBrK&WMrc_BNHhNQvbWrIfv_K*)*c%e0*MJBF=>j1ELQz3O z4-g0?gh&k#LZpNMA#mgVe&>GsJL8Uf{(ooyuE8~z<)&O2oDbr zzmee`OCBD+F&>@+U;pCe{u1##O5-l3sEV1SYm=>OeAF)+kS=}Fu&_ar=rgADD#JUnMl{CVy3${n!e z*~i0Ubmyk^%5-;jb&ulaLOn)aB*nyOtjUB-_4z|9MN#`qU2#Jw3gp{r4Ph zHej(>;Z)DGaz;2F&^q{YsaJZBvevB<#^}xfs<3h*C%_Zar$t3Y75-fEKg)%ijVICP z{?{7!^Fp7%f3H2sI@ka2RpGx-2mX8OFW&uhv8ypFLSXLA@`OqM*@7zF&-3IxUZeKt zhvQC7Fp1h&?vEc77fp{RBosTp52cU4{2MHqEpxdxHZh=zfAp2lzLwi)37!kQ)V~Hb z2UHeA{YmTZD2gQYCO{f5>Bs;7^(*J+=e?o#pVT;})1$m<4%E`y|z=7rNO&gj61ADW$HOteql zzaH-^|G}|85B8rW=y@gZXlrzu;>&iXu*Q;DSE^d+!9%mA#Gdf&Zu`T6g3yR@lSWCa zInzXKNA-^EeU@a`CWy15YqNh-uA0f2jubh1!TYb|$<39d6ADnz$;*~?A0CH?7y4!l zoJ5NUK>?A=jW1sQFm0K(vt7Z@1TJ2YO?aeJo-9!g44z89qlYUfw|#l^^r^r=M(EnM zVG|t>rD+0aq#v;tLuf<=KeCF&^7x5!=YD;I?&G-t8rT>rwXMu51}4=wt4YpOJ7Hnu zSuhL+OH}haLqKOukej0&d$YB@niw>i%%P(*CPT`FdT?s&x9~O3H|k8(D)3z(dvWdf zap)rv|2xHSsaxN8cr>2OWh?qV2P$Cd| zw;?>*v74E;PG|J0n|3hNwS&EceT&+lx|`N>J+7CA1IJO3%T$JLnmL`+PQIehi|6C{ z-7P}L1ib7Nk1Fq=@nn)JLKkW=C(vFcPYjcZOk&c*}HK74Q=}L08m6(a# z?O}_MK*B2`8_KTNvy3qa2Dy^ahz7wAHhzyNyM~S2y*dg_(thA%Jipkbew7!~w1{Qs zf;E1L%r29fq159SUiy=oyJXud)Ip<)Y(bI4K{5W|rA?!)u<<+fMILAE(T~hZQu%zk zzD$Z6F&h-h(>km?J>RezgsY+%=249H%iM0EMA*wv5XJI;P-@w_HneCeV}VFrIf*o^ zAv$|ZJd-N66ySjtWoZN&u`4(pi|h90Hz7Pc$|YMz(~P;NoJkpe-8EGxM}M|t%usTz zCmw4K-&wG=Y>Yaq=#rRGGiO2-6D%AHKf)vV1J&YXQqEgm23id~#(u(oMDoPb4fplo znHIO`fT@`nNhuF}PS5)X!wC;{baWti@)7=&wAYPEpJutv`K2)55>@f(Y&qt71mih= zW_;_Y?0xwDt>-_6hheFOBuuNvZ$C!*FE$w&l((PAFi^@7t4bKJ6fCffO)dK{En}`3 z_o-&bcK?C>R^|RYJd&>|d+G8`Y+YHZk`pS@si?LiO9nI3oG|3j-a_u8VSi`6wc1V4 zaEa9UF&7M5nrvR3`*5}!gz%bAx!7uUUS2>%>k`+x3AYZd;kE$Bj~_n~vh}WBG+E`f z)qzka)IA46!i!fgjfnD3%}-BnzIaJa{m_*r1+BQCB4c?=1PJ@S+)N)IJ*()H_~dJM zqEuhlbg~YE!Jt-GQLJ*6D(Z`wkw^IXRrq~!HWBkQXDsA)wMT5UqO3`S53XN1?Uo`5 zWP+$9+Bhge(ND3A!j>khBD|eAYh&YD+$y@lss4&>g-h)OK0!OO%)_vK?Q%P<-$UUh56Lv5J^2bQ zE=yy!=b?CK;55T@3%odZel+ISmt+Mi<%xy|hV7@ zPn(Oa!5|gwR40M@OE>F7er{u{XYG31+r7aeJ(+4gH(kifscxCmF>S#xYaGZckLFtU zxSu=p40#O#Zl%iGJ2l6HoTu74fKLOJRuV7i9T`;sGsnW}qqfY+dbC?;aK|F_q}Xv4 z5FF_n$|P@YZy7&+T(>mdG`w6(s`ZhTctsi2IHO=WBC!{2TbW0aSlR$#7EPCz**rW! z-TSu_Z{vvmqt%^#7nKb00ixwN6t{gllNVGx)@^kq*}P@WHJRP%+gRpx7G1EVs}HI_ zHq?K=eDnetfQelTLkZ-8_C@&jWJtUwfrJ_=a%i-{LI>AS>bcTc;&vj@X#ar&Cs6QS zY&p}@mVVV?`0e{QUrH$>%5+6cs~O4W7VYn4SaaRnQzE3U%eF#tbLT?UdbFZTgM4@h z-`jYL?r^=RagL~WWVE~c3nuL8&el@z0n{;SZ!Ts6pDdd+bF??>BfS7bPrG$X()R9$ zg}aGgDE(3xn1K;<7kE2+&Knvg@b}+;KUgeV86r0|H5LXdTb9Wu|2vip3l8E$M>8_}fdYiDbjDRwM3eDOH; z2A1n-abY;MfbZ^G-~o!XX!HFGDa&Q0_lqw_M}{TnG+`ZtuYf?K4Y%jIUFDCSt^DAO z-N@`;JLBL};)+3PZ5ZmR*hcwtCs(=FS26M?X_e}Hq0y8%mU<%e)lY-F_^TAZ2I)d^Cil#g&7+eN4j#nB1Fii~*Z&VJLN>|NPo zulvzBX*n?iW8WmNktC{6kMar1Z7l8H9E@R3^okiRB3gW2+7&J|*=VK@yGYsnoyFY6 zi3=ndh%cV>I&Zm(HSMLBvimojw>!SwHYkq|TkO+s7GHZy63q<_!0-!WGGB!jp8oD@ zb0vdgTe&QX*m(Gs%nF~vEY=0~h10xxGi7pPm`!2EC&(2|4NzsaQdi)WYZTyz`(X>A zEn1xKG)SDk?9k2rTfk}Uy>2(i(~Y`C(}vu>_xjT^rYUBQ=!}zYr$1=csQNfrxIErC zT%#0L7?6ZU=CwYeY}g=Ies=}Jl)pvW6owoMr!l{I>XxzJDB*T4Q#hO0TJwDNSYBm| z_eL!ovOFeNLsd*X6r0uo7mb{mD|dw4q&L+1wSRrhHl(Gp8suH4&mg9^p{;YbF&;592?lOXEq_e6<&sW z3`*acuJloUIHfvT9adCadIc2q*`fZ~`qIQbu_Y40;{ak>oj^$zGx-_CimipMZpkn@ zo<}pfi|60LUI~!JlF5-9j^YGLhXrLq7K24e^vqw=?aXPqR#_Z3C%C?Jsj_9ejEQw@ zcoe6#x`}XeqryS6MS;5Dh;qDsj9f148ZlxIk*CxJwh%PEs~NhUvHJ>T?zxpK%Ws}` z#okG1uu?*FIZ%p!aTu*rbkm(Q-KcJyqf_0p92O(WN^8{Kaiei$cy+vo6E0t*#KBw; zfY$yN@R{`3ANA9~ZpOvpz2L$YLW@^#N^p#6YM8LPBVm2ftIVnRSuuNkv47M@fLyS| z9V5Rmm)W^?_@)$M@NULVsdcGCM}Q`MWOK@k&mUKV}4cK7=H+t=%yl$x$m^CyISTyjv5Nc@rdRJIer;D z8unu47d1C|A15M{orLVZO23sVYo+#!y>2DgrDCBowSw39-SN$Q?S>!K4Ro6j@t)|n z&H~3|ompLWto#YA>lT%}UtSxNYnm~$iw zMD=wkQoV=wNvY8=Xirq%`%sdGV(@mk_j{xl1y{xjbB^YB9D{GQ;$zwuYlX|0uAE9g znJgKx$h!cjchcu-8CJ~Ys@Y+6OA?qe(yQV-R3^Q)$B`pa5^}6!))#C&Z@jRdo77ot z5qwqeL>v;K%h{Hz1Pe8x+a9hrBHiCU8mMvP1LydWo;m>kiemOyD1)};H6L$&W@3lI zr#WO|siV~x%2yiXppjprWGTW8D_4iz**9MEzEbsOu6({m#O-G9KuZperNFnV3MDXM zxaE@J@dJW7vfNZAIKo_bd;LLPTJT9^yLd~q1UD}lQBaxR8ZPJehI7-pOW-*L^|WEv zaC4XQTJ1g=TuU0o^YM<^+*psM_PSn@yI+xXrYEb8S)VJqgEVDFrdkD#DmIBF7{x>> z<;H9w4$qSYR$1q-wrepS-!)vMf(vAKM4!7m1fSzD1*>7LXEIi)@rRdJBBo}1d?t`s z<_!-M!G^JqWoh+mXS&8t_r^UwJi7q;aCe#4(b6}vZcQ6P8=p}9z6f4q#CUeKvnZ&p z>W`-(2v*mvVcJAtljQ4ysI-o%=-Nw^V)kvw%?&YUV0(JM1VTRqr*9Cx5i@fhJT+-} zV8L%VP&v&(t#q-a4B(;dQ0(qzm$;|oY;H?T)(nfsGHc@jdOfE( zg4y8n;tPMZ-!GE`dd1FO100K$DLv=qz-{dv)M(N50l=@uLna=(-#u1|G^Ng?cpWhY zyIFlbx(Lv!$PfHN=M*MPfa{v}j<{MG_<>U2Is_TBn z&r&vN1lk+=nn*1&We-Piqajcc6H^jqPoh+*!$t=BEzsN?(yOW5N@_j)x2%gnZ_2(#T_B!LZ(@4ua>H_8E%mUkyzS8xezw?Gw!)xvhY1nYUt-lOU zOiCC{qhr0JCH>w8orG59_e|}Lq`iN-I3P{ z9kWg*ibpadZ~e(`$MX&x7pcK4-l#t7dS05F79`1~Ri-NgQy-3C_mUhnZX?DL61TTR zIBE-lDvvwRu+lKF!rG#m@C;>Py6|z;rGV(UgDn zef%i(ez|3K>lkOqHH$sCmY#}%I>?EVjh8Ja_}!S;h%z=q1i0dsN!U`d<`4njqL_pP zK#Bcyw=Mo&`h-#Qj{lv+8i~i-y2%5x7$z>)$~oIn+TWS~LyOjobJ|`E$8*g$B!ZqaH)60KtaIs zg61@AT;|(Rv9~FGjC;@L0Po1?d&9vqi5rXB;fp9{^G2JQ;Wj>bM={Bb?Rz^S%|+0$ z=E-GI?cJUbpXmh!o30C;0eG?~42a-f_!vtP`{+Y*9+$1e2 z=kvnldrhjK(b0or3Zbs+J!yNfYUK_g2X|_F3&0XBsx!`5Y9woEDfzwwYhQy>&@t5o z!(+H!t5)4}QUs^>mT~31pY~jF(WEB^xwX8bOglHMYai<|w5=$(8SNp?@Qo0mpF&Eg zXw^~%7`-cYpCx2MNytH(Uu(if35LH~j$A~mx?#~R3RJ6Wt-9vbGZt6HOH_MMju@6_ zb}@x9cDV#|$Jx%*Y8Za(UiY>3irKJn9Bsuo-VvSO;#8&jVxCnAe*aTMY5$Z~(Za^p z8-71|hHBS1s|0ASlx*Cm!khx;d7N7VYyvtF+u@MUfAhdLt6HaslH?{au-wEh&)2r@ znloFn^!TX&x;>z-< zki<0$@&vIQCjWNFgrY4#c@Hr?Ti+7$aXs~{llaC*J<0_ow&jw{#p}?&#FRr{SYZo^ zfg+?(Rh5AQoLp5Oq+32N(nL7cjciLgC<6Rg@~yI&eoABrkX?wt*{C%R<=ZZAfn}9Z_&1E{0+pdoTv?(m4i?UE~({!1?pN{?$saJ zujlJO2sG9d$`V2N)}(3Zj1d}C(mAj-iq3!OY4)WLG*JPPqj5ASyL}5gw_1d802GWW zzN<6|T9svr8~bxJtCMgLdOHiPjc*sng{RE!=*7g3X^v%MB!e2R^e9uWv{x|1NvE2} zj<<0erE5`B(8}hb!iBz^W2QW0X^*=@ou>=$tbFoBxn~SW(e({|GQTJP=2%RdwT2<| zK?SxgxSr2qUA=do+qS;eEP6N`+L1ad&(~k&x@!SnraVw|joNM|!~MzIJ<>xT70SJR zjXM!pX`Q2PBo7vGHYAk7J6ac!qxu;tb147*U6II}J|(9}Jo(LgP}dlFgALFJH?~G+ zRwda;D&y{0P@ybDfbQa_6RQBq9(+}xsUn?D4I8X-Yt1cuICzrlp4wLx(zQDp_a>WY zrel_lx*X+;z3+<4quC5qz#?@(nNF(eEcik^#q2s`sv%+GJGP!YXg%OKtdt0+)2(_1 zqUm~Tb+eyjL5U6gC)Qlh917cs;^3%KNk1$KevhF#g);~XaNvg%f zBG8q7Zm^t{5Baw9T6_sKu!vo2vd94t-2eVk$i_N5rgnUD_Uo}eL9^{m_=eYhI1m*c zLhIq$}&-kA_h^B56+PR}8TEuiDT72bU|EJQ=gHm9gEdr>|j zWCx$@+Ee98q)gLuhRuxufraG`Y9;M55)}1?^&jj9f}ccSd%W+ za9vgrMaVGpwx!zJcl6t_(OzF7iXH(luO-E$vk~jS3_ImsI#&H14j+ zUCRSA=<_FqG#Zi?jN|3dxT_%Vez*0N16ii^wdVf9OT_6^^^a|O!UL*mO=5fKKqr4V zAKA^Pu{nGVxxUD*Q{(@*NX}1+Vh`br#17}heXc>qr?o~BGa_DzWr-R@%zJ|iWlw>0 zi{KNiJK7yp%b)n;BH&+QsJ`HS>)&!`_9PgM%y)yOC!w8S3z0wb^ma6jDW^0R@6^xu zwvis-^TMPv#>pxJz&jcig8{WZs)^SO_O9`FP@xp?)zY&QMA$k!U&$s#4{Hj;cu1#4 zi}mn)>pLGkIrl?iigX}12@PSs?Jp=ez#-k*oXK+{#N;&x)*#>m@uk020PX8be9O*6 zz}5+=cbWAI!{zA;Me+W%=&B`B^8iXp6P-6UR>OLdVfi!FT%v?g$$YL06P;H5a6w1N zT5PZDoYn4-iPzYn4Cc-q%(wY_65VT#LY8ib7{`^AX>XuU#O4!fYp4QC(pu+q} zamXl4=;fi_lNE{pP$G~qoi5wH5v{cX{2fZRYLPKJUKv))wI0wc#WzwcMWjW-!gm;1 z=v*`yvy1Sv9nQu)cuU^8kLjhCJrCf*obtN6%HNim>(C;q%=RdZy<&o0`%w9F*T33T z`lZa?Z6n#3VRKg2+MB#e((q7NP#Qh_CqtKnxqW@WN0tdif!CC48tKXDRsH6|l-(dN z-vfnI$F8%xqF{k4Xw%Xt`g`{9Oczw9FA(^*$wHt-+7MVwo7E3axhLKvQSLY)p`-Ic zD`9nWNa}fuv?emH1wRK7E(%pCHiEv%*Nvvk?kvLjZJQ!P)z@zLy1`wxSqsbS;UdFZ zOVg8{6S)0bYZ(V#sb?;TXf?Fb`g@um;)B=czw33;m!XWY2~}R-m$w8@DK-5rAl>Y@ zb&VKvFdD7~-hYvMZiJG#XhqdknE&vMu<|4`8BgsLuG9{ktGfr#ou%d;!_Ybk^EW}% z!nY=9fBP~_YvAjwkA=vo`+&+vv&OTtKd3oZC0-mdCx*Dl6VuH5?7rgLfF2o%}8=BVtAbpKM{-_uIRn#X<}s@%^c-R z)OGz*U&%g!h_Run5VvBq>i@*}M;hrA)0GcR6m?Q+@wH%P{>y7g54%aIl_+`15Ye|rq#^@8W`4@yb~+Y;)|3nK%#?% zspx?oK>3UE1csoR%tWS(lYt3MXSeOG6_Jx&E()B+e?KV>x-p!qenF0d8ZV%N&ldeL*|F62gSqGx)NNZXCQjHacQkCiYC7XzyJecPSF~B5t*^~F{rKdQ#7o>7tkwBMFM-QeYcsc;`XMJw zuvN`D&CPD2G2dc9J%AEYqFK+%s=3zaHxY_Kk;a!nkSGOrvgLq*QF%PAzRxXwUBzQ~ zZ^SBIFi2=fjKugHT3o;l*kwH(dLcyi#f!G?D+B3IFQ=+*RTah!!;xr;S$%b(8Q;M! zSP_R)zBDl9$3t0Rj?Z@N-SthTq~=f-14lQquUrg`uqOE&4Sn+M1*JK`_;i;&DVljr zyR!dl%S*SPLrrXfOt7dngGi~)Db-qk5U5g%bc;Ok;Y4~qxmES@?1;Cg@P~!*Nu8~k z8sHj=*6S*K(_c6(@j367D(BbM(X*ywda;$wTSn%&nsSu*;EFp-Wa!sr*vYJW>O`Cn zvDTz^VFYU^XXw_<(c=PbS6&N<*8q90H!OYVUHh6R0bvQNeV#_abv%Is(mHX%6I;g| zWc&1^NjV@hkkIuMBwWkmq!Xsc`okce+7PuN@oD{mO}ViiRxSJwI`x; z&Q1SwYx7~l_)y6b`kpiEX|$M-yS9Kh5N4?(%t5@5+Us#wcunLFjW|@E>r55lH)}6Kb9iR7J+fyVXwvHR3+XOKJ~&k z9|+mq(WAY2x6F+>zCAZEZB3bt7NOuF?v2w`?NtE}Qv>br-8ZSeFVJ$LB+9~$usF%B zig&b@@KFYND|iRwDAdk{jC>ISH%C1$oZGzXimg9}+td8fJyb#fE{!$ex9cEBLS+i0 z*=)y7MvQ4SFH1anN{^=VF|=a}x%}xVJGCe4r~Te)8zlHY-fA9(?4_TurkcW|#d&@$ zrw|@?7SM(R(;KWUk!33fLArxqE!t=0kfvyf7*gnx1lY?w=!m+{-|bbq({^4#Wl4+S z@#bzG;R+zft;wPlq}ybLHhXCdEV^L*0`TB_Vz?m0)S@vM1pGI`@?4P+eD0#5N%kMS3 z7*03A`1eX6UXIVcHSXIPBlbt;ITsSKe)Y@0_ZIizxHlLiLoD zs@|?rAo)YzPX3qg-7kV{WT>9(tQuFy)G)gHz{#LQFxsm?Gtdg)oZ}L@S0+j+wJv!u zmJLMtA?AQS7u_!PEk8V2)cqx^6l2}-Xo^O20|oWH5*AJIT@Xd495f4D8j0*Ch(*>_ z8d366&LCX3xITvg@+`pTpXN9(5&Q35M?3Yz3MF^#ULzeTLZFp=bkL;%Hqz-8pd6wYkgfhwIwoAOsr`Iy)>T~>bUR*Q?T(0 z&-lFuH?w3#8*$1TEI;1pz8rNFm(+C~`G#+xyu{{8W`6;o2mz|3&j_F#_wSB|&K)s?7nxD}6CY%6YBh;G+x640;)I#X`nttH%!G2Z# z#i2waD~pcQa$tAxgK{&6@SacZS74}7}}A{#zLO2u?BUBky&!o;q}q zx6LqQrt00BH>ezy>qI@L9>fwP%Qr_$%+Jb<3z_~bUt zrbmq0R&|UxqID-upGKfI3U{`qd|VsWwmVjUtE&7twq<)8G9bm1w{7nPbe(Ak=i z)JA<$soLYpv_BPoL0vJ07>nH>@`44Dmbl1~JFn}Hs|C%aKl@qr=~tb3$Ux^icygGj zF6Wy;{wxLc_mi!`LMO~s)M2=59YW<~ipSb~YQA7f$p`f+~@xC%&HxqRIac=GC z+X>mENLlyj=&5DLPl!t=t`u|AP*Q^L8Mse~vyJbg`~@ZyP5x%e`Q|L!!7wXq` zLVRAODmr@1<(s{q8R4y3Il)%U%2SMJSoFpXj*GqJh^f8V{ry~UL;?GtUm1Jq3!u}C z&@FS;A}TFstQox)PusZ-089k$CI1a)Ulk>%q%+6=tH#!jA8zJrDbhc^-v{9)@nlzg zx)VKVG&VXv^v$A+!y+5mB6Y;neySe(QA_DtPe@hWta9EzmC>cx42X>v4oi*K+3Hl} zDUxj$q}%%;OnOcdkZl;s>K#aY8#f5?GPAY%I~$Ttm~aL(jHe=BV+!63AHOSfXZQwo z;A4@wng<(ky(0IcTSBW_h0yZ&me~)8Jdt;qZ(DZ(BH`li6q8y?v7`R-wKQ5V37)sq1PV~@ zYm|bS!qjyO?);KMU$$VAV9PpcoLtSSikje>Xh7m_bi-(zF+w}=UbaV>(@O77%wn$n z#+tdW`V%fi%GlTFkAnug!+X`Lebx}xu_e{5I|@}*2c>Jgx_sw6`rq^FMyVK|>CKQo ziHq^VEN#bS(U&8E-7GM02T~1s{-W9TOxAN`L&$82k9TcdsO+6NrMjdO?MWmFsb&L& z%;i4hXpUh7Z_d zMr4*06SH{uFlXW^su!Z@nbJYgwJ+)mnvEOIslir$D?1!xRukE+eh3lq=)smZ%k_y$yx8+scPEam=I>f zXkBn@y>n%NnO})sY`+@}<=A56#=a5x9j><#??Y9$IL}(MwM4Ze?i~TB5uZH8+ZzOi zSLYP$$>LPE0nfPQsxY>CX4X zEu#up*zIm;arMX!G&vy2N?lY<)vRzjP`-Q*7iJFO;!_(%PPG{>T`MB;O@9iW3$f;r z#a7=&GF8J~?F6XU!1HZNhr52+Qc{!GyRG6!hD=b3Tg;JMdTQ~#p=K9#=)|L<o+|$9u`igZQ)$03_sV-_}R2UoUR3 zyCIxEBk^TinKAq+v&FyKH#~Ei;XOGu>$3=T?=Ov%!NRKnQr`w^G))QFX0o7U#?`*34SfOuLg_{eWax4evq41PUa%G}W2w%Fcu~9PxQOy2z}%Vb9RG#HzGmu-KWr zd<{Xs-)H0?5LY~unw7glFZYQO33$PTS{kFE(Yh%jw0dG=VnTtltFl(S4Sdmb$re`TFrGTx3HM2&JcR<1nMjU5gP> zYt>CVjQMv<&nDefa6QJxR>kJYeR(F3aJ*@oqj~?k$ny|*{+rzlO?-DsX^Li(oj^d& z2=C$ON~ii~OZxnN4cT5Uk6RnlcGa=#5p9Yd^)eU(* z`Cr-p?#yA`bUY|8pO_=OgLEu@#n$d4FYiuD;0nUCmX$rmOh;=YD`oR@zG(Q7Z8pZb zt8xaTT`glub3XJjL=8<;uX}98E$&)T!@=3K@`zevS(l1;r#m~E%!J3Z)i>CbCH-El zzAo4{;<$M9#XszX`<0^?_yok%*@XJFvsrjTDdnVP$*pMY^sNx^IBt(Ys5FZ<1}Zml zCUC1HrO0TjwYw+vIep|TPpH<)xht>Boa%ErneyoJB5XIHBjRfTXZsW^J8%{ zofvkmfBCWez;v$_zS*;ai6!uQXd~Nd%`In+3aH~Akp?p|LDZEZSEJR>*R-kUm_O`P z-WloM#_Pn7#&!4ipEDC8kER*}_ZD_yD;ACRWvG~76zSuNug&@U>%tArUt0xR;q5RF zYcg2Q7{=z|4W<}rNo(v)zSYyw0l+);bAHcVFca~a<`7v~F1vvK+Xt9wMCAqbBeOIO zhB#F)bDw4DqKl01*^)3%>pFB-8*xQl7J&8x4NsJgej5v0mit5c*|P9CU%O^RdCl8VcNnVd z=*{MqkrhmZnS(7}_IQo;$G^K6F{2*m!Q(AFvQIU~wC@-wUx%x1J=heMJ{>1V*;6%l ziCl}uie&~i?LK-3ABb37dF{@5b(Tv$z@gknulTuz)ITTHV{Zsdx94 zK~%GbBA)};U*l!lP3#eKr}ZXofa;;RQa}9g%067bti#QE3=?78@qIL@TlpV%RVtJy1W~hsqjWiY zZN$y2Wcp-98!SFvdjLNh1#kwEmGkkE2xWw)q&1J`han40r=;5&sNPa*rC-Cv77>(_{K7ojc;Mv);y!BtUga23=9`;NWaX>22(Ni0D(1J*$rBY}R^=lgedpp=i=87sP&ghnH&{ zxCzzFt#lXYzC1FQ7TgjA0(g&lz9QUfgG;=%N8-AV-;(pBdsPsr_|Gu+TYFqBl;w)Gn6>ZS=j)t{iE4!@XDO=^? z&&;74{p{Pb#Z;FX+SiqBNAc?lq-uMsqJ?@@$6}W$61T6%EsKhM(te@na-?cv_>>YC z{y0lpbp}iXYA`;)h3uL`$_okvSQzDe+O*Kn?vR&wrhaqXyg)dYWIi)3$XFT~`jg0} zg$$&^t=}qvKHfhC+ui^dr!k)m~ZAu#LS48EK~b_iJkf4a9l>q z(0z&>U`g^0Hvr?ZOVqBel819g!fp(7Cc{&WFO#{5IG184(eFfW6`|fV?rc^nZn3cl zqQ)F5;M?7WDeus2%b(JdiPhmJi}sS;5I!^Pz|J+iR_aIdr0sA^=7cHxtCQ$hH10= zm(JwEqa4K|3j!NkPM6Ef2g}u4vN=1Pvk+z`m(t$u$oQ#Eg*vT+N0A#fLoF79;j2}^ zI3U{il+u+R256}&!txePF~WS9l4t&V^{9nW$B7?Kimpw*Qc0HTQsGm&6KMW5wm;>Lc9E;}h{rSkPzDG~Bdp#@fkX6`&`!DQ%hX@&yjwM$O1@f)aWOHLy zbF3m@wx|jkL!1RoA4qiAeRgOqOSZu>6DC86<^(q)3OPHB?B~*OgtT=}q6S zf{HUpc9DeIfFaXAa8riu#DR&?q;wb_ezxi#b|HQ()8c{;Bh!WqfACOw^Y+3 z^wb_ycSkV`v-H6+FAv$j^bMw65Y`2h4LWg)OCNuNM5+#i3*)mj&#sX(D6Ro2uEj{p zp$A9_&V-klxYx74x@^l+j-FKxH} zO4rf>>Zvjwl(a#LZxpEw;(H=itN!^3PRo8vljf*!1e}20b2d9&zDF2SE6u7G0#RGc zdqcdbmkhSWqm~y8X7eqYt4;VTWJl%mPouiWWEUp+B7ZCe5(85STq zp$jJl3hrqq8IM;T_>g+BBGikk8p2>1jJ~|C6|j46_Gkr$y)&BRzQk(z87k0p?vf^W z?y5w27;5)NY}yiWh*#KP+u5l`^EgD_H45)__}G`o_(pZ`Rb(5D2E7WM%k?Zs=*|1g z4mC3=L0lFSA92-TPxS`t&$cf+LE-UYCbhxz&9^`+VV~%)8(88!2D~Lw;klg9_^x-G z<0==+NS7^99pa3F&xT$`Yodu|_w*tlH1D(RP_HbhP&5Of0?21AAwZ0+rO=m+>U85K2Ow!Vt|<_xrk?}2@@ z*{7k+!aZ|tdk#=!(aKDW6AN4^02_-E+>RvpoDS7J3j;bGIeq2AP|5v7z6eOL(BafU zE(Fa};f>CHJKdsE+~$S1m!xR~q^fxJX1P`HUh3J3VqO_Qhk=V7euQ~-whg$I(}Pv6 zk7zzFQWfszeO><~tVEQw{T_0Go2kWhjMqlc^$B8mxb|#KuyKm4T`xgWr|G zl+5mk%zK8zT_@7@O753mj@KQq=uB4Rju}h@833pA3Z)Bii^oZ#cMa-ut_NFIdv!^3 zS-I-W10dGhQrHLG>lSNrKs?zwas#&&$q%PSuCJ}tM!MkToH_7N2g9EB$nhS^y{IWU*Px)B>$#Xd z05V4-F6%+=!!ipcDryD24)mj-3wPAE#O=A9eGy==Wx#ly&p-*aCySIF=uP@e^r(f= z;WIme;UznN=p>qzgJp*uSQt=J-`a;4FbnjFYJXORsHt%ME;96Yv3X0-+MHR^(;Ijd zX1Iyx^`R0ytI_)KVs0*4_V*t#PmLEIw-X=Yu0g%5;5aUU7^C2}cr9PrhB~y<6MD(p zm6%62=3d0T96Hz@#2WZ$#zC?i8GiV|qTEn~fL&`)@H z!%78LaL9e)8mlP0`h6#4{FTsrQ<05V(iPHFjCiAmWr_rdBT##@(6}6kI1_dd&HI`l zvD=jsynM^_e*G`44TQixo&`SYUmbG!M9u#pI1|e?UI>_I_Sr{DrDwSW!$VjgSAEoh z8&?ST47hT*tKdm4-Ng&Cpr+QvRb3ehAu6{KEZr0_XOhXR@#*f%mqkUZBN;H2BO~hV z6$mj#U=o0E>Z`qByN5~yednc#^6In_UBA3d}e7D_z&k z)c=dCz9LUV^UZJ)paZ!$&vrR{#zO1|^{CAi@2yFyZnhX>wET8ScgG(+E)UOxn;%7a z1}-4tedj-%wZd{GtnPoMqE>Ht+*g-S@+TCq2reuFrR_C9IdrMi9|J%)Imid9-i~JxJ;R@ zhSN13+qz(PLMkYPD|V}gabCH1W$GCxUzTRlh{*v6lF9P8j~862?XM%9KIJmA__OkH zLW2p^9I?xzHC}EfxmFu5P1w^yo)rg_y^QIf zxVKtB2Uvxo4j%0LE72|?;ojUB4o7-U=_bDfaRsbLH$@|U)*rn%2e>YXaOb(uG$nE| z{2uiI8)>@(8RhC(sh6RYjl!4z((UqS{O!&6j|u*H``?AF2cAL%)74^L)^oRBSOySu z!D&@*6MsEwKMmeTx~jGUbQj*w^Gh~cLQ-iF3`D#yF$r-3v=Y*Sq>;tdSMiQQI6ZXhA)gHgsaFrJW#d&9C`fsz3NGF^c3x6&wak2 z;dgJ=6UPG6Qa*pa{Lcp4p8tF3e~j_}(dYhcreAUkZ5>u#?)c{CAzO=TE6kOI zovl^l*>@gNCJ!Do(q6rK*?}wv&Mws-P3)jp7^lcf&l;zhCWQP;udSS_D(>A9BzvcM zbMEYDV^|RZoT0pBQmIHU@EV3xRkVcua5%p<{kT4O8cCj>?Mh9w&51N!Jw!E1{U4`6 zs)~9G3JRv>F`}=?nCDz$mD&_v_z`<5kt;H<22sT{!5fV`8&uV|@82czH9&(LVwep* zs~u^&hM}RTQ*xGuCAP&ExDx3^8$XSV{*~JlajD$s>(r{`Js4I(^L{bm^1s ztku;-4HqFv{zc*!owMi9%jUk{bxEz9xM1bgowAG|y~|}F;!i@@==Yoj!ISPSMfaS~ z?=0O|S%GIHHbYq(tE*Y!?F%YY_bF7&YywQ%a(3-sy7htF^medN*iM@)*tyL6BK87y zb!DXr`MxDg&RWr_&YiV0#_u>YRrq(3%Wkm&+u`}rm6#xxh2hfJdhNi;khgE%7(TW5 zmzmrD8VLQUZLr}#H%98}nq6~ofxgR?Fo0mG>H!lEEBtIqtNtz&axm6ST! z$PLleppJjt(s`Rq0n~7%J&P;p%$?9Me(_iD4u3Rxj?yBI%E1dJ+9prUKNA9*) zot$e3EuKM zUs_!`OX+HebL{!?35F0_-LX0#=}0Uyr@egjDudRYK{se1b!K-^0!BBF32AlDPM?;u z{?Hf>?y5g}?OQaDVdb;6wh7Ju8t!<0zu-#oJKo!`xwGNgC%7$qx+^WE0>OZ0Ub?@q z#>Mke%(5^YRnsy*@OX>PG&zMoy1QSE;g6!@NtnQ{_c7ECuHIZyehF=T*;T&*>Z4R2 z(!YC%r|#%}dVhcaDm$0JZG8Kw*;y&I5}U+JFOL{{dw=E<4{i|4DDv2irL??CEd}mH zgV{_F_3?k6_z6l{`HFFcYsR8}nzlz3OoU{{; zh+GoeTVJbE-W|SXiS<;Ahe>nEkFsWbV`*h}e$SRNb~qkwU()_$Igr!ct~iT?VwJhh zHx@tJq4@^s;lm2bUcHhJF@YxDc1!s~fiGVaM$N%&Ot%c3E zgw^h@z^t;|>1Oeik70_g^;bCJF-AK(Q66KPRc~P;zfrSt!|mKkiwRdgYh7x2uOT!* zhAReW&uL=M;*AIhWcbjNt&~#RLNl7DC0E%eYyIWPix*C*8QehhrcssI25xWi`hVDatGKqd z=xdbvsd6fm0!3P^c%evfDtLhc#RF+^m*5hhF1QAVqNPZR6Ck8Uk>C_kTtf;0QUU~r zyTYm6`#s!;`~6?O-`x*~1C8vx_S$RBIp!E+c9@tKlU(3P(@4{GAAf9c{jg>8D=`?) z)T=^BnYOWni#c4*yEp}gpA|#73RBi`LR;RQ?PnWh_KZQOm&axYGQBC(hg%wvXoFQv zDsSyYL`1~OIt~ig{CLx4ut+yY=MYeiLcKH0=5~z488amLLx8%Xd#RO>T3v^r?)o^Uv{D5r{wcqrc2*Qe7cQ*n8)U zq~F|=sbeiXyPH1*z>3oXN@i1D7tZCaGC!V+ANAR;sc4;l^TxK80917EA@W}_h~bM1 z8-b4jOo6eQ_WuzB$Z`61ez*Pa-w#$b@gU#3qcEg>?_O+*6!Y2pv?u;~=U=S9(o*Wv zZ{J^T^?az#7h4x!L{}h@u;Ph6cFNA`=V`H_W`@33-TcH6gGtxz>|&O})YQkPr>BF} zCH`W6S+hhs+U8M|T~T4RzkpdyH^h74n|QO`pcs6sJjgn6)CYuq{jyb_gPQKc*>3Q> z6~;a!dwqR`)KSSU^O$yH`LTjmE+rOH|5u0?2GE=4C&XK`=YgF zBKQ;%Zd9owS?fu}6+2H0bSrFE$!kRt(*2tJI=_05CwG#ZovZBoRJH!g4*<5O%p*bU< zp`1|bZO;zS5JYm7bMu&ll1gflo(AK3XXn-a5(nf(Nw04Hya)r=gio^W(|5?E-FmR8Z%6Zuib{v7eS7Mr!Q|w)?&3VtTLkBYOZ~il?_?)wO4Ga3>8iIavr)(_ zpvNz|fyv?PYncX7Km@D2R%ZY%J#4+dQ1^|3oUhCG`#mFx=ds7G%+qAv{N${jWz9_W zg6KQB)GHyc3Z#veTgcG_JKBp+A@!GIa@*gH;DWc#f|Rm2bwEAZVM=jqutfTwe_&P>9#{TXpF?*^;z zwduRDkDo^*hA1{qC{9(-r@8ueQlnvT1W#Rk&~oBzbzf)-?JDWPqbVs7(*rN1>;oF! zo?NVdcIx)VSdVb29Z{1<_ic6E(#8^(X@D_5-~9&+MLX^OtW-iUe#IayW^7ER)O+;C zI(&=62<3;a6PYqyrw}7j#f_sr9`Q2~HUN-I!Ve+k?z0Ael$>+S95=lCN# zY6>GFs<5a*fy=wZ>*hwMXo62L*c#-hpGOF%pF*IIoNC@VpxmnB!Rq4DoGekLN}tYc za5sfmc!qr#Z+4heQ*c!juywK^!VbgxGyUbsixJ{q#fAiT1IJAD#EoBRo z5~r55M#|(60w&^;@adE%5P*)7r(nE|al1-=HXYOJ?K~jGQct2SH;Lsl?S`CTcZgR& zRk_JVAb#%3G(%AUg^clo54Owpjd|Wot#$3nhRM5;Lyy8RYpv`P1Y$B5_Skh^qtfa^ zt8~swb{qz+(hZ@3#xV|ODH$=m4ow8b0>hIjX)i7P-oXY4N&q70E^E>Hb zdn)RD*JR`e14m45TlDskh=dR_2*Ckd*b;oCCSm;zGccKI4oj|Too`LN#_1m^%VTv2 zg4(S=Pc;)tl<~}!`G(A6XmA#u9)yBNCFabz*jp_Fp*Pb-wxMN^mlKNfW<=n7Bx3pf zA3*)p2U2jW&&trIY@Tb`U`y*RcY`mjf%t#wafDLx;Bc+inEHMi4u(Htb>k$lPZSY} zH2@jHEr2xaGpcZoDKx?udCm6=LE+?ZRVOEP3+<%6dPJnimmV_1>(qUdVQc8D7fH9!_JQy0;yyAyfi=g3q@1JuSPCEo=Zz|1VGLqO_7Q1s z+fRstG%6ybY|O!*&&YdI5M-!4t`VDx+L4_a>=;Fm!R+aEz*#kuD+Ce9yOpsgf|d>) zJ@HQ@&u(i}`WFqgGBCQnTU|TU(g>$}?j2xx#FZ%NP=Bnxhh;I z-(0e+73}nF{`o9hmX z6fV*IG%8X6;Q8+q6^^|O5BEx>3^z6^AER&O1_66)$R&h@DfGY)3B8he!hiQ#;zkD~ z%2xOTsoYTieVk2Nz-^YaWHJoOVm%RLY8J5Ib~XV6mxl1h3bE&?$Bn7)qwzWM=VtGVii#rt$iQW5 zqmy*d_KHW_lTE0YBd!>V09x~=iCv}V2mxmAtM4E)(d3cN59GcV9hS49C@!|0?~i4B zERFUVh{pMY$}gwSv&~ip0^{&+)q){mSV{yjw0jo|`a$lR~7oj&`>_u{F zF!jA$%@KXSU4M8GByK&{?J>45;~&ye)MBgvbk+3YH|n8q?8hP{9{MGazX0bH-)7?7 zlDI@a)e`5xW#zSa`DwP@WKH)QGl5!amD(`EuncDYBKrJe?1pNPA!U3^_tLYF@TaK}t^fdLSjYQ0CuF$e>Cs zSJ)k&+11b&HU9u(wTUWhr3SJ+=YzWf>8s$+S-SUQ=IHOcK>huH4!x*f;14b51NjVq z&>K;q>FLO|wKYSMvXT<;wz*x1-Q3)KStS7%{&RBg&p)J`$aKwTuNTxk0XB@6$j+~(g#T4&h$8c?L;)AUWV1BZ=uN^c7zu%QgHIyIioh*r*DPc_)s zLmytkG7ME8jcHpsfUe2*(*AR{Y`1{A@u6MI8W4@E-B$?qSop>@g3rp#)X9eYc~xZE zY1@wXJE>&~H*XfJ0GhgLWucLU=>n+7#NOn5tL_@rqLXE&?z>$?JGJ#RgvVI_&wHu( zo6UXTRw9f;`+J4EqYfO2P_Z>kVCV{5{&p02t&L8l#i!!I2E}s^I2ZmpROG?i?gCgN zJ_l={{`%1&Fi$rG5x|^INU7vT094aHqq?8n7pS zeF_bYU8;*cJx#O4M`%y$KPw}$QZK8jh&$(KDf8p!d`RDRc{n_6GxaA9SHBZJ(mDG8P!BY1|#~F?e;3hr`nWceQGj_4?Ez`Oixg{FE|e4;X`@ zccKH@EHz2zt>xlvw36k$SQLOh$gVSg&Wa7r-kG+Ks%nMEcNVB;_R0Hve^Cg<^bc`4 z{RO>z1F!l39N{#q+mnQ-EFiESfno2?`PZ($^7+ zp*l}>z})8q>y=6u`41kVU`$9JI23H7kw`%f0c$c~sh~n}G@68X^~BFlPpE+;(Q?j! zb#dj-Sq2~2KhMw#m;#)-TtPEEwGJs*dnKbBeX#)_jHVQGI!)A=`;) zE%GBOaPjc$pCK7@zMUcac-?!W`k;_Wg_`1?0prgv-~c`s_oi2tu&}BZfom~e1^55? zW{yPfZs=dX2iG3?-}b*7rYHJUp;F9IX*(n50fgZJr~qm#%-ZyuI3^Q*t+9kVMPT zKy-n#o#oPu0q{$|?85xK0Nl?zPD}10*Urw4qvhVf=v<%7`e zkm-3B7onpqiR=vO$2lOCt=b}Q8V=PqhMO)U5=|Di`iSv$mnzhUN=?! z`)DpuPP@B$?j%o|1cZ{lh@jgnO(LwOHgiPm+`Y{u?|u8&A;ljRml~+i{Kt3 z1YAZC)Z$`*9_Ir@uFaO4~7?-clvc&Ub8pipW>cj&qgdq zd98|-$<+hkvkWfoIO4qjJ3yScEwn~1GC ze#D7+55cbVeXh8s)#qv;Yr7LX+y|69@dPL%Z_t*Q#t1Xzt!2mg9-jYJ;p`KnbAF&o zN>gR~p6$BJ!aWAz9qd{&`TMn7Q)@4E?TQ^2*B4h`0?Wt9a_6qk3yZt(ozzyNcXtFK zCxlavEML2QTd5^}zSgQqnw1LmrJ5C(Yvj!K9@F5dPcNv`ifiVI z13FvLuEJ5;Yd)rE&>PFX?RP!wGtxM}?p_f@6R50sQIBD-5C}wytoIG2pv?j!LNaTk zI!LD5CeJ7N)f>ee7IEo<%?uZ%PBo|U(aF(8il{Y+JogU{Dz~JB#hZ)^?}jy9sXv{~ z49GOrm{bX^g|CAD70JArsWD#gb;3s9{lZ)ZZ%*K?DZ>iv1+fIr*x{T=>=~BRQEz-Z zS1MM-%-tuaHBhjXBoibPG_l>Z`U(I}TAegXimf`aPeTwxyvYFih-hbGWfG{Z*Zdr0 z_43s#kwKAS&j}HSku!@Fax}<>-TfAN=~)os0;DJz)T6==%K;yJr@|0vXXNFfl0oCq zsO$E9<@G$cIz_%SdR%PiK{#*B&@qhrz`bIH)}4JSPQc?Q=96-`Y7f`RpP^Rdu{{%5 z^OdFQ=29w#FFePYuaA1vvxLJ{&CXPDlM#IDjOuLxJ8? zw!}%4`=Mr}g?xCGW20t@$e`0wrD-3H-=g``9Ie$;-S0qTXVti~{G1JW zxSR@;9gi=Jm6Mg?TJwAu%EqfU)zdG$RK}-qX;)zzn{Ujekcx5DP756Oq~-{;&oLjA z_R(ZJ93oprs~%v=@e&rvcBjAD>gk%6v3KJc&ahLT{zBuC!A^(}q_hRU6cYbMW@IAm znKz3`r5e|rv`cv%DKWswk`v(g)>E6*(n@cYQcGePA*Q@xmH9gVt>;1ZCrgQtNh#9W z8)RT?oP&6cU>=iIcLrQx`p&c$SS7hSix4uXNtcA{u(>C7Thq@=1AD~JT6V;Xg&a*{ zmGrwQx3<43tC%X{(b%6HfCN&{~#^1ZDOs=Id~{F>uppq5M{osA_%yN*UL0 zOx)su7K<{`Q&+y=d+(`_-M@Uy4aNJWK>?;&_rRtaKpwpI8K(}`I?kNY$#%U#oJ>mE zq9$V_UshUGY;&^8xivQpz#1kLvzsQ+A+vKovucA1(gA2E=f*H2dD&vSO=bcV!edzj zw{_2d?!NRsRRq4$Tr1)?C&Z<Y%tu~pe;!}vY2_{+J~v+s+(s!0je@}q^-!m{HsZ}2p zalq53Qr=`buTgk(_x3~uvc=woO2F&FznE2fRFqe*w%7TF#223jYm?o;oA6=s`1rVp z-G-x4wMX1hIq74&E+gd6onPnPTCcQB07uoYJPMEdmBUj#0y`}EW1fb~8T1z1i+%r| zCXc`baz_UVM>nH-Mm;_$S!)K#z{xLd2ho`D&MFTnl^2U=cjNPSTg4LhrTNwcrC{xk z1$g3TD-NP7+a*aYu?)xCZ11oM9O)lTOVY}9td8-_bb0)*Zx(wG7n`S@&09fP)_CBz zhkdh)=G)%%An@&NGHh{xML|R3#(VE7CRWI#CHpGQx5l%cOvUpSWAZe9Tn7~^FKiGQ zPuqv;7W)@@!x%c6z}pDigTZuunZq-w7d}TJ**#zwU$`QPwS1kmqtOqu3@LMbuq$AL z)w@3;Qe)av%l*X4SK|O{`si*7SCGcG!R*&UUS$1`C!WLKZf~t7EeagTw$awU{3&-D zVaG(WeENO#d-MsrAUQDaQsmtA!Tm6#nmfHO;*w^fDIM=fIk2s!=7I2RZK(kd1!vf{H$UfI{wG~ zLKC|@*QX5V4iQE2XYY&mn`$_nvoq*Xy*vrEt&v89LJ&Z8|naXue*_m_as5mB0>uJW`rXG-5T z!EvxenQL3tq^_FVwk0^6aP|7H+*?0{*u%Ip2G={a8p5QPE>*HY8+=U5)^ z;)L}?n+_=NzJ2@7L`uf2t@)^@4WY~hl5;y!OTr1r?0%nT>sGzo}SYGXWEKHr$a*La>lC;+Q6j80+OloeG42g#1`T^_DwU@gIr?>pH|3nzi zBUlg?jatoMr!Du_e7urWV}?z<2sn5Tx^bi-bO(Nk(bC`uEo%%n^ujDGk#;1|=&e4l z%c>!$WF8%H2p)+}V^L9mg+GeOGpT8x8VlEtLna@O?5o?F(9JjfFIXPiznAB*^%m%h ze$Y=n0MUJcxT60V4|~i)rX5yUq{ZPq#-DoiXe=&f#PnH|e-cH7Sb1r2AB=OKza#{~zss`BcBO4QW+rH;f5 z^x<)r=g$Xo@*|2I8DFr7O=!fV!MDFS!Dh&GXYV6H0OVmmszyFS@H94qHsUT_3+|4> zZ$RStA4dP1bhv(9_9UJZM8H10rJN09`(7PHhla+{@ufEvOjRvKCa-hLexw;}DDVO`E%lUx-5^#diJY?YPQ_^G*)dCkI8C?^Fh7AMXE|)U&Z)qOK;950{C(s^qC-(NEP4IM*wtg^VwO7$7Fx$;2-k_>=D) zk=j?4!IrIXIMcm3XtUv;h`K%t+7(I@@#nuQDDCAQKY#n*3gfM*F^c>U0?pZLh9)-n1WJix|F-V*Sze% zBv`$i+?fmijxGCu$}V`4Lii0!TNV~8m@1pTGPvVGG!au%@z%Gz_4kzk%uN802Wq{Y z`2-f6&Jma7Ba->4Xf^OCRs*O@Y8tEjA}e6jfMn#9lZi^(Y#e^N&4k=BcL)>>ZmfSV zbHOh~-O5oJf7HCC`sZV57!<*Y(YK~|Kp-fu?V5`+0bhrlXD(j|;+ZPFlPruIV!57y z@Tdb`lc>X<4owB0AOS@j*mx68$^~n^b9RXLZMJx?RLt5?@Z85*BJF|fPiswd5dD|-{x5~8~3?!aNR<`0AeO$)4~DTYyeA|`g2$Q zf==1kRS*v;x0J)1L0aE7uV2zS9=P?{4$y&AE+IOIZzwkpS#)=^$}^Pl|2-rkopM3i zt+o`Hm6{qTF`5F0$`(C$=5w*^eYj@2*d7QJ1rso8`3r7s@(JKfJsR;MGKUYBI9QZ~ zVreV>=sYTp13V84(=Vf=9n@%;^E>*;NtS8y;ww5{m619We&}iBL3I1@tN8=j@DIL^ zbZP+=^ygt`R>Ra)@3nm%1u%-E84`Ae+ULHm=boSfB{o9=BYoz*4q zUmpQ@qz7m0Z*M+Tzf|fwDCk~mbg=>1p%@OE_{xw+Ahbg5i^XiAr9M%XJVJlo$~LJg@elyj ztIhd6HX6a`2HBI7re?}@2=MS_W@h1o(t{kB(Z%{j!&d;0>{?Z=XN{22-w~`QXH8hCa|K^w14WoC9&ZK>Pl}c*ac{Ma794!FN)dn2 zAtT+gLr%|E-MnShFdH`Qh4p9R^g&N39>?afn<0Qfw+g^%!Q<3=u)pl!VU?QnX z+Id!zN6B9khsN%9&4n%&P_ZM`uRlVWR^D@~>JtE@+)y7PCl!-O0SWNd)>iiYo%`;8 zJr)YjPLv-xd{`xiY!$uCC|PQAZ7&P_IY`lu>i)upX{sM8fK~JN;ryoz1pXHMzqYpd z{Ysku|E<5q;_qQvFs2gSN55!AA7S;?SJQvPedv`z9JO7DMj^n*y9G!86qm9kaL7?= zP^k|1B;5dra4up3ghqZysaill9lrj&2B+9i9UZTFJn}RY%4#=U_j0Js{;LBIXouky zp{R5__oXM;@dgxsGjWGJ#=ET;mOfo?PGqkgd`JPUND@7_w*qtpIs@kiJawOexNCUj zAqDgT!KdY(%+XuQEd_4e48U61&Kk_$NUyzsckjZ6Dk_VimT0v1=m*gXunViy{#E5` znIZ`}8IM?WSAvUTgYO0X)OWGnLb)yF&h>k%fRIx>ziVl_nUnJsz3@I~6Ppgx$S@Nx zl?8}oj@kUIAk?A%ttV`IL&MGxnX7u1H5qg*>h~Udmm&q%uY(6D=q%VkAFaAbg<4K{=EWG z#Ke0pUUoO^O?c?5Sfi!*Z)l~gjS3C`?>M_?f$3fQr?SWJ=x}1tu6xp*WXX<$iB49{ zXRfS@xm=D|COv@9bw*8Er;%IeuwNK$cP%F5scX>L!f-ICYv0}0*Fs46Z!dMdA|zY} z6w-XRZsd*m1h|(Y&8mz7x8Rw;5oKgy(b1@^=`pQGTxHRMn^CwPfb}~tTG2hHh*m9A zLINZw4lUUM+)z5k_9?$nN#9W4e704I9;l3-we|M)2aD>E#_!Q@`Zu>df$wZyn@Cj% zJs*F=G5T{tLPC=MJC82P!ET=CH=0>|?`mk6aDr(O1wr{~P?@(N=n#46%Pd@P+i#Bo zT(xKQWRFKsm-a_F#Cez3fS=o_>|j&>5gw1)Q_r4*gF)Sy|bOMci#rI4VCAW0x3#SV3e9CNxpodcRpj)cOd47CzSW94<$yX;HUjsaLO;578KaTFwJKUn zWiBQ(MxgdhK!nv?V+P8Ag%I=yKj;CX9$%cqLoazK)OUJu$IhBM zzS;D@c^{VlfQ7rw6D`$lH+JB_0g&Z{MdRVMUez6fO~px~g%&#Ri5E%k^nzMI4((B{KH%h91TSfco!UapM zh{3bq!)5>aaGy6W68-4a;6lT)50GQ@EkblH@$tNIEvo2&UJgJjZXE6D=rDj0j=|O; zJ7aaz8=(Kk8L&86l2;>^D%w+PZBz4P<&tWKC$<{Yr)vzmZbqN4BK(dYBte;78G8b( zC^~=5TR@`_vXBqbniez^Buco%Y##}4dKqJ52|_q()(f&*je#n=g|5}*MapR~rx6|N z8zXjsVWFX~Rx9|kdYJ(0s=uruRR`@jGtoDwtu2h4pP#R}IAiz!yoDF~ORcFa@uZM; z@4z2FgpDfg(X9y;3iaE*kEppBZ@iENeL~-Qn+1s9dasWPiSpkmhM>8?ZVABpp{EI_ zXQiTLud=zIbr@h9!50zntq*kS(k0!^TH*wG@+)L}3aD#6$V%*S zmnI(Xh5;|1?Uo_uo*LBF)-WNzbZ_yLk$3>wu}+&LH0|%L~%gv?H$mdvcAoa^13vTbJ|wOy-c5(=~@~J zt32>E2a%JLXM02o2S4i$UPIR~2bz5RXf8OAs4!E3FZnSzo?p3l4!`@xu?_cRAltc% zZ`d^6TSzkA-U`aXI_Xx^3+brTq0xB2dpD|m5c0Laba-Q^&E&Pn^mDhPek4q8yQq6S zu?ZFek~L$3F{a9G`yg5q5*;z z&%QF;Ge)8&7(=K>*~WPCnG}htM8ej<3KJJi(L-w&AtAean0U|BLZdR3Z7DI!-hnXk z_?wb$^2jrtH}?ZkuH(^#kXmiPf3}{!jG<#ZPu$$tTyO3zz~K`SqfzHq#bZHbW@;+G ztn~hj+BG%Q?UKvgZ;i4J!VhbF7(v1L66M_Mw|)%XHYznQ`uf&P5wtWD#1)$&01?d+ zH?nP=wv3eX#WC6$^Og{_Oym?*RNCDGtaF^GN3fRk7@HP=Dzr6dgPIPfLtg@0 zZ9`1NvhY&AR;sWvWkKFjuu}8_=?xUwPVTa?NR1Z){7fMF;?$2auMz+H{)YF35h7_Fyu`yo9Ah;m8LQolc!(+=u50;shh9xarMlBJw}d)u$TpJnW-sOw0f-GQpU#if-v zf5u)hdS$r37YwBdBAYf9ufbI`#<{~$}FHm{_TR#fRu&v0Jua#W%5UJv}2`l&S8|M5U0VR2eem;7xnqsr^SwPrhY zlAi~YIY3uZ>PrmvP9)O&6+Iu9O@xz82zNJ_I&GljnBsV=^1<72@9&9D-{~i z10mHQ5#-XG>VM09ydykfkb+ZWxxoBGi;ISt#n`9fyNCn#^F*sN9%d5Vc=PYs5ytD_2WK>xK-NDTm@Xjf+9Gt>V9 zAtLK@))i_GcaLs2mZ7vGl00^;cI!gjQlnC*;~0QD?`$2ASbuRrZ7aQmdD!_M`^E zCTH#H`+@lYaYvy^?@!kffFjYz;PH==5*dV;Lu={pYXr|bQNe%w(VF_=)5Fl#R58|%jr=BCqtv*?vI%8VPX zfmdNyPFM%@FgODD0DO;m7=oB{@YT$_`&q8Hro|Wpz}b75&LC>K+@S? zK?p{fqHY@15~u{-pgS9&0(+g4T5163aKEhlI;nrU2Jq&ArAV>m?r#Xw|0h491S}y6 zH8Hu$6eL{7UT~PW|)G<9}|Pe#ec} zZ~8y{TYt`p|HhP|5omEO8{1`p4YI&VA}$_qRU(jQ>F7B9S6O412tt7EngZGp7%c*a z%QY#Q$mqX@YDf{Zf#wD+J8qm8H0Cm?B1Hq(a%t_Eo_h}Hy~kprtdS&gYh-&&SJHDf zbfpC7^;~m{f=VU!q#SMN841oo^N_0&NRHr_Hv`K zybEl1VzlIUzYLvfPgZswspA1etK~H*K={;4^Bq6wkO4VB1~OyBV_@!Wp@rrEWy9{0 zZMYZlPU;S_Zn1CEx#$2*{==J()loPwmuqzCDej zwn`8&(GcPnklNIw;!8klwa#Bc%J6~ag;9w;Uz#`K-0>he2R0O}zVL5H&YbrebWYWh z+Z-tJq+zT0YZ0st`TAB{5A^<1vA>7BY-VQGT-pxw8B-q{)OE^LF8O-?=;2av?kaXt zw+jm5lZ0;+@|)FsbWR1Wle6!>-hb5W2#wD1zXCCzuVKl;c6fIx5EYO%BpiW+phxUd zGg2ha)}5lsK0p1-!z1{%l*bf6_{vnlF3b*iRC{Jxx&=vJ3-x*WJa2ybXgMCOSL)Ca zh+<|DJVZW(l?@}R=AT*2b3S$QWDM9ylC;xnoB`HlJq&cJoVJ19-dLv7r!Vp0oTHg* zDJ?mXe9@9I7^6H`1I(J)#pMfZpKB3S0qloMgTuzg_PV;8D!@Nv;-@K~L2F81f7R~& zN4R~#UvK`xKWT=}0YWS^NM4f)(=HE%N`_?XhqaIgj%xujwRJ6_Jw;l`1-+X{<+m&> z0{a5h-6J)h*VCxrXVu8&o%S@${nfS6-F8GEx;9~bID%Q8b6C1+Ma38N6Sr^i1EE2l zC)cj5@+*1)4j=(@PO&=dy4pgYVvsaCXn0=02-s^bBe+@pI}f&Zq)7|OZ%cj1jx#VQ zhLoA)xn=p4n&Sr!BEiIY`U`TPskZxylW!*L5vjy{yWzqT2JJK+T`tLhZdM<_|h^cjue(@n{#m11dz`vT1=R=xzdUU|?~wqu5_&kX>6b?uc`txAh=WhQ;WQk5D`XfgI}+KG%bh%t=L8_21c;sjg~n0~ zoLqm}{@}Kq7so)A+zhdZ>@-vFQ#Qc(L2#f$q1y?7lk{tCskrFxXhIg5!qfg?)yyc@z{b_y5>p&So9Cz!mJZBtVY=wWDGxMMdXqd3&O!GPa{=Zmd zS8=A&LpJo~m7TslVy-(Ngae#YFNZO|hS`(THJ%|{pvNAbMB=8E>ukuj_f{mR8aNq? zRva(o1I6_iubmZdIf||Cwe`(pBLBmEf=~%ijaPkf z=XaG>2tlks4|X%YEfzkPc}Be{AQU2x z?AD9eEjZ{Q`JW!2_D0FkR09fOvWRV{Wk+-bEx7W1+2nGWr6A`SX(WvqOk_T z)J&Q60-I^QPt;AS8k08u+<)ufd*FcHTQDtEDN3sR_|e(K1&Rs$1R=D=VDY6OYo-zp z5}yxutq>*JJky452;x@6iPNG^8EL#OqiLzF#3EqfQ4g4H>t>&`({Z9?Du@D-ldB}A z^xqZ?ur0?{!v08MQxvTCvqeBaz#K~uF*Sl}Cjw^ryK>vU&kHCZzEA~x9!(@24s0=u zRQnKM^`R6%_5|0^u!Jiq*1eohrBmQWcLk5oHZHH-8z?IS#8?qm1)-xSn&!HGB@0d8 z0_xx#kjqphWwv3|O*cV@WV7K$3F~q{TU*0I?C7zRp4(~ho=*G6!_EDNKgS#N$YK`B zY}ekxKTXe5DInBBz|lIQ;t>V3>=f$Yp}~FG=rzDPUu0H$2kYZR!d|$%R*v#Om9+wfV14YPAs^wO+dvYo_hkV3XBPmndO^VRao6yO)dd` zij!T^WsT&iEB=&1H_!*nEEDjEIj^!Qw8RSv0%I5vSqXxGv0$ENdfe&=BY$B1jP?a@7=|7O@B*pSE)&5rXp;3#4xqT1tVGFvp?NTk*w)e1OI4x+%6VyLhnbQx0g) zo5B)AX^<$YY7AVLp8@uf(S6i9Du7dUs4c#SrKDQ6%Oaku0}Z7l+6*9zAmi@Y+h#8f z-%fxAE(kZzUe;4_EVt#QBwxbO@&hvfOsNtD$b1TRs)IFU%9By7kdMirElwT? zegIiE1gHw{73paEdCcFJAP?&gONm)6U}7vswiNgF-di=SSMHy-$@i{27(h!_gwDWI-npD@%#dDZ$5M3e)kB-%QcG4FoKQYsj+Xq z+g?leSEpyES{N-Hric}b-r)zE=$HDc97SsXdgsgps|e0p zzi4Kb_dM`ZUK_H^ZvmkQ+Fa|IM-C?aUY{XP=^1Jyc-HY7A|;T!-Gmy=h}F%UyKu7Trd--2gC%bf=Iz7 zm0cE{dPg2mV*Xm*Y;P2CQTCDE z2c%@68l0_6s05vJwuIZ$YVQ{deDE+>mNZ>x?N6}!iKei<22l;a3wk5;39uWwzusn2 zb*DBbTo8yW914jm-vcmIGTKq&(NnaDtq;C!FWgpIEw=H1lpA|R|C^m`>Q1QvjzX=meIJ%E`&a(?{Y^l_oR+$JmBNzo zaG!9;@Ubu>ak<|v4|n?8OAHaJl_=QphYzg-Gdka0r53hYO{<4^I@2U|xniVvMCZH; zkw{{rxJ?wFh0;my^))JSUyS|6KP2EVg_lgR7pPi9d!n1 zrs3(C?j1RCz|2!P_5xn*E!A153U`MuY$>7wfhCzcwW3c9L6~b10vEfj+*7Y9Q!E7l zt6=Zo&;yFF96&yR5vK(tUBAC^97KUxxeZVr&m}JMxHJ+lFIky@W8s)HYZps;@b0vs z0l-4NI_P*H3RJ`laHw5T#ezYzUya;rZKEDFLw@OkRgyrBvgJjUSu|DRF_r~ffz;yP z)9@Hq|GN*UX@bwi7ay3fU1hmSCGv5CNY<-EuK;obK*U!-F*owZi+>iEDt6y60Z(1P zXawp>K43Ma+6DUw@hO-e0)9K77WVRm8Mqc|1TedHjnJ+RV16ut${Pl9J11vcv5_J* zf=F`#_07b*ZE~E02;iFLq53@q*?WQDN>}(Vl$~uj!j^Su`#ONq3PJVOl#Fi^0NiD; z4-!jB?;ik1hSb|iN=aW;w*9@`j3odOrn|sIdVM7#pNlOBkhAmkvR(~UyXjiDnV9ci zdAM(1H^zW)Z{6|daDyMg%T%MP<5RTp0)t-7@@40%1L|F0Z;`vN9msM0uqBvI>RJ!H zmT8KE)&+#2z8#)``i~!NP8V`1jr$9Ojaq_6zrRkAu#t7H+XJ&Cx`!~ar1N%t!U7o$liCuPLn;P$zHZ%C<-(4Q(vi6MoB zwQL8}LDMUE+Ga|ke+PQCwZDs*B0<;-BUB#$zLajZ-$}8kr6Nxv=Kg~dOu}3+7|cA6 z+5h(A4NrAq_kROGb65N^1(?xZ83`a{U0(|hbo;TheQY&o5&xokrZ1?m6wen;&?~t2 zk}uF#dwG7iY}w=&v}YZ zc6rU`eTL;|nGpJPfjcZy3(2B5+ny>d&%-2Pe|KR1=PV3%di!1G2h&}&U&ZyQVDuy& zHyW1gyzPB2A0nw=WX5b1$J?O&d;yxh@FU718uom_R1MSc>q^c)qQowBNQz~$O%$!l zhwBySE0=MFO*~>vV<|Eg7b?8w-(Gh?I?!Cy?(0>lN|!Ruf#3Iv-L|8>>!aVY9aW6V!Oq+_PDTU*cEqQuVjzEv9R6Ue-= z?f29H^=k!tdVD|`-2AH1?M|g$jYH(BAupEz(XpjAyI-JF>~MxG$b;oVL7xwb|1FBi z?n?Sv#_oH|De0dV70b<(`h@f?OPzYsEJ4$FpYgb=KBm&$dFbxTRrw83MzJ@nHtBv4 zyUJhoN-dh5W=QAM>K$*)3#kf+23cB}v?nvfmR&B(AX1iA)`Q1>Zj)k|SGgLDPA|oO zp6q+^n7*L}fh4h)4#+iM5CKc>cx}RkeZCuvKAWo=MLd~1Ig32npA1^78=iT9X6v!{;T2Vai+M-iQXQOC3Ea@H*Qpt|0lz9PqyC7b)7i#~DMmeecv{ ze0R(SYZ&m|RBz5ww@-e9H9Yv}okMIzr{^73MoRp*$kD6~Z}kc$B@A{Y0bUpbC>5GR z9&N~IZPKs%ZI&aO2`@sg4so6rFgxQtvZ_)VxEY4qk}?)`F|GmdUNFpjq+4x~OkdF> zM|#IhuRJ;6%VZifwahDQsZ-I^hE+*3Xlswq^)iX2z}NL>kAK-d5^z4NF*M@Qqes_w zpWEE_@~YgkLU{AOkExuNSx$rnKHES-E5x@O^`m zpCaVKtgs0B1^La^n1g!&H{FomW6INz_|HF1TZqqIzGmI7SxHh0^Eg6tB?Ja@kI^pF zdwVZ@u23WGplz~Yy}6a8wIv7i8N?+O&vzMh*ZgA8*Gr1^*MH!D zXfLiuJ`nyMGA<1NQfRsB6oir*sB$u$s9^D#n(+a%cQS3zIdq`H-g&VS;oUYn{e@1s zgQt|#aH%5X;QaH2cO06-PwYm@_~<^1HMWJqYXn1bJz%dio27UakXxzfZ`@Lq-pW>_e zsiTB{AJ3|Ph~~Y%u;0*=HOC(ax8(;%ZpXDK=AG%F{FX#W+gap6nLiJXIZJkd3kGUO8<|@K)VAdCu7pJ9J8$KLa4tcyvGN|4S z5mNrFQMDr-F93~_K^o_%K^CzgnJ%QXoKbU{`^g)P7*Yi7X3+WF`jw3}$EJ7GmCk&j z?vQqt(#mymHm9HK%TC%(LtqGFSyf!wCf&L^G`StM`ZD4My5V%gJTm0;f z{QR6X=zQ69s|H2Pq5H}yx%;H+iwU=4BIWRPAwecBAJ)|uST*cFU7AuO_XzIY3yt>{bAi3{>l1n z{}*@f9o1yo_KV`{tA6SzgNi7190UXeMv-1(1Eebm2nfv3B^2p3i8GFc8l{NRq99F5 z=tv2%0HGrgS||}h4*^0+fRJS059<59d!28cv-Y>wK4-0yzh;&JlIOXf`?{`QyD5u( zaSqE=(^*m5cbyW}rG73E>#egr+2$t%T)_VMY{4WO75#)5f z_0AUgcsGHtcfW*qUb&1# zluP>y!tw5g_PT?Y7gXM)xKzu;kO#8^M?5E1&I%ZjLojXx!Z|&M~iq%V=WM5v~B`JDZ~7 zW%eF5WR;Zwt6i${U*2T<$m{1k!L6?fIrb*rqtCltbL1izLuOtFc&gs~`S3BcgS^@) zHjTYw8k^Iwg<&>>pjp1qtx(^kS5nOr2bP7vSqQOL9DYK>#v}CVw`W+s&8>u!NWSG_ zmZ<(!#^<9Xl{nb;GpMgRapFWHh+&!}b3;}4M{sl^F5bVtSYu1xToyq^jx5!~PD^uV zsVE`ZNR|9R4%l*<^QDCOSsNotDau}HMtfoYj5O%EPk0rY5yq)}zf!Mwh)ND6f7Wkl z=!+XmV0We}MXLGF5^%hlqU-}l|LsDlh21p^@+2!8*ybMfN_dUfY`NfjIcu5g8BBMv zN`8+IulCL!SB7?u2VN5_0Yi54V$E87nLg#Q4g;kY;cO=Qo}8io$_5ii#swLRqSJrN zzcg})$5YQlow1@i`HY?Li%i0%^O@jOCC`A=qgwYds;ml1 zrbYWq>)blgNJ9xud@7cX@H^&?y5M>JkFlDy83L$Qnf7DqzSCIwxRaf!l3CT};#DaB z`F;9A%1qJhv6&;Ul-xSqDsRJHRDooI;u^xDk~wR)3%MO21Pc0m?=;+`MJBAR7Csr_dl0eNs-oziPKH_8hzt zP9rHL_{|imJz|-Ocj=;E{S1K9IM90WFFDjkbgJk6Xp2W;^)LS{9Ff{g0Im z8rX553u#@)2}g2LIC92tlq-CB_v75`@e24d{>FI0l-S ztj}8n5J{soaNUgT9Q%Dd=UPe62kSOyBFPRx6RyE#9$p)_f48ePp>1yI%r+k#tGnJ0 zHV8B(ya(&lqb^=%-bVsQL?H%8M#B?>QdQ0>bw;%H%Keu0g=8Hyz;=`v!RW#Ah15b% zw^HckvI0vENvWrKk+-ocTW2)RM}kURtuNG`LOdqG3)>EuSOe$YM7-H-V2K&O=rB;* zZn>7}674!t7z?d@&)F;C8w>rD&-OCnk{wVg-nAR#R`&vn0xKzaq-fwAM`z)vn3|-t zIqLj=o^$%MakThuzK4&GqVFlX>O{L58x&gIy!os!y|*MKHdRQ)r5~cfaT;%|_2vS2tpNOF;i+m2#emevu-VL)fdr+P%v9oxfGgqzJ+K zIn6d22renS1W`L(saxt8NVh&<#^UU(R>SIwJFO0pRC+%>I!t*7_m-*Rl7zhFBSwqh zQdgi({_?(xo-5&PTeE3`w|)lPvlsz%XIa(>y1pNIP0^NesR9G9wA}1tL|_+EE_s|1 z8L1vV=Cz-vt(Ca;M6|+mRiZ%_Tc>8Cy|gknj57oWP%6%JI=t-2Tw6MTY5bE#gSI6( zwOk~wR{s@=oOcD6XH)5M>_u+F&o8&{#`kjjK9=^yX=U0bA1!s9e&^h|`2-VEJj6E7 zY6+enkv&L4F`w8@)|Z>ay^u?7>UiZp`B4uO8M-ELLPKh7_fEgnt2WVBa~&P@ZP~*% zC@cpZ#JO{uL0+5S>>|tNUYGU+!nba8QjAv*%~F>R4_Z#MOx;A%gs@TNj2%WhUwUdW zY06sz!Q5A>9^9F1ruCTf(&?P2}MdBtO*uM+TZ*VcFC%X#8l+ zt*iB`&Ec&tdS4@PAz!=jl7BK9*g?_I%=8N3W|!~cZPr)Oou-xB!Xdh)5C#4zWt4Ud zUpF!EiEV9dt$&9NIzaw>yfpJ1O&M#N97oFvtMs_wNF3T2P-E8{Mi)&Wx0F!Q5fSt) zwB{2PeQR~*&%nCN zV*j3n;o+%@QZBM-ggQs&npsRnyRg+eOmRbi3UBdv*9GTsXUiMhra;W2|ic`bj=e8&)$4aC}F>)93>aRwYY7zJ#T$Kt4_>aGb}EnOscjjQ>H%XetZ^Qw%nU*6`|ze8FNZW85-~iqP{mPdJzb#1_C3Y-ct`Wf|Dx%rSQRJ@xp%F<`ouZ z!$eMi;H}7vG4*6i+ua{Ir{~&X$1ml;p;vkD=keq-<=j)yvW;``_G|(NGfB93tcPWH zDCHK(`TP4v6KQ<~dS*vCjv@aueu4Ao|8*kF092oNzn;l#orQrSa}DWPg*$g_Xy1}4 zcpjATSEzUpf03T;NOyRT4J~ihSG%`L{56XN1yv?mTMKC^YVBT59-c((Yg}kwa}U@G zUV?C$^LAA`sR-jtumP8PZbiMH$}2}5pJyK19?@0ag)A$_Ynw`3CnQibqJ>71#sjLhHZvFVqR3#Ej@V9d& zoR1Q|R2nnM!|waPoJ-ym@;^*j zBOPH+(s)I;Ei^ah6d!db%iCroUVDt>PqybgIo2(G?%Y#QFPcB;L?_H5!ZyycPyPUw z7^f-}tZUCV8^65miITamaF1L2?sl=Tl(4`1;!rt!zT!K#$Wz69jUb5HP1edP$VF&- zHqM1C)L3P?F}r+Mef1JUOk7=ciG{bsZ7Mxa0(5TpqTKFO40>eR1$V4fwaGe-O%tS^ zK7E?rlq{YZo*dUx&8$j4#u?*10JVgYXclc;C2V;J@o)pAl8a$}5PYZ<;3rmrvclMF#e3PNDmz*3taTlLdZ^WUu#$v+B4I-58zC#t z-F{V3z_xH)jfDvO$#y5ev_P5!rF-}FdZ#7k;@j7+e>pgOIP;fNgI6Wf~oIoo>4*^emrp{Upe8J?E zWz;vckca2B=E%}RMw8)-<}oJhBtWI3{Y@0q_w79Hz~FzA7E;c`kE9K+)J_JJx?tWfdRKfW(dmw@Z6*=vtMxkU(uFOj4eLxH&!U7al* zH`;do`fFoA`W)c7Rh*3f^^lc|ldKBwo~|aZ-6VoYXMii8zK3xr@<_T<=@D3Xr__(l z+?)abZK1M|5Y8nguMIvq)ex4I_-6KMcpL7i4;f39cK5?Lk{~lfceI@d=VQ9+k-Q5m~ zMJyo7Nbo}FqSK#w%B8A+r@b29C(|5(aU-Ph?2jp;s}!WLq>7B zMkAQME5|9Onk>JL^y*(2qeVXJ<8N$+c>ZpMDp>7yQ=4_5 zIRRK|o?U@!z!DJ|4HO68UTw+X^O(loH1(pcEu=0>T?rpFp;4Y*W5@iYdJ-h7-m%ty=32ys4d zUw_HE9dXUfyoDT0z&K6i`CfoC(prksL_jAWIb)paSy}btr;v%muW{rh4WEE9qTr%d zU9idSpbxIfknnWaLdf^)bI$88K%(xx{*Aek`6T#5{5-9XL?VCM47eEq00E;{>T)tx z#lnDRiu=r~@f!d)?I+bxwOg1A;f&E^=>B3m(A$k+4W_l>2<;H}diK(j()$yW?qlzC zg#C$cdNM=%^4!1S&OH1^SFT){nwrwp)dl!Bqssdz&(C~54Q%AjPBK0kB7dmp$43?j z>D+BY;SvYDBOwE^lypjiLp{bS@y&YN^cTYzXS%TVSpkU%)ZAQ2&|#48_mL;7Y<?q!aje&8{v!a%iYqGmM`x5vGlvA3TMgS(+d<7r zn0TfCUgaE2VPq^Da<)}(Ix@8wZ;SO`WMc%(4n-y*rN z6}*%qanAP^*o@|e%WZ{`Eqf$PM8K$X`eHzc*-~A%Xj50d71n$4OS$b&qN^@osL>9i z4T;PA<6Gl!`*#qJ9LKwjItg_E<*wba*`PZFgB+MJ<9t8$Jkjl5N;lM81^~-sq?<$J zgm^R9n4%#L!JYUTX~r#i$cT-jjd)tJ69MZ%SX03Bi}9X10S=ChsvoU)zX|Z~-e7|2 z^rkpC(0tpr(k>y#2yKWAG$w*JQ1m@@CWMo%Lg@^KMa?U^=>=I@22gv?bg5DPo*QcL zGUKw*2bvFK>-mpM8OM&+*Ux<^4)m)BB#i{)UM)6wjG-Aa*hvSw4xzvYCuW}8_4GDz zyecqSb$;lx)3R2uwBI;+dDEw!*^}E!Nh+{xI_PG1duo&Hy=@1`uAl^zHyBZDaAAR( z>n-N3>WfmC`|Du~Q;pDie~yS1Qi`eb=E4_&_8}F4kv7kja;Z};GSsZL8?n^?lJV{v z<-pU1bFdF|nrO*RB;J?-*jCerE7vG%A03({U}{oF)m}S1_t%XR6|k~H85VZ6O0N>e zKxB`kE!Xq_L7cnAk{mSEr{QPND73Zz%*lHrX`ySBB$e@KwDy*!_a<8*3o&8|E*xAh zhP^eQk;VP+@e$2n-@vs7VLov^G&+?V;!?kMg&Q*SXz(yPx9hax5_L*xEP#ehP`K#1 zM77J+@m=#`HXmRgI9Ws9#7Yq`T%=WR)k7wi;r4Qy66#?X>*=8J7Xti+avu*a4^K9= z+G^}s%nAI?AXk7klC&aG9TPL;RGscL+klIgo{%zqb+Q|`;?a?zsM59hQHMSNH`T3{ zPgkaDx4)1xZ3BqKixBUk6gWe{uJ*jlAt;U(l>dKFM@#J%2Yr}x!$szoK)){ox{E%y zj;Ps&(?y6u1iG;p5p1MujNQP4Y3b8odf3KO|8CU_Hq{;(ix>c^L@4Ef%k<2=l1mE{lT8%1ap`@Q}qW#LnV(6xuK`r zQvK$eU`8ZPmZ3cQoyEIGbKWugN$Mz(vxmo^I~G+V_v#quz0*;&gPl#l$&^ab(64dM zV;g~Xh-nvJ(l9I%z0DhA%k^~|^(V-s24X7P32$G-OY3t~ShBqr6~EP#ybqj+F`sA} zyziaP&Q7L(!gJbQZJ}dg4DR_j?VS&wDCW7HMs3l^0Nq{f0ScgL@ic(L$@}2~sQ9fe zz1FAm^64bP5qw=?%Qhwd-Y}L!<2ET2zd!RTo?U+BniJD5h$KU^83c&=tA9?yvvu<@ z!2|lf6gjTR)g?A_TuZ*@>w!lE^S<|7vT0W&hCNez3a|$k+cf$8e7dN1g9izsFaDl* zgnbj*x4F@c6rQ!)8F6l75>Kqab1TSqQu|=k+Rd`ezMOZR^6F!0|-S#58CmuM1fFfr?XHc)gMzSK46ZKH7Hw2WmARd_M#SP z*lz-8vMg|az7o}}+?emz8dmZC^!C?4K_WIa(W%*ZG*8sH8tGxMnn( z+oJGvY@1=Q@b)UTsJ}Q+By^6`#0Xp>HmuNZJrTLN7~(-<$|T83i8p{uxe~At>nxdjg{s9mAsAR zHp+49RqVmwF*j1+7C`GgQybuEH7VxSR|L*2BEH=j9OIQ7d3^qk+wJ}r-h(#KFaV*~ zBMz+8=l0S7%$BE*rYON1-)$8*0p6$+iKAAzjpefkD}%VT%#k~(Bp5l6P&AnQo;xo< zj*~7QC}bYuiJ-2M*%y{L-Pe3q{$w?Sa`@uBRJrtH1*bm5#3NS@C@T zwbfsxRH|F(#4xfn)R_Ln@(3u)zMXHT#u%+MO$%Kp?Wyq1GA_o0IsaGM&>h=aJGhG* zf{U>zFUMm|KDy$^14s!!C7Jf+Fu+Z?cj(Ze$z(iEc?mT~SDkx+nEc4ySp&+9KQX-e zt{wB0PUv4*QW09x$gNcP;Eo9yOLir@F~_!bqh~rD*I)Yj4&r#5O}iY*ubY6^T{=rB z0WKQ6{zvQVtEkM$`d#td>!*Yar1ZgI>R^G8z|Km=+<3Or#eaaTy0m4K=5L&BblsrP zr<^Kc*c8~QqxrH35?PaE_WjEcEH#R357f~`?P19+`|Hb4%}}7u%^f?}Mq=t!(9Dj>N%>xo}X? z&$2S%RB0)pa_~6fn0Pa%}RPi{4uV2pu3-8@hpcPlW3%^o!_6`s}sbm_bT3%Qw z|9AoU>201B)})k`82o)>c|8z!vtgyaNC8_X$y!(Ybf_DROZdjT5nsTjuJW`~gi zBMi}xQ_^|0BPnQg=Y)mod;PyYMhhoY_?SPOmFp>%%f4jqQ+u&jSRF6NI=~|n1!yHH z7hsjr+}q5E;;5Zyr_zm92N*Hnb6kxx;9mkl%Z)h33- z9mngpd^^QBfWDmHe{^2 z;=qU?xo2@S^ssUbXFPl<{-538>9Q0qSr+Yir&C@*D$OQUW+yVJW2f@x$>x?=`}CL~ zL${Q{qA?iH&0E5un^$^jU80W++#4>NoBi7^cwE{(e4^*&MHibGK}C7j&WzL!ok$6)APcQ@*Mk1*0 zz3`l)G*A|SZtPdxI6(nkZH8xuLrLBp(`;RByUl2_9)2u1Eb6f`4Gp^1Jn+mO=N z-SCvDE%=N`9V*_C>^D`LdVr)afQPS9-<%Q!=N<#IOu*u3og1;pR^!Ea>lZPf1i_{p zvDxb8J1hY+(EZydHpojlbu~=IYJqp95}ZDmQ-}i0(&Se*3+N8|4-F=FcXxv}3A$X^2* z-1xL5g$b7!HYkITF>Q9xNX54PmsF9#9fliJaQP|t4B_B0=8e$}go3MV1>A7>?V#-; zGdMOE*}$v0YE)LVQJsneZAj%f^l^RA!2z@_2Q*czi3gkRA7(H3_EtDoAUtjd0@O4U(3quzOBw=kOXGsH zwvLV)y5YGwCJOo8{S6YW@fOu+v&>;pSBve^b<<( zfg9Vuccp~_NsZ9HThA)I%s~^%!>^+YMzxPs-q)>}>(M?^(~){&FBvtbxonHuC-%YW zLv2;|@C-F$Ug^e)3b+l{{ZWdJk6VK7UJ>zaS7O3qI5wo)0JKQy4~X`Sd3FL4)^?pv`+f{}`Yy3w|FRflhb- zzp%x(SAJx5@%;bkt^cvxmx8MMd;4$y%AWamN$J1d@XL|^oZ+uosweH?`Q3kd#^w6; zKjWN*8bHeoe#aX0?jet|vWrf$aCNG0RenzF!gbf2Rvh%tL%YvQRZ5i?sj;y2^ZKS_ zidOI+pnBf167$v-b9HsSo^6n{eg=APsu8%{9|jWm02EwAP`MSvNI@@e-(O?x2DT_a z**f+elK%IP^Hvf zM=~64-u!tvVnsh1-;3E7=J*h zKmwMak?i1r`5|pkEuY2Qz^K7ZAk7J+VWYPRE$gIO%1yRhV=4r3fCl9G--}7*<&?s{ zxabC;D_WhFog5Y*KjrK_K)^2{&jIG|bh<{>h$Qwq`1hd9t?npDEuKFW%%-557`cCK z*s&O3pI%3wy5t;E&qi_`udq6X{U`gY$4y1Ki<$mTcw`;FD3w%m?Jk>umpHR?4Xq`F z+8lGSo#=;_2zcETJ%RB-Xv8N~a+dcvJ3BYZYtUY5+5?cZlp=;$4V{{ug+#NCKv*Id zw3II;nxrWTR6qcd&LfU|2k$O#RoES%k=?dChV_Tbbov|#4T5UQ(yaA`SO27v`@wz5 zvoErl0!j=Z;i$d}|Gw_d@wPdG!^`{Mg=}pEClr}w8~C=-^-dNj29jbn9U zu*?)V8XU2IPx?2AK#7~A($Ji7T^zm8t^Ab^vl zlOq%#Md-hfxAiY{ja{&Pzw!OUB?`Dmx+}Ey=n20}gp^+<3->Wj`Sm$e+* z{2wCi#K4ijFUaMGXac0k0ms8e*YE6nVn?4;h@7<#6X%4^?rmKg3!WDN{iw8`_h6TZ zBjAJ*2q+HA?t=HH#wlMwve&P)+0xrmmnFJQ)h1esHlx0OmvhuIM-62#vxO`Hc&Ljd z8rjj&k!PC=-2@4E8Mgn89rgXxUt);>J`&*mhTNzVf7+^^jj(`ZQB#|H zSc`l>QsGsA`xhpnyxrZ^6~*wfC2#;K41oY;mf`wfH~5*m zxz_K?{cI}L*O9CHDfbePAwmS8>O#rhss-xzC*t+uNby4k_;HC5-fnei>uAJ8W>ag} zB519ig1+60sAOi|#+}TaZR%D~S2;+M1)V7n1^Cdbb~mGvKNM$*qY@OjA*n%q2`Cag zbzF=Gm}{u75q9|x*mcyBTydq1otd;B9G*U>@Q3pJcgn#Zb?m=c5B_bp{x?6xe_1;I z7vB24PL`jz4iM2_cljF6V^AAnYwYcfj6jaPY32f9V_f-~FEd3zy`-_ts}gCN8^=Mcy2kX&r}Fp3?_3 z#2Qn5P(M4ErU?TLP4vgTTC0wEDfld-m9tgLE6+Oh%m;w#o|IIFhl}^ku?#v30f?c-IayjZ-w09{(gFiiT}~ zvzSvr&a=YKgezt*kEcjZ+%BK=;L?=96yc~=U6TT5`LB?zVXhZZ&B|^q_v~-WBe_znyI%pID%y^kS%i_Gb6oA@IH4P9($dzajmHP5TzgKRQIb zO5`15cDs|_`mVkYEOA6^)6xe zA6d`?eH0+EvW7}+D&a->=J@e%pZHt%_WJ58tzi3~wjD#(0a*%Yv!B6cwKWb7uDg2z z=#DHxOziE)!>BpMY@itfvs)7thg#Sy;me8^5xU^v>E{U$i`KQYv@~Zjq~vueg?C!7 z*=VzOqF?^zYhcD@y)9-#4MXdPGhl;z+X5!2-`PYZbFA71$sF;U3n`|BmEnI*UTG}+ z9ae$0q4cN^hqcG4bNspD^a;JanM|PU8jRCYW(o8BPW>%#H?isF8E^oBEVP?m0Pz98 zvJ@3BI_8dRO^zy>!~}WS@T)1q8Kav_l&Hz4MZnK}VnbU_pNYD9oYv%PM`jIoB9{W8 z_swu5e;}189~1#XkLiP`Q4v7yKNYD382MsNGC7KaIT{t+4U>fgw3^#f(WMbhQBy-1 zb&SX*I}Hgm{*>}X4ht{kv@@6xo#s$y*{Qno6#?m{6@tj%+{R{Wo*VHD*PnUu#9Oqoc(O%(6F`W4zas zMHo@)fgXh3?Dlv*s-9TrjU8|}v$(jpQ#nZNt2+8^8o9YtbhlCl2~=*5U87t2_lIlD zS$XDluIFtk&Vm!X_#Zz037sBHD!uVVJ9vC&vwzv!K({MV>|#0bzF5_SEnD&E4nrTdPBn9^x8tSVE(cPM2>1Ytarv7u(nS66;Pml zYkr2)&~o@bqD_xKdm|oyS&jb`cbYd@?{nZlrOBgL`O-ZFz*sevrm=)jJveMG-hi!J z?R5vJa2br>5(bD?`cR)j6J*&6U#vBsxUtq7e`uI@9Pe;au6Dh(RZ!s-3hl#BzS=kLtdz@6pixsFUbRbON6EXuwGaO6%?#tTOA?~U^?CP! zLVeAxIjz>8{wWOs4BEJ0;I*-oB#4D==D(l&%U^j5x~Z+BQOsRBc}Nm8imwv)9ynq= zpyCjW+lY@5R+7&*uQn+ff>$?445$IyL_BZ+mS&HSgQhCoh{mzI_9WJ*P5kJ1^_`Xm z?T7ds8EN=!>`8z;mNXWvs)&yc+TjsinCyW_)7V6}zF0vm+8+ z)}=frHi2(h;vT0{?eNL-MmfKm#vWCWgX|T)C=ajAz-ue97vKYFZzDKK)eY54WiIW9 z#c~HixgJwPA!}pkNeDCqyy5j~c@an8{exVDkNUb%7!7Y+LnGT1zZqooYJn33}0eOBeL&I>;*ZN3@yX!mf7;HtFeN z3k-7^kI*C4Ui`s3wC#}j%HTQk~Ak38>pR7q9H_FK~HO@x?>x^fBhD=!lA`Ak~ z92ego^d0soj+8~Q*4&9DS1x9A6T0h91&<<7W>|7Yo!9^T!HG}5;)0e_J@ffKfiodv z6dHt|$E=$a5s)M4yp78!e|@v+pdvwiHEJr~l*B8waBs7mPhSFaeY-pEDz2(>Q%g8? zFFczsE#~Jh&UV@U>_5|imKVVvX;OJ>l1{qd!}-fOmVsc}D2{D58x*;rmQG7aj(~a9 zaq13^gH#re1&V+CBhRctB8Z$pb z?|CA2YVazg_ z%uY>R*CB=zL@)TwJrOGG15!vwR-E?gGiSpMTMyTVVBS56yx@dThU!N9@o&)H^myuW z#EEX)j_W$DUlC|2_>Kq_9=wukl*;FWV}Vu+JbaIwM(J%m~wOQ7!17_7;M=mU~j9bl_^5%b!#NDCu)fjhOAE2EhUsp!7! zWkwE?Ilp+~)il9>p+8>)-~!R1YbP76vYI7ntXT&vr2ILHrRC!5C2gubMR7rM zCqZSsk#o^>9}aN5DG|?q>DQ~Hm6_kZK0G`Oi9tHm@yLT$BJM#QtlIZ}a1>5@dkdH? zuXF!=m_Vzm4e-DiMK%=QK7|-I?T*&qtlc+ysVXr%&=0I~?&1mHVMBFFx2AAD!%>F9v9X>|13 z;4*;cCH63%Dc}wMu?8AxJwT;EO5vcn9*MAxMH}ZRW}sd!bAC)wTXZE9e0C+kXNYWo zBgJx#n#P`R!1S4_t8cF_@v#s56bB@{IzZ?F4dsr5fY;}n%~$F&Jk%H-Ks%Krbk zfl>pibXi36EEf1(i4j+RFCx~GtI2+LMnGV_AEovyU`>t#<4q)hXe7nQnV@qZ7TCIp zhl>HD9N)L~%71mbZMbvjjT^{_+d?V84ePG7}F5Y;4TO6r~uo!3jCNAXLZtjq*iBYdX@loB~TpK@Q@=o!JkP6 z>EbI3-#_}f=qhSxAL{=0LCBM>_avz3FmUKM#|kz=`dn~WtwBcaKh~B7feG&?u?v!+ zJ0pgu_`eq+_>ltXlCqA%ehcMnmMi#VTm4YJhL!-pQ^>5wFKC_g#&$JiseY>|E_cA$ zbbcWF+A$((@cEM^Od^m?!~nR>htBvR_R|JpKP{K(HVge4R~jR;+2rNRM6FtWjG1<} zPPybQy#1KZf(a?}hot@W&bjg{tjs1u$FHvy5(srmqjf_dEV zeu<^!@giGDNaNaVSMnVD+OZc_!iJBQj8}Po``90Xutst2DBu`{5Y1Z!uCcok3?Qs- zS^%JoBuI+L-msU_kS)0JFR^Q|I0+NNND5i4vC6cmeCJC`f}tg2gX^F5`E@zxgjRhRkG&GRhRKrZ5F%t^3$7=<} z0r;+|aM&N}hy~PM92yr3U&x-Kn`i_9WreZGcR=n#b88e0j9B$A{$2!^7c#NxKHR@y zkakF0Lqj%2+55J$0|jvRD-W5m$@X9v3}jw|+OxTl@F#bwB0@Epy{YtRtE~86IAEk6 ziNb!T<{p%J1L}axTIu`5Hy2Fi4uw%Gz56wn-cN5kvQX;}%^~5^E=%#nsas;i2Kb%m zD1lw#WbA6u581CK34oK}ChD4)bTJ25OYV%yvhuRYly=lfv32fw?d_>@XThz-#YG^x zGls|%D33($i^q>Gq#Pio+%YVwc6jiHu?BwGzE1K7`si8Q(b@3*A>q0G%YV{j|2?eh zzs4V}=Wp*4Ko5F*3;zI1a|ai2j_2mgzYcl%KgCM68020A&Wd{{<=bO|_4bN8$+}XW{I0%%kCK;%XGN50n{-sg@X6usu*a>Cy2XI0E{s zyiYE>9e?(;ulAP5Y*6S+&4mtx4LhLe6p`VOTHRk!!f5d#A6ZaMdb zAn&J}-3!(kC<5&nMGO_Smv=}*P|5S>f2BeE^1LPor1rT?PTlh*O3xC=-gvB0Pz#$E zCemME0Tqan7LLLbbQC1u)Z8XU>wnESG>q?t{Na!>LniC=hFu)&E(DL=U?F)QE4Vw`?8Kw7cbhPYj#na3pw4D%l>ne(#_jcY(UYo60} zBS+GQn1@`SBdNt{EPMsS-thkJ&h~B~$!Fn1Xhpp78~qBb`B2DuBm*yo?KG~n5*MWN ze-9nIzqQ&fAA_u0zUxfeOo-Du55;$Et`_u^hP3u>5-=!CVQ_=T#zF`^shaW-NM5#M zhHw46ua&9}G%3EDA6Yuh!B>|8sL5voV|leXm(&B@g&>A97`P;(UtGJ`20X}*hrV=^bm06r&ECI@%VwII=3Cb`V>`r%Y-Ze z@$%iR_GrkXbgR-M3522qk7NEh#R960RhI`Z^7*#+?SpbfobA?wVZuMN7AxV40FitP zIA`q)OEwhC;As}L^rh-H>uilHA*hwg2011})nsq7EqAiLSl$=K^d2+7c|l)-XXGy> z(w+xJ#kw0FzP+VS^uQS;3HQjhm*DY63CJcFfuuVq2+t0nK?|0AcRN82z+?qbneLp1 zMMFK?<3+<{IdnCdd$tyE*Z7+Czs|lcwl9J#335TrV})QB?#%+?newtv{!V%MFO`$qq$hCE&J0DLgax0LObNsIi%Qon z-?L?{S-*Gh)N1G!seYl!{=fM85AUuhbUoWN>zWxO! zt$6XG0e5ou%ep)R`|{8`vp>8%QZ|WI5nEBGV7`7o0S8_NYPx8SJO6r<;83ieW9lGV z=^BQst)%qi+Q*Q`?!)h*rn7JPAC4iJSn0p;iv9%J7v+u=!K#&kOdqjYxW(hXYpm zq(c3A*u)wLf{>gxH^rYyD!6STDhAvhfMTI5!^L;k3Pg?g12`E!2nn1Dxq@u<$6&yu1CnK?A&cl_8`V2b2#K-= zV`naL#0uM$@*FQth>1Q8JP5$Al<-@G6S3Gr5fXBTP;qF=JMex*6E)?axpB2Ev&%At z8JApa>RBHjOl{;=1yE7s6uPuivuVs;rtBEZAoesf(2OxRUUM63c)eh|fHI^VeW)(e z>vMnVsGJ_tUbw&yj|$^HR*H2c-%)xu__WVd&HF!1^Sj?i>E3F^DXpNZkp!D6cR@t( zLe0288ZbI}Y3U-tqbz{Rknqi7SdPUdAY;EhlAQ9Q0YH}eLa3c`AH6br@{njdKZPL4e}O)+&aMAS5)0~A=mEbtu6R~tpn8! zUjsDHz#_IA@$?*OxAh$=JbKOx4q%x$+AmtTk;0g9FV`0a)pq_n=f@n{1TW(Lq9F@I8PV!uquGvCcHPiJ(j?qXJQDD}GK#?r3)EBMmD=OTHsI%wCpMB;XSCPlf zjlR@&vb>Az#5gh1LbEyL`bEhNVD`@f`Lzm&ZjKpm`|j~oHMU5N8?Tv+ct=xyGx>36+Mm-Ugwl&yq+t1*F4eYLd! zL2r|Q1DXz}27+nJJ}h&JZ#+N02d|+Oww42OCooH#L}DTX&+U}BR}C*JEG$%A{lZT0 za(ZivLYU3s>0M9gP;SF;u870p=3(?$Izx%P^){?rA!#FfoP)xPaPI0QgS*z?S6dou+1nqPSZ?~p2BxwYG?g3W^D&5UK z3gitmwGzZ`$$0Hfv*#3y!1!@9?rjbZwe(!WaM}h|w=%jp{3zRR_h_UCeog z9sfVmm}QAl$hs}7O|6d+}kYgdMlg^Zit6NXnN zB!;Q)+}-$rEgd$3_`>-h-re|RLLA~q6Fx)~J>Edo-nznqkQ~(b#$`y?+rohqV~}J8 zm_5sNNJqZ?JkVi{nTl{n+kF~e`a0+V0qGoY(>uF(?F8@L&&!ta-lWp>TgKntdI&fw zEhn^Ha8%py@UF|9CFC6o_IB@{Z?d+MK!~4w9C6`1q!b0kN(f=^@oext-GNX7KH2c< zv~5K!&*iNWke>>hT|W-_yn0Fzsd+$XpBAm96>it?>LMlyvh}>WTnk%J~4~WJJ0P)X3(%?Jma^0P;9SM{+ z@{T6J^(`l%Aow^umzwdO+3>^(057c#K-2?~29yO^g=hwiwE5>A9v4M-*izm#<@x&V z{YPT0%V#CC22seKnnwV@0ve=*`|M(_wjk(?z0-nvdu~bl*NKhp4&k0UfIW*na}6K$ zC_4GNy*(1zu^?y?N=Ts7A6zMglSL@6*^E`i;pU8xg-7q1c=TDc5SSOBd?)ai8au50 zA+0@n!XY-w4YTP3;JFirhr3m0Cqrg&==j-A5US{EEJy$?%jBbEdo+$QqywrF$bLka zBeb3Gm@N?t^q@{+h|Xq=%Plnkp+4YHLAa)NZ7{9kh|TV`@K~;@Si1B!$kn?oEcP<66UTEuo|fcY#?us$HhU7JBQg~ zeql3uLgz0Ak*82&X~$=E(+7oXXelLjYbgP(0-$beIk@rdd?ga#_$)k8 z9{W3awM9=s5ochRl_g9p-V7fRQ;- zH}mwC69v*wfb5|rFg99D4Jj%#h-Nh#Jq5K*3ZU{Y&KIZwu>VfUecyn^FOnvz+h0D_ zg7lRb)BQ*F_y00IHFeg!DA!)E+9#)pTMXoZQ(>R=Vj@EipVaM_v54Pe1f0cjAeDo_ zSOO@3QQKYy)yf&=@ppLJl>>`*s-F;_{*DB)gpIkO-3HA2;lMw509$^9WKLE!Zf&`P zN{@-@_y(0dh?3>6${DZ+2a>Py>tvv;^x+jSkS0RM6EKGVSg5oQrQo2bhRH7v8OFuc zA+7O_vqykcB|__-fpFjuM`9e_Rll#bM%L=2*rmin!{!8l{tBVBH_VQ;VUrS4kNd8_ zd7hjsz|>v|so&YId(KVAKhfSAm2DN+gQWMZ#s7jR0{+P6N|iB?&1rJ3)EMYwqiuca zWvZ;0u+2L^ol-q#qRDBQgUmSuz!D@Rj@1UmK|-Y+T2);<-;y*G8(_#?{_Fjw)U0G* z%U+(_d9Ojlh)k)vl$2$$4Q^Z=4ol2g@Q(RUBH2vvVC^5kIsoNMSQnh^oP;J?{e;}6 zEK}SmV?iZg1xA>ahKWV&SV!*J0XE6zhiL5t&G9#S$wS~M_biX3HhO!xVO5#7TJ|YKlX^%rmN|6aU%~T)M zDaaNJF+0iJVC2MxXvF&PS!b+uR+w-F*B-K%v~+Yl1)ESa6o-o>+2b}qbCFu${j>|j z@5!NRtA^+nvV$E@w-?uTLlINf&2vFg79ZDtwi8WuV9r7h^WCjfjMu|~y*w^gzyhES zVpAjbDs8-RM?fJ{iExn~#;9S@11e3;W{ynmJH=at3G>@VgcBX87_zb_w zU^9osk)f^VL+I=bV5fI_*!Ly*F*JJnD6sVxakKKf`E|F!1_VH9$b^{XMa?wt6GOHj z>Cv*@PoV)8nyV_{<*w>GAFtsJ+S(QB&V?<1DH>wCA{AvK!q=8qVgz1ffPRI?ux18g zvLVl|&azZJALC^ptIVEl@&-{&zM$H*pgE-?=T4Uuma9FwjB`AG`YW@j)l-V@jtEH6ir2&=N z!1h0T@IR<~ z%eN@MH|!TgK}A4C=@1Yp2|>C=KpF)_y1Q%00R<(byGvS18i_$9hLDo(?#=;*VfGsR z{`Rq7?B@@7e8r(7Gjq?Hb+7BZ&hvADJ{rq`j}j&SWC27MqrcbIVgifFuajrJktl(+ z%M9t5jPpd!TCJr;&&-R z<`F(XzmftU7IJ%BbQZlnM0VW%{joZF_jo_UoA0sH(IG}T@4{34yM-!MLw>5TcG}gY zlRpP0?H{Iq=D>(fb6!s5MOrLUEanU^G407MgtsTe3@<8c~ zS6Jml!{D1+es|B{OeH$Y}Cp(W^YQh zPHa_Cx}Y0^9!Ib;pIz~pnAtg?K|w&u4ol*OLMcj!NAs^bE&_{u?kzo04DH7y3W>^w zw*wbRbaNSq2D+SK4p>pmJ$Cs0uzf2X_bz*SwxTzfZ!-LNqcYl7H;jTeh;V0$M667^ zRyozJkKA!hl&Eb1o}Q#gI+y3#lPF@UHSIWtdgj$P4JZrN+r!pNFhwjZbrH;&6xi99 znj*w>_v&9*Q#@MFIJy;)~Imx<|5 zJlPrEn&WjIoj##a_}s=)Xv(F)4ekI+cG^n;1V-Ie_E1`P%sq76t!&2VC%}pV+c(fH z|MEF7?jvZiyp-)X{}o5O|FUOVLqsg9GkK(Y1kx` zQgKmEJJ`v1w6F8fgi`j7hu1ajzOm)FO0Hd zZuiVyz35%__zz5U)|vyYIepph6~oBY3`)L z@cNmdepvXRF%#f(-2Q`@E>Qv0zA}d#1UVCJOmShyYT{7Mx4bF1?(+TS_i?N!wn56!A`S7b_-x+UR%*hKBbetGt z)HZmDRw6)~|E!=JpM&9ZAZ+=8=WUQPopwJdi@tYQ7FV~6a*M&m z$7lAyJX&73hKDPrM)qO-sHWxk!t6_Dj%Oh2MiAk_%c|hIa%iMunIcAaN_=@zx{mT0nBv( zMI0x6%B~Nizq;B(8iS}xSI|DV?_|4SV|5Gs^JZbmcVW0u2+?Dc3Ox+t8x!()W_Q_! zaVj`nUecyUD>4ybNlajp0N@WZqbw~%qVZ%F-bJxfXj_8u4G01RH1c#xNgs4qaQ+Dn zAKn};Ije*?gVY9wKk2;Mhl%JsHnE?n@BsDSx;emyz4-TuseS=n(`WdH0l36B{*R||Gy$m@WHhg3~b0#Op6!Uoh?xSwZG=yKn6&a?cSRFsPqPGhvpe> za$p|C{POu4N{(HJ|ZG+g1w-JJYbrR%#?f;p*QIqcY<{1?c17@kS}s%M@vpxvKHoLu8- z)g0Kh7@Gd|t|dMLk0+lN=prIaK{&NO=u>BPTF~|l!7S5})HQ&D+A;|8KLfOB6Ry71~Hq7b7)i=0UK5h}H z5JSfQezUp-2syyUe)b2pVLGH3BcUR&_D+my0c6^na4KFt5_T=BG@>u8B3>=-4t5gu^yWTfgb{--k*3V6?!4X@CWs4gCJ{B7B~| z&%hEkb=jI^`37)$yw()G|KfQP=U^$r#_k1DC)zMhtYt14fgHqQ7V_#LNJ9^%{Cm<^ zucWDZ%-i`H4+c7ug8SRlPP~=A=HWJenM`)P=;Em`x}hWmdP7nn9KM(_Jy09DbD#@1uaYj?4@WU3exBUt8EX5@ zGFvDr$;2x?gRbL)t!;hlzG)A~@xgrf z7%>y+t~RjxG+tZ?gR}A;yxAV9mg@3r@H0G5DT6$a#udZHZ*Z7n%J89KO7rnaby}=F z8I9W&1;3WL;i)3VBob?>?{liGv!7iV1LqV&8T?M?7MbWM<4HyVPQDB8=jmSC_gy!K!>K(XA+j`e47dI?v(0#c_oG_# zH_rCKX{V@6hLNEgu*1dx+|0Cm$vnfu>5Y$|OgACINVy;a8x!v&hS2OA#^#2_1 zZOj6x=R<|k-@Q848YCv=x>tKs;@unD`H`gTdOB4;mC;-L_Om9H7{K=b?i1bBfU3Z_ z5=vevFCvCYL0NB&eADD%G4D&(H9{rvtl6CYz1mZM&l3_!331hCq5!n10R|{EX|I8z zJ@g8ofPi6)8FE?9l&;iG4Ox!|)-X4@Q8!-%D#G~c@`m(gUS6Q@>UjbHQDZ)`e?+*< z-`2;yTAO9CE67uQ*w@Z-O!_4%koEzfgNLg1*FU<|s!{x&9D`EP*Y(lGOghH2jRg+s z81T1fA^TSt)|kM_HokEuX{!;iragm9e-hYZY7UCZ?jULKt;e)rGi`b%NB^qK<2rcT z4H`Obt-X#2->U>*Ip4q-D;*hdAH(f0THii%@BIGwvH0ev%8=9)07><|TH4Fj9I_wN zLob~qz%g54W{cF)&sG(VH;A^Sh|v5DcCl}w|2{MbIcRYJjDdump3L%cJES&u@-2nu z={g1aa*8(FIzbtzB5{fb=_v^s?u{OmPYzD4RXWbaMHm3UBVO@H51d}c?K4D|lHhwG zmHh&RohN|$L|)c>{Qj>B+SCouo?y2GGe7JjS^-k4jW5RwqxyDp{5>3GQYn0H!t7UR zWgApeMe4{mJJUMmsh%$?a&!Dn_pGK13@=wYia_V4%65o#w|R}W{S>2@mdjEnhbGTW z|6O=|e-|F+SG7ptK_7tKt-J~32^l|+res%@A!aP+)v{N?$|X= z2G&WJNkO5LX@=z5E$aMK56MtUbhQ{gJA+JEFtM^Irw38)tb(o39#!+@PBrjiyb=AU z!>bbBoTQ`2w0+^JF3KX8>qCHxdk9AE*!=@*4F*w9kQZFx$!qI@e3Sv%#J4Qe8%G&Dg;TR1M zL@dtZAi_ZL-41|x1sUVa&9KY*_Z23VR377!&n|i*q7i>&_h5HH?B+|06fD&13Mu}- zSK4uFvT*;+mbv)xevF*> z*Tq4!;{h_Q7R|T`<>7gOC{0_Ut-Yl4xSU6)4}@W^KNgm0YG&GP`{TreXTX*Kj?}K( ze`?*Lt}yZv75Li7k)c7@j@4Vg+>qA$h(-{Y^1cG5sC4*F)7b7{v1SZL_XT)DssD(S zv4(yS*2ZVXa3<(?DGYi}I)X+wWZ3^i6G=1SD7u%MKGT%&@0q)CB!wT&Uf#fu5wn;s zkPL=mC}Q*{2GCKMEJev$({SBEGUCs*J;uKe|G>@8#`?jL_(?|~N<0E+e=@-HjA_P< z+waX}0~LO%-qHpSgQSTCCU!=eGM!wfD9`U=Mej;HBGmizj?&Ld0snMcf2HXkNm(^7 z4w;EcN79C@>3bA)!ra8p9%IV9x=T{hj&RX2#TCW&wz0y}4Q$BRdQ zg#pYQ54)?Shl4drJs8_7Q0XhD>I-d*XX;(#yIZ6Jq`{Y_d?|zI>gy}f_j_K2Tgiyd zHU9t-;dtOhl>$=&LPJB>Cv&rv|6Ky+R`6U`m~Ici6JX5t7?xZ6sGJxQgu~!qct`jO zn|%6e(w8idk^qsuz(gK5JrVSuE$MKjGtW1TN3(DsWKQO)^mH=;ij1$;JmrnTS z!T|Y(RPc0}hqGg`UEyoLvak&5;-p5C^50={+#4v7v_fE}$LY=tPesKSPU*RPjR;J` z9$3IJxOv-w7d1?z=;i60%l5LeaM=uZK3KYd;H>9bs{m>nKPIZvJRM>*T&ioW%^$O) zgAo(kmcm2d7QKVBn4=!gLeWdhO~(TaD(&`V^>l^KfKzGnKZxn~q%TYZE1>MRmWj#B z3*wZ6Kp<5Xv(GW3ZVDR!d&g#lq(W41>lpBw5Mp0IlTOs<;)v}F)z)B>f{ybDXRYQG zv$kF1RW}*8%llA}6;l@3k(CZl^XYbU{8cV%Q~7U529~N2=EHOtTm*ifvR%(D;6DI) z_Oa32MR_FzbWYbd1drB-JKg6r7ZX7~oE1FFgn|w2;x%v$*=qAgaetNc_dFOaQfU|8 zJ5y;y_2G4)^#KW*CO_TVCe?BC)>OU6zb{j)pA#j(9Gj~YYULh!ZmFNIj)Q(rtqig~ zL05P2YPQK!8c-QTuBIt1Yme%S)4qNkkG-0h{5JB9)EqUxO^?*`UTS3~W<@cue&XI2 zliwK6)vk!EM&oXd6;k8vH6sDlTveG}w?+gIIhO$LY!)0MD}mw~%?&cCfGW?eK)czx z^AVBd^OU{XdUXGsntAAw(y-?+NeUhRcDCW7w)k+CT^!X=jvi&LmGsK}EgwbvBMPCE zpYWKI(^qMx_wr-Tl|gF#T;8RD*RU;x_7Fe~#8hYkYQqW4R4HjYzh1OvbEn&c*4Fgm zW@f=dkejIN{JsH3qb9`qpf_7>*&pJ`V_`-Al6xts-i1`9wc~6g#qJhE%ZyiCmJT`@C`~kF1bYE;Za0-7<`a?y3^Szp<&1@(3nk)Y!!HhMp zKwiBbdO{H61SGe=#%;g@tI=xs_i14+LkdXH6V!)suMxWsUgtX);q6n+QvpSGN1>b$!@xSg`@FKdSi<$E=-_{)*+SBPxzIkn*^ zj--DgHCZ+O?GacXZvUqxfCv8nqm%-QU05rj?jSa~I0G`03LFB$ppyhDIGlEGFfL%) z8Ex!1E{SXdcHD*$&vnqpq{b)V2yMPv_!1_zRwKL@af>*=P2awb9*7k+Q;_(8C=gN8 zMTp)Hq#ED^>1%X}r+wxn2J`;fn5nI5gD3lqnOT+P3_Tdn)-49nV0Y&@Efxb#OAb)_cvB&m!wsoNn5@Wqt z%qi@0P2snt(iL_K`u+P5A-aNTZB`D{-6CcHc4vh@=pnW#c z{uqGmIX(%fZR4_)iiXihZQD__wpz0SpGn|u8fa0)r+5?nMd(^=@@Rl)&2#HX0bp)? z&mnG~Ykabq>W)ey;(wnH>7^*QZ=M0hY$ksGfC#d`vJlACy<8*Q`Dz&>Lpc)AE9mA` z0q!wmAbkRIO>DR6zk>YQ*S8^Nlqhohwsps6Am7&sXe=P4FAa=FHCxWLH5d&R^!JPb z0IBOJn{Y;)o12^O?8j+kODmsyZ`e6v8Z&?ev4&AnfggCV(ks8lzvwv>n+WFu%)L zZ=IQGihYswfbD_h;NmXkkL2uyTV1OKH*w1Kt1`9_7jC!J`c&wtSbBQ;eQ`S~xeSiD znzHLFmN-2_=KUuVIBT_tJ@HphBgmpTTyF=HJ@czwYvZM?@j(8FVvvqq{CwmApqzbO zYHQ20{(<*7sQOB}jwjddG%Hdj(ujMoj@eI-t}PMKipy=ke^!;&({N%fiYnwGJ#M_u zBlEl~^;wCSG2V*h!j|U?w2X-R{wvA{`b|DdyP8=BCMWg16?Q{}mV-H^IVv&S63tsH zJsR~9V&2CuuMJgi@vrQn@Hmv5msY9M##Cq>d#GL3hQu2^PqY$7IK$RnzI0ffmL^B^ zD5e-_C98Mks-(|Qi7Hs|^ zfgGjwG99X!aoYUt;elE$*>H8)KeEn5jixo!)254g4KALXIT|Yjjw(Htsvkb&9rv0K zDxRs+6@Ili+pGG>URivGt$40cs%CF;&lpF`fv76m(Ck!SVP!swcoNHuKyoo&X$yG) zL!O6_a&OcMR#@*zhKyy)1V0r=t809D(8<#>tChGl}M-hbGn8*i%WvywpA^6*pP2(7Vus8(um$~Irc=p3ro9QMJwzm z(}BJS#1-l_?T(ZYs=1O|Y&*T_G=xiX{hZ{f9vxgIJ#NZvPy4dOd*kbUR(0mUAw8H= zD;zr2uqNp8BC*-x$5Xu;8UMk6fvgvD)$*pfK55?w{p)>___(HSuA-ZC2x-L`%XZqJ z-)XDtwDj+L4P}u|;jO*1p>jV4}u4+p||3o`<$ zXL4TUVN~Nvup2i$t?|M%%4S5_UcP(~N*6h0zW<)UZqZCkwnp)A0flGV`IEr4|D&Ye3|#eHff{f5`IYWmlYN<)^dMTcI@bO}ODv&!oKM{oEa)C+F_t zhx^}Y+g|psm0?q0-Fqd~p|%;@<)ZtwWwh+>_7<-}=(O+IEBa4F`BP666S$-hLVjdy?LT?`1blFf?9-?PU_P@CKnBt{dE13eMSZg=s>yiJZy zMeB)-Q0<37EQ8t7gIFRIiK>j5Yad>77ih@ZK5)KK@$iBuh1tAc|4&ErwkFS}$7`Mn zyScd6#Ccl?lI{crKOU2^Y-3#RjDfD3Ven^PW#a-PGCG4})~cjznzM*=G}hil^K0He;f9!oPmolaQ(eVQ10Vfg%y|TlB$~yM+o=jxLLj@Tp?#4N+E2^XQq>xChfi{ z%Uwg}^}eRRBW3#1noR};$As(VHu-HD-92Gr=Kb56+!kh+Ev-z;ZH3N%tmiGrg^Lqr z2=R84`M3a^R>ME8Llk=HUhaCO^U-GT(!*&SrQLBHF3j7XSotcxM60Yhd-8qs%Yzf2 zdk?WlPz_qMjUL(5EGVzZpMDT8D)R!959T^1S_~=FBLoA26k9(OG8VMrM)U&jxeV{{ z8dQnZ8qM`5u~&)>(TF-~WxNL0_K_445`S!|b|c9WXDFW9tJ;>uv#MY+i@!!zk)U{y z_c!v-Lu=llCVd2C_MMr8gdU|saa#~g@51v21aP-2OpMrvdsE{WJkI7db zpB_mz3BSq-x_5Gd7eS+d@L}K$60x4WzcId=|97Pyi<54O-lPd13<|;%{b_CC^76-_ zrSH?oE77a9qW-e?rJqcCq9tKwo%1Xj^#x`LvY~hNWJ1xs`qy?U(HFA*+wKA{C1yG) zBpC~@q~O+4cRZsb$p5JM318j{gY>n=T z8@z*GW#ZA*HaXZNTLCO4?U}E4Cr`BN?N|Gv+BI@N@#HA6v-^Hqi*QzY`O@cesM2&T z-8i3nywwbhihT{&}RJzscJa^6RI^RK|6YzCj7a*4t%PcVAR1D>Dr+~5KGDtsV!6sQV=vP)4M2{bwNmb za@d<}Cqds7zV8p&Vl_603ks+UO8V`6(Un?7VLRe3ifnF=GHuhY8(eSja=Fwj+9qE` zH7QO;Tg(J0a}ej{6ph4*-sQ>N}&PP zW(%n|&=!%cG_jK{VXvCAmU6Y5?@>IKZ2jt{gCd-ptVDmmLy0Wf9b$m1#4x~>e_WEg zcb=KA2$6Gnf2C3JmsO6Bv*xqxNXRzEA%_@0L6yRBDe%y#&Fb){PPpN8RC|deOR-79 zUJKED+>7q5b8wZU4Tq%fx?CCSw?cOOK78=yOC=tx#V0Db`TQD9X_$dP<{g*Hfa()O zi7?{Fa1wi_7(eri{hRzpA3AO@blSy!vYXkKMNt#Eum4a!U7#aw5xhQu!r!grx1TC$ zMi0+>h5QLUqq%`%H)*(=Uf`zqcJUDWq-l`fy^;`yEANjWsR9K3*3iVs9#_uM9;5gXmUPfvgQ zYQXM?VUIGZ>2*^wi7_EV5j8^lW?Lrr@2;V>8G-y3;#~rVh`TIC*P*3cM7VTPH8Ky& zB<@TeUOG=^;--lX(veC+G%}%7JC5Uc<(w2Pc0XZXpd_RAlUb5&nOwF83S{D?vFw^{ zV6dd`{ortxtZQbDd?Y`f;P^Ry80vTf`IhNpjL=?z0& zL*hLz)}Ap!W*+ULwfIJjDUjr+wV7Nc!q-~XUm4DP#P5H9VyzLiEJwVB&h`uReUVBK z*GT{JCZ}x*|964fuWvvu>Y*aC&{vsq{BX6~jMxM(b?oj3{%>mO%`WY$q8*o{EQxUX zaROKCx1;^ijmEtJwvuLADJ#bZ)XOAflohrKR~2V-=&IxS!KHlOf!T8S-M|)ziDK0j zpK4yeR;KBrXMHSJX?)fntM>RhpF{?4V?DPi(`wN4cO%=nZ~Xe}Z=7qf%n61Brf@GJ z3bt0b(ufl!87Z--5?c3%-XvtVY(>crKh*rNB(y8=S&QSF0PWhnO$eOGQ-s{%57~34 z3OnuV+;{gSCQ3gmXvp;tLaS<{PCh@Nmf#Y*B_B7BZ81{cOYu{->!BBBh5st&c8}Ev zyth3>#&BIzS|B~ub!`@EIrO7B{-&fo8sh0XVdnyxB7xcc3Cf!7Kc+93NS#4V z^2fYWAWbDU;TJ@+$u1-v=ldP7A%8rgc&w1zd3@K1kUa1vR#w&hQi9#9>CE+?bY-UP zQUR`pg0z5k$+r-PP4FYHZ<`zSK`9728n-s^$Sd<0cYbXZzk8K&8MkBpzTt{wi;N4j!6 zBI|#9(|vp^QmZj&tmy{pZG%o$bn>aSp-7P-ZB;^3kfGPFPq#{L*^b&K?k1mV8*XQu z?nbnbQF=9}9R+~9f&81i{d#_x6`adNDNT=US~AG(iP}-FclEF1edj)AOBzjPUT;Cf zb`j%sItP}cSGtnk(#wPz93k#!^p^zfWW^-;Dp3xj}E!Sm`G! z-A|nGl|5ao3e)H!N5bz5l7-;@7IvT;4~xO=cykWZ@_-U~R(yT@;M(HNaFrCUm-MpX z%O_WiY`j4s<)(vdx0(tfZ7bX-e6aRRzPmdbcZH?Yp1A+QY5$~et{HnwE$orUfpblAge{JD z<}x$1XVW!>S77-jH{~;2lJFF4(JAfFN`KoL_X!GudiAH;I>LdUZd1#1rYj_rjqH4J zWD0X+E8lgD3L>NPvm-qY@IG5Ff2s+S856%3rEfsx`f6!jJ)SL09 z49W-RRmC#)uGc4BG@%Q#{<4!M*pq^~v6fqFek@kQp3T-|Z1Y9DS-I8Rg~G9i2~wbce66(FxkUB5u=XpX~$HCJ|RuPpfKg$Mr>n=gbqeJGs6-#cI9B=n3xpF?{L@!`I!|k|IkcFxxe;t|;DY zT(VyH@XxakC%za~A7@Qv~^Ac7ge~9C^ z#LOsS_ntH4PmTbGvq{$@cHz^wlN2|rOzNKSVmF(Fk4=*)Lh2G%jeQ5zELwEs-4tYby4z(`l6sTH*St>%xv_^5-4i_WwJF{z2=3|DZ zPswd|qtZUJ+MGQLT+4cLVl5AwoSEEa42L6DL?tRn0RoWFHJ= zRSX|1b3@#tp?F#gi`eg}(#IbX+F$U%vTYmf#{t!#z%Qp`DuKyG-aCcw8f2u|k@mgb zzRoVYTyys2;oCEu4MMi6evjrV-s&y=PGF9nTixs1k<#;ZcHhQ31dDIOnRMlN0-s`! zyCRWVGFRLACe2dgbb%c~LqD{y2Al5v?)vn6isj5oOEx_D_$B~V54>kue3B&<_$TXG zafH<4FIe|^s2TSt;E(T9|?WG2&lr7)vhdbXnEDb@R7W zm(dMd$eri%vCOIZJ)TRS45=+RwZ8@RiU5U79%pBDr97%DT!N;X9J0Zqqu+|AH1;z{ zW!$$6-JQounjlNKu|)sXkm+SmWQE}op>58&^HsC~sRiQJwT^!M#$i;C2$o%_+M5z1 z@?JjpmH#7wnrEKphBlO20dI6vyW=?Mok=dY^7r;bzCG+2Z+w^g!jhnh68=SnRD)T! z@q2GwZXMYJmPpykAg;2TRkTPQvmBeQUl0Y6&yE9Aw~U7vBEzA@via}$;KJOSQG-F@ zw5UL(D#2AOZ11H+sT=q%G;$eyxzdr;`&a?DA6Bk%9yp!{2T@4|Ja^~k;1i{*eEH;|6 zq4Y9dBoEy&+paXB?Fg5$_w2>JdGE#EYc7*}6eIgo9|=FM|2(VKspI`Rl-X_X(WuVyZTF>I z`IXDYougtN;)Qj6aery)f>+eK@c?YH-D}w9qaR0(K7hS2T($5f`9P52wAa2WP&;yv z0{74CX6hN(lZUI_o6oOSEvP@=B9`e%V2~QQ6rAK7!e!)H;~0HvrOsWdS)JNDr=OH= z+A8R12PJ--N5DD|F9!LuH#r-gc%9Wi&<{$|<)oJu4ogjl(g|9d$wCQft6NJh(tVQ@ zMtYmW!Y{c?glFoZ8WE<> z5L_*qxkz3$A@&x>l}ROn50m%>xEdW-#Nws?C??p-)zT7({*`7lx7H412F&~MMkEfN zyfFVA=%#qIIq%MW?Lh}JUF3jOa(U+SxLz<)8eKg4EMAMR#GHTLj}Ctx=*G#0dqd9< zrVRpG@Yw_y9sOY!)A{ruTr!T3i`Cux@_29utSaaR+lsrU)ThhFVp?78(fVpu>oxW* zy4bo52ZyY?*Jpdi0dRJ`LHdV{U8LlDPS16p4*xEZ=S z@={b(bS2~AmPy|)U*6#5xwx%m7eDVyO&Oh4KD6eOYw`Z!GOp#{PLHhe_4)1{e%ezdSxCHh zI70b6_?vbhiO|!8uzvi_ode2WEn>^1mi}0e!wD$_<>+yG6?&p?e%9 zjBb6!$y^pbdNR4#cQit3yzU%bNcGC+HETaDC*JC8Ol#pBs*jQVP4*_s#!(mR@xj;o zT%-00v#rZ;+HZ0t-&UYaL>^73Y-E1*(w6na+3n0Ru}1fUKf80Uom$M2`$fqzkX&K< z71W`YuYcvf9K1$2M@rRUj=0sK{hg!L4~I?Z+XD&dIv#4SBBoZMSj56LY8&RLg7i5jBLuehjz<_Twz6 z!q)Fl%00OZMbM#9U6PNG>nnqH5vZP|HQ^&vKUg(tzs?-l7s-M~Y-VAyPp*U%u7G}W zpE1aO5b|*M$Y(*3GU>#r-df8G=F*(`h{2J*EQGL|K&v%C-0`BZ75`zkpH^$1P2;Y7 zM@hAb-2IJaG7apNX2)w4h$X908P!h*za9N8d+Q!$#wt1(AN#fU`t!z$%SusqdrbD6 zbqG+bB8fa7eOD1U4qlZE9g4RSz4t2*4VRsZ@|eUcE7b$2KA^ZPw?& z&8o_c3)vISSsD|@*W{a-tc`?6WXRY0FjP0fetf#MxOwB`B24>!4?{q!ai0L+YZBrq z@>Uhl%$i(964I25ir>BWW0L3L{@sPn)_z|p#(QpHotZq{Y7P9@Q0(Ds$~%Lf87p!Q znGz{?=ltur>zAgwp@B!>;P$;+hF`9 zd(=44_X-!`3prFR1F9VZH!KDi>A(J5nuazzHiOB8YnKWi8wC~LY86kN4WwK8l+AZn zf(jO;CE!MD=8JY21E(sHqScr_kH&((bTsOEvFr+Tf5H4)B^iD#sCQA!_0%t(mbdvP zIlnt|j3M>YoXi$7Ybt8S;)NH{qSi8^<)P^Ux;&=@>vSOpn^~$?I(Y~NoLksOj?I^b zLgaLjO(y3Q6K?wjm}PL4*wFdBPsjkBakX|0)%2Y>BLkEr!Js~*68Crqf6Y&$!-cyF zu88GZ9FY$uiW!qzD9~*|8$XRdX#QQb{_2Aw%EOnlGG&G6#?eY_dYbmeTmymdW(c^qG&axphXM7hLsH+w6^5rPk zXi2tpFt)M!hF8pE7vvhQbe?I`qPu3lKDfuhm=m^d3f=V_+X#nkYj=KiP#@jBT(d6s zwM(%lV3qyImPYKa6p3oZvmB{{#j+fjp0Q|H>TEc+Aj`3YKX8UHjm8=`6#KPh>QCP3OYzOoS)QZqlj%QWJX>eu~zvc zjm{)(TljZ4d-p4sJT|XUoQ2H^JQ+-y2*MZCQqETTH8WI?pU(fh(`PXh$V9Jj5gU)A z7-SCsj9Yo$$%7(6zZbM`LYu z7w566)9)_ccRz)V=J+Tj9Z%V?=kFu_yj@(lC$&2beMBqn9a&YVHXb?_YMIRcExUy` z^hUW=uURm)aNcwSLbb=#I3Lr!E*!45w_gUrI4cx4V(903{EpZ2kZagoNFthK`SE4+ zw11!%2#oQf&8pdV9uqxoDCoO3tr?!=z#i3H#!$-)l83<{Qwr&eZSzHhAegGKkrM z4*O=B_+UVg)hDvRX(5XbIe}#xLThv3RKg|buf6y?uSfeS3aq{;e_*FT8{F=ugzO3y zpk;t=^5(I^qWzE$dxd%LqoPsUQRrPH{p9@btY3_iqJ!yQK2>FM?TjsV9t;ug^+)c1 zhTau#$6u>FZz%UQ>rZVNZkZO(a}9iJ&EGSebX4u)jv!>W>5Dnd)KXhl%{um>fS0=_ zIp5*=;$_zv#j6&XI7)rjgjT>ix=$^-?Uu=_3}rreivFpt2wW4-q|&@AbM5WpCyRn@ z^&+CD;%$Unw1sb;9MxL-CsSs`IG2p^NAuhxdeKPS6C)_46$mdlfqn$1dpOFRu%gJO ziyqU5nmKSd@&KNTU<$@)mvepRu6;oRs}lF9%Rkvm{aWGoRyer`{QQ_FW{RBIdP7el z40h@an;z?z+T7<>XWr=w>M;1t&>hS$ALz@n-V&S*Ik~N6hij@Bxng_F%+sA2K{2e$ zZLV0vv<_ryD=aGHqYC+A_)V4c;iBT(@=e)>_m~upbEI3}_p;7Sq~T#Cz=3 zNtTeRgQlH`RVlOOF4`0CWA>FOI)B9!G_2XpIo)0Jg6|7fKmrPmb>()bK2q)$joDj8 zQZEf?uxh-xy@HI+jhMCeN4gCAhVA&JXOeNQBCYg0cIoL4!|rUn*@E8aLkyW+nGOz7 z5j`#VYMe^E6NU}58-i$-o5d)pK%0H+9Q=-Wgp__&Y5Ho$K5X?LvgcXb;=P~Ny3K2l zB}F=#<*`EYvl@Hsrk4Y|1fKE|k~%C5Q6j-4SuAyIz#kay`J=*obk4l*T)Rk){JZ@X z-w|h1&EhN;f_skp&bhy<#T}G!o!8#SLfIx>7j=oZ48;oSS}&wa=Ly4TWFn~v{ygj^ zar4xAzFW*yq-Y|sq*u78M=dqDFzE^9c;EN}ZYGVRe9zY%h7UG*@c1S23O(}?J2?NFNa>Kn;h;S?XunR5;! zZCwV+`@-GEs&?;GWGk6i%=Jyy>~1>;W*>2%vZ$ZAsu$Kee4M^d_riVWm8ktfGl{QM z`+748=h&PA343N5+okp7&*Q@pHnkA~*?w<~sMb6+Mbzyhb~J&4<66IaNg<@75(v*pOlbndRoNiQ1Skgp!}rT@Cxa4Q7^jJ*)s9( z2|Jo4K2R|UJzv&pZ3)3LZLBMQK44DdcaGTg;=P>59nU|!C8<KB4nCo}Y~< zvUP%Zsqi`VV8bDt@m);oC5ZV!;$SMluQN;io?<1<_(xP?-U2BSH!1QQe=l@k9WAIu z2se_31;1||=U_gFprdJ=0(SwimH|GT0Hw%?3^B`-Fh37xtH@g;A`am#=VreL(jh(X zMo@)KdR{5RWsD9n0n8f!n2g)pK51ozF2n&wt7|9-{6UgH=ACh@auyVBu4|Q%&lF!*?NY+ z@HB1|I-pm$DwRUW)&NRzxZG)^c2#B#yd8#kD@{|V>CQ!aU}=jE7Y(#vdNb|Cu!NoT z;(g#6htB9voeVqMv~Jd$n|FWCc+K}Xhzq;X4s_!cRH#mFOWsdHjUpan+P|WB;TCIC zv$|97MMA&+B01F!g@+>_Q4I+M47Ok+AG=tLyq6bxZOU9n1$R$peO>jT6_jjoBYAJ& z%a+b2y#XbY2OR}mrV=7`*xMip(NiF z(JH#VYU_18d#6Xt=Ot~!NYXT~t$}u>VG86vf42mucD0_bZ1v3Dj48LJwl6Fm&K_Yu z?AiprT@JiJWGRyWSx1p`%O5EE)nKW6Rzv80RpvQVB(Unq`EwM9X9Iyf)6Z4Jc6s<#K= z$Csb$aD;BqiFsU1-`IKSkD|E6XB>D^6Kwg_igNT>dimgKU^tHBQpX>@^c`V@Wy&>H zc(vVWcCTB5wU+grsv-U@ZGtM|d-u*qfD36QVZ*s9J6iCL?cfhxwXWnYpS*8wA0f@& zb`GBc+78sNzIVNVw>kk^AoZz^+XFDV(>ephL{%#TK;2bj8xo-3i^9KFZ~UZx{{ULx z}**OJ1l0-F{JFJEHgBklkL#tRkufI5P7JIP6^~ zmh@`IpyVLMH9w%YCA_BzXM+NpjSkPJov?-3O0baotgm=&RX>IrnBXE!X7`r1A@6>Q zQlytes+$mEm)PxSTRfV+=cYuNH~Ye24RVVAKvieon}w?rI20ZjJ)f5B#+at=W0$l z*jrVa*^_TdW0wp|F>AWRk4iK@xf0N(>awj4~WQ5MOr1BTR`gEkYJ>mHeIbGYiO1!jsnNVc!^WhIa{j5MT}liJTG#&E{TO)9Mhzuie7eaZ z7xQ=+Pkpa6qH>jeRB-|0nG5yZN$4jiRkgg{B5$mC+IDME4kj{DZQkrg^xi#4vh}k^ zz=uKik67Fli?6Xamf$M)8DyTzB?x`y>-JW{W;992#;aB7PGB)pyMevJckm&sc+2SX z;ba;ic}?dj%yS3yQ}>4&)IkURm*K0sb6$a3BglEp#KkhZy6Qk+xCr08>B^t zAcAyviZqe}(n@!SG>VEyw{%G<(j5bc(v8yHFu>5m05h}C@qgdzy`TL&``cdc-XGps zOP8@2F0Q%G^E{4U9j5$aG**TA5AaqeSDHseDA@JXBE z8Ql+n-&wWL0oFuT`A~YQ(9yWZzGF}0IXymssA);0}MFz;J*p|0rnkPMMW${oS8p+MBm=oOy{!<^|bnv z2(-Vin!0}5oW{#0yOztii7_lDb8I_bA)3RW`b$bpi8k(e4Hgl+wxP{SA(QGJEk4KL zRV{2pt1;jgbdE}zYv}tl4Mdt%jhbR4{QB%XOLW$7YVizr2jwWyJsUeaOh-Zc*3*n&v-W?pXihMa``M%I0Y2lsaN^ zq4n6)H*L~o;p65dTnKi0a*_y|Z!tmcY^lJi^Zgil=^dX|-^+;&{=+5&d|BKS;%5c- zI7}TH!jYPlUR3-q*NMgQljSL8IJUtrSh@<}=dE6rpYZ|Me;O#tTLR(gKz1F=wyvt2 zbhCg~oxedtF;&ziZg<8<$+Rtq>3iBl%8}J4JJO(z_jov`yDv&vPrvGO zvCH3Ft2{@k>X$+NW3t8A0h^e3eYsN^G@%7RA+BBODs|%?TcRqA=|AsvUOSmsEF`R_ ztJw$kWwM>ueXom?=spuq5TKQbtfd34o~HdSxAIBL!cpk|0~Kbr?1{Y%WMVh&JgTwI zhHpu0bD2U4VV#=~_>q0J>3qI3G>rkLA8M{jvbH^9-!Ig{t~Um5W}$*XtGtHlRHS;% z9m^6{Z{z^vHB@V1uFF4E-W@ET!f&ONDZykd1fjG%X0XNJSn&5g(2%bUWnLZyp5D9= z4sZbiLVR)9UerTVl1Z3i#^k(Q}w@lJFCc>J8*}0ZLeeT~# z|0@lkT;a}wNmwB_Xpg$*ejN>e^3pEVlv0l4s2mhcTOq;A@w!w$)iZb?H4f8CV_#Ib zzzE8CAVzd3Z1N$sz19 z$10q?N_1;m=HVR1O)9Wg(8)43FZ<_-IX$BiV$}hEm&KYXXr90Ly{Brx zoWa=w6aMnHK$Jv)-XPohi3^aG@wEo`N4KVtErPNY!?wJ;NvWOpG@3*Dcj-fbx*U6F z=@h9I9y{SqFZWdHxI@oOO3qDga9YrF8H;dQ#k`aj!(H-_4mfAHB%vLLQ?D4f@FIm-V<)#;F` z>q+pzK_x!qI~u=-(X7I5_coLM@s1`5s=Am+8``P93x&Yq(UT*^bx8z1)pZyYGbHTv|{U zr$!NQ@rxkEa%%JG)G%^x)19{E&i4tLxisXTuHPGDmym}oQ7Kr(9*Mb+?g3DJU!>&_4F;|EIA_#Hx)I0wvAP-Uj;&P`hmhR1k z1}}Znme&Pn^ja@}qx)XrHjIqv-Q=qeInNK@#a28EG*E2TS{CQVo z7boK2FyUteb^pF(Jv)p)EA_{R8~N~qVhQL?bQ(D!L4-eA1Gt?4^cs*f=M8XX-_xC_ zpV9Z^va!B{LqQLN8^7xOkc@k{;&*8fWWxP)jSjQ=8E(-M8B9zs+e(D?#5r-RJ@|Rl zO8QtE7Yw~chIzMjzn(vrNK6b1BcY$oE>}D=6CMoxbLAIUH~CDlbw>A#;V-_G$ajaU z_6Q$Pn#5&D`N$@hFR+4YJ4MQ5R%yX_+SrdC^Yb9+_GQ7*^%$6UQmBQU`@1!=PrL+r z)&`SwsV2TXa5JAKxxJd@-I<(OwbB;l4!{Fs@W;E4^X>xm_E&v2cDL1^V@rYQLrAr8 zPNP=YU_#L<>%=#$VpFte8h8$nRgbrObqh%=hchDfhWZb6Yzn}>p7#D|^Y^ItrpNTx z)B9Mio&v}+PpuN&?=79}Es_(*TYX0e?;z^|gNL&|dnxyy?d${|0q;{f+pF9MUjgni z)?^phy4mK3ECij0Omvorh2MTp7ISEg%b%&jg#CW~a?7E&!B)WHa&%OnuZSfRYE=0I z@x>hxUznTmw*r9F9`+Os8Zv-tDZUHnhNAjRErzJD2k~?Tt-V0SoT&4 zx-XCEj`QpMW{oSb9L3g#U-sR85b(NaW0?E+?NBRJ7&c;aQ=&OHpO2xDQ*lCVtk~I0 z*|F$m8B-fdN+T>{@An4FW9LoEy_I!`vG1_!-${9_24tGEm(8H>KhA}knD;o1q~9V+ zop8+!{rxyz^i*`~@qKAi8sDq6l3nrFzc*kGLXM{wEs3VtdhqBr% zk5N(97LDTUa!t=ok0Ewpk1ZDHR-f2iImk?3GU6Tepo5%?(JqN)KL1*7!n5A{DS%5# zaF}#a{#^0uVbI)^*l)6Z1Kq4g)MoPBV|fRw2V`RGniCUZ)e@#IVHeL=Dy3Wb<*|2;H1GWmxY)1{;%)&?^N6L+m;2&MpM(uau9KS_P;y!+x}E@bC9y409u z{xa&M&SbuAV|^s<4j*u^5QT&h-&4t+KWJG3^!XW`2xF!nQG$bT3tKoIKH3eH@fX?zh!wU)zavcL$$0*;aqHAdsL5sQ8UE7kiGx7I%Q@_%t_O^fGL^+UEq8F|~U zKC061OJsHQ3obFotJPnK9o6|_*9l`bNlP?D1^$`g^po|{1zsM3*qB@9fCv7f_va{H z!oYkmm^U&5|5?pMVgf>6 zjpRPv8*+$i3|Jn!mnz{Djqi*x6!700K(=}&Su^BleR2NYr`;Ivm@|&JA!fEJFy?90 zj_wK#^Jv^J9=9D1gL=EL=OvmdFzU?r13S{2oTG$$iZR(!mR4>4n~$5%=*AXR@Aldp z3gbqQy;Bgqw2s=b-S8hC&GzKn^++dlM!V_-md^TDAcNYGl)zr}K%?~2goOT<`3YMEDw=5CAwLUXS8Wzu+ zP&8PFxQy}ossfCq&j16PdA*p4N6P(5AOfDuuKT=t$%HcE4!Na3-zlYp9(NEB<=0+9 zVjr|nku3T`Ch^6uUfd$*Diu;`I{I4~9#P}^?E!o>^<;lWA+2VYt@(^5@r;u?mG4)a z!2u}$F8o7*C?=ooX-}@GKi}ddZrb$T6LSY772!3f_WBZsw?jfy&!`) z2MUUVfIC+1@+{#j?9XD(PtJ#lzN{%Vt^YOo#@hzbef9I`e&e-!0QX+a^k>VV>L0Wd1*@2laDocy~g78qn=F3YRmqNg%$tUB-rkV=HG>9#y;n> zIh-*6?YrMBoiZxl%Jp(>`&jH9yV%$YWL9DeIk|H0;u38rnEV9cZlD&1+hCe$WS0sD z+7XA_nZ8zB=nV2z103lIO)iK}%ETBjo&_#E3HMc_*Tgpe_s0-gnvEqn<6xka7@*ii zQlIf6W&d9LHsezzq8i2U6~X-`tks&AbToKNtDxBej;_9)Q&WS( z&tb6-9c(#meT@7-S)+xcsz&lk87#q4q33p{(MD^69jBkv7Dgo{LtKiasH2kaBYHz z%#GsHCr{tRyzzEt7l;-_R3@X&EdklEfTG1ZP+3>mis1(aRc^rcPSSf|1WxEewd9{G zr=T~5#%T11;%h(^G#r+Hd`JR@@$H!4mKe;beB}%p9I@Szg8s?1e6m;dZ?9D@Ph|RE zxExt3*D%?MB`MbR#F!qnzrAheM?`r_aq}kgW_s=3W6JTOWVP*ar4KbFHK7hKos2MB zN?s0!!^R2Kc_1kLXn^av%BuHeQ~g3CJZD~eud7vBjJF;t*4_;x!t z76ACikpU;t#V%xWpYAsphWI_z&6wr*K)mHI96krRi_m?EK0?b@B6xqB!(H#3( z@|f>;^C1vLvG(hIpIjs>jRrh9zLngEqrS)nk4;F+@kt`wnuynB5%MRvGwc?%?d*+A z`Qy!@vVv7$*Wv_R#_!4cy}6>}WKj#l^vKf%vZimT z7|1)qAoH_*ZS0d!k#O7M&KTa9#Uw7a)3)-ykiQs6;P{jGge7vk$ZGdY>jdhKy+DGP zUNHLXq1KPr@nY*K4@w7M$T-dW!ez*ziVaZY-g$Y99g%`$0q7quG8j|@HGO);Bc{V7 zARizRRFsAiaJ+XRhTKo@o>SNsxlDYYgX-*(CIuNvtM42uET-%E` zXRVz;sjh{NqMns|6muqO#F&Fui+FRe{6iUlU8kEsWbJ-7$1Li8l~}e39;+QvbtfYM zSGXhQTaz{3%aSLZ1E@f3jlG^3&abX&r5nw|jHbhCWQ^|;9$Vlyd6|N#yqw1<)55lV zMyJ8Ai^`x3gQ2l?U9h|nSS|VPo_WI zsjxHfp>eaf&c5{PwG;UJ>&vGU8#$a^iairDU6Ngq(m_Y1a}W;=Pz8Q5Abx+e?kSNy z>0(DJnuDkbcq1uY_URt~hr0tQI)=iRTKK0sIpt0>b^Wgll>a{AE_hj~JbEJ7J^p93 z$)I=Z1DOKSB(JQ*uk8AU;V6wPsmaRWc~S0^GE;JmVK@eupzWIw>xsTZItcH^B<-90 z;(dKJ6|@OupLCuIlw_E^MTWr->h!mS|44Et9)hYgemI^6GA2f9wj{tZ{6?!;;t4ViZ+Dtj}h(wLr8lT zefDcO?P$K@pWsR)tyMlL=5`0F#wruPCZ2{|hDZ@usbw3k{H&w-(YN(84Y;3?fh)fa zj?U)AgInP~R;Evag7!N__9upI6%2fFAtrG=vc^cC;Wtb|!Tx7g?};_6G-S%j8E;B( z&}#fW_!etC?X*<@tR%o6aM06>k+2RrHw3~VSM@;=86sZlVdNLK_ueUnKzvu}zrdC% ztI#97+jIV(L5Kb%{-HW6$YzL3uix(j49r8HGY-S#kyub|Abp@ituwv+z>amB&yp3W z##-j_!TgiEz2#9t;&7V8VlAe0(`tg_=^Y@80S;KvfZasTE#4OZL+9#Ecr@1b;2Vv_ z2y8m|MV@>FW2;#GS?A1&F9O>?D4+_6)J$UVjy7U}aHjKK9@tqdfQ#Xq=DE3~+(6i7C^cXnbJL+XP?fmpk)A-9F0;!n3WzJB=Q& zH=t>GF0ts_*CRWR<7le=P=k?->1ixL<5d=OSHIZ5A6p^g|1j5zmJxEA|9<=u);KKw z`ojwVa%715XemUdvlXb`O$IBtuYi4W9Tu$TP%%#nS|w~C?E8DGPo5orq%_$9SP}38 zjE0B?4d`B!eZimQHfwd}cd8rMF3@SVPJ8bP+BQnzb&_=W-GTJ+LSAdo5!N1Bo;Q>f zP6FDL6-PRS4X*sjQduA;@bu!D@99PD(+Dd0BAeo{w}Je_MR?K@YNVMdh3K${1{P<&f$lKnC@Bq zY*<{|*lRwTr4ODCG@BUM#_=@$5q67Zka{vMylrS~Q~c^PJ8%Kvcahcj=mEsH0C;h# zJQ0S2#oaMjnf1Z;2d$VK))CAq15S{?81`0rSfq!gNa>WYLV9d`H>GO z%M1<|6-0dwt6E+oRH`4X{EW8pi&2c+GBlO+t6?RsC*p(Ym(71^BL}e$Pu|~L-}4>- zz|+8I31cDWS#7Zm=;?lh4oASKcBc@ zN1()hlpA_)0J;H8Cu}iUucK76m@UNUVJU}d0)bIn0c3w@K+8-`J{JZS0u#8v7E& zshq&p=r|402OBRdI0aG&?R4zIFwM`7=W8LtT};)(^pl0A_V0!Luiw@ZV~1L!Fnim7 zy#70tK*s<7CI#^y-wOQk|0lf2|F;)q{(onSSYPtw|L-M`o-`j>|J@Nn0CIHC7mW(V z>?JBEiWdTy8zc~>vt>KMC0_*EoE1ET&=5ZdN#B=bEN~6*QOo-{E9sqW0 z@?}fqqL@Pm{hia}P%+O(GDUBnKaN+~8tFh+9(IP2j5We}l+(rDM^cWBcf+NF549Jw znH&@-r@qOB1cU`23PE|qZ@1)H_{4JGoP$aGCE=JO~ht82NHsa ztryj67QR16hHHb`#B+Dr0mjPgY1Z@vT(DW|+ua#|B_YRY zX8Bv|Ng%an2a3V#DXYottwPYqIVGSx$SnXD$jzX6ZgH#*iqjgGIt!dQCQv#j=2{f? z%+|vm`cFiT9faA(v&r*Cz+3!~3dA&0kH840BR6>=knXTii>IpBH}@W`(x@$8pdR%} z*4_eyT6E7>AL!ov1Qlc*`iEdfNdI%(_Bin@?5_l)6F?x8QXXy)wb}Vzvyc1cejwsR z`Jrr!$Za7<3i=}7lk@Q8`-)vR>_#6yPcS0-vP3q1Yd`9m_koC$ES|32jw{iy*;OZ zi^cCNQW#Na-k#~X0F77pFxI3vVl-nvEsy0>m`Zjd{_AT zEB+yo=sN*rK9^C;gR}J)L9QoA6mPN!1g9JPI^LUg@&W?K?`ve@dTPy`cV76P_dq4y z4^fvrCvUIxTVA=B=|l5v0!k)_gDC{~o!xfL7vXkSd!TykX}yf8X=(np$?_SPH};Eq zo-5w)@lSfhsBoPdaLuFHz|VS4!M-qs){>&l?2a{XXv%9h=I5MB=_G%jmPotXQdFuYnG=N!&d9? zJhoF&1F^$jw>QHAY}v^=He9wcH#F<0En0p%>D)SVc2 zd)ht=5V2hK&WnPrA95Z;NRHPCp$Io`QDa09F8P6ID&Lp5LLx{?W!u2-VSHSbbm3A? z=7@WVpy}9JB&~VR^Xds;!M-VBFvWpmVDPg%tP-5r?g&{@_;a(MK;`{d;bdg{Jr&ti z7}kxB6qUuKM5}MYJs|x?32^!~b6%kG7E~1~mI?74GDG(?4T`JD((iguju#|@XuEC5 zwX7^y!Eme_l~Sh2*KB#$4Pa6OIb0OVq|(um^b?AU4v=g@3aQZ(@h%5?>6RGzYu61> zpzcbJ$&QiqtYJ0Dixns{8_yemuVwAqwYJQ6Szb^T2*=uc`NplmErV(`QuwRBTo>{v z%oyTlQY07hH6Yb|3^;R%M$%_n0%HKuX`r5s9D9R%X@C(p*oDqOh(8#Kdu_BtqP>#0 zr`p{?AMsA$Jhy3l)sE?P{ma9f^NXXeI;Zpu_zDuT(|-$k@$l5ULAeYXJd=L$hKIbK zH%uycy&NF5PYph+lvaXhrKF<*Ph={Ak<9?YNIgr;X2uz?qT|b?R9BjVdHHX7Q9>U` z&w&%T3YZ@<3)RA(0E%8=<)$akkC9t1Y=BdMbghdK9tFzOEmUMQ6RD)(bGwO(y8Skh za-#itEr3NWxz0JO*#xzm0+0HM_)ul%rO)ZvCD{Se=9k!XD>|00d1Es~Z*XT8xAT%D zJ*WG3wvOU-z{ICN)a%(vP)pmkp?FX?r)&~kp8lphw(8zT0yn76g}~T)@L(-YQtkXD zaX{n_0`AV!MC_VqAqnJhw%EWF+Vl>ss&Jg@v%{4)P z`w~{yf9Y^{L=aFbgN5?enCz;fvnYLU=m$>mb)diiPG^<~IJa?= z_6EnfelQ93qe$z{++~{a0U*Q+?HkQ|I`NYI?N17Esou?y03Djpp|@m|6E7lXEizui;BmCq4D8B~zyIb8I56y%zFH@_j}@CfOPRo_@}tEwiQ zE>IZFN(?H0?KHjqHt!SM@H^s#_(l73{tWDsj$?$2ZzN}DJ%uPi3YMN;d(i!Urox9G z$tUwJu##_N;#VxT1Rkr$ymG|xEGsrGDIF9!n%W%cz?se|o(Wlf4HAmd@D@NI#34Hk z4)AvUZWr9xMjWR!EmUx?G}r<|lTJ22DkXJ>_gbD}1S9C~#s<7SwYO$$8X~C|jH=;8 zB8tn#R=8)YT~P&#QEUviQc6#RMxqhaBIfZK&8`MPpvBcAi7#A z>GJ6cUGdm!WzuwXycoF-pWhfP{*kVGk<^_(xCf2J=OIYC>0t!|T$`$Cv{9S9czSN= zjk96W+fT~FL0PLxOd@*l0d1-+5E z>Wb_}5Os5%!CqMusepL5FTn2-G(&ReGteez`DOJgUBb>h;&43EBqeCx8e-zDKk=_q z;$ICs>qzdvIjA9-0T68XWQv>Al`T;^MJO#jrJYtZC=I_-J-eB8Tbe*;lt;>CnU5%0 z;3~K2`S%wycb`5Du^XS5zYymm<+o8|*A24G?Xc7Dno-E1dvgYe#%bJ{xZIn}pYHRt zeh69H1>s~oH6G2@Gqx^SP7;m-qf zJoqkO$e+W9efBt@kc`H--mY0Kb`4}Q%xj7rH7M=$v1j=nARE( zN4@IY$n=YB_#tsgNegVIn}0-a{GqOJF>lSo=}g@Ac(&=6{qz%sR!sdqQqlewfn@^I z$DLAv5^W`vglNriGTE;DCQC-OTN%;?srUW&3U%4gEMO=scSlSM)$xI;VT@QBRFr?9 zk(KkKm8+NpykjK**A&QR+WVno2p)Q#U-+#1A;pvXn?e)WFJDc^%b#@9K*qOlg(~#@ zBfj1oDpPH9u7;-kg$1z5|H1I?Zjrjm;|r)fy8Ht;Q((77Q(}2rrsk1}o9#QM zAwEz+MZ9bn?o`sU|LRI_?a z+QUCQg9ZsxoS{+IS|7NteJpfu-iY0v129m;WW``W!XJ_i%CIt?E+oFC(j_?))7U-U zJR|FZvc;x=c4MXR=$PJ(hc`|JYl@b3yQbD{z6-ya2LdJ4rBt$C^oZZ>U(ib#g!$;xaKem%F{OTm$7$6(yL)of8TEU*eWmY}q{DRAKL5Af9)J|(`hFjMAHAqK5 zQ+3!o6vw`Sz7@@nYz+2D+}q70eb8oE&D0S7iA1#8CY~y(Kh?DROsz>dflSZd zBy&Y=cf-6`ug{ajWLAloURg+I0KqFU4xhWGPdf3_HbFZLn6( zll%9nJ_Hx*DytS=+M3-nZf;c5;QWz(kYm)MB!##HV1UE0ENwXr zFmD!6miAQelL(kJxj#e8D}3{bEH%SCKi!|l6k_L>fN<^|A=wl2U!O`k+&w7Zl+4wG z(naQtx6^3nzQdFrYxBea3%B9e7d31C?`A|h^P`Nfp@g{QS`fPRTCzEllVSVEpSe>W zsva@>&(s5{;jVTPuK&FUq&#>qzSX!_5$YrgEI7CuT8t2j;Rl8giq1zh$i_GQi^1EU zCeOEl_5TmOV%F5Ojn=CUvQi`e*~iX~B((kK0FSx-$lY=r4teMZSa-0Quz%xjHKx;i)ourl{$G!WUt|P|91l=-U zZH3}PH6a>_P&Ro#a1b$jvUehI|H1&3@{Z3PD%XNq?1i^S{f=e<4e6Zd$lov_K=b$6nxe zyDXg9hbYvMn<{S9YcQwA{rUqK5U0C}XI`MWy_mkvk+)1~m94cf6B4Qak?PmC0dE;u zS6a*|>64DV$I47n@G0;yHAzo<*xlaS4k2A+(44y9!#(&c^|5g`H!<0d<=0yvZs7N! z8`0x<<~Xx%qIe8|toZH^R0vx4l=9ebQUC0?0vqz4WATQL&9>Cu03gUfw)Ned+?OIopzFpx(Culxs?E{Gq)JD}kfp@lq-~QZ4U6|O7 z%680@w;wtTT1`^EI$TXD=(vFL+m_NT_Ao{R3%OoQd!4M;E!?kC0To2ajeW&UOIuz{ zsxHGL%U2-HZ?gs=cI8UXtP}fs$YU$UdS4SR; zirrhH;HQVt#J+do`M`Dkt3{2)M|{`cZ@Ne8&NM!Unc^rooC%7F{EDJgg{xXn5O~V) zwxRH9uAi0N$R19bTKM22MMKgB?YacZaSmu}2Mi8kPM{O&+6Cu)5-KmtRsW|=zqo9^ z;rnX7d4P_;rZbha0e?}-gFrFgO9QCT-bgz~mtps3=*F8BknP+%GODwiyr`hp8f^JU z{H>#CV!eD9XWz79G=&yMlXbZIi#JNc`Q!aT_ARqEZ(Yt`v{Q3PBn8}9HDvO8F4k5x zIahzFZ-mv>XEVX#J$qjp^ugi&0dk-xNyxe ziRZ_9Cq zd7(Xl8g9Ob?~5uYpz^Aw$ju;r%wAOJ;r)tx<wd2be!wDOO0si0)u zESJhBzMOA&d-e_M-I@n4qj)asUW8HH$QJwp2EhlWTK9)X-o7OuG_gY)O*0=Kj8lxn@li`{A3tq1H>oXd32* zSpl@RUaQskg`w3eK+-h~@ z)R`p3{;S!t!29Ka=jU4D#vxwal>U4HK_qoaUK`s$@F~C=(RWQD{DLF3CAt^b9cCpT z4G!yURAdwG%^D292d32D#Ht%r)+*(c)9H2l@kTzo1Q#RIzZM!trC*Bca+#?FqpdC+ zbsqcs@O0Vwtak@Qo_X4}h+X@>*fMS{t*FT5Y3}yk4e7q_XJL%S*h$BiE%v(hysI;` zdap4fcw;7Hy`jio+{YWVtDmP^oMHxJAD##AW^Z&yndOKL9aQS97tuQ?$6@6S{b2FFLgr^?0s-~h$!B@OE#_^v={=ZnQTi~v zj9&vgg+otfSiC3gmnOND>fMQ43QX2TH{El=Z}B0gt(Mcl?f}ELn{RA;jw<^? zmr(r9^=QCX6c{AcC`Kn=DHi4;n=mdF%aI=Cv3v_vor0S*7R(VFDZuQZ+)XF<1>j$4 zR?*8i`_chl7Q^hXVIsml(G1C?`Vo%4+NGAl=&kWR$*%29c>qCotPu#LD1HM2U~gYS z(J+X~uXaB=@ePz7x*_Wp=}qYJ=H(U;qCkRaRbI z19G~JKiwb5Pdz4MQiy%HK$vZQ<c->2VY zT3YaB5R2KjV&$MFcmKQ_NNE9wWmt)?D_5}(Dc3UvRVW-W*JdbAiWxfrw`yxYAj45q z8*`fn#tduT=TWDw*B}Nd5CiTvwYwzrl5rqHMdf5mi`kTBmW>KSu>hCBKlDT1%Uf^q z&=W3y26{%V3-WlAw?XgGSL&|)x>><;Re!tywk`R(w6iVMWf(AO4hK|kdT+}+ zipzuG>KAqUbKj=F0#)&ePR>v{X{6oa*Ge^%Q(baRcg&#wSkY)Oy+ptl+Uv1#i>JYZ z&sK?<&VK_lO)BrD%UG&{^7vOWpj@-MI*_`!)F6UP%KV!x+)2P9D8{8Am`yF4^4P+7Xl=Y+2RL@V;uE9YB6L*NHkCK#E&Sb)Oh*)Pcsp?``O8bkmXS3ehq7}v^d1HkL5LPPjx7Tow z{Ks^7^EN2szf48RQ-s`)YBJ8qg&+~YHbHMocVHN*!3(g>N3nmVK-$Zo>q6yaNHU{Pmt!5*t(YZy89AG4o=Z~= zRz(?-!3HT<1-Wj7_xe(FvlkWVKt^$~gV4?YOX;Q&P7A79>=!5uK&Fk(esFQo#6gW;(uTnRcUou}`k8hQ*{=83Wyw@5OFo5FKCSGY}Lu8UM(e>(=CF zmH{6Sq-k~9#5vP}y?HM~jxB}TPCTedcri{~@jk+9j)BM$#Ix%H-4ccvqnhjdyggAt ze25}S_vrzim34r2`VlgHgM2X{AW?)8cPxM*8}VZ>E#{aoLztr{5imot_g$Yb>X5A#PAfPNhT{rqkQ!4OCkM2#w<450$?U+dr6GCK|L+1h z3Vt9pP3`#UFZ9EDeS6I2vmVzukZ;P|FuxNw{O_?2p>V{pCAv^AwWz2FUxp0C*r=`7 znOr!udjiM3BkEr+UNaKwe_X#>v1dT0CmMYC)ta5ZNByoR!z`Wd(<4y>uT_V)Y{VGL ziJTHc0;MQnO1W`Lu>SYS9U=fZTE_mP0@-5k{+H($&yJ#q%4&&MN}LRjhyowG^RYm& z&NDP};vGm%b)IYB!q0Pg zI1>GVbzFm#uzdjKPy)OBrcp`2c;u4K8fA4;(_9N78FeIP86Ui<_Y zfnLC^^aAn3DO&FRg?BR2Xes6Ig+lIwq5+KBD(VMP(;7LNx>p}bZ*%0xT|1iuSJv#W z$NITb3RVjvNXT3q>>jCag8chhKJMP&#JV2ksvtLaNkYGNF*wWTv)twm6t3G75P6v( zL9GxY$jcew+GG+MDK3nl%dOK-^^;}zyrlTucL5IFKPFcIeUdgh1JShrUT6Dj|go_W|@w&3$+bFIC1Q*<(!jj*GQs`*W?_vQgrFc?4ZPL5M4j zAhu1|_@)S(ER}J0fw#5qV11^~>{$Kkqe}w)x`(eNg(50Oqjv@fQAd;GLx0d*=E^_ceV#&7@~wx7QUKMB@gszr}q&_L0Ro3k9n zhcO;l|Nf!yq_NUQow>|^UI*TK)4`nC2Co$BdLSg7J@FA(KsLL#md>`AI4?GGzv_IK z>}jW?+-jJFvG3i__5zI>kkI0b2JK|b1s^D3SWu9`n(#sTc}hpbEUge+_zHU&vwtJTd7y-Y{tn zkN^NOlsQ1=o*YITSH3ZPC6001^B-iUlq;|GkQF?Q!`23H*bQo-$QbB-T5^;@qq_54 zOFoqM)!PQ`vgcZlOV2GpJ2iO?)GK(msp+INj6oP48RpV<5Pj$C!@_(9yE~;!A%r;7PkHY=&*V~F9V-E zbsn1bT1^S;74Bx^3ob%8$^c6Q2XKgE8iYO#>4~ayE}LQT=<`iPPpa4UIBMT^5lH3o zPqt^eR$jPz84;YFRSTm*@1!gUzmX)vII?QW`L^%Jj4JcKn~Iev{lT;Q#|p& zU|{#6{=-8R(%}vNNyjNR)#UOdFj!h?Zi-I(o1rfzRWRoG5(qI%CTl{s(ZTlPCg{Of zp`=Kf-zJ7WyX16NhVA&$j43iAqEvCyX z+tP5~f)b1cdoojJ%Yf0|V`4i1GKm>>IaaDg+nX%fI@jo9CXJp7=TCbE=DGyg%~aIs z4IL17DQ1CCQ)D3MgD!9|ilJp>^A1cgI&&<3>IEDaLlUMc&37t|n*H!1kT+#{X{ez> zu-IlFdRrTr-J&G+#CU99_t)a|pmY6=XcyWaD=P^qbCJ45OOlQ*PkGL2WE=#N(Klcq zfjB8tF}fd!Q_`bdY*M%x>O*}3_=CX|!G!5x&LrTfRsxQq0u_Wd25g8J4g{`?fsEt& z-36Z)5|WFmz-G6$nY|Nk_QH&X_sfsrJ%1=7&o^I82u3f85(Yw;j{~VfGoCHWbMGz4 zFCPtAz3^$60BS0EfSvBGN}O+cyIY$EhDjaJ*Hr{RjmZ zRK&2W73xIXWj;;*f>^-BOj=2v(rNg*n5Y&_WpdOwITk7tC_){<~av zQO?aSA{zsB#JhVoE+{YvmI<%}lP}Yk*ZcLUl)6rBetd)gJ4us>1uE~}-F*piIIS=rd~|s5@qysCfX&XUFoC`Dspziu zqb~OxCRtdA1!tkG4eoBtd42vgFUF9QD(Dja%RO5pP(64%pYaiq+-o|t{A!s zL3DpBGAd1`7n*vqMG%NIKC^#;heKNNO7sFkyp6S`Gg3 zGrgy2IF(?PHwU)HJi5!ai0OD6z$)v{!zgShCwD*{V$Z^MEF&Z)lFOu)m0`dsRvVdh z-t0JAJA8ts;MxWd=9ovjFA!0OdVhBN5bVfev+=9QqG#a>OyLkmMSF;r$>DpxYY)5v z;XqdVp8d#<@^SI%Hj!qoD={?6m+P-r8x-n!ODr-td$uUPJ_p8p(x`-4pZ)Q!6@kYI z=?gw!ur5&*)4^DiLEs=?0Ai<>0|A$IQpyms$O1C(*pfFSL^(ZcK-*maS8T1=6+adjHwIFBe;9&FMgn@$AbG9w&W zoU?G-0TY%QC zXxKP5ybW1`bzk67nh#$^$|w#kFfw2HdZ+-c?ONA&@hWSi&|q;#3r7|*;(7U^c3(0F zBbfeSdrUbrP|lyd%8|V>U7ukBODQ*jk@i21c`5jI7ohqHuy09|IV@%R z*Qflw?!Vx2N)vW<_NNu)+5K zIFxvo4S6Bvgej;GP@QAbx3E=hQ!-HM+5SPt@IHx;qcehqw|bu%IjFpEQA>5a=68 zf>LAu4$HN}KlGf|K4gS9vJ)qZu3gNzIJ{1un$;Q!>_Ag?| z!jZ0(nJ3=FqbODLGFCFni4>lnBA|UXMM`X2+PXTa+La?~V5wzv+tWW=5lEYF;o)Ny zJ4dQ1;V5M@3ZBP$PbHTFZFL3*W4HdOtHfL@4%Rzf1xTuQOkQ4ce0OVKDm+{vrMqAY zMr#S1GR5H%%&;jeq`}N=nvn4M8rt zf?0S38(<;b1kg!E-?Ow4Zt72YI8Q_jF)Wt(Df$KCD$n{C$1wiK`((9%oO9{XL9}kx zNiH;}Qq-|ubDvVV%d!?P&?WLBoOGhNgLI-yWl+!B!bt&FYrmcmBAD;5zB4{I@qAyq zksl71%hLWT36bTl#7T(nRYQ_vz;G4Plr*4mis=JWGz&oVer}dB%d`b9m|`hi)a>&1 z)~oQ|np>&PlenGCmanC^Mc)BEEk>NIg?P`$T&3TFj7a7^MKlCYiY1If5L1Pbad!;X zGWt5SQE|X~5jG9!SPXURPB&$(8zY#a8fH>yPj-0h1`eu2t*Y8uN1NoxpDW01!;i#C z`;kW2;4@bi`zH<+xl}~~dYnfvI8Ulz-tlOm-^Cdj-gK_NzjdwtBu9s0(C`>p@+~!t z7cq{OUR-}{`4jivq9Gdz5_r~iyC5Sv(Y?PTB9BiSnlVe|mAD{#G(BAV@+I&WZ+5In zkgKLQuAOmRI|Fn&KI^~XeP4_+TtaVwV~pAdDwjq0ftqQX71L^jO`SN@wzytNqSD@m za8%6js#I7^Gy=cNL={`lot%cFGQ-Kb5_PI|z5qJ`7j$K7{<&$(ub!G>cS9!J>s&+f z53|;`-C+o#1X`|a#69L^H~di*nppX+qYW{X86d5NQc=P zl@&+J!w)Uz*GOw0Y3~8up9~zoZR^|mncSy*y=9D@qOzuQjsTP>Z-w*BkYDw$QV0oz z7ofs@cH0|HiprE+moGzI7fW}|55ZPUcI%K1RAviY>?nROW_VVG6B2pKIB^pX=oq*U zX7;G*jGfL`{5d^Ant^RO7m)dj3v(I}- zpCcq}oH0eVAx|}fno*`*++d$!{MDfC-|?NK{4;JuPWUZU0P5OpE1~0h zd!V!ek1p>d=bqnZAw8U3L%rKJmV8fLHEvU1qRSSNWL;){BMGG`*Ec~&+K-?2a1JFV zE}a2Zk7-j<+9pkIFa2(-JhlwxIH~1bu4w1J%fjFi3b7F5FZ@_#FXP_?n7J!T$Ihn4 zEwu>9@im5u@lWOe17W?ZR*_lOm%_xLJx5A;sc>sKtIfvFgLhUFZ0Yl#)}Tk7ez+5I z0!4n2H@kG`=LtWAabD!M>97%Ma@Dwnmzd5XkK55YDeX~C#S94BQ$I= zE%`Nz&#tZ@JuW#nJ=SAiqAx;s6QRq46@t2Fq+O<;dmwc2dW%M=HyC!&&T$%Sjs42| zhczn|`if1mqzgMks&)tdwJ%b>fUl8d{LN==&oIn7m&;uFw)ac>=Fn>}WO|3Dkh!rN zC|eIUBh0#=pDS=Pael%PQrHC|N~ghWG;Df|Xh*_q%6ow7Jg|;IPpZZ8>2k0?IH^~A z_`&M0m9+*BSUog+X{fL|n2YetREh|ScIs8=`4<}U9MGQn`(Cs|@1thIPK@KpfO&RfEdhw(VKhmJGXnQc`*I&4I<%dY`5D>x&Re zlXpUiZTwNk?pfDlPt(sgW5_1^I8-oeo^bclSwhd`bd+%y(kiX z#}*0{9U7!TFZa10APj(GOjXJ1n{m01Jg?CDwkM9OycrD@5|!=LaXF=g%UKN7B1Sa`(D3XnO;E z>lDy?*Hj@(I3=PI@xEIVvvbvB+UP7 z4E$t821S@?w;_Zp`U~Cpu6yt}c5<(^ zjdqWib*U}rYm`o*9kA|3YOq)#-GEx}G`#3!`yLTbuUT>E*}wWQ=T^Xq)?3s9V8`YU zyZ_W8y3V7t@L@reZ^0vf{<)pQewbSyBITHmwe^h(_M5vA+WddgGhlsW@sl?%GJtA$ z(IqqQ4RB0uxQs7zF%R~P0bFr+s>}4FZMwkOQ4zA^*N;uu6v*mQF-#5B1+yKez|i>6kX28fkP`l0idyRAS&&8> z5Dx-^pbK8(25$IUvLL^G=2W6n2`VB}j!K>)Nlbpx92$&pRS%{&_fUVx+N82_T77&{1NJT26( z6xtz565X>+!;BLiZ8_%F)i!;|wYV>87wa6m&bQ}Q%7NT7;u!|Iqnxi2A&*r$3k`?w z&+&GYh-d@lcGraqET0lc{tmjtCjSMtP*dIZTA_tmWmC0SRi(TKs^!jIp2>oLxjuiE z#4Fwhy&=&5`Y1K(*!e@B82ig8KOqgqexSXL!*F*fIm z8^Tb%hnxaZ`iy|obme@;a?KCD;MHc0nq?nJ=0ZxjpeI%Kj@8Mj;e{ifEUMx4m^*uh zbl<3;Td}ikzJ3c6Tsx+FF|xt~1qK62eh1O>gfsES?~On7i=V^31aQ5cA8=>%rFvL{ zisH1WK4x=T3zB@#NpMoLvUr31=j*-okFU2;1juV~Ju}sTyerYe=nxguT5_1B*1)aO z12@*_n+g2P(qP6K!(PL8Ih#pxUMo2v#RbFUx1o;7X|>P zOzvyyUNdOnCJO0X^&-@Y*i5z7tAw*BR%zV%=6xR!2vOLbxFn|N{ib}hBv>Q#TXuA< zf)VOfs+pk=8O+x=Vg<}o#kgopEd4JTSr@bBlY%T&0Pa0{y1)hw86Zvkd#Bd$N4@NU#RPLNbyY6LdE-fW*92D(nE zHz{uVPdm&eQIjOg!9n{fg|6OI7X6VfXK|s`6!LKHfd+7%1HqeRc%>Lq-&Dg41GAKaWTbokZFA;GKdU zsE1GI&mW*CgHC{Z#7Rb0EfYKo*mK{PX<3a39&~5b{*Be~)#!sCLDW|dv}dKK7|AaH z454A$F5S!sI}+&nl$oY+{T>5f!wp$|Nvm6VJNIR+L0`8|99LGaC;%neXn+cuBi<*Q zTTshj8bg~3TAS80N&_6dlxfqQ8RnYok{EM zd8x|!KOfF=*&8wyyD-lm)-490|MB21w!hfm8|N1| zcmlj>k{|{PHpHT@rb!pNG0LXuuNo$OkcS9KU%AKhJQ?Wg?1zBu$UB`_fOCfH4m9#| zMiwy(Sb)`bwW5ISF2;V`1*}$fF`?zX-~l|o18dfulaxxe0`agcNq|)L_0-et?(9KC;KEGMCQ>7cmm{;tpH_!<7h(9isk}EVa zvhUkZU#pd{B(qoVQ4YyCFMh(oC$!ZsjlKb*zyI^WTfGaeX)C?~H#r`!*nGI|su8@^ z?4`JETT@;=cNodB`mOFAr+~6WSeC}Euq-$uNA`pXxwY#GHdJz7j5zCO7(PXzw)6K8 zGw>8kb6>oBnkQMDO8!iDU`Rx#Fp&J&iR%^sHPUurk>`S1 zUpNw=JDCh?ns*mkZ+yWby3wN}m*v?zEokivW^JFzEJ!utd}oP4t}lKaTn(%PEuOHq zcE-Ydk}*fmE)5kS+aE#QH_d<*X6(Ca3*y0$oSYnT1w1C#^>eN_JLWS$<+<)_=QT}siJ+w7Ro5TIu zjww)r6bokCaz8jGpq*Ee8l-y542PqVM=Sh=kVCZik{*jk6O4&(XvHCcxa|^7=E8BsCUavx#E)$mY8`5-I`4Noe-GhoZR4cSh5>Vri z&>H6DuwzSLx-|g=#`GIt&{WbZhEn*=c>=4 zHGVsm_{DE-+fq8{Gg?4lsN?|Y+pu99y>VezZ}d=|BJl+HK(mx?aIt-sePm-fTHw3W zxhc$uu}CFMSLcQPPQjps zo?nB(TI9gOgZxpE@V^!wVi!a27b!pv97I}{57ndGL?D5u)2U_f{`{9N;BEu0MJ%t{ zodTN?K7rkt?-^7Tk7XIr!N|XdN1$yj(iJU0b=k=%CirGr?0~Uy(IZwnIVUFvT8aZa zLjNbC?BoA~sgF`Cx@?hxlsCR;cY(nlh-U{+n-L7q^<|!l?TSIIWdM-cxEyFl|NE6M zkrWvKS%ZIb25)R_JqBoHtb5V|uRbssD*&K}|AzeE2cCd-e^mY7q2YhLb>#ne>;ElB c5KrAbIyWmAx!v(TdLMWh8<<`yzvvY8U&3_5!2kdN literal 111302 zcmaI8XIN89*fxrlt%$g7NJjw?Y0{+xRHRoa(gaj`36U-Vf{MykdhejpLDUHYg4}~bp9HwEnLG)4<_ik)^?E8C;ua9-1@cpn zk&}~=mlF023VIf(CMyg4pZ|~n1$fB5h^Mj|VLSCq-!71ijr;uXpCcZHgBEN@*w_s2 z-?e(4w?bqOveM!Hww7Da@3V-s`8!EQ2b`W@v2y9sCBB=VuHA%;Np7F~WO3R4{(p+Zufm$n|G8p) zcp37)t|sm${(mm-d=U6wcU8`L(qA-k*krS_DsL$KZuW!IB{sGl)VFRYHs@z+>*?K2 z$Jy9kNIh_W|1Q2~rLAMov_R9Npp|2;Z$+oGX6&u6Ht3swN62xupEH6l<@FxwW+wo1 zt^HSOfVE#zoor=w|9{`V)5a=|3diysW8>ueo_xIOE693VahI{b|M|<8yOa$}#>V*O zRfS({*Z8(}xB7P|Z(g3vQa;BvB}n~ykb6!U0eaY0jot3~(xBG~A!PEZ{jbqY{9l7T z-lQNm5xTvi+88nSp6JsUIG3`&yHP@(*AVo-_o=8T--&)f>(nXEWR1o^qt9H$P3Y%K zHF;W5lY3EBd8#FMj?fHvUam~9P8XdQKUMTB?WWzm{r!FU;Ov_72%#iE{_QZsPoI9p zko=`8j>jC&tufRs&Z1|`t%m3&@TcmI z9MWUIEoJrf!)(Xj^WQ;7*uJTI9W1xSX7=>wFWJGs)QPahpxqiQ{ne{iqBd1|rG6?E z&5(np`+2v$4zzCF(tpn_nNp6OjpwIvw{mcAzf%hasV1ZlP9y2G`*TRcfgN}IlaLTbUr~l4<(Bnb!^=l$%N#sQtJME_~E>NbB z`qy!v(?q)`4DiM__DjtVh~BoV`r%WI-+|i}p3}h88Lt1tYv?O#B9tyLSZUB2WLG+y zp^Y_ei@lVdA)V7P6+=t*w!Uc)!e3V9o9;0CtIcSHN%wV-IKeUBCtnDPtv(z?9T3Og z@-%{K&xayDDjP-|KO1ozn~hKrVF1KWf?^CHX(jAL zhp+>J`s{Q~1!pdzs{qLms}0xYSBtQM@Y9>&o;d9cDTnlO>@k}B(^++kvb%g_m%&*c z%{9mIzX>YgPtmP5EU6H~9p;G>m=#Qv zvNvy^;rQ#Xui-R%*J2IzmAW&IEKO%~z)R|2Z|?O6#1^Sc?%gphP|JUUeNgGyaQq}b zu_ovM3_lY&t#l!x&(AR9hC5MfZ#5_}e)BLwxWfLcXEwiPKzBKJ{#<>Eh?HKbG#?w= z()5D!^-h6k!DP#_zAO4mHGO7j|+V-p6^sEsW>Jn>2Y9>^n;cx>`FMe@RRoQ zqpK~etM@|a?>>G!L!PJnXyvHwOQPJ;4Obgb?7)+S<=+w2bE~m4CpePgU5)jfIlOFZbPzi_>lg?fgSzB3wKQH6tJX2zYJS%U>l~ws|l$C7hXl zP+*TPZZ91t zy)*f`Wp(-nHw)t?GMA8%TeU%PiVo!hsRp2LN|BVUMbE(Kfa#YJc7|7iXvct-B#)ndNn(GBb9HIzUKYo{i{hdKScI=o%MAb-O zOM#7&iHBj|8v0Oi#P19Uv;M$h<{3GpC1|Ne6(kmNXiV!1<=0zXY2>hIFo&Evdp0Bt zIy~j}@!kmwxHOlLYR+udYXr7H)>*SNReieYGZr%6J$^C2PdSOHXvz_phX#f z2h4Xbt%v(!&+8>zKaM*1X=a#y^NC7Nn^ET6{_g9V*o-&6n3Unj3j!Ckd=pB&gX%y> zqmPQIso5i#wODGcdZ?!eD8cx)N`{Q}pa|OErtl*{WO*;RzRQ?ez+hv0)_rs@PP=tv zqA{>@^@*7Q$%$Aw3U|Snb>H4OJ#u2PPRV()#c$MWu>}l+X=i$e4dnY&%G76M!i`fz^labH5BS&DO{{zo z)Wxb*%gKez@i~-E&Vk|O&4yk)KZsjK-u#}>81<;>d+%vn7DniW?(q$a&yOLe{`vc?MqYKe#Qg#z_p-W+U^!SlCO&6k8Gf3u!nWO+3 z*>>~`VF*Drr=`bUK4rFjd+>>7ZyrEgHSl%zR+!IhtSRog9_?k-qsE+Hq{=xJLH>)b z%ea@_x*q`fQnv7kycbe_U#y8%K`WEBefu!v`G=K{Bu#P^lj2IbTQNe&KAHZiYa${S z0RHj0w+tZpJ23_^uoL$!pbqu#0b_KdSd%BG0YD!)!ISJCN(`6|%)%d<_go-Uc|au*q?o`c-IsDi#)%#zfMyx0oIU z5kK+YM3o34b@0bUt@5JvLLNPK!e$x=HNmJOq1!2DlTo;^+_*!1)c(9Undveqc$H4V zX?41FbVOzs5N!I29KmPOA{op&R7m*f+4ZdjS`%^A02AU+52-EJx3L#)DSyZrc(^>; zoQqSHs{40uNmw{3P!DPER$ZSib9RUk9iZT0bjtj`n`)^=E$IkocxKyleauo8Z(ZYN zbxel?7)PI-;liK5r`}Xp3cIJk8((-!N7)zu)GyB3*7NfOzVf5S*3Z`YN6UuM4(d7Y z&&CoRY_hsq%}_hN$LZG4oZ;DY?wFj7uLUBSyZa8i%Y;Ix8!uq=Y}A@pU{x?Y5+jqn zML(Kd|LmPF0byC?)a=eZk}lZcjm&l0=-|SPZq+qDOxu~bPvturERYb49JyM4?<1&ayAy*dkC}T9?sS3hbRFC zN>neg^QgCX>`eBysnv}eGT zNoY|7eU9C(hV_&?x;;n@3YO(m9Y6I z{_CsCff~hIDpi}e4_w+w^3+U&8h4_uPmYr#CP(Q@c8wWYlbGM$Z-Q;)Vp$QeMDQq> z28G9MEEEgPmEx6FN^x)F%{0L1GmE2^kU|YWtMaD|A$F~Z#5lo?YR6;rpH!|P=i2#B zy4JzXic#OMBYK~^(r;!3Fp9V)=j6?cIsKVA08*OcC}`s8QP0cZ_+junb;efG%g}8@ z#T(H4f1LQXs*H0rvBw9bwm!&eFd9yIin;Yinsw}P=2F`2M*8}jP8crZ< z{IT-}^Jgc{1exU4Va2(lxxI*7-w~mQMH~n|cm0gJ-H+d~CHKSmi{^L%5Vc@WX&8Dv zq9z|ZQSIC2&7hq+m24`IdRQ5;cr z-vPy;X1rV5>@~h^ychmK4Ar4G=AUK8-`=_n{%s)Rg=#CY1%op;pn|;nl~_4^owfl{ zmr}_ittdhq*^13exD-FIDX{usD=Obct2JZ91;R3fYXY{^e0FW3$oP)=X?9u))lt3J zdmP75NyE>Ls#y@iNk6Fy%fm>6`NJ)TPnhIMOZ3KMS*iBx?D@&Vb#%RML*B^n6$;j- zQqG?@67loNCaN@|QGT%A#wM`Phw#Fl@2`lxwov(CyzzMGlrtGM8h#1j$EW;RQUH%g zIZLr07~|$ixD&3n3rvappnXi67tXtJxG*vb$=}o~J&sS`-m5MF)NlQyei@#~4+UH! zAZS5gGs(TprympjFI>mUYDHe1e#0?tqOjA+4ZY-<&|8QWu3eB)Oyn-)yVl1ku%g)W zMTxJtFQ04d(w)t4Wk(v z%0x4X(LIa=?l$^CU1rk5a?Nl@EFIDPhA(y@_D z+ghXNy{Q3b2Z}ivhJpEt)0)3$#Nh?0D;OtYbC#iPq}jpRf)o#6*{xZM!~XH8KUyjX zH3?|a2kT=|#)*`Q6T7S_cj0!W;c6=)4%a-(3MSqSI2A-w!zU)5-yXcPw_L~9UF1Z> zTOhsN67w0pVb^Zm_T9NrcBjy96cn}7wpqNtwIGYz7r5h&yWE{Ig5#ETQcgdIxEsP5 z;q~E0E!G%($lAu>*o~fV;X(=v7xnB<<|%-xyYtcaox;|K&|Jf30)d{h}zpHTG3aZ4c{-CYx#Ek3 zjAa=z0)(v73Pv+XyQn@Dqzr#}2Ggtyy2@sj4Ud%uR?~;bph?Y}=iOs&~&M z^3=L9-nK1)^F!-EKd-my82;YFa1#gL&Xt83_C-U=LYQS$)!2 z3kvm*3Az8Qd-97wX92|T^`nO81~Pf*#~I`PJi0B1T8$w}gXMw}Q1ZfJ{ky2FUL(Yu zgLE!g9`>lt^e{)+sVhTbZbYqYC;ZcSW&vD~Jz>y(Gf#sei(4E<1jdw9MKThIHNM}4 zLdM%PBKgDKyC?XveCEbNOZ(E%7cG>7v>f>!R2*O`p&*2Al?d4VWSsYF!d9<7aqh}~ zGo7;D$qp)2Rex0$rOkufM?iLi-^Yey(|Rh7tb4>Dsq1tYxfp#47+0X>tl3DXs#n5% zqR&(A%q{0ta$eZ{xOF5Ue*0zI9l`pb59M0!Jx)^>Jx61tk>9+%=a# zPcE#G8`7P|6BJ_!GX}ieLNgwe-LXvgdTZlZOT%O^^Mb*{5wEsP#E-!=vjkl<Rb#>9Y-|}gy0XQR8cT)ao|$s#n92dOGYI{ z6&DTczmT>8X-!{M7_{$wWpyo%V8mO_AYDQ0{4{ifgnwa)#>1AO>?tl6Z2 zf084I-vKRb$SujSUi)eMLybu4#{Cm5Vp+v1>IsEqm&SNI)_y|RmXAGz2=oC%NzB(u9aW>egpF4a3>k#Vob_<%U8{PF8PI)#S7^0dkS2yEjNYTD zek3RtX*x?Lz^#$MB*!IHN7xCl3w7-CcY%a-)ryBOGsUk}M|^-S?h?*bj_A(40%iOc zi^f;MT0Yux7~y6a*o>+Z%mz%Xp&ZP;FS&M~j!ux{jY9rqv|dm;n$KmcaPo5j%g_pS zj!Y_5e4i1JlHtJ*TWUhVkbf7}=}Nw$3J&${6;Q=7hQ(1KlGD>>Yq*0>uEmk!&wke6qEZ zh=XIAwK}?vDs*;sCcu}96oTxl`ea)JoSbT+d~kLvjx$gd!B% znkbRa!Tfp!t`22iq~Z&KeDWBQ9CHm^ds84~qWd&gV*9BWZbE9e?$ ztWSXMd;kb!u{D`=eZVVK7&X?En1DJBFJTE=NrQ@Sc&5p;_T2Ymhn$ck$T1;#Bv?vZ zo6{+te>(52BTmerwD!?`3}jtvH?&V}V`mZhAVj=I$wwu}DjEU*P%a@+4|*geeioP> zr7W8s6$(q;l3dWAd3#FYO5KZJW2X%*OU6!K*0Df5FxP6 z<&HJ(z7W3I=`7VrQpAbZn_hs>uZQLZ>O@z8sY4sXeWb``EBAri87*<9XO)ry^9qQL zQnncMmjn(sH)gb`?^&ms%Kjor&-23;(oB5jRozyyXFut*GzaZM_~<3OXOUH7886dI z7VhdspthfnOdejgKO6>rTnLi3MD49Bc-dbz(E?!+FEdN7rcsoJ&ZiBrH3HjTyPI5) zcU%f?+vY93Bn^(&Yzd9F^OL?nIN7|_RC4Kt!N7Y(6BF1d8cOEQA>XClzX^jmA#Wa> z)NnO9_L;gE*c`FDl%+_2-ETOQjcdog%=a-?)rR#X008j%#=(p45#(|t9q_<3K&shO zr)W}SWgIoAp?GP(=1o+W4@^Ux6J($^*C<}KQMDaU1w}%rUW@|CCLv$gQifpze7{{O zy_~f-^o2T%NGmR6i7-^U-dYEmfXxZ*JWKx6V{ER@YG-n#rH_r3cJxYtwLZ|`5?!w zc6R4`&xl$E(IGRpNfe`I<)9u)QPS=j8+>wqi%}P`icKt34;Ix=IGK)Nt~S~orpJ1W zoygugykGwvLqmH_lHl$r$LJ?e@z9my$r)RLZ5E(PH#cboG}U|*8myADP8|GT#b?$4 zUi@l6jYxDU6(-Que)>->)s#`iP5y?t)g4nGRpU(e4i`!n#3YLcQ~TSi9xN9|9RO`c zSTkC;CFRJD&u9}p)lIbvv}ck;&j3GcLVT9?wSLjA zhCtKv9Wt^RIoEW~A3Z<_RHq4st3f9>7LkdQ6)ntOAC|Yxaq2JO*>`c62bpIfxr_m}0k z&p;J%a?dwkkCtUdwiUS026j0z@CT%34{l4I&nHeJ%{4ZLMEuxw3Zv%be6BiX603#x zSC2(kmTLy@<9?l$xGf<2*~?DLh7}SvP-u@IT$9j>2rYIrlA*ML)MZR80v39`4%GL+ z@Kg5QH~mUxZjMwwur3f>|GIQF?{dN9Nr^eCjzSS4rkPP~yp%GC2o5BN+wWGr5wk&; zr-;2bign;K0vljao3pH=KGLMwN4;t(oYpcQQsF2N-JMaIwznBwU+kIuYy(|j6tr(G z&}WNbuFW>~Wr&}CZpIbXA%zwLWXL4CyaxzZ)KW^8#-cY8mugK)GD=ObQfht(M^^kg%w2=z8Qx0Ll2T%uvo z+pTVrYQXExW)S-38;(nG6l^A3c2{<>f`RZ^K!nUC_9(j@C%pGNh}#BU5Q>^ec9VL3 z=>ne!fjRCrxFt!{#p{DRrWlS193U*|*l1Yjw#n1Mb-qw3g{9|l_|NyS{Dx5;_zv0| z55dgk;V;GHe=T^&YMzY=i_EDW_*DXJ1{IKPvH-ISgwC)#v`>$`UCWS1X1w9Y-Ra9p zzvcU3P;RjDjOAyo`iQ);z^t_}K$;kd_I;SGG3SF%y)I=z^34e-t z(01BOD?vj7V%yt0!tBd$7Utl_gx;?%?lhAEwkRvs5ml^j+9-0KlUm4^$;abM+D6$> zGYgH_RP@JgTZjWVssd9(0azO@u6{fz|KW4qc|{e70n?u z1-crT1jW=$R!jtr3b=ckVD?0XMQ-Uf$9S1{RS$M63*(SJ;bHXX*Z%meSy;L-BHd`| zZgfNrWOZl>cb)R7t;{(>93x`dF|oYG#X>BTmX{njdDuy4NBsS&4y> zwRvgy$3h_Q>J}g3Hu1LjL-&C|v`i{6$A|EtvZVWI^F;)tAbAiJCfDp|Z4tL{Dqf!L9|DK93F&^0=XsgeG28Z)cSC@Ri|;;~^Ctq0 zPS_!v>Y%MxZxGHL=bBLX<|oy%Zi}`BOUopG^GxrjLMYPH)JnBzaKD! zAb9ZE&$g&aiDu6!R1dHxRVpg%X>&s@=zMAhslhV4Jbd9|x+-pFf}O1cwXS+5J7{

1r`%0ld%&&s1v9NgeovpjtgdCIMOC}0;5&pXB`;F^x9$CPE}uWVx1eq1V|r5@|nd7-F~1kFCy`!ztHR4W4O3_|_6j z-ik+QXIJkd{Aj*~1_~wonax|k`qCS>ssO_=Xl49Fy@Y^_nGhH<@|ITTZKV+0>!@D8 zv8QXlYH}P8SfO{~DDBHob5z{%lP7H@`Qg8G!%OV#W4x~3#qp^l1UKO9G=ay1p`IUg zFgAyaI2c9d4)|JAEr_c*49=Y(Z8P&}UiL9F2l&zwhYh02eX~-w_ndAp|9JHLm2;0X z+6H~@x2eTRhEZ}3=&V#|#9(k}-t><*6b`PtXHO^+(nP!bC)@~3Toa_K!CnXx8$Sav zFi!LnaAO`u%|$1o(<+n+7JQdI-7+=d!k?u3AqQT*U%>sv7-jwP+uV1jd`7P+1cw)g zXd(QDS^4N~N}|xl;wY)C6DzCXHEM)Cv(ewDS@%V}B-puhbbHvzrwIzjIM_S1atH53 zieHB=px}ZPRTm{MlQ>8}*5>8kUH$b^sv-osT&hx@;N?N^R!1JAWV}E>Twc23RfLIn z8fk``Uus-=8O|;mHrZc2xS5=W^Y~@yQU>q-wbLxStedUF2dtw7#3%{W&uyG0t7_ir zpQyyUbFBW{YO<}q7Q$pdp;}P?OMYcTi4zV;Wbnkc#CnsKMu9tlR~eQosf2IaJK zk~@D&U|ezYz4b!Sa( za>{*GgYZ4i5OyQwd;?2nO0tO4GhXtbe_Z>lr|#N1Yh992q)L01zbrF-e&&|WuN)Z?oXVZrDpHU zf-3>{*tvMzfyRx2qk=q#%j)X`;h+Xb*wmMZvbl58fnMjDT3* z)8k8bLX9_Q9Vv|pbsD>TU2${u)-|s0ta=u#Y;u3zUPTyY-jpCkZ6#}L{HQqZe{CA2 zORi9Q4of77GbKQYyS9f}o6*^V9s0!y^sgj=cHN@O~HzND{XafFZMA5v3Br+-EMm>13Y zYD_Ij;yk0@1bnCbUca0bEtW(8hNbv%vrRXLM0Igw+E~j_9Cp1sHBrBhnm@}zrQl0= z>|0#l@FXfL4Z84&wwtZw`sy}$E`31?C6oQrXojDU!9v8y>iaEfCEpN1x@q zieUB*k5mt7$oU!uEJ&NV()B3|>dGTQO2@zZ$H-BLB}gpL3;f*Z{d3BZlDk2ia7j+^ zvbiOK@zekCqGK$-Q=ql~La$un81C4+4A42-`cm7BjgUnDRXj9wn2DdS=!?Q+1GmeOLUyK_v~-CtZ< z2kzXjM+&~GD@)6{aFgpIr%9=)k9u3eF1(dBuH5IFp8xu&b2z)nvcQ16S5hZqiVGwp z=l~i0;J{Gyg6lQ6aaK&UYB`EuAyd`;f?q%)A5n0dnf`)4?IU&wbMY!vS z!mTj?H;Wk+FDvJOIb+Xr2egkO>+_3=7InUN>{&p*55auOgc8*taitlp<`tZKv5J5; zKBM%qKYv+KuZa@nuy8NtU@X!bUs!t~-{U}+5hThoqHD^L+)P z=tO3hAHI&Pf&m_dP~s^6(oqA_YIW!c!*;Xp9Wj5Fixfa9oy7WoOsQbvUPeC0Th+4p<6JXSDNAX(zB~jrUi!h3OY7E+f%klo zm5!R~Ho zXNY4B_HqM^i(r5;;MUUkaMj;bRzBYgoab;xUq`jmB4mK{>LI>mbN}4#@!@4& znyGd%4*Jfr6x*k)J$Ku_Iskem^3b|dKT!)@=-7B4n^G8mFzJ_*>qE%I9v0O0m$NU! zm$tV~=iu@3A#`xe6Zz&)vL`R76rRxTzv2P4@!vnx;zwzTlms7GShNqFyrL!;G+u)Yj`ZCk;$(cr=@OCXH9goedg7bwKD}>;` zke|_pYQk`$*lIqTJ_Ui;mN()VTEgE@esfLi)U6atnmeP(9%<&!|MdHCWX#;82|aMb zJ>GROHL=4(C_aP764RLx%t4Vat2&@*@^|R743>^+kEYUhW|b$BfkwGIVa&9+AQrA6 z!013N1h$j~K~Pyy>vHq$%|H?3=%auqfbH2^c1Y!`b0y283n66E;o%We!aAQ-hQw&4i&-ChhWbnFWFqX^*C5mxje(`96_r!+w>mG_Sx z)IE(Fd=2UbJ9xtCu4+4nNDi-lQx7K`L`_XyZG>urvNeL=uIzYZ+UX*nfQGBeR%&=$ z0DtV7f~3KNv(GPZn%X5i07Na{1|_&4ha1av-^$u-d0Q8|ZCnq5#UJd6Hy0>`Zr2Jy z7J5Q8Q~nlVic8u+Q<03bSsP2FK<=EaC05p`+#H%r48Dp~Etp}_Cp}6vN;r9(B3SS!hpEbt-dB2{GJ_5!k{Phvd)jzp1LvR;8J%&*kbj<;DnDcO?8s#S-I3! zOzy8|Muz|b;VSJ=lM9HF!S*P%!%r0f7A$6@)g0o8$t3^5M+uKgyZV7R zMin7+|2QKc5f~x}{W(&*1yC3l?x>cg+aUudz~I{v8-&S*HL+V;3JqxYN>Ii`bmJ4Bls)} zJw?HYnS$_HCbgKcoZE9|%`5WuF)n<I50`eR#;0*@g7dSF4F0wdSjfHl3KbON@TzGxr=>%E z#t?miF_mgcKc7J}Ap3u~`S1qI{rxA|r5iF-LyqLq0*>ws;aU!E*W9qZg@)Jjggf!A z;WJd(Czd8|)cY)KUk`DSVm54&D`b1Tj5ZUkO3>tf^71pRKCK?<4#edcW0` zR82KcT$qy_$#R;umrb>8CBPR)8uJ|^VeahYiO3fbyFF&}Dka#@@n$C{6l~i~^gDb% zZjpm|;mW~7mWNqhNK!xfJC_#IJ=3jwZU3n2@pe!8IZEf$7;YAtESqhmSjCc{+hNml z4??6Box|3a&QMCW7PaXe_+T?T=I907Co##&^iHU~qEdNbnQ^XH{yfGXof{)*-vJj; zDc9L(5Bc7h%}UI2aX`BbZhDq->vQT>+2!^#tT6^ddWKpe0M=IaXu#@|h!fNK6OAj} z?;pn_;u3pA7CPrv;w(Jqis85F(D#?k?GZ*d(z*+k28?Ik0h0!47I;g1NDB{!h9j9h zO-Fl;3sMa!kcd^(OBoDqDJ)$46bFaNaED9~NvxZd<5;Gj-qam=2WGVO$U5pQH zqB=f!Le~TcovC*K*Zqb>%ujRLD}$t(fxH)&dg_5mBcm}z{p6Pk7(giGp7P&4hL3Zz zzg=)M>Ptq`Umk=LA}3jp+*@2Wn@w|L!lF=? z;E1CW7#2-*6h`gQtB1v2Z+h98eA&x0x=}*-J7z*v8jPXGnR5l~RDs95jkl@d$a(z3 zk>WH+Lc*LUPoJDbLaSdeWn!f2YSIkoCK9$$a2W4%@`Ond$YG#D`vYM8=Y~P=0%bvy z?aqkhr{@oN=&c^5FEfDdw<$U+W2v_lybj_sPDp*%fY$eur@wj>-eUUD#{ZV(s;giO zpScUznfi+js(7SuF>xx zyfTSr!@gG?!NkH^<`jXs4}0?8E)4pX?CDcklCGE&`urcnIkmv;ysapcwLz#6<9-=! z0Gdi{(8l{bURobcNoMuQ2Rtf1TycQonp94YFc`S7evH1IDF+hI8yo)Z!)68yW^7v- z$N)cFzKCM1YnNq&Urv{_P7*&!lFb%lrRcjD%y7WKojukx1bEBkru8SiR@63VEb5~S zR_NU`^DdbgSl;gW=eI=HoC%v6i|+U|EgjqwZwmHJjKe3gf^*5boWC%jIwj%clj;)5 z4;b>!}Q3&|~sWgd-qSMW4OF8qLa8eKJydTyck`~eac*;U8t;cU|a?xh) zJL9Bic%`T6U6{jCI-bUWY1K%M#ueVOa|#12$)Qs}k==6feIp+ZBWIJ@FWe0EO_y}A zXJyE`UFM3@G8{Jx{F1b@r^d?!)7R&^8uC!46yED4zPjr_L66FKwkIx_y3DAZ4tP$= z?fbp$8`M6NDNxYCPq+)Y8W^)k8-tPAk0Q%lXp1yR*5k|)JsP9#!o#O?f*dB%`wUs< zaV5o}rVBQ6f+eHNp(ImHM^wn-pkU{_bS}kzGVC;}Aqh?bhm{v^dKL;!+{5@o7N9m0 zhsI_RX7&lkhU-J?@AC~^mVEg0xC1U#MMw{^8SzB@ObPI0*Uzr{P_7k8Ar#}SB^Y7@ zujJ#;+tlp#^m07>)fSs`wMjmQnhMk8OZ9bG`Y`o6yg$t^9LbGS^Aj!_C?aYrLf-=W z?ZfXc+qS(MxC{g|1*cj2j1Wq6Lsz7e?R8>b6H27IbNe(e{WJItmxK{aO4MSs@8MaL zgP!MuVoV2a;H^~_*<*H^!JaLs$fxYbLe=+rzBV|`eN%XgJS-=UVNowX&QZFZ30vKr>ht@?(2#3M>BVmwQQ?m%7QrqIM*|2`ugSjbIF%msNJ*^YLj4 z0XlR>)6d?T=^{Yh0>MVmpZ>o3Q?}JbFK-dn(vI=J0wxe1VBT)c?ltmS>V~gHZT0lk z_~VotFY|ca?4k0G>EY`P^k1f)SJOqDR8%TSP9R+Q+g~0+%ASFJ;DK3LmV4jAB~$6F zBw@DjkHJe7bDK-GArhfdZ?w&PqthUngOOTM+xK^8S;)tw8tc-~bgH%h*1$o=``pMj z!#9xu_f4Gje1Q+irY+(C!;H0;jFV&e++QA+&niOV)#bgNOWhc1ep&=l`L;!P=w`(_ z3mR$foO^rMtlYlw-ut85-~zt2b%g+$V`vBL0`pdRU4lSjQdX92piAY<`10-2g!H*T zN@e1gJ9gIO>4VRY53QJ0FS;#?^`DfH;S`XKr4E?}hY;1=D|HELFkGN2+&@|5fb(qF z3gll`+#=e?P}vA{!qFX=7DRJA_|>Q4qtdnZ&jt6-Cj~67wU}Qmx2?{bz{5Wph*8Jb z^fRxP2YIlTqmH39_VI4!4EXiixA?k{LUo@{aR+;to%W9&y#_2MjER{K1sXNqZPy%D z_oF?T8f&Di@*@G8Il*8duE!(wp5)41QTVkHEVK`pvBBZ=G3G;oO{HSPx)-?sxAZl4 zZcM8ct)H!`C640B#I=S|w`V%h1X|1P`-$+;eD{BDngmDv zn)RZ*2UbCYK~-Xr;lZ^Wv(2?V7HUu#k~j@50YatrAB+4Aaf%C|f1BldsRltfan$+0 z+%|PaKdI$esE*i`muImN=9>yH0-t}te0@{Ey{VGUYX(?L@@gq^%M2!~ro#!?YXMRfemshHg`)hb~A#S;rpRqXA#7-)?d7 z`9h`emz9p_2iLBdX8Tl~yw-CNk#9d(O$1srXOP>|%1cJ=31s)@4S}nsUdma)XI+kv z&D9ZeJit3+NC)=RMxVLmty0ZR`GibYAinF@NQ z<&Mow8bPExpxGQp7|3ye2-QB!oPXW)^eN zpQhiDoNbp8KvF%_Ik-+>s~OM;NnoA3*Cn}`V$Km;(NcD(F4z10>$?<3^{jQffcR4v ziWsC2fZ|3Pg`TI*%>*NUjklm1a_01Dc(^l7(osN9O{&;^7O*n!pO!b9A*<11o{&GPUr%T#Y<-nKUaBwTq`RrA`xY)8FevKFI-;bVDX-mc>dbdv?SHLzO@88sAN+3 zxO2l*dQSVZh&*|>{cj$Irl}-0>UrUs-b_rE0O))j7a{(WNd%GD=51%6wy*tZTkqMV zPG?p=v5>OoXOZ2Gcd)>G{iP2!Oh=e?Zp$u-4W<$&*z9V$FLGxtPv81bo;`y*Do@Q~ zv})La-PHZGIX4lW1Y&U)Vu$1a%d(nVt(q~HwDKM{@`Qf|d{w%gcWKRGJ1(_srnc)1 zcMT}$FSZV~20eXel>Pi7J}^njxp8EVnMOiw?3f#lhcUp%5YhydieJAuWX@ptu{+ms zPI{TJM{+LHabU)==FOa(ITqilI6lGV0RNd@Y}O}ZQLC$j<1h@5*{}6UTsq;)=Q^sJ zn6KvAd+Wrr5EkJO7fu;ks+9+Ki-#Z|rjS1#{g+yqRPfjD*7@Yt1yZ}P)_X!GA}oUF zpx-DJDeoP0Z*Sx08J3~VgZgW7nj$BJRUt8dZGuTojtXwZ5C3p8IgV1;rrrc;e3@=7 z=v`@JWBVpERCZ%ZQ6z27rNBi$d6AUlE>aRE5PG|nSv*&M5cKHWO7Gk2;uea}^ z3k!u9B78T^h0!by=0x#VX{X9RK>x$VNEXkkx7$gPjV+O)9i=|C;M7?6Wt$X)pRHM;WE_R%83cY-eh& z32SgkCU0W6mEG>g_V@Q2vuJB~w&%E6C3Bo3dHDa~t!5){u(~?w=>9%7_3N#hcP%89 zx8H`H{cDvzRnY15A5C_VN8GGXp>O()r|2;@yG!)HmSW}Ef3Mn@8^hgtqR16ZG2;B+ z{9Hfb{6V({nbW8K`m19_=LnmyrGS=zSmax24_Vi{7xYwmR&>}{m8|t-_4)0)cgU-U z=fzcin&B@Yg=L5SNn%q`3vO?J*HVl8-2JY@N7Lgz)Mc=y!>vAJ`o8N2jk>Bd{=_pG zPd|>%weH1~=zB((t)p~IsffA8=LEc<9HU?Iq9R~75vAs6Jgb*jJw-%%oD-K_J7UR*<*^WAJC(l%wUOJ z%a^&DWeO*LLCiy++?NS+afmR_-0A=`L{@bpo;gOZ+zk@dE&Z7MXGZPvqugbGPf-mw z#Xm1cm_Z=C?^>Dh;#k(KMx^mKXCOXMeH>zqd><`GuT;j+S6JMI-$l6Y--Q*{F1hZg z@!gAd%9PP{_~#be;Hx|8tdH_mA!n9r1SjjX{?xyKU;9@yVE3Qln_uBj@c+Ah#^|o& zNNe*pCQCqvxM%w3sc)D5oq*Mw$IFniw<$*PC+71z$RqSWs?ApQuV}m&U9I~dL!ORg zw3>AY{ORHgxX7Q^-Fb!&kH(>f7N0!taq)}%@2yj(|0SKfsyTE1?qt1|{p;q?NB^;A>LOjr~axD7w-5^)Ioaxa@4b535Os3_tqbo7nb~2 z8D5;xO9C9ZyM6rl_@8!?tdF~I{~mxaOiev5w?)MTk;m+phpGxbl79~@IYVdhh-qR3xaPm%`)uspVGQvE&Ndf{ z8$NoJMsqtenVGM-Wo`)I z;YoTlp~s~Rk5SFWXKN|MmY78T)HdvB7P>HOAL144BccErxc0_v_1LizgI4f7(jRSt zeXYc5{f@<7rs2-4W#k2D)S}WHe(#OHi=9lr;H6-qMy8D2qk#F2=}kyo*QLrRAK6d` z>_eyuU^;m660uG=-yy~M;iB6-p^sPb``In1L=oBwqRD9 z4Y5=3(3546#P#G(0MJP7S^wJnFRgm7aJ`+h3u zC@)7bs3@l8B~@wi|FHL-VNGpq*C-;WsEC5VMx{xUE+8Ns3(`SR=@6ABT{s>$g7RV}Dv)%W&#~73R zjlg4=hIo2RFEh%xtyOs(_tXqEv(*KiNvFTs`jm%4zx-ipp5bVY%NH*pnJ2|}W~%b>-%X9Bsc;_1*Yw}?h$mh2F97;@trv>^ zHMdWib+xKCa+}9_D#$9?{y$EEgA>-Dd3*LIb+kapAuD8cKzb&Q8HbQRDtY$zBAvIdqOs_doWC>e@m>U+n zmGcU&$mDXqqg~09WJxNj5A<;_hvuV8lx&dK zcfrNQMF-EWz{Cl9*aPWYm987Bt%S!&x)z+{?&UT)k4-GDT<>{--leKX6++NFpQr$> zzLiWu^>ukIe5L22&%YOQa^!u|HO!c}7PYjQRs(FKk$q@J)hVe6S+)E=PzAD>!XZ~d_3{o za01RHULOltH8}h#Q#rw^C~)P<6}@5&mFfQSBrvH)PI;Nu_?uel<7Ao zs^LqWRuacE=R*?~&?nMf_PH@(7<6ldY`!NKBDR0tvasA7PZHeb^ROAJW7 z3~!!x1Io`E!-z@AX;7NWT-QlbjMgMl2>c6}4U9vro#r%w>m*II03$7vB)JN zCp{y4$SE1*_DyTFfjr0^Ur!VreD1)s+z4k}&7-W=O$6HPogc7HDR|g)l;4n}rj_Ev#Yd`{=Iz%v*{@=*JN zaUp@YMUh=IDH4tZ!;_LoXxbh2uNT4OxlXhKiuHjKH{_N&B*RHlijKV<*H3^V#f}7c&OkDduQY0$r^YHfglUBK~-drLW zx}AZ+LH@$x8dsk@O?(o(@MG!cY9VA}qiXnZye+~6y)i<|@OGudN@cD1(UL1+u-c^Y zQl?|?%F^xKMBCt{0b{9Rgfo~1D?%bjC{kUpkGl0bnnNf zBY{w?wzbySb(aeYsspr2?c0HnZ25TKTDwy@aDX9gleADn@jnGM8%u=Ktc#oZPxvFZ z$DEfZTj3coi

{=z?@jZ}(OYU17m@W1q8c(9Sf+NlaRjP!U$`=!4d8*Gv+mBBe?t zb&FguVO)*lL8VHeXI-R0#NbxoIe&w&K6Oj9rOm^xPZzOL1{X2ONb8z>_KZriji>6-6_lyIvhBf>j4U2z@g*|? ze7Z>`PO-W*8#+)*+pxRwoMo-H19-exjz&)R>`HR%E*|t!Zy%78(?(1*-?bO;F>sxD zgWLw=913x5ukHonZ#WQOS!Rha;^_aNDhd|4Encl;C(7QSncgtO;Z?CKuJdtukFU>5 zq8;ihES0###WjX~@NRvd;gtaQ*EdHjsYi$mpN(Ps{CPzmIUghzF6SizUn=LMXERp= zwJ|=WkdzB826<7FFRDdg{!$ILAEE?|OdEhU=ymle!BeOBta=iu#n)!ptpM~VP6&ae zdFP++u{}shG7BDZI<*C>JJsS2Qdp+H6N;rH1qg#p5P1fyxX*2<6G*i~+ki6vwTB`t z!{2#0?J=BNnr+M88OdF_&K8Pn+sqz|S}9LQ{ei-5~= z&AMU~9oDaI(|lpDe8T&iI^dmFoff!LYbzr{>Mc~%b!wN^IY414dZ`bjFUlDl7_d}B zDG!+5rNlxvEmc1P7ukKvia?SW%yg5{321doBEG0Hb!M zCyNL0voBV+t+Q`PtDME}AT|>wC1<7jkhNIFNCec|XyAUL z9&@6!scwDPR=a-Qv1Mk8u&!F+>$ki@bltd4^6p>sgRm8s8ZB{)50ww#3YWPrtE710 zT4iET(pBkst$>b_7K87E^j5${<5D}VQ+Z!o`eBM%mc0i2sKV#kbap^5r6&P|$< zS&C((b9EV=4WD{w9gB#qU97{*5`w~Oi6*CJ^`OJ@glow5v6kAmc=a(yml2r=4K zRkK2hp#LlOWJxTjHNYP(7V2d^fOmc55)!%-%&NqhAmuC(qV6e60>8MH-6*G?{Z;J7t-fA3zOdx1mE5 z?J)*tq|1yJ&UUM!Nsl7$y(Itc3GNRtjkr@`weGyU4A zmkaDrdC>gowX2q30Y`$_vRNiSACHJLKeJh(CBrl=s!OtN*krhRy9DXw(Jv#Bd+^`F#vDy+l|}UVmNbvB8^pn92~swXG}kQ zSz(!m+=~T8M7~?~uLlHB@EcXVnj5Uv{hlb7GDSKw;y)IZY4z24#v+FNL>m%hj?X>zf+$v3tRxz2S=Z5Ibdi|2nU(P^|krkUM*CRqMbfU;Q3KT zHTQ$FVw*h}=80)ey0}`>>THKzt{qR7;rnxP)D$CP`yImkiLO_sh)*>==+x zCj@M4d#$cEgTCBxX93V5w3izA%|4~U%d#K6f6|F|6!fH|;4mXYH_9&_2)Q*x+h{?p z=13$q0S2Yz<{UNdw+0CB)8G*%;oBEzABA+@+xoa}HS<=6p3h~u)@-}V)d(!qzNg5; z&pX|)L45sFeyg7?1P)k6Qy`JV`*VL8m0N#s7_2fouag8a4UxFWJ}b?rP+Z#?wXPSd zbDhP9sWvo0WYJo8EB&aIRkvU}7$$xq$Cyx&^5w4T*yN-(lC0m&o1sH13suSSR$!6C z=#X)*+S$D*C2w2ITJ=^bngPTTT}Q{;`H}`o#X}-5fMk!fJpb}F8kXS zkq&8r5Bd!a?Olf?fgtA}CjY+!wrVASW03VwiX2+h)p&39bxJte?$g)-54XK_RF8wkwYf`$+CEULj z3SudBqN^2s5^_Bmm>9)iTYwZPG9R`Z8>|U}=K{I^E7-P^6@SdAp^PS?m5{XU!!0vJ z5Dki?uMcdB2vG{Xq6re+g6vaNnYKd8q5IIe(K&^Fzu1hoDS*u}e&krh!jLSt?9h0Y zgk@8~-Q}$fl=e}8!33;y7aUQjoetC0{e$2+t~^}7k8DfkoBVDT3S>?i3#R3!I&L7# zflsV@r2#LcPF#tJgf9!|YvEVpJ@6WZ3;#hG1;HC~AU$is5rfe~3`w~8T|c21QCL{% z4RqvV=s83Eo>TzTcYYs>;mAEwwDploHKb)UU9hce7ATTkW!g$W+MIxIY;kLs=jD0B z+_q>1DiCVN0V;es_KnDcpD%yQ@zCFzfKi z(IQcTz;=zjO73>_De%$Wr{sHwy!v>r6bB}F=rHAUPNnPQ*c7eNRu^|9fnp~Hm|s{e zqp%vFd%-_`obAo+x_;xv>y(s~XU}f30Qg`05O|W~I$L@lUZ!&P(v<97iK7jCm3Qea z1}!x&gVpLxW;C8JTbbz=Iz+)2z*-}MRq`i2J1y!J0rDM87S}+R2OOGjBtN$Jcw=|| zawFVl`JDR1B!xVNDJU+3PW!Y$H`XB;KiA9EVf*9>uzAR0V`Ri;n_=mWkDLUbn-2LO z;U-hgH!~I@=2hGh*59K(|F}6loNJNV<+u681owLi(2(`=><;ql=ez<0{mV%)OrRTZ zR0=~;5#?Z)q~rr(JS;ZoU~leTiM=w zhX8Oa+KN@bZ4(=b-^NeHeIjm?NUfyzoSbYhDew^!D(Lees(*J*>Z~!pR#;a+e$c%JgYSH@Kz0j;R{WKI{QYhvvD zFVrRN^nT=&*%K8$7Dnaa#{&RxGLQ@|RrXYDZ=w9Zfw(_Dc;6=nwMFj^`PLK^dQbPN z4&5(TX>_^_1OA4Pr1qio6^cw%MTNe-1`GX3uIB{wo^Px0N;g{n8w^UjdoNxI`i$D) zVYY7J?g^EpW5;#YB&hDmyM{oe^oue_?|wMFD_0<>Fas08jts^S zN${Okd6ww36(4SOO`heEa*8QxL+Tkv(S=`SHP<^^J>X*UNRd$Ufw&?xx08iSUr|PRqb#oHD|7HmuUzi@$rGkw(y@cUt0sNQ%_swVcmNdRKcbx=TjG zlT;cvndpTIB*_~eN$mPrS*ITo%2p+$gpveBkrJw# ziL~LQQ-X6o15TP~e#j`=EE5!7@bA>rURh89&@=j=fSNzcg0118(>0xJ#mWchNdmLApu5MZ*bw{X)euE)?dyltJU z2eU4Sf(OWGPQhCUjW4zj(|f0k=05l>`LiYQ3~{_A#S$t&+?HGSp||M~&)z;v*K%9* zZRAV*tSAUw3eae)EuBK0St%#U11rF_nc*$UsJh_z6r%D@wjOxb8Evs#g_FY3C`Et4 zx+n8Sk{{2y?ZhU=ulW?ro?>R@Z>*_=H6J(O_TSmzy=gwTkncJ^#x}})TA-eNLx)5< z*lJ^1?tbMEExayAb-o&=p4SpH5!vR2a}@s}^22Fsdm*Q>e)aam*Jc>^GB$|*&d0X^ zLTaxM6)IOc*S8QNr6Pi_>FuI?(LBMK=8B6)cvfP>RHSciIEjndFsFA z^OsjsDFFlj-pa?{P}aa(>og0wtVXmSkAe9o2%ti|%|RqPTz4LNWfY+uqlf*$+rv=$ z|D~k9fb@+8HmzC(Z{C{YoRK?bN10kh=Z&#;>zliZBn|6!lk=JFi4q5pw3u(-8k#e5 zcQK!P8!+*!2}z3Cz4NT&H%t@X?rygqH(JX%W9j9kkhfqgFo{E2Z!76mFN9~hAZhTc zFP@#wb@221^~!5}{58Wgnyq&&=i(Jl18cUHS+}jR@|jbh0J2_$^5~=L`z^ydfW@7* zGyvL{Xb?uIySo>ilZp4p3nV}N#!)hsweiYP&a(vvp2qLs*5%b?l-Y+*z*|pcrZ-oZ znlQ-#aAp~oMvJJNyf&G zqq(YM`Z_=b?;8NgjW5|xtqc=%6~Y5j+=u+IpcdQR|N1U zo)k5uM-xuD zRir~w4KV3l9{VNMbFbvkb7o-^VLqKiaocvu4Orz3#&YMIFn*JYXI-15LPU!1>NZth zA^sh)=FxjfsCfE1uSO0VX5KQ_{4=}PvOb;`WQZzVN6x2T+CV9AhOjFA?k>A>7a%K8 zqC;R3){RlF6s79Bdvj;B^WG25&f4Fr7^HOh@uKW(Ip!uJ4_0j_dM$`!*n3t1{&D1m zmQhGlrmF@|!tMZL<%LNv%j)g0p0T5go|!r+VU5W?w_Kk+ZyMl&k)sw4?Kg}^!(P)} zvJ%XA8j;x^H@SF~*oOvb!$D{{=eUbLD}AfSd5{+rPQoQ@PL#V&d?HEOynpuWP{Q!9 z8f{_=6}C9D^3ac%Tw^}?@_6_cy5mLu4NyI}{P%~6Z&_V9WKvIOq2DrK+Tb4}tpFMT zN?kQH0ocp%{*{xWp-QOYb>AIdcy3DRy{4ma7i}wuazdA`KY9AJ&BpOpm6_)Pn=t#X z^Rv^!dcS117pf0iZeZ5Aff25K0IcyMa<|JleL-8&G_MrAQ?{1G*m=2ikO95DlXW`f zjeNRJih_+EhFMVWmk^4J7HyDwXP9a!L^Uz;fl6KdZC@Tt9$?-zhm84`u34_{;#Xsj zslPqEFv-WsV=q_jt4k0c^%|^R{oyc`d_)I{$a^(tRrAWo33<};C8{fj8JNy+Lk51% z4QSE3*>_s!R3Nqi1&0ZlfuX0sw&8Y-Ez(-sN$l~W$Jx?mOb2_*Tccy#I-}!DFAv3; zPxn{)HGWrEZU)H1!dEL-uifOH(=)OnWrz9@*x@Ta?vBpd1O9+`=DR$hYdyDUCD;>x zXDS7~RpBeYb~xfIshBw>5|o3p|}2U=fS30!lUl6BJ< z9nZ5JBn24k!aeN&ry#>^2^q8eVHER(d(TbFj+f%#dD(qfCU;T zIrh*t6E93AJDO*x(i3ol_Fm#ZfqMxE5FcsQB%5hbn1V@Qr>!_mXMV<2LDpEd)R_m6 z0l-v#x7v=!w*O=Db=5}9(yQCJVxT^2{#2aMiZGKMv~J+*75nO8N_aQIF^!d0)F#H$ zn&?$&<;F`#Cz}l<1g!qF55p1i%mO3xSd9~gXG}`*1v=T)WnS}rqR)>pg-*VZw|p`u zVDSB%_;?tCUb}DMyO+05T!bdq+%FSF8Ei-jcXp4XIPH1UpjoFdB^ars!;0`sh0m7@ zlh<-TO!`7k?&=W{J@upOf7xy$fsMZF{8f_To%)5X8>FkQ6~#iJUzHxInd+$;E>b#D zP}iFz=Ve~8Ual(T)Gq*$y9fKEm+tRzbd)zgsDkKBEh@oI-@`|^>e2h%Va=&>hg;lL zp^|1*YAnDXGj#e^ZuJlzzCyfS!{@&l*>P|Cn{A=x&L|^q=)#J>*42@Pk@sAqMFwwu63&!0;OEK#-`mh(yjxtz3WP^C(KK6t|wS{UsLk z^#0gSs#wHCN)lpxd$(EyEXm*h?il2PjvkL{@WsqtN=j0Mwl0} zfU)Mv*(&YI-t|vdjv3S|Es@{tuIqs_Fc7ddywYYG#H*oZPb<6zs&G1ASA88{s(o*6 z%h`3_^myBiuFw$4Gv;A-p5}tjtY1P+);NAPHI=vpVv%mp8N8Cuc2p;( zsbGt%n1!Cg*-L)(nLFMMH8Wj!n^#oy71y24d)mvk<)!&=vQDjmH1Mj4{fup!z}|D_P`?oJ9(q)- zY%5!2vb@Ol#dWr~&RVB#mQ|geB1W%PcWsqwL@ zpP%%@UqxhEu2iO6*6u1bkmiSpOwd}9Uzu`$2UNPSeA67&oWM4CR?0c%;4WLVtCdE1 zq7K5{#K+qt`9ta`;==%>zk%m@xj{9I)FFgP3H}oQ)asUyK%}6>lYaXi$8ccO|0~Eb z&mvvayhNZ4`+M3-N8|OVN-N=YMBdpG594?8u{-CHZAK5#_;edK?OUN{NhFlm)z=S} zuwnY5-I%@CN#r};$rf2tc6uS~RWg>!uG?=P{Odti||F-qq(%T6=v zvy^uI?b-7@>@3bZbRv4Uz9NFc85ojXj6;-_60VGPeP#ygkXO1xnan$b~qZBnl>kAkN@%ap?s%RYLM@BFo_q8gKg;?6yX)^%w~-%1_{9}*zVsVEHE{6t;e(y;wcRq; zm59rHafSD??p!c1W~8#(ctcqUuY52|OUd(dmMj5(Ce*8`ZAZ8MsJ4s29C( z=e$V$uUysfuOZ{zNZRX8*UfZReXcKw>u1S@rA>&H0;6#hjp>T3nq+odtCDNEnUPB$% zbdu${0Fx{t1(}3%jaN=^`Es_W#t4wCGHGgOuQ3XR!u-$4HN4Fl2H`0Yu|rSvcJTGF z599@YA7K`aP01QbM`lk>lNqeiJ`)90buD|aWXhIOm?J!wBXbX^ySeSpDY zocjiFt(sr*9#lAYf(yWa6%HPgGtvN4}@QL+ZHK zs^iGutHZVg;ILIdLq>ZRuR`UG0Fz6<&X||IGou(Ri77)fpBo<(l(qSO4p$BK*QhL- zse7S#-d@@vOr=F1V1gkUTJCR83&g^jBLz6YvK$TwamJaVAbPHnb$CS+06L_Pdy!f5 z`WLzrkzgn{zxjDqtA7<(q-qr2SWUMWU|OCTU{`n0${7-$uEN)>7HB2P-j;IQRt2zj zx^JHzL<%buf@qGPFK(U2Ecn07jFAzGKeSWAq&*%|gKHk%K0IHN(CL_R&S1-Nb+gTN z-Z;(nWN4T#{~qr3SK)_`CfGfGGey}bbLK(M^V+<+DbC3V56AhnDu1j}f@6vQU^U)) z4dl5aUBPThoFR(aZ>EEG7puK|#wAqIgxAwNyW7kbS{1uR(Q%#&{eouFO6MQ%yw+8H z)V~Etx;=)YRzL0gqJZn@4}BWf4h#*y+wzJG;hgyerG(=Y1sAGS zU%d{B@*f(oT1(yF&Wm!u=De|pnTNJkvaXq*{i)FZArBTaH5GZc*0vs1gQS-%*-clZ zJ#s|N6YPN$kn{0+{`t<5yAlz+JHh5=oryEsASB=&0CjLgh72Y-34(*Y=INAPM_hKX zu-@y>9NE_iDc}Qib;Kfbr_Yq=UHBJ?G5B-pJ2Gvc+NgrBWI% z(*dlFkw3UKBr@PR7fuW`P;1Yd?^-c(V-|`4ZLC%r?8wb^GEqm_;_-_8%A#J7#d3qx zHM zpDqvcVKKvghbO)u7Swt@z$GwhvO8Nb?A6m@?adfhdNMD#OS({b>}t#etu9*?`92hw z=qlMCJBxnXZCqmQRix;-{z4c?Z!cVV{5^ZQgIOt-1M!$dw454SZ_$>_*08)2sg)pR z05K#y!tmqnVZ&rRr$HD70u_g&|6s)XR;X?fk1<;fy%5IXrE=qWWrx82G=y`f zW4%qDjJ$idVT)=cb1_#`Lx;S;uq)VuZ7BlU*B`>l+@k={qyd^6a@gUPHB);O3f0rx z-L(q7Y=nl6oc3G)FS=+%!y5Ri&9DNmJq8+i5(EKAAlp1)NAoPULWKdKhPqP1Rcisd zZzrk;ceNoR=!*1{kxN0;5VHb0I09Sr6#Bj9heUVT6zz7Lx(d|8yP}0|-O)*u7Y3Aq zGG8ST5gkytzn{beKF;#N>Xc7F7gtI$V1rP+@*=rMuCOp+%bOW^gLQn%Zh;#(415^s zOsMwxsR1}^onMw_?h3?D2AO!3`69tG3vGf0R_KUNQG#i8ogmlZ$o(gs4|=3NUSO)_ zQzT@R-$NS8uv*TgPmucL{A5Zs_y(p~@@&#@%4LrXc%FH2>S=*cCD!P_z&W z`(?pUN6L9Y0G0z{qy|{42K&?e3%X*2%PMbJ$NK!P zfh+Dv1n4g5-ioxCUFY#*%e~lRR@2QMu0V50bK+?DQSyA1n{K_@JY6veF8GMvHy+6z z2%_#R?YK6qyne=c?~fFD8okwkb)kN(Jqdzf7gbn_b3MZ|7Q3e z-7C*L`ghT3uVP_&;oqEM-iw9MtE1>^xjn1Vw1m0G&*o{SSU&OjG)<6;z5lO3kpD^lK*3lwcARe${r#PKu6XUGKGv~rf&1UOSEUnp{nJD|8#Qj zY0M1y>(<(bSr1iq=KMnh^M+iGstQL@?XhbDb#botf9BqOQH$)J>_c-4Dw#O7?bkTV z_DF2z^)Y+J*O1FDWB2x7Fa6%rnS(C4PsUF;&pa4Uej}jy$dg49iUTx(Wg}vYJ^FIT$RG?Dce^e?X^MAJ_s09K6%l>q4`Ja{qRBteo zgMbn`3jWiQW2d0P;Gf#I`n`RLQOq7m>W!vDVv73(&nxCRESv}Y`BPHKK=pKL3&61mx_!Z}B^ospLZ6-%soKKTunsn;89R&jxWD z2Y3U-9|niA{+^V_oIg6hwc1X|4MyfgL4PscuNW?+L&6ZVEsNIRR3|?^@@W+Naw2a= z(Eped*U}p(2|D8*vwIW|3JQb$Ub11l$8G)1Hy*ry!~VU?8gN{mul`~Svs<}Xg%c}9 zLu@QGek%V6(Ux?*d=|}@=3D$iFUP@=I3DO+{Q#^FBh{4q6p}v@rZdbHLe6Eyr@^Ut zW@ulP=o~{qPR=i#V7@Pvy4?-<$;b{Mb6bpA%T2Ip=0*Dm>oEzN@jQNfpmU+bgh#Hj z{QlY3uk%H-8ydd()Po9dEn-jgwl9R5#!hi|RvQAMhH@$ZFhx_45;{=f&GX$hknmU! zzk64!b*u=89zkeB1( zC?n`K_BG;)kL4io{nhQ+I4~Iw?+;RHa<2q6Y4FPWuBbiw?EbLpYnmhYNxA3x-e?7^ zz`vDOx)5<6u&7|)ljWTCJ}`^iQErYBwxS#?fmW_rX~&2s)XVC1n;xeAJ>iV@wqFEn}d5>K$no_`s_1|rTSAj3Er zo3J69OOAJRmvK$=Yim%=X^&9EQTKGtq9ju|SNb2JgHg(%IOLAUC#@v0Myfc+iPpGl z{aDZJLFtbntHq#nm*CQDua8Q6_-;D}yBi5&9Iel?`N~_6A|f^f9eqtTxUbsZgwU|O zam`DHx-sJH2fvB)n7Pr6pq=;-7mA0Oah9XuR@sZwcCVEZfh&71I3&i)seMa;ux zi$AHq2g>Vv=7_hVK(Uf%9vJz%T_Jk3Jwd@a-*;{GGQ+qC*elPseJS*w%m)4xaUB@S z&dsK@d3oYHzdNC)3z;!7*zE^9-+?@esgL*{qP&cTKyqeZYeAXl4Pok?O>?Ih+DZ?1`yb_{;LuejPQ z)NH7KoGF6RywQ6XvD0~)y5FYN`~uKu@k;t;dgdk$_i}7*2AM#iMZtsBL(_wb{_#pS zZ85)o8AdGoVYs{kYqTGkPnXSP8F}eCw*QT?pZ6~xG#3GH4{WOd&GsyIR9|e*93v0Z z=``QIPS@pbK!3}8?TW7**5j_a`NCngB-iwC;=piEH&wsaTz^7pdtZ^qOL%Y~12i~V zYynxTjR0-{1QgjI z+SX8ad0a;}-Dl#BRW1@+c51h8XUFMuPL(?XtofFVJr9iG2BIGw ze3CpX_L>=@VDiW@ZO=Z!WSfOgrZhS|xX5n!a^G|=VWP!euS)fLVz>Zfi>9?>d(KSW zR&r?hF{S~@SKhNo_7*w3pWV*Fj|C~>PBxu{6?r9+m%ax@!6Te|WJbXtwqzq%b1od3 zp)(Ykfxt_YZg@!7c(7k>bad3R4maPk?uL7->*v3G$9Kj<6`D3%48nCC(XMSbVo)f< z>M!yJIX8J#T1=|#R0vh>6S6L<#R5wNVRJNZJ%Nhqn*r^t)Kd3a>!&W3%tKJFI&yapF2U}+dZpoGdIvP4hwTr8Muy&Zg%$!{+PVHSHNz7n z{DUpg0-QTLJFd7n8KY{ah)X%J#?4_+gjps^kLj%p)o91vZxHHDo?rsvu;Sq2;R2%% zr~Py$*K)N!`?hVii90r&n;uS2YMF*sf#*k=LQQuiPfz_Sc@E~A{#I$f9^dKy{NgV8 zeuiowTI|_%n`pHyi3n!Je^T|gE3wgEC3vOqma|3QpotR$uX53RuiYhqHq_=(f6N($ z)3oI}pTQtwPyd3QRITEoJ~kc$3NBHLU#Z(JPXeG_?7&0XOIuq+b-Hh1#jj}g4B(O+U0pX%}B^c)b7MM1?oX{BwE!06m)prM6aSH?TjTU<9= zW%}~%LsyS!kfhx+aHvxSoSizDOQnBv?QWMM#1tR8&sK5sfH7!*$y!^Xs?C);zd zjAzcAPuFCMLDI2Dh2eeA2b~=hlKJ|z6;TniQaCh}Y;+e8aH*)_lmCfF0t^=G_r$pt zt!*!@YTq1n@`F&FGey0{hdy}$6zf=Yj=PVNP4Gqd_|5^n{3^dQ>#BfRq#Dc`bM0kF z(;QrLE`xn-rEa-yqsevL@OmmX`^#J-`x$P$-)62Bk>Fgp?oL_;n#$P7S^iZ(H%?RD z@m-!xV=3_3t-<~N=nxkkE@J%^MUK-j`#lO8cmW3bkx;e_Nqj6eLK>JLO=@~VH; z01Z@x)b99nAtY3jwL}b8mEW!=Pt%ZF255~t?derp&Dno)m^)2|WF>jzt3o41jIzZp&($?T>tJ^k5y_ZmIUpIS^oFM z!-Y;ggnhP5_g$%{gG9tT?J1%tK$wqDo(F6!$uG-bJ{rFS`_^=Od9KA7OxVozKr(-n zu~ihz+K>!%d8nPiYWMNA<+;$q68D3^O$t!NQ{@wFj`ia&HqaOnMObeEh?pAC2SM+1LXqjN*C^_m?jG;$in+OB#F?7=hD1g8@Lj2B2Nxsxg=gMg@)5YOi7A+h0R8vT;EY-nq~>$gHJ60?(4G zyyt1Z&u|rA`ncKOLx0}nc8%7V!THjht20Ugs~FcUE^f87o$M&U4p25`6i$peZVyfd zqvk%)C->%eD{al#Fj&?UyZFIDMS50UGJKp&)lGD(d8AM^?bxW3TG&@= zam1|xq2O!n4%d)KC14)2Sepqt5g6yKBR?On1(|qa!Yoc@MD1>St-INqvqqTR&DOgD zD!KD3{!rNPZX45a1W1BoBQ-4FdV%x2LsQgl-aMIbDaw5#ZNsL7a{%n_ch1M@EE)O< z$Z3I9{n`v#s!p=Ilg<77=(sczhIYLRrL74D882^BgtMyZCUz^2t`z8}fk zyFT7cqk2%1Lg^iTUaw2#;`Xnl z2wXXs@m~A;FQxu=T^odhl=yl@?q1`QuT@(LYy!4P^?^9zN}t>E*H^phfZhbe3lg?# zfRw=v7%Sm1=17BN_~-%Kj-tmyRR$;;bTRGG?|||iz;z|13m1Qo%^Ln(t-$@A0{7)#o9jOs@uixlUl@@g|4a7%aOl0RgTwSWUOa%S{Y>6YMgtcg>Zq23T0b!54fkv-Va0X z2*IST{c+KOt~PSEO7TY?8yZz}$;r?FCu+-fM~`(HVDFj=m*cQT8(dj4(-wFsUJS{y~;Yx%1XdW z3|tf*_<_ZSSe)EDcVFVpf~$S}R4>h4xa6|<-HxwttJwxd`jaPXllMNQ{ApiZ3FPQ} z`_5tx9+q!n4VAKX?X1dO#t!}I4{2xikuTbU!yRFN#l5((Kd#l}kaWKLb#L-j`~E(M z{Nw-jp}lua@&DaWJ300J+x|@eCcp~Z5uzOjynu;>OJ~|Gzosy*)-l3hnaefL`98i( zog~N>Lw(gQ3W(~kV+0E28~>PCLW8a+UgigTAzs0W8lF!mT?hsG4#H*yBT^H{8|9@! z;xEH1s&;x1u5*UZR*O?rj<$U4g+(Tpdb!H9udpCn4yU~J05U$z!@uX1T20^0pY0i$ zQ`H9%n&sl@;s_UtjgMg0E)0U*9}xFNP1W+i7MQ@(adkt!h9so|Z6#p;&h+3i@DL04 z+uDsf%Ae8kN|2u5Oh&& zd`?xVo2Z|Bfgu?&Or*4FraSa?vZe4zNde=AF*CyY=gH(yunT9?+pICm`Q@64NxWiI zKmY~Y<(mmTaMW^2_EOqt1>h7%;Xh_f%#E2LQC=@%f z_55f<4I1pwkqKX(Y>63#ZtcDO*ZhFjg8K6IyISQe(|pr{6iBuy1D|>OjOuXQ3a>0w z0)mdT`kkjQEs39KUFW=?0XnhT@2k|~(LU1ZLl7H47f?T1VMSMgTPYXFKGF@t9@zqa zo;w#}d}nYVj)<##SU$a)MPQ;8b-?8sqP5;MczJojEgn3uB+k$lI5vtvOtxn)s1o21 z)eg6`o+-*q(F^RJY?Hm_289vptf6lhoBOD{Cxs2|c7{uL-*w3YNx$ZjJVmHcm27{G zg-g}N(IoE7j_8|8!K@KeTdKh-5Ec(Q*$$WV=xD3Gj2kXA*wF+Npg07qZZKml`)ylW zs-mjeVgn0_fCom)RZIJ=QDy=5Bm8%kD&(~?ggC{g+a&lwF?8${-@JK>ltu9IibB>xS=DmGVb{@a+sfWu5C9yn`tXtkyA3#k5HK1dga_$TTLIg~`sXhBh@DtrE6tct zU|*onWOXRikC8@HsI=ID7uv8zMBz{q*umV2auVv3c>lzEoz2;R{@vHsm1BF&sKnRf z%_QiKqd$Ss&|QJ37z*ZGE&{GCX-Xd0x<*XCs1XKaqlj7{-8cDEM%?<=)Fp1)FGO7b zaf@Y#G$x|7Uzw$#Cje$aQzxT64^&$zwzKw$eEIAd<}oq^1I+g@n?VEPkWn4;OGUdR zK3U!y!`_oXN!AXE8H2;+xsigm^+4i!L1I!UoE;f={>qiSYzN7kipxp5TA>QM3#Ogb zZYKNyx{h1N6E(*(XuWaINO#CO(jb{3d2h23vgAan>I2}xZ?rzJ+12PMDzOx-q3KgV2jcWY75sPcYbYg|qDiqn3S6A)=wW1B!@ zY+t)IBn^uM@RA|czNZXbRGph8PyBke!1t$5U*+yQ5TEr_l|E|HfT}wX!9nbUd88iC z8>314q14xauqjtylQQd$Auyvtxm3Bd-jS-MVc`H4%EcE0eAC=#+YCUEDAgD*LJ6{! z@cDg+2Jl3iS1dFrJ>d#WEl&JJp)O#a^GYaQLWNg5iBnKNFAO*-#j(xM4R*nzb4@KW zniBXzhU8V=JZp)>B17vnqcgq1oOj#^uv+tk2R!4iW&tl(GXI*R(ysG!^2%I~xqpi6 zZQgCMy=o;GumPtNN*~3+7ogJ1kN&HT+P$^J6oAhIAT9Uttd(GI+@xo&bMdi)?2!?U z^4affymu=FWpyfMepHl$S`GD~L-9ofQM6RUhnBPS=NyvVa7WyTXEpi~G#S!##`Zz$Uo9!$7ptIfn43VkrjK)v?e~k7f z7GH0yQjR2un3i1ta-vAEVpR6R;Pb4I`%xsnnVxZJ#m|LmNWZ4Te2s<=5NIr42r)dY z!RIq`%gQ!5nT9MR4X6Q&CV1Y7+qhodlM5H}+qLJH?J6dL*^h%yF>0Xf zE->@cg$(_uhF9M!48mBYxK+M*QVc_a`ht5Mu9+{CK?W7EeRx#jVRz0gl{ZW@x?S=u z`Z-|+*uS`d+c#&FKy(D84#k7rHBk18+d#*#aQYdn(DUhA`0i~MZyWsS_Jb60oGn!S z5CrYsMCrOTYSr~%`WLXH1LY8_u4v@=^Yv z)I=CCbUn8u4p`;^h#(ywa5%{vIXWGaiwxNwM|Ohk3NE!p0M|7c_l=KL#P2UX2@gfr zfoL{TUpQXY&m1UfC#LepAOcR@m&$|l{{Ee%s(iLsFdExez^-B~?KL`KK#dd|9U)=e zCT&60SMb&%@=i9dBek8}-;JvJ^yitf0sKTcRG95}G!YSLn@ zu5-KSp?CbGZGYP19N1DRX?g1o(Pom322?cri$7_mB%96qpa<+8X_^tYP_M;p7M_sX z?hD9~`!%-#yIry63?z@bo=WcN^5x}(v*sqy?pTF$+= zZo{i@lHaLYZ0+^ZT4wjz!3L_5Vw=qYc3;jAh0AJR!4|~{H=d3{_3j?w1VijBK+Z@C z$SjDQ>c8U=zk(lVi_f}^c&WQ%;hBo<1H^fVGC%jLT-85GzRvw-L7D@$!Xk#GI3XG$ zg35TcbDv~s+h3M^C(Cfy$(Z$xTV2LH4)04`F00a1f(dVD7^)wkJ4*|NGE`z zGNOPYy(66fp@kk|qxTX*OF)58LkkIz0QpW-9Dncot!J%gt?zl?@BW7+m^*hl=Q`K5 zuf6y29lWUOPS+?1=a+vqK~;_5mr=E}TvC^?_%@dtER`Wj3e3r1U%QQ4g=Uy*>q>IB z?sBVW&iK#uJKSjEYcuLo_n!3{>9 z#&PX357H(8_X_5Vs`XxHcXmf$wR%&T?O>x5*=c&1Ul?lI;%3XD4OW=Qzy7 zB;l-9ysLoCRfYD_bwwgk&WB!3Bg^OZmtQGMWke4duzWM34pd;>xFv~18;R6Z%W46` zk~)Zq`YD@p5QyC zd%w0lAikcpu8hg7EJDW^Rd`p`PW#mvrtWOJ>X8$0Tqk%?AR$R8Sse0HJ`x`82)aZ> zLFrrdid>fgZNE3~j#2s2V09B;q2-?#%+4W#=;676UH}84L9;k3Guu||UDRFR#2Ko6 z>t)&N*K@ccx7$tMGR(9MB%g&YN|71&dE%JH9f1=(=LBKm`sRz|sptt08S^oHJgu(x z@xZg&x7pd)uW(PqGzrY)RBh~o<{4aW^Eh(kNKF<@+Y@s)Q1 zw;+92CGIf(P@rsV+b->by~1;zi^Lqkz~oE`o2!~rH8~+6NBE!VtPKfGAPv~{Q!Nz4 z)-M4-P3_4k{&FSCAfjfa23h`Klb2ui!C#k~!V@tA!9Q<9LEjYZ&|^?kBnGj7QPGP% zXceap%ew1FU+KELx#{Lqv`9l zvyTGBL9bEyJln8n^m~yw)c^n&;&&0;w5ohVAv0J#{l&eB&)mnkGXPwg{~{;k{RP86 z{{|Uw)yGi(FI-yI_iRMugw%?Jm^R$ryr2V+ePtqqG-2}X3RpkX_#gd<@hvES|0wul z5dSYMTa?|Sc<;4ugu}^AA&1z+kPa-LPd9v2AkOpw!p2zjuFp%i)7w-m-Bgf^auTa94j?K zau%(gyaob9cStItaIk9Q{U4IM9|J#cDA_CdpV2#bgkRVG*30WEk0eL%u+jYH+&|FH z#5<{V@^?-v`fGwWKQo!vXHw(uaX0M5zaI~J!(Xm?vnkH%zD|%Pqvmj8M~PF8G->AZ z=Qp6}jhi0vB8xVceCgtR{g96TBU)dMRY%Xzkf*7a(|I3t%rH~*+>&Kc0VG8Zk)77~ zP1VEt@5(Fe!wCPn_LWR6tg{>#h&d%S4K+PL4RNO5R27tJN>gS&5)l94Q?b!xxn-X5+1d0( z^I@mqHHfd@MyxyB1A6`5o22BVex#~hVm;6#@gCyo7sse=*`r?<%FvWghTF0|Ni$qTquR%LN#hsfB)KZKK`=9L!v1Hi}`#O z*LJ41B-Y)30M*m__mdMV$^AS`#z`YYq4LfLZS24!p$n^$(m$Lp6xQv!X-D(lUT$0B zIXn~+xD!SaHL(>L`|<8>yC*-EDKJt8H$eCJd_{8yZX14*p_&?)QGF;zq#K(=C%7Z7 zg|q1J-;SWq$I2PeRNOr3s;H$P3?W4_I+mS!f$H46v-Bv7eR;jlR3!y^>;B*W`7qiV zi_nA%Q5*x^Kt5lah*YG1HZ3%%UBSNGeh$-YUbGZbuof5OG`M5+11dXS>Q=G-h!-a^ zKV>I~>-JK_jR=N!pFsWEUmm!FOWIU7=tLE;4XLO+I%KyyN`g2u{DTr331c*Oj*P!T zUnts!U;8N_)=yZ7lX7TJG4hRq%EVI-YP;3?dGIp3I0B5|BvC_%k|joA(pdj{)d=oZ z`L>PM4Zo)}m!fW9IeBGuB(-CQ3_I+j#eH3c-4n1L_9(L*g%98E8`u~`5TJ*(MVikE z;kRW3-TUS+0%}@rmy~L2Gv`P#=N1CYx0>jWR^AA;p*=>lva+&fcl=V!`HQH~-M8(TGLSI-JJ-VN9ChHbV>&QTzs127}uk;?V`FM@)6ji4I*KWTL*zT?U>KsGgfARBkpq%3AMdKRmY&O9>(>@H{jz2{rBz*qJyok6KG7uN zl`0E7Z%REn=1kwmIAgg+eAaS=x3ZwX$3pVVwUygPhI!yCOv3nkL)1_jmKlHA$cV?o zdCUa(jOaR#KEbYi9dbc>0_Aw5xq0w`BhC1Zz1wY5^FV8$;|pA@O}pw>D~f7LGX(sK z$8InLpS|b})~<7|s01)u$m*!Y=-v&$C7v&_3zyZXEgvM7#IziE-M5c=3fK9EEA@_RYvWu8!*|yv zo+HD1x7zafuL1I&@L4{7erRLg>9J26^?MC<6#g=oujuKEpc%PQ$Q%4&!$g(t5Es-aNMiFqRx9h8`rxt-Gbj&eXH2i z+*vEDPSWFj`@!f*&AT!JDw7zLNlvXR`L#_o!j8vI3NW=h60Yb_8mZKl*bKD8eE1=v zcv?~_=v(GFFt?B=P}l^QhhJa8p0u=JT1Hd2({{Ftnwpy4Yp?~Jr5GaO78sF-N&>v@ zySbG(MHZ+nNS4ep&e$znw)joG{@mHK=fxC>SGBdY24*Eo;-^OmW>X6bNo3;6fOmnk zj92AFMzZO8$y}p(5`5h$7PH+&on8yn%<|1SQdpXVw9Lt#!C# ziY%8;gTT^21jMK-n_cn4f#b!TuOl_G9$<#0`qzgA@{KXA`%!2bFLQ+8H3Z$U{+DVeQh${P{jzD~)E0Lba;akbUPBPhMlC z!FeJe*OA3%<}qgg$ch8oTxc}ALZ1i%(K28Pb~jlDc)e?pJ9Yy34i;d2ohI*EuN9%- zZ4|@4x^>^f<1+W^?{??Tl}-CKPHeGLP9J^cyMyyW`;u=$WSiGYXwQ^<=E_riHbuiy z*J9Z^3SW-;@`eA?yKZ%Ej?95AA38xG?m&jhZ9+IXk3nl~q1`}9%p_;U+A&{{SVOy6 zbu5tI+J7Lq%yD|JjFAP7%gT<>YCPt+HU_`l@3?!j2iqO%2fq9D|1)*6CzlMDxE!MR zGm0?#A%@@%f4<|FYl<6FsPSwbRVUz^M)tL6EBb3!;1`Cx-d@BD9Oo{4Cw|pJ_+%}n zV10SDEFxDetGN9oAzv#MYhP`=JzZbxss|RZhK6&(ii+2`RaDlN5nge?o*dPp&ph-@ z*3`w7kdYQmxsbTS`)amr_*-(U><~L2%1YMy#ARVa1|U~^jz&2iV7ku+zEJR>Yg`Kh zI*6#>%;hRBLq`WVmj@_-escbd@TI05d9Nkhq2MtoOOl`83|rv-QL8 z4KV;xEKW?=gsYx>%|1$M+C zwN=9n_4G=?3_{|#H6OsooVd8gVSP$+q`1zHjd^NWd-flQ($_sYz7(59UW;E7KVDNM z+lFfpYI2h792#L>m&l*f*Vl)3>*?vaSnO5~YPDtu$vBaaM5UhrtLWCD*aCwQ!;UpZ ze_&KxRAzSA+|XKBxE87gY@fHS0wQ`)gUMNWqKPtwcb@#l%9t=~!v}e3bIP01q8n2^xSMI;PvGg~Ld1z%t zi>Ws`d;N~CuvPRM%7qQx@)zc2hPThpGR3@Z*4S2@J0m~G*v~{<&h!h`XJ65*zi4&Q zwSdeYz$<@}GCf$eCe9To?^|GO^P!iKu$X@*eaq&=>93#}NW7}Eb5dh?>KpWlia7DLnt1Uo#AZjAuiFg67rtB;Z;In;6j+V< z`0=k=RG~2@bdz8|HGEjkTlPA4>p{!=M;6F|CkNP7Vw%e=y)WK6u8y~>I}n=9!8~GM zc5eWBu%N;#XM24JcO>kbeOJ~*Q^fq#Ntm63pxi=m`?QB6ezzPHe&WtwZn}tN{C-|D z0pgX;$;k;-l-1!N`X#;>d1_00_9C>0mh8vH_lJ%(O$Q8}8%-Hqzv?E&XyS6O@t4We z24@6&E)AiED8DCCw6uoAwZY13r+dUnJ%_1-d-}57ZC+n7D~))+pQsiD}@$+tQHY@x11ju8Ji+E?mS!t>Ec>{+TD2aK<^^iL<&1YU8mL> zgShU98xG^$BzFzbN*+<_XkE7p=b%%wpVBo=E2amhD|WPPhCU#=U#?4}D!DV0=Y}w^ zJ2t^`=rU5T4t7z#cuJiXdyy(~CfDLg?$6bDNh4LBX2i0f61LQr5f&Tw)t+b1q_}<8 z_{>b4OS9|B+I^#`WpGF^(BV>N&^00e;&q zxYI1{h1yb5WAr;5?uLq`zaFrK!(%D2V15UK97uYpMDI>6Sd$8GQk|x&1GNemEN_s$ zo%jp@_~p8LH-N{@PLGO;;#<1gF~12Iv}uwhZZ#Qj8g%vBBB8?z$utgx+^pA#lW60E z%@A-q9OJci>KzL?1jd*}48WVljZbb{bsd3%)D?#v)7eu^NPmQ-+c!)Ta5fN{t6C|h~tJ;3|-QNN~rEQe$ zC|w80nrn4EV@g2n*~h>O@nfGu18du1ZVgyMuME~F-=4MZ8=^)sf!|-Ef6t}w>7 zc9|-pp&SO^&q8(H3O!hNL6rNF4;-r)^WMg7*5CyLI}O0hqSF?2|DO03gM;*V%C+OR zDg)?b4ocmPvYWQBJ|G=SR2V|p58pab>5R`79hv&{sc7EUA}=5M{*cosc(Bbkg90Jj z8#vYd5;75^f#={r-}L3!WE%##KXqg&FkyNxeXPjB$fIZUNt;HufZ=5WuQ4J%DRC8l zF^uk7K;xG!kcU+(R#yI4bzAQz%X^V=@scI1Di=qPpBhB1H3)B za>#IVTsiq!Sf5cwAhf+_!UPBccj6Vsl7LOB+NUtS@87g1=nY@FysT`lInowPgn`lh z>XKmJU~~jTi(wbOD6m}cw2oQ=IGWrhig);jD}XSC$CJt!7nJ?zomLAV~RAF-x@8_ z+BAaHY5W1b5FTO|VzaCbD1*=$o4;>htR_B7O%_IQ^TXS>Mwy+7pzKHrrh&%$9EFKs zbqQ=MAP`6%oDt)@oH8;wJF7cx^%WlYa)nXv(-0ua(gPMgwx@y~wp0_&NoFqZ8qC^H z6EsWig$v+Uj-Bi1(ma6^U;Zg@eM0tFscG~q+p}jp3ry+ zXt)LuYr9i>E98jkT*tevW}p>#+w=oECv0<#0o+B)t?j?s66Yg~F^Ekh%bPircsym={1usF|gw3|UOf%>0Rxu;BuIol(N5 z!As6#uh+oJyecx0cW|*9iaNrWd9-a*IF!aRc(M6>P%*&Lw%Ik$wDzg%>`0rzf_8$m z3c}YbHaJ-2VfYWjK&SPQC$_ZSabN?hFi`lXZ%2)0zA3UMK}u{x=#w~UdgZmp^8Lpt zf838~rPIyInWfQA9VNQk!=PU4a<4a?t~hJ2iU)YZHzxqeGFZO%c2cF8y6X7DC zS@U7GxyRl8L4O7Zt&QpdYAMnmd~xUm1(oh0_|aEA`!DHze}teD^Acq8Z8l&Zq3k!? zMW&40>zm_dmS9(0Shzrn0ED)(c1wweCzZHU;m${3*0PZ4b6h z;rnn4I)%DrRBq~53Yvm9063frUi(6gw2)K-U!mGYk}L#bYq;y_z`72df(yfay}hh= zqXx?T2@ZmgogL%p+#)GEGt;`^+!@l~ImMIj_!LG7aF4F<$v^3q(9Dy|Ky<%*s69G$ zqFmg(fQU~E zz}KoEF{1PO_Des+8l6(Ep|ov~R>4OrtSX7e6(yPifW@a+-p!TbcG@y~`fw%l zrRS2uOR~qaXSe0RI_Bdls`*3MQ7$<$3|cO@Y@0H^E|fX;HbCkPAa#$fw}mug&5RQ% zV0WF%(Tl*OTu0Bn)U4CgH=rSbYU|(wnoGDsb4)k!#{gx0zVw>k2VZo@>2ncE>j|Dq z1e6&6FzfM_Cc*kj^Xj)h$t27As9`mevO%pbMll1hc)*rNqz|zIT4uJ?vrB6MPfYpM z;bBS}9_$ZE95fx>++LQ~}%6$JpWM+xZiHA6)uuZBT64~u2TB~Vl zMrUMLrCYp?#aD5m;3Yc{vXd)X{!SRr%S6Tf5CL^9RBTdGLFVLSd+M=cubn$@bu_O=IOY^GAHDYV!iN+)^Jv*sh(~Mu#LG02T3%b6!Ob) zrbkV3ZN^gixI1`R7%BI-4|lAf-g*1 ztjD18jS8LVUFKC9ubD}C3Vs~I5FzlMAA`U)hi~!JA|tTV;nNIV!5P$XI2GSsEq8ys zxF8qf?!BMjO~5&hk=PUG6$Mt3cg&0XuSRI!(ixK%tbj0kMo_so(XBo^ASqXV2evTj z=TM4}VHA(`S9%>mM6RvDgI$In{k-?e-1Bu@31Z$K+X?(E+Dp{3mYq?BdXV;gWJ9Iz zty>|Yq&PO(xZ1izBwiateZVwyY{>WUT}WpY)AET<(lA1LtR+(iupOiWM-}daIDXNj zMolS@;+DcNEufu*1r+>~YoAzhqk#{X3h-@!z}gQYyK)_;iq++cUWAr-g+%nfu|Y~S z+3X)c;$3=4h4R#QiQ*2lx*LE$?o0BSPXIRbiR){$?&)@(C`6p{9R~;ok4Nxg^UCR6a@@r_JKp<9FozWgH!y>ph#FCA`unAkK%LiTnOT5rK?uNqNZ4GL#pOXyTcY43jRMfJ^t16RtTggv%=XzC znCR>-1X1^Zs#zB}T^L}s4(~reZvzkw$mr~QhaAZOxhw#5z=ZtSUzK)fA#9e3SMqcG zT(Q?a-A&I{kPx|fLIx(8zFfUPPJJ1ddv^)pdVLy1z8P9#kyLbFQBl!w?NU(Vi*16z zIF`|KD{N!tFeKT$v#z))`V-J@n$)-#y4IFdvU=I>HV8CgS^%-JDI>>uH0$nCpli61 z`Oq~4ikv?Fb`QW=|4Q}WOrDjoUHA(lxwh$#9s8BlfO^!>=7B)D&fzki|Gc*~9xu2_ zxC1A|rJl=I4;dhpp|4V#F}4fq3evYgZ=9F;foG_`tF99)&%&}W%zBhr+iEDmD(Qot4kC8&L-u>0JVW z@_E{#leo15lFRTKfL$Zv{BJc~7om~tCqp|rSiM!q; zzh4gbz_qWY?L^f59phga6V!F5SFv(cQfb=PI*lw*aU5(ncp74D$qknG-VaM0ioEGa zzXv_JU1M#RKPoC&S9<&U&UD}{44PaK<;x5CGMhfhFG4-$*FzWhltxnY8~kn?_}1@c zljCnd?%BRNK9J#%QIqY`RD_X6Y_&y8HMz*sCLM2*$uQB|t|c+@+Un|-MO&k)CXNXv z(oWX&JL%LcUVL@YY!5`;w1XEanYvz+uO34gC|z+^T~ih*D7U^ID? zY&U{5=nUXkkNhVTbO%^R8rSo%u)ayGt;1}n6sxb3SKjHGw+1&pxw_ni?O~B3X3I!< z#u@Z{b7X=e6ctGpv?G(Tc4I%mx3&Vu5}Tgz$>-$fgF&4Rh-#a3Z##!|i0Q1&c|p!7 zGTOqzChGnbUAAYpHv?FKa5QVXE4n9Oaem6sVtU&D^iyn#3*+(mvT~(U2&@rMvZP)^ z)#Wh}l;9cEZ)yzj3Gi{XgKKl63>5qKYLEF0LJSy z_W&CfZzA8da4!32IV;P~>O--!+uA#^R=+qY6l^!kfB>#(>J^v20RbQzwMLWv`2*fL zzn(n1H2cYw7E(z&3ma~xMKtp0$4!V_-1NadDHLW6o~(v{bTc$J$di+qZddWna)3Vtg;NthM1dylx zwwv9WN&-q4wdXT&8Kk(#$Tua}ofH>Aq}=xCp1LV+P2A_uvnCb(Xi#Rl^i?88Vt99m zPVAzBNlW;WHB3>?y?fz~%_o+edLp?=ciA|QX(-hrMo+qbi$44B7>X`AQwL3tBGfMg zc5L8_7}Yg30GzyW?Cnj!--I&(yW!G13Yb0l>P-XeNF8F48`nS;yZSZk*l2M~VC9*! zXA>M{{9=4C{a^)Cc5v&!8Hv>0dvb`Dcc+K8{`mKf#F?cQD|8_9n#)jy3Md9_(eXDA z&x>3wxc3mF_4FWpEH4pOWeD^reE^;Sh5V!8az~5&m=|3-bmA_>DFbZ$YdgDvYRnYp zH!1VwZfI<9pv3UgcVfpp)&`$ll(c~=9`>d~(M~us;0glONyGZrVIHimIf$Cq-(-s2 zxkrbhI8$7?5553Ps9=A3**wA^*m;->9ptS+H&)#rJ|G+y7|i(yFtkgYd!AruN4e%3Wy6>g*CEvslioCJVkALWm;q8IUk>ftdeVQD(O`+ zuoB2ssg3cOhdr|3IqdU^?XVvcP6oSTX4G#RXNb`N-xECi6e8@WVwy#z;@D*+?JE(# z3SzcH^9|X=C0I#>4%g9LdRtZ$`mfCp7^a|P;w)&*7luC9)-e`@N#z=u%2)(O(-!1& zmF%`I*iAEIV!QyBV3b;)^+ess5lk*ypWb8Vv|2xWFpqwJAs<*epMC4`;6b}S^Vf$z zO|5+bX;0^tkV&!i6WW4;eLlG`e{N5p5)Nsu;$kY*Rj^pkHR1l?!6h7Vpm%ntq(Y&x zGjX~#fa8G?k6qvNmO?K`8mo^-G|(3c6sNmTQ^2e)D%R+@`7;B$L2+HzfebWmWB1NtdA-mF=wq%|i3P zr@a+=6?IA~AS>n+txzV~s@T%{0bQ{G(0nv31ulo|5BAv?6uH7YN>lyl_$k~SaDM=h z8tS%GZey#y%se*s6F+bST|Lbrm%h!K__MmI>e7VIoZTMedi{($iPf5=U>^OM`HadmWB4^}yhZo9;HU;phmqMa}J=dCzPQstdb zV`3CjvlUq7TKXH772WRfpZJJQZ%jCgY6G_G;R1+&evWn4;}=Ig+6PYWE|mZrP0BbyN-hl%ea>c!rfzLBf!+Nx~9q=kgR<@ z2JbgCG+C~${{Sfo*M~m1;inUXwLt7#X7QNB#KfehSy-kQ303Q8Y}GwZiye^d+Eb*Ar6oSz1kJdYNZqbh;1>lUVQh7f=YD-nt5SZ<+Onz7&96enJNT5}UE_6P?)&G5t@$Y%| zMDg8_v&a6y;UxpVI5GQbYBUr#-%)Zq!*><1TixsR88Pnn?{~N8=Kw}vr~0KIXwu5J zb)17eonX{>rr4D(Q}doPZJPu?jyS6@$LKN)aimQha8*c~DTrCb4`Whzz^^>j2?T99 z#971J&u#2XbJ*tO17XH2L zpva8DePpawT>jpWaO<6gnens{SdC#Nv0FVn-1rb>;xlJ^&j*4+lb^45X6p0j&%1Dc z{BMx5M`~~A|0|>nPMthl*;_Jkj!Q5#4LnDe;TjdCh{OW${^tP2^OdEsW#SH44oEZO zFDVNAe0(nSAGyW%9(f|&x6MMr|C%90+;nKGNTz##^ztQVWMV2AJ;1qkuxoKM zrRQ7xVSOfulfPV0jx(>njDci$$Cg0w<#)~DX4SSBHe6Sw_okyzPG<85f~JQP6#Ey8 zowj>$d4ZwB(h`BJx>UjIthAw}KUIa)yA=K5!f+Cd@FuSg(>kN!oK(+nsOzW(l>FHy z%d_#*{3GZ{BvL&c^8?VU-152L)IzOblM}EZcLY1GJ^FcjYI+(8;380y-)S=}!hs%7 zd+!Pd6{{H@UNLrjd2Ov*!D}?F>m~x_&^K)lc-D?U&k6wLK9JCI!`I?iIdpg;3gf&4 zvhUqPXNzk#4WrdDH(c$60+9VH^5)BNWrYVzE;Vl)J~(%+&kVmaC=XPkd3~u@Bp%0b z3;?h>j^+EedjxSFEgwVpSrR-!E(O@I0LKspuc77I>iPvFpaCPb3g^JAsTx;c0cC)I z?%1-53;pLs3>b(2#mXdUo@j2UzJ*2Xo)FCEqXv_&ZJn5FNg@OAHh|Zg_hrBzhW*gy z((xT~b^alE?GgLuKbEk_@&~)1b(*ytUvgUF0nxewt3TGYM%w6fP=9*A`7fgK|0(ss z|IA|k_J7e?ycH0}uq2??btNQ7+0tEj3T+&TPpvi@oLZFycGVpQLK zIvnR_W_a>py$RO$RZ>#Y#+k2g=jFLCMM{;mo?c)1XS9pEr<&D(oAH?(S=N#c@LHj7 ziVx+KFA48>Wa)KJ9`k+ zP*ATJDC&XEX8;)eU$GjvN+y)%7-(sWYG(zKmc6}WRaE5T(=31X z5^SLSRb#}qiBn8R{{psY9<0U^)0TlwRd3!{fz5P4-hUZi9IJ+DG2afbZk0Qd8{3@# zMCQKWMDPRq2Lf>3IvfQ{w!}I(#lRCQoMonu2tK=_bWTVpzT%0xVS-hR-;FFRYr7VE zU_O=2Y`C*-QR0hXk!AR6X(Ccm)JNNeekMypzDomq3m_z)=cztmY}|%1g)1yqpNwzA z03P*RVvmy3&Lbhm(c%s{~4*Qo^5V!gzffLY-tGyS4K>t8VolL>w6TTVn)b1Bzoh_-)TD7Vpn|&x zPzI8HnXj374jI0VI+ZaU)wla>9eaACeJ6M1(bf&TVW1^k;^sy&OZ;6ETImV4fCDD6 z^u;E(VbJaFpehnAYHCbXskLbB7`!MAJwe9;!EOq3g zgWR!1#izecZMEjxt!vd#pIR6fU8AM=(T_spA+v4FFx|ofE>oa;(U;Q|7fL1KAK)3V z!=R(4@@6-?V$RAK8R38c9f(tbD9$CngBd3+sou>}!0_&D%4Ew8*Vtt(vy+A~ z42}ed&(Z|*Jdjkd7`}D3tz%_WLc~M|B~3XHfS}^42URIk!73rAS-kXpWY6OI+`9;) z4>(-ImjCC4(1V?wTJCP1hVE+a56!@o5)4KPWl1XJc}`G z{j}hCTYcaPj^nH~Y$QKk0Fjo;)+w2E)x!i%5^pIr*tW}PjB^|FFtw!4;xxs?X4lJy zv$v$wEyFgUmSH>4Lenc9bz@3pq{IpHN?K0`0##C-BPCAZ5f$%VH|tfPj{d#9LGeaA z_GqLY3Dn0Sp}OroR&4#LO5vQ#?W8TgjWwQ9L|z*AGCrwi8I2~vb|;G-&P zTt9r#G!Pff??Nij4>2@uoHiOT!rPWP2Y7x@dT$+JB5tRh87cF^0_N?tskVboql2h+ z_I)vGpSW@5ZhHZil}J(Ax?;hhE|gYV4`HC=z0Hr3v)261F?K2rvct z#L9(B2o{^!rOkGTR-Ns44nA7!H>bRYvXJHD+8iC` zRToD>=7-qV;p?!jI7z2eZ={}pPo>SgCuJ!QYzYwmaI<#x7!CjgnR&VEeN?w(|-DWeWHT4ydvy4Ve}N7lxn zpr@q?M9wU|oPy5^*`;Tt z8F+)j6ZgjW^f6v=eoY{CxDvR7E+>jg_JA;pA8XBUbd! z%#&>Aqz**V&E?8Fsj_Niwwfne`aYIyF_?br$^DiY-o^cS0vPqS2N4d-xWrX;FQHKKca#ojj-QF-wKG%<@BwavR26qw(MtzWR|yD0QD!huD(Z_b@* z4{4yUHdedbwquIgcyGL`l_Je+d)%NjlAa$wr=21Jzf-LjcR*piDqwph(Ae-sbK93; zp0=BxD?jr(BcW!4@SjZS2pBMYQY<7Sg}d0{x<3JzZ=l!E{hK~@r(+!(e~H*Ki=7&- zbRNkAy@K!P^0^F8j@0LLEl#CG zMP>Hpx_$5+uJhd4&)A5iwDk@E-ahGWMc5 zdY#>&Q+AUIJzw|e=dC`KZXkJcA|*W``G6@lxfmnx(e%^C-u2P^bXedUR;4}56B57O zjKN^at)>MuQ>DfXem8B)PQ_~bOdlzf@}Tp9J)kv%pVsG8h$hCr3>XO~oK5nAd+D2* z`KJ(7f%dDv8gT)h8d1MtNcmk%5=sKgpvHa-dc6*MS;@71Ivx)kKUBeL*A1K?SUk3b z+WUFcja_(E;Ju@((SF%ww3P5-scbzYAD}A~q5#77pZ6D@=IaK_v9qtqsK=XLTj~Xb zJ07+I2;g-bi@pf}9`+;b?Re@`Dck2mbdas-c>g=GyidM+=>Fh-H)!6sPxzx94W!iH zq_(kV98?^(JYBzeo8?{}0mq^A^;u%Dj-3UtcN>z`V$_$M^KFK*5u^W`nW!s zF_fxBJJ&;p0u))M0)gut`TCNG^M0&!{ON8{qWlbB2$w&0A7xS5vyY_PqZ$oBIY0t( z1OhN#9LlBaBcmEOt9b&C2gpsiQ`b*mGx+?ZRFtvYXg*h03(KOSuaR1A-y*`V5OAGn zVbfo1A&RyiXf{YP^-Hqj`G+Nw)Y1xPSFoF1rx?H63EHWgA5JkxTMzN_U*)?i(p{KV z=SrS)G*FtaO5)?kEe*`$=JPvEYVL&r3S1fxdg^DnLf64*KKNhzGZ?+J&Ldc#4Qy=f z!9HQ2&}Jo6lb=4hiV#gJ?L`2AvN{RW+Ch;a-|sMQd~WUykaQ)6DEK4la6n5@Yl$+X z=9AnMk?dD<)RyOqX0zwreVE-Z|E6LpLd>m7->)Ex7UJf2fULHiRWnrZbun1L@9tr? zXE)3H@|*t_hWtIHXMOa)Rf`9gg+63Wq%Pv#&``zv%*iu=UZW2b8Qh#)ame|ukBxwT zLS=yuNF)nFyPc}<1=Fq>FZ$!`4wu3Rk}AhW0KnZ9N*DA4x;~Uu}e;Zl^ME~Bz69D# zfG~SY5K9RKUwvfmUp#D&u=F*ChXSe}2-Vv>pKSo8W!!HB_IA5Y<6LB|M>#|=^<^3} zYzKIuuym@gImS+_Hm!-y!WR4V_geb)T@+%rJ$CRNUo;gPVr}iQdvL*KTi{1=Rd)$| z>vxw1s*`!8a;;KOBgK%m&8#;349j)tzE_{hIx*&Wsp-#KZL^)G^tfqY1qZm4?*9vr z1>8nx&)HBqM81h*oMlO-CAPs3X%{m~P}&-IRyyI@A1?uFX1ULN`2Py076j8kA|7vb zr(xB&i!A?lI1XvpN;A3OkJ3@xa2^YOBB!-`#UQ-4=sE`jkLwCnM=e`q8zhEhNf8j= ztMK@Zq<#AU>KP6({O9x!cy!d6PLZbt7`)QP+6AEP&Vwxpu%rG+$kGPXO-M~`^^~_D zhnH7kJrAY%?Yp50e4)PT!-qSeFU}TYet78{LMbXMlKOlv|o1xGfe$Hlf2e8x<78$t)?KFd%7680>{R#wuK!^go zo{fzSz9k;O@EMYkx%q3L!U&B`-!}eqbQ(|N*1xwc%*OXzzgZCw)XTw^!nk~0DA2sD zKKjWgD3HF|q{}lHs30QO$hWs^5%aO&6}AuHFaOxTHbxs~4}Rce^@(^6CI2m7jn3Y{ODWv#^Bm88Zn(J8zE!PzCrRYs z+2J*|WwrVl%&|!RHS*O!N}d1E8TeKI(X61G7QUk9HEP&!rV6l%Vj)iz&Lim?8;{!* zE`A+ETtBhSK3+;+O-#K4dSHl43pa2@#^g6-9tp(2RfP1MpTbwQU!Ra5T=)K&Q)+ut z0-%*+gQha;n6qUHQc_SBPlnYQvjR*rTeb!5u@eO>DNeURF6ZJL|?a zcB94MowzZK@Ze*%ORJ+`V%m0%?YRf)>LZsFWz3Q?b5NmrO(r?e(`R(W!w&TbBK;3p zk7gMs%SlOne(lwdR`=U~U<2v-!E^V0UgGOx!N{^Qe=mAawujMnmL-W%JxZD{fr*i? zIRTfA0M1^%q^Zg)K*4sb%Jd|oVYIh0_MX+m+DN9-A-9jhzZ7*k)LH=%@p zJ~Vx~FxFj)HlDTz?L;tpozu||Fh_Q`$*)fuQ1HKGGEu+vWiy?^S#_v;(cSG~`f^9C zMz;mh7YdP+ot#TamM=F1;wPYmG#K9{;Rk&VRRA<-<#3|+5?s{0q|rdx^wYv73xJeS z+PID>ntSupHcZB4WrdAJj3J5L$p>PZdEf>l0{TQimNs8;!Vh##hWwX;0|1!TxOG(* zZ0&8R<{p|4J&fmUB6lcR7IpsKx9lATD0ucwz^p#*i*s}oCvU5N!e2u z@aIW`Zu-U+IffpD(VUz-VI_R6U_>t~S(#FW+2a=x39>ddG(p*m+S*NJKd7#6WaJWvneoQ0g&$T!Xu<&@|aE#Svs~I4_KC*p68% zJCcT=)vbw47x9l+S7>hf#jw5WVmVm>@gfe4ytNg16v*kdQ|z|KKmDSx-a2yjVpp*n z#S^pIcpnXe->ATI(c&bQj`L?sTOw3@E;j*Tx|<2S>K@e>v)mu-*jH*CGrtoj-SE%IZo$AD#`V?nvNFw}I$$3A#OPNkX#er6qXf+-eodaDfQWiVM#hN$ zDp&egKx#I?m>stS17~+?U0Wyx5pYqe~*> z+{<_3G0Lg+GIs#lG41Pox}M(icmcq~v*~^?eVoWAfF@|L7<-Mhm4A91?_jlJL_97S zREo@_7Wxep6)lE~9THjYJ**dx=AayLd6jp-f28z-F>;aT3SKH8d+Ja+DJ;xFDOH0$h z9MEH1iqKvSq}S7P)Cd4%Syh{fb-*HCWPyKWV|VkeC{F{{)IJ1;SYh^-f+>K^^hOMs zqJD~gi>E+qAs_BHlBOUk+L4qf4S3NMhYw0xvMLxvDsw3=J zj#hj|rO|QjD==|!_|Y!oht!=Memcc!kW%AsUH+SLC`#;xYBS@ zM(FAu=2}TI{@uZGwym3SV+?n6;Z#7+#5 z3#I9_kTr#FRBd*m?RKqj+4fczAxAUSDnDOvF0j^Ek-+{+lIFR@oH=Csj9LdYOLxNu z>ZgZViZI1ZQ6A1sWePAd!FA>82r6yUHz$NTAi>-(I+0SZI2%nCPRK5z8R|LGV^7i= zs;n!{ot2+kk5&ZsLr9;S4~d(W4e3{85u_JqwUbje;b#7I5MWJtqsX<0s3e@(cgJ@;|wz0WFnto=xxdCY4IP9UAgukn)KI$og;r~dqN63kt8v0wZ(i6B_pe zcUD)9TWaN6NSQr337DgNR`TjG;JG{aPPqo-}&E8nW&<6if#Yoiqle)$f zinSjSjg1V!F1`^0eOB6-s4T?zAr;9~?JcS)D}V&~%YJQ6mphzT*D@t>0n(e#mj(Jw zL3b5s=9ttwCIe^ayCO}fn$lo1b0#tYy>I_U;sg*kqHVWrS->n<1X$h`^$a)vdb^q= zOJgapcL);i+TUHN)WgD%ih*CJ>?ysBzy!eGn7N3bw9yE#=|cwch?(at7wZ5~%(SC3 zKrs4RaVM^9bVN{fm?H+jv4$-#@c&`%J)@f3qIX~HTP)ZR5doDZAYF`fPyqqyO=?t{ z^xjKU1XOyL4$`Dc384ixF7EQa`H(BsqeeWTys9p zZ>Bs@1|rz#prBg72rvGw21-Z3m**T_2L zKk~sdHEt?YmFw2tuMXlV5B08812gfC#mc0ThUd)r|LnGk9KV-PZX+?cfD94PX>!Ou zQ)QcRK;=RqZh~gPnYvTEy$gnu`a=}E?)xu=FRCO??lc(nAh#;-+fHTx(SF?M+T!Au zFsy#!Ja$`*-FkfwAb(!{np;|kEF}{^MXo~~Y{Dnpb5p%>jM%AKsDEIcg>yHsTd1C| zk_R$zt!n@z%>`Uowqs>1Et06I{pWEK=roT*Tqgz?(R01r)=ixy*@U%0PFWE8>aRQT z^*!R$Dzl24R)$MY-b`3ONqc{^n&xnJeLa^?(ldzj??^)Wsvas1=+Ii_Iu=5^Brza` zP4VFY+=_*m>M`a9sINCc4Or*(q9h*hOH=$7D5#%)n93e`DD&aobKi}HU~^6n>pd^9 zXmZ|63t0J$lqL-b`?R~8lc4i3NUbBc$F{9d zH4gcsJUXpYz}$Cvi}`^Te`Apuu*;f<;{hmwc7)OvkMc;ba(+^NpGWrp$TIj_#>qAY z=tMhnXG&J$>QP_NMZVN?sq3))*sFX-+3WQ_s`Zjndxt5O9m9;v7>G2l$_YX7Z*~ft z8ne1qoG;N{)m0;Qus0JQ{zkeLIz{UK51_&K!}rg4Ce*ugQY6OCoFo_BHSFYbrikX^ zt9goV2T!9}ewUK(+$xlq8Lza;ks|NumO5_k)u+KRr0-4KF19_zd*}ZdPE_fvwPF#X@~p%%2q+S=O2>Va_xz? zuH^IP>x+o%qU7b8UV&ow--*2f%7%Hhi=nM+2>!n(_I!E|$HdYSUrPXAa}McaPN)-Y zML3{j`R%$!fm)-@n7{wgoqnP4FZ!_)0k}W69<0j2*|VrZbXL}!DJ9JUF?NqN(q5#H zjQXTh`EPlb10ov;-zX9*O(1m7vt!;GmKXI}^~ebOH)I6w2E5ICAMv;RQ?;bc9IVe1q zlUSEd5z-t`vAYvmv1<(DPT2Py_nW{1BY1pZ;g4kFv8I{y4 z*oxvbhD3RA?Y-_-7#QMb;U7MH@Uq{f%OCiH)UT-3XRO7kXJ7Af?K!4ug<3<#?@+SL zzp`6(D=Dp zsGy+zEMRNdeFX}%Qhc$(AeJt^F90S zE42G`x`1_{7DlU35? z`i2gt=ROL8)I0*3snFUR$1md@xOq8uF_Hs1Sa(9{o^SIFqakE)mn*Z6c>lZAsVU>@ zxIOO#3~=Ay!g;DTMWIw)RFWzV+nYMyiBJ~dqn}SaU!j4bW zi;~Ei0;^aD3gWS4Jh+YXtzAuqdnFcv#};{>iC=C>Yq1_mJO>iQK&&eKa?X8WPkz0@ z&bfPk!W*~hc3KVucpu9zF;uj)aD2+FH5fLV6{@PP9^FCycL1P1yQ6EmQz{+`uXyhF z`ZM31*Y>}Dhlmc-y%#ly36%scSFM>zlU-j6nw1a!{YU;b$$B>VaU{R->3o45eEiuK�T@t+SF%; zz|q#ixeFl`G7#DNYSM8VPAS)QNKnQaxN`aWvHX)DA<$L*K{Lm?qtXbdUQP+;L**&Z zuQ6N_FmEYnm*^;f?YVW;d!zRv59rQ*ibnnpCuiGTxE(TF)FQq~3TkDQUyTM_r7O;> zRijZt{OQTyypp%?YL)*bN!X%Z#BFJ?J7+TA0bw|r?Q0%DEH?56Xpp?+A+Cc9 ztKd3tfa^eTaMuOsYk*yGUX6C-MR;TX&+=Dkb@PCt;i(mR;2$16@Tp(LzanD^Qn=ZY z=%>;tGd(~X4G0|hEwPMUB$ZUr*!h~FmHU>2Dblj5HpT|Z&9(5U`o5&(3Rwa45(<8v>fCE<)|sA%5}(+>bC4_K;W}~yK^>IC2bhQkwe4S z+A>B8-QnIrQM|TR=;0w9k6xt)WT#q=;y4qbyGU;q1{hz1LCoQU@@Zj$v}^PRVUoCA zf3LX=?>qfqif8k^qM_dT4Lm`}IbWjp9lYUx>=^G%>p#od71S`-cA4-(8t_N!hC&Hn*c7ZLQ|1eKUlK*U>r-K#pR9(zt} z#MBF|xJ3L1mP&O83F=C~{^soI>6MsPWQHzE>p^(;9Nx7ZPF*wcQ=}Oc1KDE>rdq!3 zWSldo5c5`@^_6)K@X=>fC9;4W?5iad@gn&2Xa}f&)|MdC+h;04@_G^(o$BZKX`vJP zxgLdAi&b3em?_b(?YnX6oX@jRhQYQiFZphr06Jmk1>Qj;f<(^Oj~V+Ug#G;NY)W=G zZdAVzWaxWku5WxZGOzZqW_zm+C8twl2C5CaiZu8J%_B$Vd4JS`+b3)E;oKMiJgJd8 zQ|;pn%kUao+$+}3e%*CV3-M5Z94`O)FPma+s4}IN_9CqYFz^%CE;!>3Z^f<%`n%C#cBGi&x64W6R)G=e z-wZ=nBH#FzGux+#M)}86_lM{?yAnvMcOKkPgM}~@ZQW`d#h*EMPM~F>>v{s)FQI2I zh!o)@SQxFBwzTz2twM{Jr?RD6z+qA`@Y0P6h}si}bfAG)n!Si#$_4^=z_=GdSoipq z??lfhGNsJVEqWx#pZR;xrau^miy_-5smqRa^P>FfLF$dpuGhF0XxNhw*$oRHg&8VL z2=z8Z)51efyI8O2Kv!3=d?NP_>1t+b3I;12NjK02>ioDK`Wo`g8+Ow)ala5QowHq~ z_F$BE34r!gU-Xy*aEZMGPsiu^UVg1gf+&TVZRu6jc-i(4DK}wJmIZ^8ZXb#YD|ql_Znt z6LIu3Tyt!##}3LH8sNWjL0c6p9a7%M=jFn-ACQK<`Ee1*Ml%5_h<%%+2eZHeG8=%j zrWH7b{;S#V2LV0pyVKD*OEWD{wY`X&$_cy3@81EqKCQqSgLeY#Qtg!%(DXbFI@;;n z5}wxki;cT;8*CnDOnty{WR7!7rWOwC1xI&tOK7=Yc~?)kWG|mmTJyOTtBo_^D-`I` zm&HsF5a=ybQy`?kgdJ5|YjEpd)Rna5%wml3@tBAWsG4v6yPwHDMq2t~O)U{W+EeJU zBs-)rZ8vX1YV}#dhNX~%=#xNRrq5$K(_=4 zhd_)M`Q|@&rqxz>Z{|HY&R7Z^`F`<6kKV0PHc81_$!Pp55{R6ZH>5h*}?< z%$4t0V_oZC4BQ_oF&J3z0r?YvlwfUH0dlS`M~V%huQ0N!x}xG9kHH@Lej?s#>`UtD zF>>N&iINE%#a2Li3e^Q#Y2MRdj7){v`Ui04gheaxZ5I!ju^2Tq$b48KF? zF0Io$gC3MDU_F7``UcoFz)R|ob<~c7l!Gky)}J!i6<3)zhvwW$)>7LS(Fb2p#y(um z4Q||~%zM~pj1qD6x?Um<%XF*Gb#Qfx$_r$E0>G(Qe-xNU+CY zPndo7W0^dsRe;L2bm=mqv(}lbCHy6}+L+v&oaOk_nq57hwr{)Nd_$rb&d`s*@8Eee zJ8Y+;k9D$?$m&BTd&en0gjJb;HE>iXo_|9kET@(16Qzlb98isvM3CHA$On|VmE8}T z;5hRXC|}|;9}U~`*8)TX2vfKh76l~sJ1&Imhe2h^bH`W zQK!=Uiex)jV(t-SPV)(@Ce&0we>_MrEjlrG6;Y7mzPzuMVt%Y|Oab zYEeWwmG8Y@VgM#?Q9TT*oPlS(9y0sj;kR#R!a*{Azz*jN)CF2#hD?#?V4h4#a7Zm& zw^p06!EaUv%(ae%I@&CdmeYE3#4jOs3W0Vh8V#EC;saf6dRDD@BGI6Ter^#~dwK@s zD-?fc+<(uE2lDfZw8F2Zslaw=F2)CNb!xHgFMt4}+o?PMG5NqBMG;5K$~CnP4E(47 z>8+Jq&}R}L5zLrXPJON_Z5iL1`F={yi++m_wTDP z!$kB!vj_p&86f91a|2iFqsSkOPWtNu`yD=z$#pbuzym`t0DHXxpdsKp=EtlzteBT4 zQ-p$`ry98Sq(GK8 zBA>|0GHZjt`1I%x?!flO#y6fI_kQQrp9_&rRsBFL>G>q3PM&V+^uJ(&sL~(&Qy7kr zRaE?_w~Y5uP-wFVSBSU82UP&JVi~)_V;QYsP2J<@K@Y z2+^Ps<}xXypoA$6?#ZBq?wq74M8nn}_WQN3JN{{>;-Ie6dWa9aK=OS6yG5Ms_~CMg zH~dy@7L28idS^iK$&*25%|j4KSp$>d=@Yxe0;OEHOMy{&GOszO!lV`6mtZ~(q@?ogEfENZD zrVTYdlAl{yp;%{@16RD;CMU6zxNS_#|&1$O0Wn3>InZjS~SY;7UB2-4Y0&si~<4dv#OcE2Q6N<&RhY zo>tPOe~n^pZ1%6LJUX*FFji$*uvUe+2zHp>9N$3{ni3uD1F)9zP3wSMpp4u{L!{cr z$Cd2QdfqBu2Mi`u5Hl<<3_$11f^(K;58U6Je$Jy5SJb)ceW*cz`VupZeX&MBc{p0p zAm#IJ%5Ro!61L8mU<({Y0!n=d>w$;Ne&QOD>U}v#0M)op3<=-cgh+@j3TFe@wu}Gj>!J&O@BbaK)ytME5JzhH-@6Y+I9%(@{h7VIfm~c~X3x?(tZc2}a6ydO8oM>B zSI_TQ++~1RSr}&?+??gu!IC8_PmfCN9dK{YQ*-Z1p(ysVFvcCHU>yL|Gf#o zt<8L~?w4iKqVfr(vR3AoegD3?y#HO@pkf;B{2SDB1EOxLRd)5WM;Jf6?-$lJ$~%uW z$SW)?gI7Hw|5ej;b1Z6&`i{mxOzL+UjK+#GM!!s#h7doh?rp%io zlq8Ll%TuTY8T*1S>DxH0Y3?d>6?yHRe9t5thj zn>sKv4fixW<=5WS!MO%6DBCoX#=g=}B1T=P&zM%Afi|AL@S_j6I3j2<93yDA!H$l_4u1+f zPn1p(s~j4VsQJW4bs%6?1G4fNAXd$7+CJ0@=#2=T!qf+)-oI5Vj3TX9E7NkvpuiF8 z9v<~!-+BM;IYxZ_6g^(ioGhQ_1M0~-yLKng02Bwn>`_4I{%V1hjZIr*e)S9;&p--^ zIU1aU*j2A(vWGI5rj{=(z#&pvfZAKGAzcJwnCY8EdbR3OWI=URDad1hR}?C7=Uq8N zn~~Az^(&G{%Vew$dg5#e6;YO}*SG&64n!1%NQ?$6wx&icq_5rL=w#a5!vQDz64*|B zce~Xly$MD4F>C9vAW_$q0%pJKxhP}bVJ*qpnwsS#b>=)xLcQ1A2bn}k9Ni>j+7j$+ zi!FW7PgD>5;X8G3&}h~y*u17J*p{t&ezr%ht)<1lIY(8^V+7U_iuMa0thIpP>SuV% zS9kYhyw%iRFLWgW>4I6=#6W%s;w=e?>SIUUFt(5{wz*R|->3A~>YM1~D%*_S{@L zGxZ~p(l$E5!Rv5U>dLig`>~SMu~DCy3_$mLwL{ae9ALo|od(t0HhYr?>sABZNV*QJ zldXmd&7B)`PfD=qqEM!7wzRs0wz%ie+ow3L@OP%|3e7)zjSbkte!d%%WixHaKR7L+OL+1I zFx;AIs>yW$`d>Zt7RnZe`@EhV*-RLD28JaR_zs0Jh4Vj(9QNPHr)sxw6F+V1e4WgD z#87KY?44m5D|JS^^*@RmOZs7J44JJ4N2vRDl}7B7ThqRzGSDDtSXfoutwA}dyu5r2 zZfp70&AJEg>~F#i+VMRhd+*x-u!^ol_fKH5(>{Ir^g^9*^V^N@-{fumD!IsM47=}7 zkJ0fO`}M69eLM)^lQw;~N}mX^w#g@0bf@N;aqASwBf1kw>qni}-W%vkpkM1>Fd{@jMb>&-$)I>G$p8dU#B#54YnH2=RmI$#=hv^=jY$33AG` z2Or|;R+#%l#(gjyZvaZTFW;v=Vr1l5QkwnZfRa>LN_zD+U4eP1)%Bf;S+|BtS`Pyd z|Lzc(VaC(&*|G|4sag(ix5AJ1_IxD($va=ic1u8@!+8FO6r)+W`+l`bnq=ldS$;b= zIP~}1bpbbt;ZwP^;qNzhc2uJ#N$_AOCO&aUoEfM4I7X z5hC!H7}VzY_#o$3SF-?&W+~4C+9mKB5k$Mj!NHN#zQ0=kc-qO>+fA?-&b3w^-IrQ= zqYggUFh7_#L9T^!&D%F*hf5wFa8Hhs)yr+_!V%c`tI5gz7=(7Px!2+wKanW_N99mT zN;2q9`>ljX_qhGHpVZXa;o_OmGg0uYU2q@fW@obtYTVm~gej%$OIfb> zd40dVTpwrIdLogFj3M#qNL#wh`DsKH~5=cTCCBB{yA3Fz;3e)BiQDU0YD7b z(^PKK#qU0=8*wQn5c_=R%5XfZZg2x-}w6RcjmN^ zAhA<1MVfXF&^|kq(=`sVzh1A&`_mYq_z*MQyQv;Ke&tz*!6o(CQ;>;lP}bd`Jh8ZH`p`TBm*n5S&WJBE`iWgma`pra>4@~ECEA;J%rFC)o)@z()G4lMQI zo~B{3gN%Z*DKgeO)}hkeA7@d?Z<%~Wq9oM8t9eg(<}3szGVk(T!3H34?Z z;|8N5pVHlX}JTf7NEesHA1~nM7?}g=lMjrysoPiS}vf8dO*sLd`{E09yklP zzdmcM*Um#%YOWvkWGpB*NADopPhjr&$qjQUf8g>xO%+y0Ir)Rq_hf?1ZubO z`T_B~Ane##u;yD1ltkL>OZAdf_B^~n1Som0;5?|JCb@A-H>{_Q3lW@q+WZMP&_^`s zq@>hiBoRN5ezH2l{qyINDh6i0kmI^| z728kkTTU*YxZmjN2PkeTgccImCLFbv({k>cT2}h*j;o=AT zp-I}3tj~B7A*CV+ck~h4$;wAS5d@GM{4zFxu~m2S#BM?+Tzs#&%Yc@F!Fo+STaj%q z*=S7SM_{Nyg)@N(ZaZ|l!*3&QNoe7dNw{4~d2)Gv6=HoP`+iX|poL*XL0gIYwxAzM zu4)0G`7V(3wl`5y<+L}hebM1bGQ+0`xn}^C3%T^Lb)&o}VP;x*fe3Q8zoAqC1n-lKkY!}0u}<5W5j{fw^?&QtBKlUY{7-w%Cf ztR&k~R8%=%sFKNIr>L)P8+d#g|K4;Uh-lI_y?C&G9rSv;`Ksr4LsLbN@3?Z}IJVm2 zM9eVL2qrj0(hf8oG;^(fHnezxwhVe+b~F4~)5Yz-sG68%yk~wj#QtuzVYGm#`h~Vz zN`l6P{f?weUz`{n4HyN^YBD5>_8<9=<$_1`DWL0HJNh)g?9B9T1|Z#tWf@$#dsgkD z`f>FuZ)lDLU)8$6YIFRU>;+n_$7&ast}X$5^MZG`-r55pv2w9;WCyvWyoXna&^_U4 z--#sDVe0-({(uQBE&lLd`xa$>zrK0l4|Dy;j^c1v(Fd})e;%i=gP(H5v!dx(f2DGV zTDX)4?6T$9>ghd;Jubb9U1raUy<-)3{(UyprHY@Wz&-IRsKT;;*GF9a^J9&w=SPTQ z+W9d8by|O(Cf&)B$ta!GnbPnqha{j`X4=h)TOLdo26yw)`xR74M<@g2p@bi98E()} zqG+R#{GgOoA?oq3=(lWZ5fjx;?Vm+7O}y2Azn}D#o4x06-r2r-Hxi(c+i?t#xs91j zr@2J|hH}zuq(7_gzDor=)lEZGF~_CIq>3vq9f^-_%HXNBms{Wb;FvWvneUEN&>`2} z)Mx~vT!bQ0j5#VUT!NMPJj~$pIL;b1zYmX4q5l10HxF4TAI71Y5~XcDTzi-%?kSk3 z1CKK8NsR^ivv=GIzaDGAp^p00rCb)3XX7rOBj1ZQ@?Y&hPuy0Bo(OSHVcU7|x#L=b zG_flQ@~8}70uAq<2U|vCTdZ)v!ASl6E|A6h#R%^2Ta5-45{{0)K@sGiHP!lxoycx?C=S(8cm=jvbHH1jV%YHGc^5L;l_H|E# zk5U~du}0n169jJ?#db!-H}@P1Iq)|{LTKvNf_e97|M3!u{D^6H@gR3SH*?8^W9qaT3a}W zLIYhwl)6bh{*ybfiQt92*B%Oc>_wBc9?>5%| zJ%8RaT>ErEIt$p=JN$B=CPY!@iF7i-3K4HQ)d^i)j|-Q)Babh3Cf%zo`5+Nzt0^k9Ty@HCqpCHI&_{jhd=7jKNQ5#cv^2fA;L0G zy1r5dSK+7UhszvsKNTMG23=-TG@Tz!VJ0=~<7*Z0wM4RUOmRwD{K4LqV%X~FTw7e1 z*_`(=Y_)wedC{xuG&;z%jDQ_dPPInqJ7L;yd2Y`|#T&n}e&Y-M-@ZfMfO#vDrZQ?G zfUA`ep7aDRs+_I%!K#04uh?mA!ZXik`@AJ?_Xn6gU|zl(vK}5^*WqD4jhpziceMZi`dbfujU$^wm)pRsHY0{jme~?{xnT(q zRlB9VWwL3%JM`H}inA}`Lb8Z+ciFh}eCOKso_Mvh9z0jbqNgTDHChyI+8QA_Sdo~$ zW}S?Cj{Y5FQT=3m+(dPJE|ooWRlcK%f(GX`th%XJZXUThTGsy-yyX*z{#;dV`7owv zu+-_(d)S|%VXQ`$ZaV{4x@UUGke z_c5EV4wHCg0(P_q>?X};HL5i{M*+`V>x)~iX_uyc**p)6W^UJ43+Dj~6- zggew|Fji)0m^Aph{`aQ+>%3wM_)1=7%JO<}a|v6p!d`l3jdJz)oaE?`@VR+0JgA+) zcPf&t<1n_|qG!|uVt1$;`f4`a{l6zxe zFCUgth8`_9j~V$a<3@Pxp@lTpI=)W&usr{_Ft#*UT8uI#zgavV(RFyBR~FInL^1LA z40t`Rx8sWXFEF6GdF1w=$|q$QUjTQV^4h;27mgprzK*v>6}Qc>$Wtb6#qadR??1j) z15ZWVkn+Go`}b`C*aAHFwiA9r#4~+wKoD#M zPD@*6ZK*s3lwBIM^J685k0@5!q7}sczSWdh@`Hobuo2Tsw-b!cNwCYSyRlw~fur5>6>Zem{X-51|)o{SqFi zxWBVxx=`o|cE^nI`l1W<3=q3kxHW1O(;S*qY}@I%x{q7A$MWy{NsEI(2!RBK9Ln#{ zcE~uYk*(nwj(?KITXYXukU^4wHAek@ADKsz5vzOcV?EJ5E)^|)dpLKLNv!pRtk9o` z@Tc?fF~{-oGS^zT44C3Rm9hzfY^#J}TQNIgR5xd8c#dMn-BzOjpQK}krG?dV%T%Mn z;fSRgA~5#Esba(Gm!Z>q0}nhu&M^(#Ivnd|kDTCDgfR+AQYwspKRz43rKt#~7gNu9 znzpn#Qs^0M_G=D3hl__sY?&!FkColBdjj~cqq_vS?p)PrRETq@O;Sa1~MV_`{_(Oxk%sBC+n6OM^v& z1gv7tQ~LQeF@|TA-uYUbCf*nSo1H%|g3DMuTV4>u8AVKCZz{s@8dKYZ;KL^O_k94$ zoEn6J8heNF^fG*yBn9p^?Csmw%L7r_iX1!!5;L14C9v*jk6Ud-CxiDt-CU3hw*7Lq z%kIC$#)v%fsxEjDdI!oxrH1K-Mo^~cTpn!T=0z9mzAG{1p2DTEi2EVJHbE6H3>z!& zL`m9S??xZ_X>G{zn6IAO621G;HXa{ZGU9ewg?Q4dDE9d=timOZ$O0$jVziYDv>4iO zQP7rE_x%+AFd?}0BTuwI?$0bh;}}^t6e(|6BsN1f%Tfg6KJ+1i!);^k(byiRDqZ^3 zq!b0#cb$i<^=QeZi2$q5EQ*ql)}^v#w2Tg{_b4z4KFa zN?+f&@!*Q|uynlpJvt2zo1aN4jL-EpR0C-ucDvi!*LRXFhVQ@PsAR)KsNUOI4;L%* zW3);GoJGd?d7q1Aotveul%;x4u9ACi?4#RXzezjJi)Db8VyFDDNzXMuIxVzXHPA4N zEVy6CR40p+HH|p^cY~;YA&xV=;xoM2K$@yBtL?#rr#+abuAt-n#3vX3-!e{h0S2a- zH$0zqm9XrZYV7rE+=Mb!JI)54B|Q&G z7V!o(Nigco`uQ5h(KM5=cvi)jJ6GfXI}=h}NMnsX&gJ;)=T-w>sidR*ci@BIKAg426P#BSz5$;!XPVDpkqr=O;+~b6 zD$IB9!?Pwe!xfgn&~)AV?=@7LN?-{SJI^S|6%=+Odaf->@y&rJtMc+-De}pGPkZgL zSb6$?+3jdA$+s2!_XH}3Bydvv@BjaY&GElKQXRi#$$y!}?vtQzPG>6OnWzLg<_eD? zeMh{qY2Z(eTowHVIzrx9y0Vqy2;1Yt)9}_Gn`-zC7 zdrnGwf^3;{aQC$-?pn9?a#UA}P;Jaz@?V?=6<#fMBSEk;*3xYC+*#KwHaEE`WnM+WJtZx z!!SdjqI#U_7ql#b0$Vk!ZuhE2&Evk~p7pdL+`7f>Y2^z+BFg&}mRWu&%P&+w2M}RB z#EbG95^fJQa-QC53ceBrrpAeivyll9-fEY7h7Jfa1V?nZn`_v%}79ALMNBuK@k2P7Z^nNGm?3~xy z#NL@o-&9dw9l4vRp8!Hf}IaGr;rdfyqK0 zKNMn{IN1NyE|fpF52e@eBKU@go1sQluZgt`aqY}w(}hq4%B2S~kulsRNNa#g{nRau zyAY2{g@KM&Zia2ioc=M~n%|2{XSeFOUKnMd*~1vcUO=w!8G>eX!EZfG_d`YnO@@yy z^%lp*a2d1eqgM{V?+ay;a#VcTpNb^TZ}geSbd}qaZy5|HL3U!q)_5(nsvKrM`wq8G zh@YZ@1Jg47=4O1xTg#^4t5Itysj=W#)|`h54c__gIT8dN36*ufZn!+PH~N<=@9riB z{Ro+g5Eag6eJ@P}78kSOx~8%h13J|nOJysL*lD^^Y2uqlLAGKORi3piv(-+f(eY3# zj%=l?(eIuEaqt!fy?7*J}x$vmb#)udAD{(r%m$RSb2P zn1Y*(zdnk!9=45&O!G5)qv8(6QoJ3X1UER{eQ@4xzH)9$Ir*OUa54JsDH>*+$J$iw zvQiYQ!a;YfGpgPJf&E&Hskr?89WyLeLG$qI^9mNrDAuiwW6D{s9YF^tU0Hn<+I_dz!Q}3t{wa9*fOc z$wS@8`Uv&442ml)%u#b~(E*8st$FSlyIN0Y%vqt|!PHh82CbitHVVzB&E7s^TR1E4 z(#fE?z@yKunbTFF@+)Hu;_0N+YBP{u-Gg@(P4M09qS+#LR#!IEEGS1#40rq>ps@s` ziM6piB~HG+u@*EOY}@)8HRqNtW51H_IP=nmkDBTx(}1*H8Y0lZK~r^sTlX$Es>BGv zK=1W%()k-dy}*}v)9x4nl{z~^m&FbBrM(4jw{0%v#6bWkBRy zyQiD?kY0>eUv9YZhu{uL0AKHFr4_sTv>XLD2{f0q25LTxl zP@LW^4#DkaHceO>8}Oc#Vhf%w{VD^k`@*}cU-WGk`T5JW4;7$R$I3yV5Jyj5Z+9)T z`*C$icv^y9@=JWr`BSCB_D?D8?oM#0IPZg)Yjo?xqjFhmor2ZvYQgLScdNsf6XAkp zV^BOohoK`45}(3nb>cNMEP)p!6UxG3)Ften_7U+lX_joqkj+C+t-H)o%O}S~^vCyN zCH;%I3QEqijHx3~N$aDo1$IPjuDEjRG2piu6ILY4tapg{dOyj#Bs<$#Ct%tAHrvgf z(ahbEO9^AtcKu?RQeQ4fA6L}#G%f#{Zr$*cyN4CjJteD6oZ#d zSEO=#43}NPfS0OC+m^PIPcNq@p$p}6GJ`VF^y_T0jT)?K7ur%Dm6>(jrsEA{B`pz< zI!hErCA@_1ht&S}djAS+M?4RwNVx^PJzzicjYE<7$gf84UxRgPzwn6L7eeuCV$(uI zaKP9lEMaD`>;dKpG^?btEQORYCMmaHS#?wP?@vSmk?h?`Q3(bPV+il{Z5+Yv{Aaor zugmIuJ28?u5JtG+-eL*6A{lZ>mw{?%uw==5R$swtn%Fq7Q@kH%lq#yN=;=OZp?>KV z8E6T4T8rs|uB*A1$yG#Yk=^CSLa$*JK$pzaOK8h!4cN{@qmBv71r4o}W7fcyO>@td z^|UrPkrwvL4YkidoZJ}9;PV1CMmx~)+v=^)y$YCGi3B!W&n^0)_hG|8Nf38$`j2aM z=NUO4_D)y>1X@c z7ThCAAE0!O>iylWk-_cz=PJ+9Z`2grC_71e>g=J~a3PwNIh>wjytb7Y;r zD-h*;{9Me+o)aQ|tXeQcICwks@DcXLqaI? zrK7(*N|IeMUGWVoAsV9*^3VFyr|6cpt#Z}2SMV9*AEuz!1UH0rTSy(c#jW4mrD6bN za?B0i7~z-EU0`B8^(kxU(paV4i$^c2V2DNGI6pp-SpKWz@_sIl-EE)oxEDHh8#!p( zG{4QF6YVR*HIKYp?A8;Z{?#P}Z&X;RCVAHY=yI+DtXIWJ*Lx8K@#*8W=tcZn?NR3t@{5l?5X*cYBI!z=!S?yKND1rYPbho<>N zczJ}S@as)^IDgr>{-CoxDY;<-swWS1S(U^xejXi@@S-V0@5_D`%SL1{heQ^a&{Q;_ z>_%+JuI|od&qM0Q2u2;P`v*zuKNy1riu9KC3t$9gn(K0LmQd%(2BvZ)b~5y7Fl(}>qcjy z7DJo99w#zggwLhL3clLStyHF=7tz-#8ojV%y(rt<)6f+GIR$%8958q!8HGOlWJV7? z#GmW>%7CB2RN60gHI{H3n3`NY@sMz9676ctnAR5qmf?hAzXhwMvUzc6@)AK8|H z?l{fXTP)*tT-E>Rxs{=v1nsOJhT?>$N<;&(?&Jsb6Ap?EB$07HDUjKazn>OYzB*D5 zFA`$Zbvpb!zv(q(=|CzDboDWZy^=pkDqJ4fjE*xp3eK53_4Az~XRzxQsWvkL3t!FQ zD|>M#Usecq?uQzfI4aC?NI-g>-`&g89CZ!w=6@n&nk04Y()&JdPjwM{!^{*3_@m;J zqy8C$7Mx$Qoqr9FXro>Bc)nfva*5uVyYH{jEreQnI?YG0yAKCD;Nb83iOF87dIjdy zr`ltA!%yZu%1L6(-fm`P3^F?ATQAzbEZlgRhv_Rw=or)G-$U5Z)%bq7U&OTeOlr)N z0OPBK`|2ZvCZ~RVuPeRy*KpXCDaNXu>jk)38cGDeO4e^=;j&Ct=Jw21jMWg8B3HDO zYIDy^>~HuMWAAZcLM;KW@GPyVu9x8eRh?#v(@L08Mvh}#gE7{MsnKp|xz8M5bA5dm zMBWdC?7#HTuU&ahD|+7)qr2ac1@%Mp`vZbUg|s&UGKFWbL4|zaEjbq+|Ks)P*Fu@v1^=j{9L{B~) z{S?w8HVm^m;b2(#QaOYVjeXjq;Br?EvLFkb5zv&q3ysBZ3Hj`0ec=|$T|peB<|xVPgMbyC_IwNaN>bkC%4r)cZ@X_7au z=mV7Y^^J|8Tksdi2s|1r5vp838a)`# zE!j4i4+$lOhtJA^qn6UgE9XTZ!$1x)tJ2X9R_q@tmY{DMBljd{M}y7Y@&f^T!_SR` z+W}j$9j-y;K$)T;qmipZySyid2rIfiKA>W*4tnTCFYBDTjqRLQK`@8HxS%EzD>Z#w z-1XgOURoPN?=UjUhwknwoe8m0=(|cYX>uwX)F}0#M{5fnnKOBR>dZ_r{{Nl^$A38~V>? zqG1O8Im(v>NptAC?+Z*itV&q1ve~2T(K!cp^U^-2@e;Rxjlow09s9M!q=)nLhP;$5 zF_HA;3mp(NiIuiJ)@hQJCOFJDkB&wZ{{er#BY~x06$tb39LRJJ@<1sQn{N%*%n}d?S;!nM9?R6Z~c9d#SXW*-SbxW_> z3+Z|zs-~1!Mx7QRccFnM*DFn zKDd|xt5#*L`g6DoJJ*>|UiNCKVR%%nK|wq-drl`-{L z*2#vSe!+MQ9%?&wEa-0JfQZh|;kx<@ufqNB@#pGQiR$!yb@!_Fy10Ce$4n_cBHljM z+Z|zoVp@^OlBu8Z_o_ZRDH@LVSn7qlw;r)2OsOTj)ZfSC#nIFrLQ4%fiJu2-p7G45 z%ZOv1OWMx}<{4&$oVYgGGT6I$q};sAX0%IR)S%^LXo*S1`c!$3osDJA*OLN_UV`PX zmZ+Ed=0)7OCSgP$f+i6xu%RpM;=UHYdJRC1W^yNR-1oSa8bSM4OWTeS%I;^ks?RfEXGEBCaYR;dXw!@-p;(M;Y`6J z0z{C#lsGIwpV{}_`CeYMdXoO2C~3B@-p-C=OlBm>ZNK-{<)ceG*U+S$6=r;`8?N2Y zUNpd0yz|C4w`&dIL2oKwWN&3ZN?|#rSnzczTnxzsJf8uBRhb$wcR8qTI&{0opTGEv z24g^W3_Ngog?U%IUTDph2&axF(HJ_|y(lDFG*K)5XBN zgjp1smI!Zgc>-@qLfe~>k=6G0hhdCs<3*K`CEN4QF`-KLS|Miotq(w3_Mx*lpXcLh ztlffYD z|KaY;!=e7)e_x3fOZ90HLM2JIB4ihpoyxvfc4NslvR5iac1CuxZzIdt$0#xw`!)1zM+rCWceU_7TN*mH!*jL{0TE?>LD%|uPRJ0_eZHI}KT}ANEVX+K5PS=3 zbxeOp(pjCuaCa#A4|5tM%Svl`?lduwbU3lh2$sBXc7$?cfS*f`Emk=LCJ~s|_dG!R zko@J;w^`w$)%Jy`GxsFS(RKqK=Qi7QzdihH_jj2Q8vmDgDt^{7OF^FV4&dlDxmvFr zedUlA7Y53t${ruuz|}gq4T(@-Gx2)*)`(a_zVAr9$N7-Y!vPQ5RjJe5OBE-`jhJJk z-sXSsNhJ*>FQ%T%!j_UAAk9XEBh-@&8AuSfL=GoU2g^+UjJTaeQm)Mq(Dg>-{2 z3jzgsjI_V!jE+7CRXmC46J-4@M@uvM-rviYr+H!RWTg=D$k6GFHUipg)D4C=N-5XM zesZ##%W;oaC#mE|_5jT=jc|romESCRbK;lzywQ*Ti1*h&oL1K@ll3NqvtG=dcf&o< z4!g>^JIYmSXyoy5G_w5e+Lc~^uNu?sd>gcryFZ3WJ)xubsdvY9WKll$9PZi^xjQ!v z^z--m0Z*jXt4^vc<%DE8@%7~vXd(NaG2DVRaUkxjLn5`Fio3sS;yIbi+5|uG5`y1W7rN-g zRNa0D7_&spYgWFyTAbfzyK?^qHx9A>6(1C^fqe{YTvh9Yl9?#)+ZQZB2*G zI_rL^>ZpQ_w2w&Y^*1)eJo*hJn>rX4vRyyrj{W-HsAXNP)>H?^yy#7t_ArfS-t?!Y z2P?=IBOL!^8-=n_c6*VLU>Ui{zFPfG!mM~V1|f(W@TGH#eat_%zpT(!7;F1D%r|yb zvJ}{4gZ`H8L)+&m`3kE@22!v5fu{LfFSn=aCMCI_W3t)bOXdWch&>SUC4E^dQ*ipZ z^fp)8XI4%j#%E=BAB^s%;tthfvJNso|Ckvm(c;f%y~>yKxt)9bsfl%4v*zUC_J%i?&El^uy{rQ zXiYJrw)QpGpsfAjKJPV$mR>z&T{1Z1g9N>l7bUyds?Q@iXh_47myST^FrPQ?^*ii2 zzRk^k66rfKKcZ`tK7dSF0DS|(e369hBo$@yPp25DUC})CeVa`I&py)X-yK3CQ|S-Y zq+;vWn`>oKuiUG~(xeqTQw0jXT$wZvzqF-|8+?<;=33I4sk3hb8-_m}`B>)jI&f~a zrjM9`9iX>KeE#hF>}qR$^CuD5q=b!c)harErMX{Mns@@X)f0|?7Fw?HcKv8o%k%Dh zx!xV$ET=)-r{T++ZulA$JwFw?g0?HWm(@D5gbd*RyZQOjnYs+>cH7bFSX$7>j&810 zx_y%>>Ey_(9RC{e&-WU#5b{K)zMi~rw4%5B``YZ7PWQHhCdC^~a$Kk0Y~hUOd(){q zO)!g-l6_@7UmRMK1;oeMdYjCC<~`jv3oeK_Gg}p=hBjXnZT%)^wD#?TNIzWQ$SJ}8 z6f{ILgd^v2nt#XGp^e|a7nPwlp)yc0|Oqxg= zA2X4Dn{*5{wQxtj8MIm+`ftA0klkDm2#e;4-{h)M|FxdSe{Dcl1V|gMA-<%s(Jkcj zJZ+9jZY#Fmm$ToVw6?REtwL}~P4vJ^jLSQ{`l_qLUC(7tq@viWf~-iwBsruTmW7#J zthUSe$eXV-fqj+@-u#mr2MG@{m<2B2oJO%S@9n3YpIe4G55G8x*o_FtvdqL5tC+2| z%NU=e1WJi-;N7qc2Q__TDd71@5=iPpBU}Qhy(GK7=Fm$CsgmxoYFSfGyz?1v%h$g9 zOpI)(k%!U4*9r-IWJ1qLtQ1$R8tdCSZZq8^q`tG~c}c$R0nI zAYjZEeg9*-Gtktp^9ITqHD|2G&-J8ZX7PsM<`a*uQahTI99(*cn2;&4vkr3@uJybe z0uypE+!}Q9rp0cHFfZvy+t05(7k5Km?9Rjn6YEZ!9hjd{ZVrCegZmWWE$^l9gYq}< znxkOycKae@iNG0r^sAeWjT z@LP-yho>W|l=iE{ZIdUR`r0p+s3>N_WS*IcKL)U>pY7#m670vDMiHTD@_uGkMCrCJ zVk0JAE-WZ}Wy)6>^!?_1GrhKSL?ViareT?&x%RSORhbC?k<)Omqays#pWvzbJVV#-Gj7a#7%~AEh4MS*$}^mbUA>HQ zK8~|q%WYXvVMnjBW-Cd zy|na=viF+`?1rGjwp={;Cn#Ev?9h-Yblj`#T(aN!@Rr`zK=(|b>8oU8izQbI7dG~~ z_i>o%idH9HvQ&6?6PTd%MP_}G{rociRk^WMTsT2K?4{4VFty^O1*Uzpse;=!#B22g z>{FI%2J7->C0X<6(PBf9xw_Kii4?%#gDOb@LqP{TGJU3`;7}4h?=bvqrTGuX%CrKf zdDqvwL8W~}=C@n-)qyfSN&?kGXEaLNH&J-N;|QQy2t|7gP0ed0NkuY_HScYyVe4d# z%Pp&{^uTcpqbv^MwpY$s!7KSiA_7|fx*7!!?`w6){1_7F!2eKbW9>DW#XbF*8XRj- z3}!J4lY{QGi*Iq`RE^LjAOd6Lqf^eI&UuG-ZOs@S8p~8EaJU;LO!M{0o#qp7-UZ+6 zK2fvw@UFF9HHyfzx|=8jOB^nG)(JRfu2+2Ke?CID>`@}tj7F5jltDM)l6cSu@vRW< z2)jA7wZHd`e(n7=CGE9jLb)@x@VY{PclEqAS-Q(Hh-h*^IUlXQVV{l8_yQxv%w9Xv zc5A*7Ld^&Yj{eb$jDAm1VCSBG;orPu`4{IT5rQ@J*Bd`I_S)JQEb#X~okF>T;Ukni z!}5!h1(77b^Z%3`6g5S&nnjBAlxcg1OcdbbeTt(RjI$%bE8f?@PrPew&&O5})6dR*Q1sxj6M-lFW%kO$aDZ0=LJd zjdW+iGV?WH!pU>}Gd7;HyFP+XOJ5bPrTApbXF56kkmP25iQ}!}H?QgToa^uqN-E>h z34@kd#a@gI0 z-YdIy3ZMtG^_o_|TcvwJ2MW#?qH21L40oqToA1y``5k;TWxBZU>s$cH`@Y zQu}N11;Q~_#ShO3HU1F+D&PV%MMnQI7gaUj=HlhYX4j3KwtQ!gbm=*_Gz6uh=R%)a zW2j0x@z(kKb^`Xj2f4_vR@S)W_h}v0Ya) zg+KlxJqH6WS^S5k2?CjxHm-qx=A7z*L27#Q*pat`Z2C%j@>5Y7{(+caqRIT&cc&6= zFBjap-OGKj zNX5>F7@3-PSgQGCBHKQ0p+~gGHe|ix1fe1f*-nahiQ@>`&yTt7apla>tM*REVxR}Z z0iVnwpP;W%**F~sRaJH~BAiP`U8<-3`?Sbve%262Lw(Zmv24B@BbYaC-Y6M$qXY!!pR#UXUD9i9 zkW^71yJRMIB|?4fM2Gv?v;tunQz#~2vbc)1UOBsTe=))cqM35zbGjZec-iu=MHt73 zmz!a|$~dS8T>7ul-u&Rji4*#g%<(Hj3d0ANl3k63 zdJ`WD1Ttw`ym4S_?GQQ)u+p49{na00Ql8&5Kw=a^7ZSw0d00<7cdl3|$eP)^7#xSF zu`KaR78@nBX~dc6Jw|erVhRaulD`X3=co&BXV=tLsyrpk+8m>{j7{W8Nv`i$YZEx3@V53Wp8^JAbyAgGc(J z7{`6C?4C%Gij$Q*zzkyXVD9{)Jt}?ko5BO1rmtw81NUXoBcDDE@<)vOec&Ohh3urO z>3Jca58RgRy*q%t{2NeEl8QpQPmky;RhDyQF07f#=@bvxpf+v^<+JN%2Fd(r=0kXW zE$1@9Wou3JBRwMq$xQoEtB+93)eHq22M6u#jh2oA4_sAAKUQ?*mJG`bdJhgfC(jt^ zbY)Ncb+8V0m@e~o%bFLjmEc}io9O4*MSE;_+%=B&cgH<_P{QeC?wrS8)^TxqQRj?6 z2V6Rr3>m!h+i&x4uR(3sLpEeRw(f;G`U&Y3X)5EqfDWRTjqg$r&L}q~vdO9S>vm1+q%5ta8FeM! z-;t)7$3JOe2nU5rc7N#$?uMte50Feg<(~ico;3S-nosV-{CRxj+Z%ag3$JxDRi@CV@>6%}9MzUQWT+a>~7^ z6e`kjN)0*ap+cL>xV@XIX2I7TW2q!6#QN@TeGeeMwesDT6D3OJT@2ngTq-jsvS}pV z&hyFTIouZmO6IQx^38qckZDsC%m@Cd*w1Eghnd~NP{(4%FjA%DeWI!jx=iE7g1@@k|s%p!>i;?x-VFU50@ddr=8Iv+18>7}W z9mx{&AiY){H`2Gw&UqEfkdE)z%$F-?y5$9&qHh(g?|luKbL@!ZRXEC49;jlmVKFbg z{E#SKh!xPqRw)EmWK*U>79_Q%B1+lH;z~wa#*hgaogyp2H@o4*Uc* z0({zFD8fj&KTc*QU74t3IK@-xfO$FbmdE1E75SIarTdX*{#gzZH$H#+z+6{j zvp-G!1y*>#+9#MYn?U*Wk0x2f#hOb`YLX;R+ zT#Yx{=(qX8;|^>-Lk{D`Mtx8O84oSvO;k&f{9K8Rbf;V=A%azfTP@^D7k|YEpKxk% z1!Gc11znz?c;L?WR;(6q6U~ig@?!=Wew81H9^6Z|>j_H2TIzC^&XPA2`X-RnH+Eon z#3yLYP_dw$*2Sc4x5Kxpl`ngku!j!4R?i8;)k7a^3UbT{>)Bn0NSwoW!?A&NyrtFa z_St&i&r`3>a5ootra&G2a-!$LpwwigYXRfm-KJ5QtkQRNx(sz%Nr%#teLsI;%GTc+ znbu)QD$gWn$UY4`=X$KLX>*tC^j0t;9{sD(Xl&5c7;S^yxH?tf!#m2x)}T|IBv+Ft z$51GK!FtOEnc-}Jf=xS>7?+(beuh@dL3h%9U1+X?p9!k&Pxi?pM@OCX`>2$U(2p#O zrOvTlUf+SIwu|q~YaX>(t}?gPb)x%teHnDLxm<$}H3Xz%96V%d(L=99&uKpCE-#y> z0zAFGp-Hwy^T8#tXB}=90(vlQHf|Yb+IrE@HN$L5fI2Gr$dS9mmh4vOo>Iunx490O zA?Ikx3;t9|8@-a%MZYz!DQlL*->(`M@Zd3+#*&*|#w*_&G|2e0A6xZjE1~Q`x8;z5 zOrNUL)zxB)A z$KMFRG7{%H6-tiLtp!Zcm+eB5gss|oXDL^pTTT2>xBSeYxyGPCBZu%ZKN}o)b4&$1 z3Rox!P9fvdX0Fl$Wwb_+wSg@esil{)v~N+|PMf#5Dh#jMPuoekRj_u3yWE^S?T)K#-Q?KxP!J{Fok(5b(>?YXdVQ$t zfC6fzhMcy*u(($7}3gHvb%~*eJ;eM$JR3YHq z(*NfeGJAxa9y_C3Fp)#;mpeE{0_tA;3~$hjk@&!Y-rzoj-(-%&LzQW!L?grXm3m&I ztf@t9x(Os(yb9q}Y*aKDu2GATDs1reAH7}2_>Utts>Gdjzk_V@QI;|e{6SSwrw?Rt z$BukU%8+2;kiivK*A2aCH2Qq%Mu2ekAj^}UJulf?{zGqmYxN;urKWqvyD6))<*;kZ z5Eg0%c(6zOM2i1L3f8y-PwMz0P4_bjI zhJyecqEqDN%`UU8`Jrk1**Jqx% zt1WkR@$N661($CEySn$+iIgr6}b!v~>G86o&RF0knxXdap z6aTDrOTpizbG^x};!#JB)LA;&#r?t<;qF}h{0~3u*v`fnqqE6$v<+IXQhNq;t;0AE zWyQ<6tex`Cu^P$RRL*14^}(DfTLoHcL(5>qWu(VU#s{P zy-uk9&1;Ma_Zlq%UWLu$8@b*Wzsn*>Z5}hPf^!WQt#~Q+d>?TkPX%msZW5($fX?~o z&b9tQ&ozR@cI^>2WoPs;hdKtT@Ffw}0@YKJt`=g$@|?dX8-fZLe)|i(t8y9?@-&@23N}sgpFk%1Q#4p4$S${7II)|Gk+n4 zJ5|o1V+U65(lQa55dpVq94J2KG_Qy3EKly14U9D^Km9)V(21X2fp>_ku+K9B`J~+Y z3#@L7tnnF~pOn07rppX9Owq54p7E8)vrneV`&1PTJ%JPF-i{b2OMa=@U(G~qVm;N zW!%~GT(xF-3T2|D6S|4vgiz95aKZO1t|vQ9;(}PMK zqTUagdDn*>!LoRBP!Ktuvj6?_4G0YAM2a7~yL^DgX#vlZt#Ub>zLQpZ@7%5rB!t`c z!5;B6)mF&Ap0*60_bbtp-tnA_u5=tfUH^J5&f8_$j9$Pq&VWtzwiml_Jg_`yg4SWK za`U8T$k5|Kz`lM`?z=whwDTvG*LF=whyZNgaUxpwPEc!`B7^2?@1+6jl~QXhtKgz^ z$gbefLEK&I5HYZ`CxR422i=Pf{3@EjXE8UwmgAD5sutDNrlD^)j)Kf}k9)=4e|lZe z;ueY?z#A?x?3vy7UB*F%;8WjCdo$$3ik{`ifXF%szl}lEaNs=cOZc75w{@7S5!W#= zDgcu<25$5;FvIDBrdP(Y?>WSoFafp|wOD6G~0YIC#(fp$~sk<~C|xtBc$p<$$m zhC{By_|pW<&JiJF08VBGTm^y7K79xEP+3W!H3Vk=?Ti&EVF3nWdeSP6U&}rkFzNIz zfu}^W%G)nHGC{=hrRR^=IOc0J2>5llS*2~6Qn>0)}Ki|c9QPu%ZXN+Khy~6_QPV2hnlA?m0cSdZoB*G!k*>Vo9in{AuD zGpLR)rp&oIC2Dg6VJn+5JwT+Y@){eBR=OdQIxXfAi?U;x+n@*EK3Sp;C5=H!gT207 zLpAz_WJvUMujBb62X`LJy&M7l;>X2Cw*@-mZUPln-?Vevo+?%2F*M(8{l_a~B|S3{ zbw=0v^s9}4u@!pd{%eyJR6%TFgB?ulA~5 zM(U6g0?JGZzZmyRo@c0{Kbl*~>8(3?KEXa#FKsV%c@lw^M8fL3GsSA>`cma6G;HCY)04Dk zWtxmaUSu~+P5KrsY5Jg&H>6q~Ik>!gN@?#lJy<8-Ptg^$u6ciB!zq`_27WV*LJgl1+?;Nrv>$MV?c6+JzPER{tZv zs8u8N;N^|vmKQRUgOE-?84yhD~+2 z?TYKS*8be)WtMnT=Qb?}WY03mv+^rV8L$Lnfb*vyEd&CUR?L{n39NpknD{6G-jF9e z&4+HM9%fj`lOD5{72wjgV4l-e*9;fY@v2?6e~2<9U?R3#B8OnWV6x>ydj4iGOV`J5?(NMuz5pX; zyzkOLpX7AHq67*3v|fdSqn~J(Ih#q8SqR|$Qq$s7tJ|ju_KhG>430H55Buf??8EwI zAvI8R9@J!h$6F9Ni$L#AQsZsLx9(dsdplygdgD>M(72L+ zVE9kPK!out3WCVZPXk@<3&^#DF_&`%>ATYJTF4|G|8P&18qcyj;6BBhz z)4fizP+v7=Dxd&zO91@fzoA^wft0b&m0)0l6CkABbH5~RP*UEq6*0AImbE=`w^daE z^}N$FS)Oo)zW){9umrE{J0Ij7-DnZb(qj}avNqN&ruET=25+^(BV`c<-^wSaI!SF$Bw&Wm6UEV-9!G`%ECWQwO6F5I z7e?QdF@2jCF|!|Xwr|2eAJ}-Bcvp2dX6B3AKK-=mw)1wI>}H$20+d1itwt--5;)t{ zCHeEDWAL<9FLPa1XJVz}suJL#KzS6vq>8-5EGhw@8P2)<_^oJdo8++5B{_JQ9(&0a zn^g~W?#HGzJ}+vLq}ViSj3DRbDxFD>xq>yZVHxJ-CP>9$;*#t#ZA*Oz1W{^wDG0#0 zc|kk`ky9g6XCTJ&O^Sk#SUHfr1JO}xj=erYE?m%rdO4mL0}h8yFji+dMNqA#6grhP zQWSjP`VpjV)W1?ka-2<#O3eg7!BTVN_23PSWt>hqJKPF+6X4O}R**~}i+e4x?WKxA z4fC6^2lN9-Ei*EHV_r-%?{pquX$XmX^&Z3A7!;EsRa`6to_K^<#|z$|Yv8_q?II?r z5=#r{RcWj*8W-}FE+(cldxYXfg~%kTV(gYHPkh|32$LPEQ%G-x*zog2G5x_~JG&pY z%}~Asa7HGoFm9MKCBUKqT%aHbF5Q57&j0dp$Yy#XdU)-(+s(??YEtvnQcg z*sS5_ftx}3N3<8G@aWZQSXIbMG7PqpK4&EOGZ8Xs-V(5hi(JQ4qxS4SychV|$39Nj z@g>L>8DxL)UhHolO&?s|j?E0@2a}THtQf*f7x(CCP;iq%%u9ACzUl}Yz4meuRg!0f z=-S$DU`;J!fHRGO@X9y3+sN?O((s}LjIh!~kF>v%8}ypjdlm>uw?$Mzz~xy!1FKsG z2IeAA$P)K`1>ac12F*nA5Arn1tfxb$8NACdwZ^5^>osd&2(49Y@L&<0<6|mruPvrm z{knw*d$rLEU(T_b8F<$UWHeoVkl!L3HM}F1{^ur5KUKTxL*(r(3g@erBkYS#Xkrk= z$onE38v_gODhh`Nn6TW}11j)C$c`>+A>?^rN5=rqkd&*%bmi|*jtG5J%x1m?h(1mv zP5(ZZ6R4YX|7k~0EAX(4v&!wS!9aStQIMni-^20>D`gj8KOQUs_k|PnPf4X?Cu(r> zmxD#}pA^&8jbV;0>JV_-i10L_ZLE48)}fqoZkbDRc8FTnbV%q7hzY*-H3oi$F5)8q|n2+hfhmURupZiw+Ek9(j4mzuo)k?K@4lIuH2xSXgvyXx^W%T}vSlkAw@I;UbH*uHZ z*2;Iz^W+(#)4lMESk>Tdrv~8zhYx9TzCXHD6}_!(G!kxAFye;tm^pl>vS+Io#J8DS z96%9<52-EJJ`}nkWR-Ny4dkfQw;CZy+|p8dm>+l-hjkUFs+M25ur zBTD>^zeljpvRlHZXEK!rtolj!HUh>4eIB(E(aY)ZS%^=^QMhL1u%$ENIs1Xx-K+)GbTzxXS? zqS8#d<;-ni<2*%wl08g6c;*SO2JE$Im{z&P8wgwDeB6+)7Ey{?2)X$K+~;~siTg(y z4`AX%*4TMCKAR8&Y&Do`;~Y5VW~``>!WO`Be>YlL&duB%MXfL>GBMJ;Amk#w!p5uM zmM6FwVT5*Q%JJM8vw8@=U0V#j@CLqCyAa5CSx|9ZO3+EWaYN!ye|C;1Vac%mFO=`^ zIsvmKXBR$qwy$`NEMO&7PvH!fi$^lHAN=S3{+F-CCh&rdkUlf*s<7Z`<|wk=(| zQLE_X{W+YVaYc8$?RnSjrQH>pzrzlp7Uh!m5c#w~saG8DWgO(RKc;$o5)=%t%mF8* zg7p_=!(Bo+b@G&(7|k~_O)2ET`rl5$U*UVw?-ObDsGA9dj+sV@(ni4Mcr;O2sG%R6 z1g_cuk9b#AwxlxghEvC8>p%{g@Fx$$gAE$E)_`LK2+01ZiP5*AYPM^Y{j;r*e6_?LGP!S6xK%jdC%a%c1`4gMD)IaP>0)z>HkG4H{{F>~nmj zABN2o0S!fL=%EgrRPO^DtM`8xJ!eZOW)DLrB4A}PQGJLSE1>Vr7>XG zhV8~EH;nM4;PJ&nOg;4tV{DH3>rAoyBmoms35PhGWO0X9p8`Ka0MaG8C#AN--Bbg7 z8?2-sVSUl&?+mzLUM{en~Oxd8(cE#u(yYY&Eatoxtj<`dY zqw2W-iSfq+`?6{PtSbGXt9Pu;vOGa%ytg{A2tCNXZsGNVfPJQW%sj{NAwy54!0uOar?zvJkFab|lIkS^6bQP=Zgs#+9F@nnuo3StTK9LPM z7Y(%!sxmhA^KEmW8^w2)OUQw`#@QA{j+U?l!Dk&CuI61QGNTYgxqS~~6jYM=3#4E1 z$%oT(@nO7m=E1w0dYD2iZx|FGytRn7?H>Dl7Z0vNZ2w^n*DCbf$FhW>fZ>E8EK0r7 zw)2Mj%9X(^Hy!4Dl*F=reei6ufw0DIw1O`t~|09_~n2;L^64*|2za>%wbVW6voo;xEq40O|`57O*HutiS|LDl!q zU;k2I9}8R?c)k;We+kaYoGWs{gW^&Yr00C^HQ)@3}oOU!Uu`HssJr;m7$lbB0`K{!1woN+ zC#!XEh+uA!yY-^tJqAOCwB`>%g{@PERmdix(Lns7SZd`rlqb zr=k1`R5Xb7fKhaD+8h$8r|QGqUZCEP{Q`5uw~Ig2HLz?XS`5Y8z4oU>Dt*~&ozmCM_-B>C!wGb0{%G; zs0(d4@wQ*k$OsI5^S-)tj!rLEE~WX7$SzWQ)oG0n(w7EvFkb(#zh}P}Ko6s>v}W($ z5M8a6zstQBA+&qO-&!|a@%-O=NdKlAk|ln2<*!D6=i)JzJ*(EzD7 z_wz#pz3HKG1ugsMeXR?h($1i4fcDNxl?n4L8U#D$k~C)WFRt;f*unH&u!$=-K|utb zW0P=P184}~P&WM&uKdoP4)nSrQ>kGw1Mz#wx6O`qe4LH?ZP;KwzN{&CU4FM;VSOsM zX#JP8k~%huB%hyt7xI>ITqr?^pE_wcefy+iqrbyt_vw1{yP-#?31>EgHR9UJuBIsM z?f@X!Q$_^D&P*F%wP841k0H7&d$F1J`<}MjXFG#N_#a5uox}CqAA1-jnU&jVX{NW~p^W zGqq%bDN`7^R>fFMQn`BGX}jE>^JlccjaVTcIELfs6Bnf1$IrQ`);Mjd7ou!8kbws78joY28NJeZMZv2uM3TU=$w3-sB$AF*d<2W?4B%j4Cyi z&>A&2jlI+?OX(cH402BMY?!u6i^Nf$cf1Df%mlK?=|oyiKRJ+pMQ{oGuUgw&QNxUSeB zy4?T(C=bKzI+utKoy+NK??kuatJlVq_LjLYy33S3{Wt~lOP{r>oqquRQsi}!X%n8r zjvw3L+#9aksrOo}!#cUh4OVqm@dW?g`@l?R)s=8&(X7BNU=L;r zZ7;-@5EPkH>{ZFX%s6 z{pV4KJPwk4K$rcZrBQlDaQxY9?AKbyWDwY{aT%-bIi{(Jff7x~eM&=v4l!;k@AB5C z=87r-Fb`|P`6loi7>iK%z}2+gK9nUWq*|fbfqYa5x#GCLtqK>`u(tU|1?(KnYau&p zjoEI$OCGc;C2}`zbs1-}#8z8j?j4}53BdIXN8fWSCKq5XoFo%YQM`_C9B^9nV+G9{ z>Z(I_?{a|j#}-bLA+N92&tk>pi8-hTZ2O%0jrg@-1{s0N51iJ1EAU|UK)EdL)Wtix z_#^vAB>gtc(?k}<+UWk))OtYN7|;k6J2UA8i1&*6?K*p=0B7e)fD6Hd>`W^?J}`Yn z2T`lKzam>MQU+Y%Koz7{X|Co^=yp#<_vr7+OD9xme#ad~-NuBGOf2!B^1yUF2y!u_ zlID7T@YWrSuHn?!R%Xu~85BMofk<~nn$t3%)%1HBy}_bZz+BM1&d$pUqx1W~h-5cz zVecz$Q+nwh?U{a+_25P=G+rN2-cxY27|20aK~mT4sK0O3yjbj_ggvIc=g{Zs0BM;{ z2;CowR=a!@=Rt+I6 zJ^#Hzu8)|bW#;=W$8(-_r_0_ZwU6s2+WV>KjzL1n+>Zb*b6WigbJS&OXzBF#UkE%X zG+wk`lizP^<>9`VwpQzN@~xf{ub9?>snf>fRAH32-wem8kI*}V+3eLlveiM?dm|t5 zoTJ|&Nv3j9goqa&EP~^UbKuE2|kqcZGJgl|$#RFLk;9& zx*@C-L@N?f=W(Pjh!>Ic1)(?-5&Hlk2G+!zum7>%RJ;V46On~MT4IuL$N8u9mWYd4 zj|&DS5@FGNA#O_6t5-$FfA4+MR6!$}GJQx%{gS#Jv7at)DvtR$#5FJ52dpB9EYFYL zP9kH)4_n>UGoP-g%4dyF?h;p~Th#&EmE!df^rg~e`GTf-oH&{}-}Em^5{($*I6wpW9B+sq zy^;j1)GpAX7Ey;2HA63fTsi{V?{m+8`611GT7QDhCB0;D2yk>fZ?{0RrLx5^;zVi} z5X?>%bKfh$TKA81{zKUsK7R%SYIn=y@>ZhcX3X{aD>9xIqpp+J%eR7zAG0mFMnil)BxM9>(#co*O;fdDjAfGX&q6CS z;%~%a8=&8%8vIEqt`ao1E0r$VJ%J-pLbt6m(%{tT8k?l1fsjJ_1%qB216E|sZ$S1z z&($i~n|>s}$*@%Qk>c3>ak->4Yr>Qc9|VF5&&U~ygj2!MDK}B;hL^POa0ZgaS_gxU zN+e(TL>hFp-{naeE{nEC_rhL!jlwpn+b$~7;e?m!M(tz+wwe&BZV?UEpwW6e3(K*y zmNa@-*{(ZAcQ9YWCYglOm>E$FwQKv0hat)RHVHf>X)VD~S4>Gr#OV%dd%B70Z-V2{T$H*vm2USlX98DYBk(Q-El1mlPWCEe_&i9?1VczVvsKjy$TLbw ze;zS2Qc}!xl&y`06c?c7<9)g(-YZv*ha5tpe58ihZZ-tnjC;5hRERz@rG&U1#$(QR zeatRk9;vq+l zeL!+DD6ltdbXu`G<_p?Bm)dZEJ(19GQG0}Ib{3WS@<*JMp8Rz>T3P5*;n94$+%cW< z)&#D7%5U{E+od!%`%+idVMx|KStde-5htl?C|PWc+5F0M1C_j8#i$zK?bqvuYjof` z!?{&tr;Q)aGV164=h)7bb3q}jZAN7Vx(z=5qvG3=aywOGsCXv{d-Irb+qIwTONX*` zvKQ#LW;=H0v!kb~#mt*=!H}Y@My?SbWDRU@f0K$*fJvB-3SXM+pZ>A5Kl%#Qx4+do zugHt-E4KSXYm)r=uG2_d@fevD0abq0_le~_lQ*v;u46N#r)l_VqrH1=z|vFW_FSgl zAU#7H^L~7HDwe^egW!Y7O;- z&?Tvy%Y7Sei4q0YGXxg8#mD;G`A!H>^(zAjW)tWs(>q0nLeD^GDO%ehFY%mr>LAwp zT_|14IP*A2B&0Wu{dPaT?uF~bVS0B^-cg}4&TI^rX~vVuUm2Q>{sQoWUBoG?nC@hi z8Uj$#Ow|A+F3oj(Yv<=FxKjZaYbA($5OeGiF*X@B`@d&$Y(bQC)T!mvE21YP#xm5G zWLEoq>9p?RB10G(rs>1sc90nBS89+$E}#@!SnW@57~O4AD6x|3U05Rn+XnQvY3 zaYyB~sn|@@{IV753K8vhfUL!@PTb``iyMEu!)2r@=MIX?soA^M{N$XU-Q&!zyr%%t zrwZD%?Hs5+f1?baHIZLa0)4)1+g%H{33@%4tLsXeAnu2m{5sKd6MNM;3&<0~pfs!L zHP`)aET^=1*E|{l98}obh=&iqW1!m$;4YynW-Jg2A`2#AnLsR_pnjs-A9Noi{p zX~bmCK#8h$(r@e{H4JIbgY3xWYi+FQ!^+*nfc6)6(mYU!_UdXE1^C1-CyV>r=N(8o zWS30VGG(UOjcghf1v)Xalvz~aZ8F$M88JN*P?N{i_Nva0J_FFP+4uXGCwy06mW@;m z#O;g19Y{c2nwY0#IzK7+$e2~_Qt6vz8Fprs?T_p|2`jmPT%GfkmbXZ6vd3qo;5!Hi z&+I29(5T-Z4Zpx?GI|xv!?f-cH}aW|cFD4=+Lye)a(vuyA2(u)8ReF$Zh6Ol*mYjy zCtJ0cNz-JobjbH2p$<4j60z(RcDNZwcNjk~>#_2ES8cYwf34QVJMZF=$*{mz1$pJu zSN57MkWfn|iocA393;R|Ev!MZM03SL#um(`=OOASB|-81y8qxeoHd9Y$Hzw!ON>di z=S#QpFN#wFex&QS%NOrppln*-+79iiAg}a=6b*_Di~w?S$=?E-3GEu_y=a!q*C6S7 zCSjA(bS%Q<4M=M8aa`wBLlz-e?-S9T8 zp0nmEwjwiki*$5rVdZM~xBf3$a=QB8Gwm{(CyX)20>f+!$LRjPCp6a@r9=}o1V zfRxb7tEgasP=iVpq)Cai0D&Z;zO+b(1PBliX$BH{3xV0k_ujkitXVT_&DUAO2ZaS9 z<(#wk|Jl#;d)%KNh^&Hzmlgml^?*eaB0JGHCYb{H>EY&XFX7R>gP(x5qk*u* zWVwE}B-iHOe^oKRP*4I|Z)=`@x_*zKu`7J;fVk5|2Iv6m_(8^_$Hkgu?5U&-;$%S>osjeh+dNL4d6H>i8wcg&1SN zgtaGtgFM}fsv6?WuB3{7!!etvs)vVKO?MD9vD?R@lh1Xg2&XwE=!$gF_rDyG!(uo2 zH3wB+_hszPUl2*!Xo}s{h(+6HJ^2RPB})GE9F(sr`8w!}^n8+5m zJB;7Wu@<+d9IA;+!QceUzYd;FZ_b#_d8C@I;9U;({FN)ArLT`mZBrI5UKv3qY|cGF z|1(G2_+#V0;tt3niW;Bdynnw&rV+CsGc!3H6TN8#d_|wXl*Cn+Ih572fo0439KB)v zW#~xiJdg#Uvs5j|+7p1xy9%;? zHy=gCjg&o+5!iQ~&|ROJO!8y`^zF;Ub>*NL*S83lTW2E*1YGzzf4ZI6>;1kIC7W+V zho`H!=KW+z5DVlAUM-)7g{|Sd5%C%*IG3sK$RG*C_$~rz!Q|0_U7q+P zA7Jom?$o@YOR07=)AMYEAVLqM#;ahT0G_7@7-SHzMIi3Q14k+QIx~$^tJVJ0K0|RNx9t<592nnO4M;@)YvyF#eGc%7&@C0`3RWX* zWV%V&`;nsE&MYzWTGZs5X)-v`LV%52VEaaBMB!}4l_q?eT7QcAZ$p;&IMDA@D?1^4 zU5*4-cJ8#6Tt2z$Ko4{Y@AM2Xh2VRo6RD~|JX^ULsq=hYIfudg_LGTuuU;sc3Wm5A zy-bh4MDl??*4LKcafoKvxod95{ClmmwCyV}YzDm<5qx)LC~DK^Z0^WVk-v{bC3Cq? zlf{ctPBE#Oj(1R(Md^5uWOd2 z9An9aGWLp&S?M70rLlf{eIi+P#Bg+${4MUeH`mq|70J5dyXa&m$Ikclx46zM3C#3% zNaNySjjxgJRfe7>Q_dnbn@rm@6t6l|u`N$%+zf5~c~0OjucpSH+`$Ih~Gr zfb&kLt zRst>y3&u-x_Yiu3*OmizUp>-SHV6MKha9qhblco59C2`Z0#X6Mh&9k8J>NZ9_m`bS zl)|1%iFi(frtop~-z&XVQ zaLL6Rz2JO&2bEV2=a>Q*Wzpw{rz(FuJRPa#m;X4m)T>n0amAdcG%T!XWmfV_e_=%X z-Cl(f>w6c##$U>N+I%!Ql;$M_Z1DcEJh%X+vvt`a`!~rkhpB$wzcU!U)1^mwyNF{i zoVW6CB@9=DMs9BgA`*x6NX5=yom!kE1sfj^bx_&y% zPC|U z{mD1Sv3TyuST7ljxao{WfD7|IB+SpQHIaSkBb<^1n`rKc987Xh!@@hGsEH*j28KF3W-L5~1IJI(xMm@}A^ScHeuvOQ6@KydstxWVvs~zY)+J zbO_<(bMS=VG<$i=0Pg&63U)POUHS_1L{-2JqJP36PY z3mDd`>8F=z05G+ygt5?iK-tj-p{?(CdtHx+%YNNDb7gU7=c8eYiCLMQx8|sPS6A(L zC+c0wTK= zys<-pEx;{XdOm8W5XAReS{lRW=q_0iKlT}e6Qs(qWYnH&d)jt)dyjD?Z_EM zt}Vi&s}}evtk|9v-o+UlVn_bonJ|p`w@?+{7?dFj6Uoor+f^%5)7uSl*xn4F>RdidRqB1qPY8@FS;D^EW0kd4}<_F<@tN1wBB%uCFW*0M3 zwN-E2FE$xfkry%I9gs@hf`4kwitO}bUS({I?Vz?YbZc1VXFg=OH{c0F>Z^sZw`ud^ z3_4PIIkgSP674t;ofUe3~L2Gdo&@)gc0U zg48``)(MX}v6AsU=X;PxVX1F7<|?P8DjXhw>bg8YgS)Y{3a+p{8=zni{N-~UQ{^UE zZ_s>&JDIgo^Hh&f#3t;8)2!Q|L%U8ro%RCis$;(}S)1~XDrm6lzoEsR?EXdH`@6>r zFHmLitL!FREkAsz+!@*0@Tgx!Iee{TxYnmvc*dr};$iE!!w;us^*~bUH(l+vHx$ap z50qK|GgI8rYpKCN%;?$#_BP(RHV_Vrt+Zjrz1mwHjh{S3ht3sLq-#^w)|09W2xE@g zp}h~{kZS=PGpiLPb~VOGX8lbrL6t{Tqts%(zkN-LL!8)&D!>8njK5c~BSTeDq@y<1UEYd7ujTnxtPI8+16ZbO8CBDw!l5 zNG3{wq9kQv<8_2Y2!~q$J>x^O5%+{TtYYQCmLwgnb zW&s#N$V15!393oS8YPyiff)G00x5uhE_}V+yp-i{U;wpJ34q~Q&-E~Np&=s6X6f^ALg z!?O|NekCw_az)R?+?P=q0>zKl625nYvaj_N>Ci0)>;sRgxB zC`}a;^3rRI-}6te_Ro0pEslz|16h$*FW!7Qu;;7NPV^<{?3}j&Wm(z->vsePQPq12 zSqDbk@_hla-};OH_O$SIb#CacyLGkYPZ|4>0$ZH#4Pc*y4Yutcd{V1jIbZ$iR_vZ) zFA))|zsG@mTysuD9=OX+1<-~h@CFRRCnYE)_jC^WEb(9dm?HKZDVV+5Z^dfY?80_L zY3A|~mGrN@kdRfA=GdCpiVJ6(B=~5)2r-O457HBr7M4)jWRKR551`8T&uj5(bE-n% z;eh^&QMY_j$!Q$TSw*S!(FR?Zg5=jvf(&ZTj>LJ9nbV5QxrBtf2-TWp z+z}Ee8|)=U@5Hr)El8b*z3=)ysd#WowUsm-L&%bne$~GAY94-qO#gFBf#z3{t}>+@ z8%>|fdzRY+HCv;M{Ns}f3<>i>pY}bhvDT6*=`1F9>gM7T$j?VDcD#D-yRG+lEOqe+ z{2Dit`M!4kZ&e>=El)R-5kNlqvqE&Yq+<00o~8W3qVMcLDQdn`g#xlzBDk2MzewP? zs1)Y>*GvLU(KVRsvS4=i;LKk;a3-Ae*Xs4Hb10tYZ7uQ}4c4uvRZouyN zcRx&fv?pWHaRM*cGP4U3k&jIEftvY5Gx z-D$s-vdm3~Z>I>wq9#n(mV!q}wiZ;`oQVs%&mRz9qdLdJ)&L&h_KTHS2oF#*a%7)) zzNVi6-u^SWo}sq*F!Z3urZ9oDF|{@16MV+rcGN-4aELOgmwIrMqho93$|zh1CznHg zTY#D^sZXnkb0}9Zh^;mi(R2x-aO2RZXY%>}HWnJ|PBFCAqb;&JHCk)u=#xHKV$vm* zwHL(W{`Eb@MCK^uj@bV0EBXi=Xvk~p>%%vYBbmU0Nm>qBrh)KqB46b+i}iUOX>ADzh3J#(W?Ccq-UxeZ_lpsh1e`5x<^Jlt?L8Y(eudc(8> z{Yxfgc6H)`_bM*~xvm%>-)MXYg)S-^Y&3nHY;yM0$AJV3Ko1AF%pl`iMRh*S5y=cNZc+LH-v z?*}jk>C56$bs>SdDm|0WU4AX4NGt&qQJzKj)1`nc+J{m7XoIc_>bFtaCDB(W0@QI( zzEyT!#+nrt4R9?jx!fZX0D;OmBNt7IYb-grK5wkr8U`|!6>zfjB?}>x=ODUQAuGab zb%7a3s5H{#&6}r5ccEYYKp#&CHLQ_LXOIg8$P9Po zppC+hw=xYN7YNKsA+@2-??ng&glSV+@jI8I5oeNs)(O6EiUg$22L>AfB7+e+a&UFm z1;Cc^`l%PiwOMh}GH|MdThB)%h+SUH#lOEOK`N|A^yP1ilSi!Xier^rc}rsXWkA>O0e79LiLnDq1Qs zGgdwH7>Wq%$v2;8Hd@X`ti>|bA)EU{xw6DgSmkvG!q%(VU7k%%?E94?}oI7 z-{Eg#pSOA*643?3B>JanD2r{w2+YosQ6XttoUaIbC+i#!xpdZOnj+wqN(HKrBtqWf z=OP9w{FZ5E3PJi3z?|F$v14$hWhJ*P>}2+xI{{(V;!XlNQ?_qK`)0(2*SnUY6F2eG zLzI<)Jm3vW3z6(a_tLij>ER-1{A3U0%kfn|P$4-!uT3K)8RFR@woO4r-bFAMKw&h7 z8menOL(w7bM>+_A<=)Cvt2-`Zb<0W-Z{N1Ri?y)>KS${w_mXKWNo|pgyf5%C*Y%Oz z?-to3J`-m7y04D6stZPfMj75@Uz>)Ak*7-@ByZ%7_jae3tjZQvuE%Rw1%wl%pfl2^ z(%O`eZ^DnSZ0E{2?);n-bTC;94}As*K10(>^O`}owi?v0OM_;eUD9G9d=SdMJn3n- z4h-6#uyWI8-LNyH;pSz63q5!&eueLo)@#VSa%PiH2E z(mtJ)m@i~#b?&7PYQ$S;)^C$|X|S>FxVC^ml}G!B6cb=k57tdN!x|Q+?m`e5i5WTI z01yRTqr_REoMwT@oS#9*Z&|Zqu+gK6+w6|E%`02W)!Qef?GnuyT<8aB69CT7dy7cV z0(1un0QOJTGT86|$>1%J@tAc9gT$R7aLP&UCh$g?GLl-@eV|>#wXU6U3HOM%jI?3+C+-gEt8%Y`fT}X+GPsi^d*iTm z<|s>asAnKB{nVg)zWiPY41tW>>Z4Rj3bjNoCTX%ryP4z6(uW4dl!q+9t@&=7nPJ~= zHA{lNg`}#Wh0MoD&2o>>{?!={NqQL9{H-oV>A2vstIPfo!PP*ensCP@6!3*ECaG(z zV`8o6T!v>J@@}tAg6MRs5>F(0EGQanG!Sa^ekt^soj%#12F}cMC2biaxB@WYz*eCL zka)9)E7+DdbNtjdcoHV9s{?YO0Tv!tNXX4F?sX(XveB!xqD|Rc$1SV;Pz79r7Daau{yYd3bT3_MRDw$#LD(s|pA_>|* z*3u)8b?m$Ki&NqEj4Ll$E8;sVf~J1hKe803$DI-dmr&w47Q3WJHx64hf&M;CB(z&w z?o^}sd{gL}^utbnf}7Mhn-Z*kBwvd19Im>fu`%bt{&}WdXWft$CmYfS9m=);HU-7o zA{F#6=Qbx$!ARp=g6wP>Vlsg!qX&^~3#PLUukt^U^cF(#6fPi@bN0d3GwRo^7K~NT zxlVbsyQh@C`T|015GN9*`nqMhUwl7wX7%8uGrOvy*PvJ9Ct#GA3o)q7lRr{qDzMVz z3B+AdaF&FYasBK$w{aT;Zs5Qx?)0W6m&OA37X#^8&RBE+HyU=H z%-TpR=9h;JXCtdT7ylCAmfxe|Iv@rWA?LxUyhO93%Cu+I(&zou4p&LBjEpj2r$_hM zlCEHP!ldx5#R@H>pS?JP=f*>&W{N20{GYjJB^QeQLw07r5J{h^!A?MtcaHdGOTxU& z3SAo^-F}}qc{@oZr8L#pK0*wv5s6Ux)-p*iwa9@%GyztJR&bn}t`>eXe^=Q$Vj~y> z?ZwrW9of?iQ;k%OT-maK2jcVLz5K^}^!ATth=+CS9Z>oBx@;G;6vxYG3ZD7o*Ht#c z@Fe$)1%utKXSi#cYrML5W;*f0mU;vLVs8TX9XL(Zy9wgw>UO$O{vZ#0K2V`V7`iWC zCgoe{R@kwOCgt2X+y*L)KoQ!RHEhA&zky!LrSzq^f6w6)EfP?YO^vj62ujHzrr``s zu5T~_-9KBk^gV7QRO)mPK>jQREi*)o%Hro3Aa_>I@e}Zlg&?#iRjG`;mTh`CtozF^ zfXx*H#d=8K9sQ2x)BTK^G*oSHz3V-mILUiIaHcMSm5q+AzxGVS-oK1360D6wmfFQ| zS5?;oBAh{SAO%LAO8+uzUpZwRk98g27x1R1yni)l5n|)N%_qa}ch515@Ta_|UHI?s zdMl?|7u-^(U9CFyg4Ity<>GmW3~}S|$#^Qsdq88c8+3}!pv2mAwHa(HxCC3HC(`;s zjCgHC9w<=I6?6?LDE^iZS|Ab~u|A_h0O5S#4feTYp8rg0cdC?Zd&1=Fu|Ib>SPo+( zBmMug9O8ATr{)0-tUB@QR<4t~<|zfDR}2=YwxUjL$IE7s+dPSl>tSTEk7ldlyjq0! z%1ID$#VcSowlAE)jmamO;8ik4S9%SEv9klKc>zdMOu#p1l0UzzafS-0O#MsYB4|nl z=K*;EiU=A0yIl|PtOEWaoGNa!o}An3Gd-hvcYe1kN$m{=DPCfNRuqD^&sHOv)f@x& z@@~8wZq)AC>zm;|t10o#(_ub69+*grj7t+oeh{4`?h9o#4D8&N*vlXDAXTX%=g~Xi zW#uWU?dFQqOpjg@n)gskJ}E%Pxa#xD+|HwC-b4>AQ_|#Iu)dmLmO@~{0YU!0;&C(w z!IKYI&>}s^2F?UaoWX7Of!}NBhrMVEH{B5~1qmybxF`1|V}1v4=l{7My%-^qT}v|| zXKo|qi{MY|DzCvHlsU+;$$kYZ;`vauSFh$J&y#OvQ&sXwS*S^HVpde`Mv8;MB1X)6 z^X@oYqS8r@%T z8MdpI{K7d$-hAO@&$OdIVie#|5y+h9hg)8EzDRbmpGL!6*To0ozWNRMX3@a}e1UCw1 z#C_~O`8MJNfX6ruR<7f-%L0gfq@FP+5cIHZW{26A)*eau-H-c*W&rzRs9Kr8ffYH^ zA`9Y#jXAkuT_U4o86g|}@Vw(v z`kZwTa3 zD=8@w&!l;5@~&G-P>Oqh5z@#h=e7Z~N~+ROUrAL|vA;E`K+Vov{}EEc_UnYg>Qs?N zzGc%Wuq&MTB%ja9IWPf!*e9GQJCzFu_m@qB@hOu`#rgySeE5288N6nvNTj9)f z4hB>KD7Q57M#enqUc!kKrSK4W)VLQrnz7mAH*& zDiUM+Ij9bPsx?#LwTG2}#C@!<*i(SuVLNrA`}(nSfM)q zWz10ZWf33_cx#SVb_Q_73E2C-I&KB$+CK#r z(9^)YOdS|EWU(E(a=kd%)f&@jo&)A`w`ykfy}wv1(oJT#+AqM-57#H0bX+gcyz!lh zpgqev`dA?54OgkiGBF<$o3s`mZcMks?0M( zmKWYScm!36$7G`1*DB58yA*q_^Z=x^b4K(qLO3M7|JAXf4}E8)U}ep;1)Tu(8L!)*%JC0^QVU$y+`v8;2w^)PNJ75a7Pvms>WQzVY1p&G*Ut z6WP0+Xo334*a+>Cz2yi$iV$|u?MQlw71$fe_8ROob@s#SKMBp ztIcE=rl(?C1kw~#;ovw?UC@uBI9y4CtI-hA+@JS+G7H=Y><@wTmy&&8_cpJ-c1>X4 z!^MgdOB2bIQthZfXZOpf$=7Bj8#@JVOVcHrie~lBmDIZ|?`;LTJ5Qdbp8JiIS@B?O zNx#O?m*4+s;Iu4rn+P#0y-8`tdWK~;4^m0{R6B0|`3>(6{ehTMIq1Bjg-(4yKL2(d zL34o4g~7{DgezxNJw>F?fJbUJbgb*>JK6C3;r_xnkjx}x2392bOBx7>kuo%#dw(5# z=iKpBl-D8{bUcGavn_zs+aHiyeM)c?&c$*M)gS7K(HJI7GYKn0*mKYoX?4pvn5@c1WY3{2| zr;P`<@V8vDqx<8ITFR7<(2lyL#dzu;9W-Y-db&BI(3KG5Ngv002T@hfq^x_swv`;w z*pYiV?|ZqdcI=}HA6k}&U=Iw^>+3(rM-~FV$Z-=cdw6(wwukG!6giic^{39bP;>&HXkhD(DX#?)F4()Qi!W&3?PuQ*SaavtR#V3lhHIGw;>0_h?2g~&ViZu$Bk`b9*?O%|JFBq zVc}*Of;GlFkqu8HOqzXg5fZOu-%fH;;quwVDTjA5>syfam2%adOECz1mCxlUTjxCT zOPsEd<;}*hIH0Tya!!FxyKDL20PJbk5Dim`Ro-B+&Ro#&wx|kgK>GH}w(%!_vc&op zrZs>*yBg2_n&=3}L0>D4t)4sr(K$p854zg7ozvNRfl`7s< zX*O0l>C^<;gH^}pAroX5dZ$9^QCBL0$y|^rhv}*s21uoA*yF}l;ucXdS&>Yv0qCwv z$t13kI`bL1NM7Qe)lzQ9V`8^G9S(Ukb1)4zK(P{0Ll9++PA)CZ%HIKS+H}~&7wS5M z`m=bfh`CDXhvrAfCJdg1lgMaR<;J7d(C>pye~ocL(MhLubhCc3jDA)?dY2K;{-I1u6WG) zZjozfb@e!4jxl@DplMZRgB4#NJ%P#9yPPGXs6$$sEjR6CALUys0{OvyT&F)sTmcbe zX)-2O>@4s6j0q|L9*TlEJwOqhV~mil=n!Jofz$-&QipH7;S|kj+=ZA6OxD=sx5s`| zJgeqDP#lr*do2_Qb^)|pib9Cs$Gmm62FAmx&a;E)m==ca`hRvh+CPSm2IfvyQTs99 z^j?=);#-@lq%0d^oEXf_bu~WOp2}g@3dh@dVAvB>+ERGD3>dBe+nD#rlKo-`Wq+?$ zZ~6g!r?Gg14Z`D42fwLA8wf(>Y%d?E4CCT7EU~G-hV);y$;4FC4k}Ecl(4}Cdn*A^ z@X%_y-$Yj{uKRRsXT!oPrhTf@*OS6gc6O1d-1j473n;+L{!tlZnw;q6L4$4Cshdfi z%>DaXqaQFc<|MbOMo2M>@_|U>V0|Fmdcu zRDW)}BOrRCaVyPffhj2uT!Y6zV1STDj}QXwE7vp_vMci3qmijh;tzjYFU~!)2<>K7 z5D9+UywO{9uK)yg6l+L*S`9n3k=_uNQYX5J)1AE4v-Vs4@!Mv{T@>otPnGIVTVeELe;+0Q zTkP$unwu7X&4UWh5_h9))M`*#N209LVW2cW&dQL3AcxwQW~RHt2dw{np{+s9B2Lsb z+a&-uwnexA&)ix97~1si_tNS11)a%8xKrL4NPZ%xfGf%r`-*yo$dHo|I4qpMzf$ty zymn*HL`RO3#N2?~+_z5k%w_3fK>_GN2t3i?N%0fbZotsK5Gn_et1Vy7Q3h;3O#XsW~0Vg?+bm6SQ|#J>CoTr8u@H-iEN% z4c*KeS@gkb2kT&vn+?#uV%m1x1k_nUAZv-;S&dTGwChvc0wWx)ZF1(eg+6U+7JU+{C3pT@^v%q=YhNAuw9*=PcVE&&~5R-J;GRD=ZB2_toY4ODM4`W00PZSa0o1JK5eoJuyMSYh=}b4iDc=FA7#EiF z>I$f&^|JTy=5%)p_K-3~Wk_!|4Bc)^mYh0#pL=fPj(Ca;j1r0AfVaCooHiG!al{yL@e-96CsCj7b{~m-GABTHmyixo%)ULHd9z3>qObJH1 zTKKPZ^oJ+qY_z)Gq3+F{{)rkd9z)9r`>Y#;XgtV4+F<-|hj5;9Ha)yAxHLXs1S1-{?3G--yb9UENN!Ts(ecGq z<*!?wAqHI!kF78ofL$@7Q&%N0$jWYZDLq-@-q*OSp?7yWUCx61b8w~0hPI{77;dlT2q zwO@}w_5b4?l-=X The release state is machine-readable. As of 2026-07-31, `tasks.yaml` records -78 roadmap tasks: 74 completed milestones +79 roadmap tasks: 75 completed milestones and 4 open items across metadata, skilltree, validation. `TODO.md` is rendered from the same source. [@fig:roadmap_risk] shows the current distribution of open versus completed work by surface and release class, keeping release-readiness diff --git a/output/pdf/Active_Skillference_combined.pdf b/output/pdf/Active_Skillference_combined.pdf index b7a8b9024bb64656e3a0b3ff0b671f07bcd015c7..4b31f4648843b28c4de1e8d25ed91e5bd0f150d1 100644 GIT binary patch delta 105878 zcmYIvbyO4q*EeDyA|NfTAYIZ@O9@DKcXxLV$N~xi(hbtx-O}CN-7MWpeB<*x-+AYc z(3j9xP$cMUC^Ga76b1SgiV8)8qC?+7 zF`$@GEGRY<2Z{^DgT9C2LkXaSP$DQXlmto&{Q&(4C4-VfKS3#=lu#-tHIxQQ3#Eh7 zLm8lqP$uYSC^M7=$_izJvO_tboX{^&E+{vY2g(cOtAd1@KI5znGljgs0#NbS-Ji(_ zkvhmU`qH247Rj1dQe{=~g%Csu27TSku&{`x)K{%_d%MCG(USDzmvbcZDv=`GVDPB+ zV`>DM_g=$at@OVcR0VBzn3NO2Id3s#qQ_ci2_P&_Dxwn@kyjO;Ww^B(X=Qu@l%f-^ zJcanKb_t89e@{%)EB(Iyqz0tl5xY6-VhP#Cl|qngOz2>^m{hoF<3(?d)NMj!5@*x% zj&N(V)zc?aj5>)z6{Ce`P&9BpvxI+6etWRbG_9NVYD7C$aO|yiD>1CgI2|XgvG8?<0;MZaFKrX_yQeknRSK%T7Gf-_*N+@O*rH&T zV2YvGlb#sZMsHQ6|7^|#$SVY6-Lh+?ou!eCkoXwJ)z^GnZGXxqBmZFU_-S^T6|1bg z9a9|i=@-j&K$LOJorAnb&PeIYV1cdVIw6=^HQyZ@MrIGBRn-H#FbuY|W;^BFKRe=i zyxce2Tv9TDn}9rh+wz#A88|(?oEzg*CU6EmekQcHVc6Mt({B$v))pQotF{3(CDJHA z{JTU?Kd*(wsI8fDORHIsMBt)6ieL48+Rx-J?%s`YOAu$d9vNh@dXpUTNXu<>Mi>Zc)O{V3UQJZOar$~V)$oOb(>%gE86RR9&)1dp z<1h2R=pJqC*6$25vNwlNJ7ep7vEhn+OluuIjPiSDViweEZCCjpaQ44fT1@lLYV_`B zY8CbUmmJ{}rG>MthGD2|^ao#m85j&eObVb4+H-zX8Y!?ofXsWBM50QCOSe&TS@{{| z=Oq7hJXktVKHrOie|Oh~clA*hhfZnN_=6VHH=}r zaF{-AIETf&Qs`Bm#aAZU`&AMDj)5a78qCjlA44K{O>L@?M~rZ2JG-%J%jv?+qq1`zjszMWmHjX_sbgvjVa zFr$e(S$(!XiV=D3f9zo|b`WeP4UQTL-W@g|3d)+1$!fqNmEh~}6$h?Stx|Ib@k#g9 zUmsYE(Z5u9&s(}&vD?1%1r3kioxv}JwZZ=MT=j8m54Z}M;JfMG+{-!Y@Tr@E=a{5^ z_ntY-&X~QjQsmIewfx#b>NqBLK383UGa4V`GLu-Z(lr z80%ZVahpF}dZ~sj{T|7#fk));5iU+yBZ}aQXN};oQQ7$NxmDNEkqy0&QiJfSr@p<~ zQP>9B7y0|AR8P0q=H}43rTmnX2w8FQi026Zcc?1}y!*dr2)zUa8_)jtJT7UCpG*;s zSGX~v-5^bx?1_LNz=8z6=v#z=fbiEg^8%g3@v9P>3@-u#_Nvj5slqtFbMK13&|c}! z2nZT}&@bU``R529ETOM)*L(#8zI=H1xM?R~${t6z#!tuu$Kxq9pJ*ca&)YYR7%8^W zxQ&FpmjD?3#&7%H|9ng&4)i(Vvo!>OyejTtm7fSm#{BM!g^5XkLp=02H9bB)Gcypy z4xrx0#>e-jkuy=17UUFU78mE_+~o!ce>GB4Q+v-@oSkp3_b1#kbM^G{QZ_;o(3}`= z*V54;=|PeE;0=Eyhd^{bJX4TYPJi)DTN`#fFX3}~j`B@Ld^?9>5QUDKdh^)hO#v@* z8o&YkR#Z^7a?LPDZHU`PtZLT0ve_6EJk=a$cuq0sKnEk8hs+jP=q{0exf z)RTKKTHiCEYCG4eGDVwq4T-0+ba5a&x$goj*KW_w&SG%`2sw**jPyzsC_NVDUIaib>@lr%j<2B z7uPbPmw1P6mgx$H>k%LJt@1}jL(QbbX3sA(g@)vr#d0jFvqf^5fzGecchy_0_FCD1 zw+HO7_ZS#`(yg981EX^Mz>K2YmiW!Rzu)1(0l(#AMdc2om`G1q!UAXM;cN*GBP{{~ zvw`8p;x#F^^Idy0Qg~*@cxAtaM&eYVip5lh_za_qhX>c8yhmH4vSv+V$bEnB=-=ra z(G{H%)cv`~k83CETS+;^#PxgQ6d@}>{gL+XVTTpJOxwkV9%<6~)?#^o?w1>NS$vl2 z0m4shhPKO$zH8($dAqy6e}5L~n#A=?BuM1y<-?sEcx!|@b()_F@$4%Np+YC+_q=`- z-cz73svV1tj%I3k`c1=3$HAYOa&1l3TC?fbZD_xP3=9kw=3LJ&Q3T`= z5Wa0K2G9wM$vk3U(|_itrM;UE<+-SL+UoXePCD~y6m0uQC6kgFe(0c!@FAu{tXV8P zZCILTzILU~1-bo&4|)?_ibd;=Jg}C4#2dbK8|dUz#lY|ekA+=DNy&)SaAUn&Ci!P2 zSeq4@+1cXA@4Trg)<@DCL7qBBK7Mo5~bUQ!!)vrn#=pjn-v*^>EadoIJC-TI6D}=W?$NSmOD__KUbKtfuBK zVaOcG_fK?LQG_S#C#s@yW9^H-7>GznFi0@kq!0FYU)u@|mT&q3y>eUO`FeVPO0)C( zp2X)vw9^->;_({Kc=gU{iFSk1)AjN9@;Zr|3IS3g1r zC=2FlEuSBRC8w4AJj_WB<7QzAKKi-_=PRA=`1I)$6dFiKxL25$_xfuu!9Ibg=qm%X z#hDR)KD_w)R{0uM4-XIS`8{S@n3SE}k9ar#kX^HC>@njQAnu)eJhpv6K;Qx4E^?Y+ zVC!X87Iv_6byZfOWzO%l46(=TuH)G_po~ZZLj6T8@-OzRvIGGeJNwJ&ZcH@(r;7qd zcP@8HNyjT3V3y=@Rs`GZC%>YnhN4n5-fK1r7h*&=9+ z&VFg5<(*WY0HdTqaffq6gnKR$h!02L%vS+TP0Rhx2yG&6yD`JRGqyTnGDRAd6r~)h z69xOH!j1K_4jZvcil&T4;yXhN*TjeOCd$))z9WdVdo(meV#}K8_n#CN<|`1SSSv|+ z3Yim8aITBtBi~=8I1nIaf08JR^m)WS(&w)>ivImA01!MC5Bvn)RfIOQ*?|E;AtQL= zVbi!om?lPfrFnT|EuC>N^%L~hQWAH4(n&Zh#`PcOJFiLk>&%svl@+lJnZ@u(Ww{#U zo_l+P1ypXz@M)ovvbdg=>I@!+*Nd26Sg6SHp>Xim_onXwBKm9h&tDSYY*rTQ-tomH z-#o0b03-Kro^gArsO7wM`KTCGG^}A|rDgXmFDH;2ws1ceb%XLF-|G)={>?yqgKZL)MN;K*baa5OE&czvS;V&1PoJ(zpE!C}}4aU>`Ju3HastB26 zZd4XG*6NQ}$8tw>6uvXWaUe>?^y6ryaMyoXa0I$+w@;&XazQ{~w|6%XXKMfy2aj`U zS{C{=FA;DS(KVqn={6^TEjGromf9hgozq;&b$ilYU71Fu(k50u@5;T(I0zmn&4bvQ z9hjtC-Cvm#n(3Pnxpt;ZSZm{oEh+1WCaVf23rI$ z*81ws!-g@XC!B;#Jn^gXd^X)oTBF^TrQTgU6*yJ0C2@0?Y*uPSiV3DRS+4im5^BvT zyzBNpKKC3-U=}P8{%sWSO!v9^>^Md<;GV?Lid&kt%({B9vtLBsDB7`_4E(!iu3E*rCYsSq9g_iXKYbkfAdaMu2;A< zj}ld+7p(2xch=;zUK*m1#r`Oj!WK;S;?YB=$!o)Vd;L~YHH{x2#v)A887RuM)~@te z(P3=o+c%mbuK#;aSbN0y1lVp&>@d|E9vjV=EEoN}H-)-aldPYwO^0`AC4%u9MYHid~r}EP$mI}O88Gle?!Y{T-@Mz@yvI?9Pj|%?WBpW!} z+=eE(4r$jZd@0B!i-JFPkRYM)`<#6ztVor>Q>KkM9yl{6Xy+7jY?A+k8$7X*2 znelU}R_jtD;E`o9lYUtfN_WGrRc^6S@x);_!I~RJ49O~uFZM#c7tQ?@DS2ce6&lRW zE)h$uvQVP}=NPhF0H!xh0s{jpXH)wZbou#=EobXZpMK5l*@~j?E~M~rU}2e6jfjOA z?^9oG?wwz(<+Q>fxsD%U*~7-jVX0p#dzQ|R9>la{`>P6LXk&9b2l~ds_*p8IIGGUs zrWsBzW@~LHOTGwqKPuA5CWF_MCG3?6dTXT|qPr1~EWi{z06sT?kpM}R@<5jM+l?%LWOgBvw%4yKrDu8fT~ zG4!AjoRucS+8&#Au2NF;s=;ZWUGgH*cg2R6JkUK)PLMgh&7L^!ka5WBu;<1zPIfV# zd)7Ip@A>NhJ^5d+jvkN_IoO#=Sn_5@{hA#s(-4>J=xAsZBBB&1B+_@is?DdIrmIfX zs|^OHFOytZrgIABbEY-}$v@GR~zsY;4+Rbh68 zxYz?`vh~K(tK_uY9z%E5LE*_~w>JqWUj4pfF?xp9Qw8oDDhvdfUC#?PP;wqp@u!rD z7ixn;Lo_SAv^@le&+~p2Y=|2fb-NwV()PzDkTTpMy>Nj+Oh5Ivz87ImxyS-_*$-C&x?%T2JzImv>T585wokBp4X-O5K2-Gl&8Srr8SzxDztrg=*zwh}Y4Yr5_Ou|0-a5 zUERkx;+4oGu~}@EJ*pSGw@=pJ8|}`Ws}3){I$UmXxn6#btzC9|>;<_xqDnm>Mix^ffO9rSE<{QAGoycVp<8zg826QDWNVsRw zQasBLtrZl;w+cD)*=BQ;S^GG@x)YXmP}P4sJe|N>d?|qc+(lV#FMz@8+@+h1@a7sG z{vqzjpZfVnC>gjj4!@_ z{l;c`K7tSDQ8_2r{p^s!8|*s|>uWK*KVdHsrDYi5C23jBlY50E-a+TA4i4thS9P6* z23-z^T&^}YItadAUnzcAavqfdL2nO-N<&QFzP&)YI!N8o%ze)e0X{}NXjXi+f2;-h z>_gBm9+&lBe@j6?pkzcZPq6ELJ(?lBRz1b>N}lNV2yl+5j&dd7b*TDQ{j zWM;W@wYFjEL$lmxEdX-AxuaVWbXgx)#1@}tncdRp;XmG@qvS5F+8W#yb(J&Q-zy|l z!4vW`wpYiV>-4Ar5I+`%DiA5G@Vgt?mS@B&sKQ?tG3a#8XYrvVV7cyH^&Ir}c_W3D zsitqfNsSn_zI99&8w4$-4iJMv(&YiR z(|7fl;#Z0YswKE?)v)OlI7)00W@`sw?cw^&0s+1PuB5C{_rQ3@=hkClZ*T2sl=zm> zgIkadRnbLfc`>e)f0OabE6C4paB*Dl&gj7Dr$IC1>5xY)#%_h-%Gh`Oiy2?o}r&5C!t z|L>shp%vj%MM-11f~<|O#w&Z)JyFe8k8+=*3=Bxgcv@tANL*=n!k3UC*CT@-;JD4- z=2-*A&6NATg~ffYcjXQfzt^Cq7t7Zf)+b8XnCW{}JXs8aMshRd%54U$sJF-$g*9NK zYrH1iI+V%!yD646W4oL!(7SLcCtrO0cdUppHAJ79EQ6ls>Go^Q4MpLCD@j>=+K|`f zowk~Wm}Yd?iswmulDUDEQjL^2z*yCINzz{$lIy)H#*U4oAk~{l`erql8T)ZNU}$`u z+Bss5HK@e9zV^kBX6+>FB$aQM)2wj~^7EbH%jRjO!{Xi;!NK}`4P51j=&gfS`$Ie> zeK_XRSrE@Qt*m06#arjoz*l#Hm`z&rOjOQROz+IN=kpCQP93$Z>|Wo!2C^tIb@tnx zPpf>K(;4L)-9aj}nR(FE;8tIS%T=*#pyh?Q6$0_T${xJ+P2Z@C>-$zsi(_qJZ1I-a zexKJzJb9eP=ad!{FiMtQz4C!GR=jA#3Ajl>qD4C)J&79IFg>w!rV-zg&_pa95AIa{ zHU4no&#-!FHG4I_wL_u}?Xy;Dl!;Yv;LeHuEO$qJyK=@caz(HpenTUn8fV2uC z&X@b^<`3=K+-Y2{3A%W)C~`HZLu*ejpF9q3Cf{7^mi|=ES4wx&SZE&4mawBJ@9m{N zuEc}!)bte&SufYbK@iJbKtxoTYr0QiveK40lG}HkbmsTDV{zIvDshrVMi(K%p^M(ah3q;fD`$=0&6FhD z+Rnsc`YT)3R)o=a{>k_=(^70K5eLdkzp{0d_HzstT9yclELB>awgUYfNmVYg$@u7) zC92a9srwh2y^BF&6d^B9-XmkGjK1LF_6wg`PD=ICb_MX(dpUrzj8tDqL`WESrM3A6 z?lSxf`2wHCfJ|~ZN9KJOU-@R#{kuA>P;T#$n;Wgg6kcObXk#&Jm=tCdxwF)cgW#8b$Gph)a11G%ZgPtHArT$@hSe{VBLz}dA(OC zhZ38?6Jt2x>Jv2SAzb?iPTpj;5U5E#8Bw<*bILxaZ4u{ygza zkEufC;SQXJW1g?1ehE%^^a=ba{6S>ze726ib49_g9I}5-9n=h`rY#TjU#N#HRNGKi zyqfNu?VT``%nsg?@kl$K_N0~!q*jHsIBGX@100}GX|>e0rRubXb#h5~@UdC+vR}J# zw%Ter9<@+r%P9Dv&PBld8g6f{i}wd+vsBik9`$Mkc;Rff+eYhXo+3>+GtzNM#N&RI zG9nfWs8#5Bpw}{D^C{4$OerEo?9Xf|_pEsTb?$EG>o$xjzz)Qg0Evx>E!SjW(qjwo zp-YcmBI~nOjag1%B_C`got!G@SMuN#yPVXwmn0GS&vAHmh7vcYs~uImv${JwFOJ#6 zJW=Am@VIC-8$hh7X~cv5F7pOkEe|X-lL$aLXd`z}KW>WzeFWJsh=34ZVU$6$-fv+6 z1cDE6rzj5R4JuKkn~r}q}`y9#Sy7kVN-v_19{r7KO>kf^a&_Pc+U6)AsZMQ;t(OM;VFpE%JtATrgO zP_@Qp4(dROnmA{lf~3$Ag0|Lb?5GWb1JwKFKk7ZBqmxLt<@WZ`)q2_);$S7HNxzIq zpSE4@<95kxp9F~nb!!hm>P11$nVnBlv10u=KW)9p?0BT$SG~1Lze#DIJDPbZXU@ia zYC7UWp7(aZ!J9nrYT!YApuDCyQ64|aGS>cZGokA|*B<$9?mSLCb+oO7VTKXbJknFl z!#iucr=)PI2pRyL`HtSN!nPgMEjg@?mnK9C7y})=`oydGuZV$#QPZBI=b1gZJAdY@ zEE?=#n5S?KP=`M?3^d}gSPBYHaFFw)RwShVQf&5Rs8`roHoXgnsGb!o2VNxAykY=c7n8>bf|OmDb( z=AHKz1(~Hes`?Gud>%uAv z@?f|MW6IF}N~WQKTAV4Fxn*uRMqK>w8&HJ*^Jg*(h$h>Tz!?Z7fapE!izuX2{D&KK zR=RuAM?x*e$Z$c`X4Pc5{`%yhH1UJWt$IA3jasYA(eX==6ZEXDZq7e}@VK>P_Gkci zC_3w7N5lp`>I1=KrFNwOgtG)?=?jGWargCXIhuv z0?}^d?$6f>bV`bmT18-E8@UTkPWsRDdWF(b83|RGaNS~{EZ0^IpnBrO>fqYntN&2! z!$)$0e&g}5tsQZY)wOGqi`<`cKC@Vvgf(-NRMC_zX&hZX*PCs$wwNEyo=g7)r<=L} z^nX^5RCJWiMduB(5H-(pP-$40*_mAcNBiE>lz5oOux?$yNsY}|R!|bw%f!e?Oq>Oa zg~@0yVQk|BJPtz)PYDSLDXGH=(|qYM!pD8xu)>s`YbC|jl485JqV80|4d$HGkgUTO zt(JHZ7{q%kxrEv{2HkWUoGnM&>vbkuh^I`Sf3mquegL}!TY7qWi&NWM4qSCWNN9bk z%ER={=9P6)<;7NCX_g;*NM>4<)iQO`*IQNnL(}mQP(PKs4W<)Tp;k4yS%n~KkDD5` zWQW5lQQ7nrwD<3gN&=OqLM>zMghM8#)Qa-9pv=N_(M(BHpb8-X2@V@}<4NWx={}y! z6HIyT$${MD;XDo}nz0aLEC!O!xL7Ie<_INvZ<_6rPar)xL8$wz9-V(dBMOu*zkG?V zcq2bHrnIhy(&UyP@5BwzkUt3o#3d;9D149Pj(xjD(cs+Yk_Lm#J3KtfZx z+*Q{Puh0~)pgM}cg@lAP~8k_T(1&cnINW`~z2>pBmc0AOM=Cg9gv1L}E> z2`?V!9qe#f@V2rB3gtvgbZt53@f)y_E|1#=E?>S*t5%@MhFEgNP8(X^edtI}UyCbQ zT+L|?GpdR<<(o>rfQ%p(TO}xjY)fw)6}ea^R+2blVqx93#(WK9;=+fa3*Q7xBb4Cheu7e z*P?t)lR5$q6U7_e-Ve81QYz{uGc8-=ID1Pyiig6+@BJ{c9i_y?G{rRGVsh^Y!)-u4 zLHBv9wN4z(klXncU+G0s{`uXVbCEjS_|cF}Vis(6_K(1N3_59R+3O!nzpicIaP+iU zr=|Jon~<^AB0de3{fB+owDIQ*4ANJ>s@h7wH)zen=dGkF)s{cAE5LF`U94|jt`@}5 zm9;!A-+E!Qe*LyEJybuFl3{3&_j^Rd9TZPXb&4GdaYK{X3aF-%jO)3eR^Q49!Rp2g zv`9-WF$IRTw@r?gT+O zoEFn9SDkxy1Hy|ctLb>@^GuouL*X|5;i)-4?QRmOtiFkIZY4Hu6;}WJZnm~R$}fw3v!f|O8`ZzgEgv^OE-qaiGHe@ zdyav*jMo;r8vf5|&6<7Sk3HdfC0%kmGS=Q+5PA9EnuI_~ku)EF#s(FcIY!D5iR*|i zfRVC`7sFj;N#nDk_H}8f%YJKoj*HNuj>PtmlvZuJtf0LcN6+^C1#GY1gL0{Hq|hQJ z7(xoMCpW|r3%Z)CIkuU}nX|c`jCDdr{^|Yd*Sd|L0!6UA!DNMdrinDc4=GYdyRUPv z5PYilz1c`(}kqvcsW2iuGQEo~wuM7^~~c*_{eFmtrb&W2aKzN%)Vn7b~S zYptzx%=3IYD9z_I3fad&maDcmkNPM`>Ggbt^)X0Uby|D|_Hl1AG7Nnl@B1zzU>i{p zKv{K3Tlgjov2Lq*?xr|xt8C4yChv`r9lWRPdD%AD^srKVfo4ylvE*{=6E)!6Eud#z zGhNgQv9ad*f>KY@SG~%2wiCBu0BmWnrm!a^j;B^H+S%`a#dourKRD^b9U>U}Sa}8C z8i+ArnmKcP$x&TU33>m`emPIR-G8okH!^DM(P$FC#yZ0P4Kzo3KO|qlcC2V^G64l^ z{8el`yDh#q^)j`5li61ARuaHgFwdVtlT6T9pHBskGbK-uSubVIJ=4& z>rIyM!S0;wCD~_)@b2)jn~#)5ib2z)5HFo_3v$nFI?#0^vBHjRTsRt00*n3&OWtpD*mzX%#TA0_S5`-0QBB}E z`4VK@(c^a80uqY7N?QEw|*wkz`08L_wP-5dDbN-s0{ZB~|mt zjYnV@0!|?(gH}ai(w_G=*;2e#IBvUZe>6UdS`Z=P_wW9d(t$UsKIw;%>9T3w znl+A~Lp|tYDJ35oiMi!FM~BGuqhrYD5*mf;P4jaR2;x)I8O2u+u8j_cVKKx!-Ia?qS{Vb)Ugyh9tn*yEB|zXV+Jd zr`rd-AimtlW!kOuo`Gl{5Fqh!oO@2~e3B$xkR&nVYMi|7OV01?(?a`?kfoHNChwzk zzuV`CA8Il(SykY1FJ}>i13H9gB#;QtvxmE(#(Y-Xhgz$!PZY2HoMIL5C2&FDqv2|& zUyAn6^2R5Sn+rNaDYACfP!`3ElU@g?*btkwC6(?uS^Cvab>|GSqY)fF*PR9;O^ zPh-lERuXiI=vm*JvM&CUq=bHY20Lxl3FW9qZzw=jBmFC52|-F!BG>YEMxYHo_;Nlf zDtEYYn&H0W{>HDy^0=)?6mfOgI8N)oBgjx+pK&cRf%aV|+E=B{aRUIJ^VBUBk8#7m z7lccicCvrTb4PT-U8C1mkeSM3hW~Z-5PpTE$EFr3)dtlLwY7=|5&Xsqt<5wk+jJ6LcxFobgO;NfQ6= zK5uRaZcJ-)tSN!$a>tfU5#uZa}qTCtfC@bpNrqQvbw3B-J`{y?0xhK2^EzTXBeeWMxKmTE7JT~&MPD=96^bFRFc zZ}+7=bCbMXd=7r0Xv_B|L%ompjocT4%Y5a)?OF>%BO+U-!LF777un|u6 z>C+Lv8}ua};QqLW^=sGd$%Or@ol36kmkghaP4^j_=jaW#eS(!A=A+t8a>yz>%RYA(TOGMfpj@#|t3{nY zS2JNa`v-!!!>V&P3t12t*vcN}&n2yt+;(vMCl8U+)>G5alt*-f{_02pH^lZ_#KD}2 zq}lUH|FoidvGoX!j*34EPVA9V0$j!K-$I~@6Vsr>P|1Dd;apT}r6xo3MH&eyD7HjN z6Hq$UPufzC zg@)=ou`!)-8!Rc+whC$0S`PLVJHb!#W6YqBY|z{&N7))8EQ*(ZFEBAPv&BNV9xj5( z)v{?+#UH8a{&NwlpD0+j&Ut`xC~*>Kb>3-vKnnMUcxC*Wo-k#q&mm#svt0w`uFTMZ zOrVzg@v(#I(AxMkmDlqYRM(ItjU$tI^8}{03{T-6uwbCEbl9KKo2&Q$^8wy`aan(R zP(3ulw7QmQ+C)xEwH*XX;k2~0-18c+dA3}1&M5xnLony0_JR6@+gtLRRPV#bAG+x| zCZd`3Cnv8t4z0g^+pRxWb#SM5e_v2Ce07fZ(8Dg0R;l;05kg^0kXHLFX~S-|)-yB3 z5WPfadwuKR;Z^ujjlD<>-va;|R_$o+6VrT4+KsO#Q7I@W?CBz|wqvdkd!L)prZjH%Gf|$G! zYH!5}hU6$H$w@Fi=W#F1w38*VgjUzgiao#hc?gKel7lt8FYr zaYXqDecpDB1!I+71w%CW!aD~Y_@hQayn&{i4^i>n@x<@UQ9 zSaJu#R}RoPPY`;>S{^`!pkt!JTD=7vmX)*06_cTj5_>`!4%jn&H1(cO@`YD>nd9Zm zw(RaN+dDhIax?*UMoQQz5((RNpXw(q)s&5QoEA#Fy*(TE%uFq=dpt0EOaBQS*I+z~ zn0Q_>5DLFuP2afOpFe>!?|^xSW8azb7|ESY*mMxuezX064P#e2)Dsp6EAsQYq>WKi zwFNz|_11RJO*c{G46pieE!%B$x8{L)XS1Ox`w$v{`_ywK2lE~9Am)A$Cus0#$>$B# z7T@AHkGQOi48(pv8QKN>oZe=r<1il)vREJXe0CI4s#{-?UAM=J@B15=X4Go9S{Nqi z$#?8gKaaH^LaDJ@@H`gu@X0Z*6a{lQR?LM3__3AAhP>6X+;mB!nUmF@yo{Fqh_(p&cS@>YrksY2>Ys@ANtF|R!!m& zTY`~GXSy;0Gqd-emp8sU<@k!-n=rJ=0{LQA%)Cd-RC(p)J96RL%LNKz+s+X)rrqY<`Eq2=b}?12Lmr!R=B?B*8UJP zmdW@X);g9-$y_I3Nk0NwjfkUU>gNJ-HqIc`#TAr%zDtHoUX~w&&`dh!NH1|B%sSj7rcLi`WCoAy|^?3 zHpiFC3Ai23;v=35ceK--VsU9SDD^}Jr56+yPRz9sU)SIj7XKl#_WbE7JFwt8Wd8-b zobCEzcD#F-fs9On(K@SbEi<=By)rk-xU2Ji6^K=A9-Bm-#u!E$M6e7EJM!{&!BOUQa}(er60jC-dlc;D z&Lta}4hB(VQa=mH3$BfhDJIpz9_VBYdR&TUOBANPZsKX08+d@H`-CcwrP%Pf>P8He z)>PO|l>_2|S`^`@Pze;Rrq1ry5d(|dsbmerjn|G7*p;XR9OS!=z<=j2`-)^|C$NayX1w5}PSD9*Do$ztkM*pc+nX9nX?)=3S|Z z%`pK-^;9c~Gbyi=)qYllR_>ZoRwBgH-bo|Q2@F6k`cFk2Tb=f!H`%*p=KL^Y2RBRo2Xg!Ux(O^n!Sie6eckr8qy>eu>bFwH<&0n=79NJ$N-W5B`K6nPe`?EkYA6 z(JdN@*L7F2N|@<&PV2M!Y}Q6bpS}CdRBr9guO4Ky0T1N+8j0*T-kuGQ5?cFi#E`x> zl*ByO*ra`6Xyx*FqnqsZfb)2K#})f2dr|gT)%(_*?|buIVVCFR^;R>KF!gWJa|=~^ zw-gdCJYOz)JEHl(3fTbdcYs`wu{f$BC-mTa=RLvxNCLCp!L=--^QTm-i@qL0UL?3; zu>=KxoBA(Aa6Fol#BV#DwRoXkTv}Rdb~Gv2L1p(82o01@POC@~H_yJ_sk3N(pDTnw-kknXfsTUYA``v=Z9uY@f z_~CbLkPnu5ozkyaTt9j@IJ-JyKMAudD>JWf^SC(ePKG@!}rJ1GvMxGYjm6SYPv>cR}lU#|fqitybmxJ1*V&=gOEJkLmVE# zd4Vfa{{?g(N$gfSNvIU{#p)g)BCC>kgS{3L6O^Igc0sva!)M#+PjfGw(XOIwyC=&< z17CgL`m&H1vMoE%HvFxEBYzliZ@k{$ZsR#C>jWrmZ;B&q52$Y#ZxCYl^ENa_8H%lW z34sP7UEDARNnI+S3%}r5>5Fv%-rzBAnyu*IdJ45wo5LcDG~B442)xC-R^>$P`=J;= zYEiO)D(8DJ*Qn0twh`zo(gHOwG8)YnK3H%iwR_U3_oz&dhJ|qjNJ-CH_BMDVh!QC) zDG@=kjtiwuN!r_~)iy@(pnIw~lP9U>mhT>xPhfKJNya7wNwqq98Q`^O~I!bugy z$;eiv_u_0$7)Qsl~Un=q$}o``=yPoSdTY3`nRx6Srh{K7rP+?+ z)2GSK**uZ+?gx*04`Qx9pY~T2xhyzO<+BKKKMUn2$6wB}3{wmZmLdY{hyy z<$i;B8@6xNYJNacBjd95&6!@E#=v~or3087lR1-vEiA+=VQlrsmdG~;Z*mGPzbaB% z`^&H6a#>?J;kDO}Pp4wN`N+5HT%?r{X+V3rXOjXkmMT3OhNJzXz4Cu&{-*IqQA=t! zTTtMcditP-C~=ln;R&b&k;hH%;c71T%ITiby?gUnQPI%naEQ?a71g=H!%{=p5pXg~ zgOA1EYL3l}Or8^JXLJNzs8>0mUY7!YXTUV0o%C-rPLr%(#V z-O@65yJmaK)T8bm;4Ye{E21^OO2WcEQH7D|&<5PyWxJ*cnLk50xpTIDEJv=(K%;=ie&K#f^bpi>_rjHFVW2GbAL^{r_HBhcak~j8u}M6S``3)J;Z;Tgetv~*@&B}r;_B~im+h#^Hve5Q7*;!;u zfzYlQq#8~B6{u|1)(pThIxA{&lD}li(e!C%kR0_GSF`u^$ED(*$UnEYO_U3j4(GG6 zT8T5N{*$Qf9v{E|yDZJ5+jBy?2Irk;G3?Y9Qxg+~fL1vX1#T$eG)~^wa;xbE z2!6sy_@yN+EX-8$719CQS}9Q5~HCV^oAdmeiFt=4kqh=3X)FS*^5)REj#?l`PerpH9TZ$G-=6ANq&Jy^d0p#+%Ij@xn{w zOwK@Y;PDx_6!-A0dEd#I`e3Taydw;zK_;s#_YfAR^~hFk9{y))O3D82$*UU|Cn_?s z4%qVzmz}hmKAQr&s1;$=TF%a3{AVLQCMfb>(!{~+tZ&P4O3!{rN7G;1Iu*x*OM&d~ zcSs$iq^vr&T_B|6a3yVCfPj!4&r}^#=_X$*6N8h_ez6R^l~&t_#a8?kboJPo$Hk1$ zz(vRh%RUKDxDBVx(nbzCY$Y}+?D#j(ME33-TCc}L-p7 zNtqcLhm-Z>Laz{3IUC(>@4$F25o|%ZKyA5wQA$v7Z7RtXOmZ4IpMpKzT7z9!K%DG9 ztAhJta=q;oh4oy)!NKa)_lW=5zwE`z3q$g?lKSr!+T*{VyQnDe>NP%s@5(V4UTi0i zX2I$~WukAh91y^zSFbj14A^l3B%Y(VM z#FLtmidaA!BEkur-CXJT$oUwIC-HmTv#lP~=H;1LGYqYFh0JSca?Mt=vVC=3P*{ES zEOWB>cfL62PzIv!6=r6J>Vd1Jy-agxNyrFtpY^C!!5vS;xIuktRU#H}UG(0y-Ob?N za_RrJK7ET++t7l5hvomfFyP|P|sgZeI z@N3lnA+GrTop+t}-%sNE{l(NN`1j>d?ns(+h?3|RgnQT<`~!%mffKmsfFbw{;WweK z|2i?%vcQ+kpZ+KZWZ*Kb!}I^WX(X-h+O%0&K}t{#;l7L0A<}CI-#9WVN}d?~-`VB# z{&16kU*ZGhNia=+Ufc7V#HC;&=w3{ioxpVTexfm9U z0*0!`m6J-B z8i3uKgkhKBn3(Zv2FOSK4&eV0_a0DDE#I3c>QxjJt{@;;0YOB9_AXG@&CFZ#X5Kq%=@RL#Q>SXz4&VOvKAZ;t zmGZL97QoEHgbuxUt+Oy%zG?pbC{q>Qxsb48oh6$GUP)t?G# z42((HFrmJ!lKwtWj3wl{@4R`)6CD{7s zMZ~Fx{m<<&Fi`p1X&Vms)KFc1-3b${(6s-<**kS5dR;YcK_>K8p-MFxCK>vW*W@WK zy7t-k#NC0xS1<>G22n3J^m`!(fsucY^*&_VxX#w(-Qv>n<0JE(F-+OTy0)ThCD$Jb z6vOK3s0TCC(@VKgQBj)`y^2bMm(_^*oyu!su;yAwAyQH|T&(&f%){h}yGzu+i^j3M zK7al>lh!WijOL}M&k!zS-3KfKEqweaXssa@z#DE`dn(GxOe!UcCnbzhzJ4AmpH>iAyNww43qcA%UhYM`q>$OQ+r zCKF%9ZPFVu`PDb|E89F1&wBBBT7DD>g7>A-;tj-x&>=GFiE~qd2wN-}-?jnzI+mTq z`VOkznA2dW`IfVN4%|oyBAB{~0mY&+{E;rF?eFW-6qOX`r70D1q+W)7-56c51@h`B zIRNy!titn;(YdLqB=KMWCSp#y;bNjmhH#HC&DIdqc7Z{>GuwQ#jbXPee#b@ zOoJC1>{vbV@$?k4SrHOdQ(M{`zZKN+9ywf0ig4Z~4+PMcYO$54jGgJ;o(&A?h}f9C z?PgvFjD8(8m3pzEW|_R>POUyDK&7kf5OO$tJGe*>+@Z%RC3iHeYQ9>6lq-|2(Rf9g zPf6CnUl4xJQ!-(MM(RI0#fwDM-O)`Y)f5E?Zgca z{YG%rA@+=pHZJx~Y(UY+BJ6tR@=BfG?Coc7+uvK@jyl*`=m4-PEiEmt?KS{4eEw6gh5-I-qqC<(QGc_6QEXyS+jiJb^bEkA8EZY1u7|S-P-Ml8|~P8 z(=*oA)tLg&A*c3`I(g3V-3`)if z-g~~$ok1bpBsFTt=cc;J+peBz3A74`F?N07QxoitaHaeQK{MH6fwE=ka z#m`4YNp5}suFpWl>g|@Em6es4lL}9+a;ql;@;*^^gyWi z#fVM1p>(F`quC|wV+~{+Bh!o40?)`AaprjNENXKsIOW*G&0%dj`vc@>{nUxi> zTky}$8LK41C*^b6u2ff6dd%25Jv|+dblF+*5ptEZZr(W#QBhD>8>lmC%4{Rb8Aa{Z zcVxaxS&X=Q?D$r(=u3WL@N>QxkzbGL&cC~Rw6e6LjFOU&$mV_}FaI?h7w1M5u!+fZ z%%g~$*5&hhbj^IojE_;@OrO_a+3-=}**v+s&@)wszfS#pWYb4VPeGaeXe-AGqAy#^ zc%yxfcQ`Vt^^=z;fTUxh!@|QUI}sJGjU}k7Ny%?!yLroR4|<0wFUJ6lrPz3-X5s>et?^IvkA_WQE?3q^At_3;YGA8^uN7nN+uc^JhI9m?4?R4jM-Gl$P{NdE`d@Uz z@I`U{{@nOmzV3zdSNfkxNfczOWvg(BS+n&30qM-!T*w^miMlJC1lex@Y?KS{v(O(6M^G!7?Jdk-QBGA+= zI1$|gx~tdR4gk1bHy@iO6E9@j8%wJ8im5p5uvZZ8u&T*AsGFkCVgeLsi)?2{hXZa? zwg%@(+X?Q*IOOf?t*rQpg_AQ&Ge0Zvef7^d4fYIJV`{^xSVy(%Sz=a}X6B17AX*6@ zS^jKJqu)wD;*$bsOaYY@VJQPayJnlUL~V8S`Dc10Hrn|F{buhG0c`+zR!-`Ew7ol< z!>-MH9)jlZ@L(~v>IrXWJkSlqcCYq!zpIPEi5=qz4<)bDyQU9Iyp>`cPEO@oy!Ml+ zVz($+ESEt`?#B%w0$@= zlqoO1=yI7)n2(ebLl0GeC29$yKLP;_=t!XgSEq+h>@O`97_G!rAbI7)wKTS<%qKwP zy#gsjskq)DA=y6_`})2*E-sGgG8qP%>F!?yziyAgnr}SfGNlo(L~7?0U!kj1EPqt+ zfFXo2B{7ky@~x&O>d*-L)X#qU@XR8*$&dG)HUxU7VA5Jt?{qCRXO?#6UZGy%XN zpu#{=v8cwy%-=iHJ7VG*<|nMy{0u$F_IULHsGk@qHfq<86SV(=Rv*ap@4V)Y#QCRX z%)MZzVq+O)GMCNG412-&#pl)`cL)G(veZ=^oFbxw4g$R+x}M~-C#I%Gsn~AGuVJ8! zhKo!xQ&N^37hl*Tm_WM|YqkN?(XILc7z~DJ%@th-be(QQy$=DTBec7>7td`Qr>X8K z7;OaVcR@~#e3y5oE*Z{-a(<+bA#VM;+n&zv9Q zp52We#`{k!r>6Trk;Lpga8jixgeqlaiAsEvpMP@Corps^wDtAkU%l!H)bQg7wjrt= z?7w^H0%W96WgzPVs>Q9X4fFTlgk9U$-(eiCT!beVGh%)D5_)-i(NR=Rt}7{TX$2wp zJyEB&C7eOQs=Y|AXCa5%X4O~ITXVzAAL)oWDa&T_dLwQuW_=7IBc#yLeQT&xNR^r=%yL8qv%UsbC$LD=d;?Qv<6 z*%L&u;DW%PpCGdDHQw|n=Tz>?RAhO=+1u6c@9SOey0MjFY`-|dn(|`6BaBf=x4pNk zXFi+T=3J>u#II|)?cmhJi%KdKTGo|MhXX9}8Rg4ke1a>i^BjTt$r1LrT2F?fyahCLX3GWR@)Kfj)U$N%vZmTTJ5uzS1EH>E7{q(8KF z2e|&_rfXFDxp6uEHM&8qfp0J7#Cz9+Bz$U8$OfXnfBh{;5#E3BKx9^>xmf~W45K#k z^76XVvQv5>UqEnY^V@&IorKA!wv|$(5VEeRXW{wtx(aTa z&i-yZv|U$sU!9or7g=#|SVuEOg1e}%f!;P$o&S6f0}HRn|BM_YG`e65OmK9*o`s_ph7VE;o*&e93290B2lXaghG9$-Baj~bmrs7A&d%miU0a8OF|*Zki^TEmJ^UT(@NaCzf{$uGzPL{ z@He1w=;`8r;?X;I)BVG?pObkGl{~Ypyy-|Gn3$P4J#E6gt#RebX^R1F*@lqI!um&MH}tmEcD)ag=~$BG z#JHK7(!>?{e{NL+!|K~&wb}kjv4yReq4{3$`on*I=YAQtZ)@o5gy8gwrKV5{1C;I=B88>=fq6S)aTFQoMO+O zeXk#KTl%mBxM@#6#(7b+_eS$k_w}!Ues71lQ(2N?l>PJR$ph87 zA2YiUc0>6Ko3`ovRrqv+8mAxfOl}6uut3+UinXiW7k}bvrX~az*_pVOxj&@4#DfYh zL(BeUd&zov_vrBO5ENqj5ES=zb-5j{&h@Pz4oaOLJ$jg>%1SzQ0n=?_3fqFVazFdy zPQ_&7%8G2(zX51UpF`{$oNGO2ywn)7fFU>;NnTc$0Z(vnko(FNf%ossZg52*1YNet zd35&O$C*quIOpg2fjj2Ef=)L0eqRT4k^=fK^ihcAT6ez7kUWZ#g2!@y_lO}{xlPOu zg?uU|;3!`BdViysizJJ8JoRmB4eforoHQfPpI_FHME#Pvb`us|ulIOndZB148Osx; zj~V1h>U8V~&a=MT}aPi&jsLlIA&8+EaiJk zuUEpgQM*I%Iol*;$OZDh_8y` zwr&D|&7qLZ;@~45=z-qG`!}WPPle5FpR3qngT@Ho#HwC79^ElJ&dBBckUnWNR&WE! z6lY_-WTHLz_{o#54k3(x7Iy_0mBTU33Y%}T?_!p-CPo%=@G0w-jU67^P~r7ATRw{|ANue)ZYcDhX`zqE`>$DJh**L}Svy zv=c@%q|bRTy?K6gmZRCRH!FAmsdaq$F)rZRp0pI)SjNk@^k{=(J%?GKP?a{A@D3q1 zb6qQ_=)`s&#m8$;yRAZdUqd_lc;#%@$!Y81z>4sTw^9=Cjt8GmKfHjY=f6~dlPY~U z-#AfMn>?YpFLjpiwC?cxPinZ}(RP1y+C;^qvX58hL5{0V;#8NT<5h?xTlrDw{z}gd z+7oefa#WPJMVn?RE6#g1VeZCbFQ~LwQO3($=_)frpfNT!BMHmS%7oVEj5l1ed`h|b zToLP4^L9~jk&H1*Tftl9A1WY&2|n{!4y(a?uC?X4kl-E1w`Qt7m;O5w+N zd)Ui+Kd2AND~c7@GNH^&e_tNLL3p%J;hUD{rpd2L%gWy_%r2Sx4J?fnMi&OpRCj*n zTNwL+i5D)JoUdBkBRxR$rO5s!@@IsN!&Og_0KvVa+Tr6h0_G^iebxIQ#1F5RJdygef3AFBxOvn%1$4SdMKxRb3EZ#qJt8*;QJONuP0BrdefZ`ba4kP?;#t5L7SNQLHTaig#*%xX%7S+#?;CXuR!+R_Kl`Ui9}4|W&(Q-X65Ow-HIs?Axgvofnv_mpyk%Y5YI zBrG<7u^hBgI6+0jLW+y2y>J|M6FvQL_6b6jU~UI1)uPCohPL*`A)s-tp7_g2ab)Nx5=^5ea0CUp(t*XPr5Gt>fHw=u76?6WW zvN7JUohgk%18IY&=XR~IeryW^a!;+M_%VIsjcD(~(7sdzaGX6?^99KQ-`}#NvQ)LSfg#xy;`D-W5EzDB<_TSs;kA~<7d3Xo0Z)=d&;|Gk6 zI5b24)0m%jv<%&s#oiwaFqW1xe^kMh(3k$?3#E$1{SR_b*6|iV1T~ z_d5rs7jPNW1o;AVmIfsxnA+p)H?~R@D67wo$q{Wj7Y}LO03OmEBgLH$>XK;v z&WtnLzUDoQl;XAP{Tp)imq;1Tb6h~@{j5X@*V&F7N~y&5?n)(+-z9r%MBjs1ue+7^ z{d>leTDZx6{9vq%y69=x$CG{Khgr8~lD%-f1xXBwCN7;ikyiJO!L`-Y`fB%Eni(FG zxaD>Vx67^|OpdJ&J||wG&?w4l`ipPmxJgvqtz>PukHMD;^a6vJ85w7;xR7_hV+=Fj z9(35v6xo-0=r1?0OcaWn@L9b3;AUUXNozpQ}d8BsVmIw6wYI}g%T%kmo z4hz10ja-OxQVk!CDA(DW54-HePU52%&42d!bOo1{(EpK}j{c>(?)Ttp1}d4 z;}(WT(VEA72PjN=tYxvsuUmI9 zhrTTGfoNG)ej*EtEwx7h`Wh!3_i@scLqyPDQ!L6V3P+)XsKg7-0u7Go-Gxh0K&u}T<7~_tFo@oojw?_C1*>m1ZYIc5wGDJPC%Vfg(>EiAx zg+=J|g_$)SIoK!{n*8}`p$76;ZP3Sp3K+*yPxZ=s$SIXOg4z@=ce*mO@FZQ437X}0 zNcfbzvE;3}Lw`DfWVh5@51>@TyOO?N`AC0ZAoV|=OO6^RR{iGvD>$MJF3{OdYQeVm z5L?<^nV9LZSG0n6Ex(B&nBEJND-S#bJ(VhST4VynZPc9e5ryO&E}e-#sXLgVgLS@n zo9(UUHipA31Y?R`_~_1f!&n*jfNV^awRuEwjRaHRXDzK5(v9Bse!JG6_jRVam@ALR z1?<~zZ5}W@VmfkdY40tP5`!>LfIifo>)c!F(=rSWQb?7hFur>>yR&iNLAE?xzme!{ z#>_;|p*cId_?p|f`rZ>B5$+1X_X39xwN+4>ijuUad-kLsD-;oy*-fs^rgUJ6Cm@ zfC1MuTz@johirUwbO+lM_sSN>M>=9<+uEC)4fc4oen>DOEr&f!D}sD!sy2X-}jOEU5D@c2>) z9(vKwgdtDmVvje(B8EqX$}7r#zAYf6lJUoZPdGV3b(DP5G7ozD`mP?@M{ya5s*5U1 zt==#;hPMpIBI7%@iQ)!J7-;CYkjg3D9d11iX*m0K@Cg^klejBly0V0@$HgT<2kS#? z^N|YtV`De1J&M1D3rh8M%hUd4YI29sH(KgZEVtQg*qU=N^EW%}&4ETwf@qC;dAp|U zt*(4JC=)b1l_qKVbpBu%?*7ld^=aEfm6et&Zi!WT{ogomb#h7OW*7F19|%$!W3XVV z@i_XNZG@7hNjn@?^X0tWBVZ8X3I4j%Z?}Ui&W9@T}Dy z5EoJ>XXc*h)h-I$|0L1U%xpi_yW0xL8tu_+Ggijud$Ic ze~k-%j35J>{QHtRSNOj#S}~xDeqX`KWPhT1_okO4dbvpRabkSuBu|etN~i|p{~kI_ z+th+1>v-fD78W*knO8QF1tjVtC9nS;0b+sH6#P5d+k9lD%=RN_@H~r)bzUTQxm2!? zxF!Z3i|=-oN^95)o&0$I9CL5s(VqyuRr4fd#yr-w?%P%}Nh9)T>!9%E!l#?d2IdCf z_=00`B1~OR)s;m`d+R0+APAtX7pgse>@Wx&qU!9?Jxv4A*cS=1b2yU408HUvfA?=~ z(eK;d-aGV0!Uo$Zp~urfK9I>6qYXu+0a}`kcICsl4`w#h{B1Ub(rpaY-TgpQ{Ngb> zRb>*b#<6d!l&pK%i#~D3aEHDU3p8`=zABPPv9q$d%b$>|!_U8Lm@P3ky=pVF`{g)6 z@sXCrNSiQ3N4x(*M2~>zccv5Lb$e?%8S>z1X{XFxK255?yA87SbXaQ%lY)3#55z`< z(!)bFb8_H(!C!xP{=xeq*sSRrVdB!ruTl4HQrV*v1gY7?4EgN0pLk2A$Wv|{6PvRZ zSe1VLz2+F!`b=|&Rri6DbpQc}_pmIBNV3JFzo{v4OC5My!W!!8?f)7)gNMvRA{)o& z^KD)QTJ|m9J9?(soz0Alq|60HAva9@c_S%@J5hwmhN!#IF6w6pE0j&dJHyO5(MR75pQH7=?O-_KhnP z)4W!LPf0zKw&a06SSHaE(X*Wq3GLf^=2=_r>6fcjRr|0298@2FXG*36qTjqAcj9+v z6*JgAlK%0&M7+V#Q5m9>OH)imKm`4W=H?05X-X~c|GCpVly%y&bw4qX#IK>Qtr!-s zElq2)HceNw^<}|#`p8oRxSh9d#FqVM>H5)ghpsw-!hA+3@j&t3x3!!1zY9Cl3evOj z4X{)jn@4mym6lyQw4WCKM^x4L!e?aZX!I6lzr09#IU~v$-jgevYj!D?LwyaZpmfWr zAlLqzn7jheWD~7h+ImX8|7;Z7Kzo4I*uduB18&c%_(!#(TW)`OA zTE7wdfrynuOemZAB5&F6Qzxi@8$(k{<&;uJMn>}-mvf$i<=v;mUb!F{W zsx-k(l>%d)VJm;Y%Tn4R10Lzi}U_v@;e{S|(n|McIC|34cZ!uap%v;QsQ|M!P~NqjCKK>xTigoVadoYf?{p5S_k z)q_9_xr+)8V0NqNH3GKw_HQ+ugH1pXib);tb1p2n!Ow(4O--Xhh+UrQHYo4uf6=I* zq=oOan5XzClGn18Lm=;pm$Wjx?-*gbTn^p8g<-*%^h;IO*e2#3P7gCfHa2ku&gd8F z`{gpa-A54PzXr$fZDM3hlck^wmADH4d^G*<&$GZnm6QyEneK#Z76rA8X~tF@+6QT< zFNe=Vb3_ddv38brWSrcChV{TNA|Ox#O_0xLSE-h*932u96dOzzChlanF}<{|A962H z;xdZwCCxPLfIxf@Sh+NLxaV`M)vmT~`Tp7O*pV4ak;G~|k3xsX+C&Bu{Dc!VY=8|K z@~S|tu-o$Odn~IEbGUp3P=iK0IZ)KbIKS7oHBcFTIyE4K9xM?{IX?Lw$I0n!gx@%d zw1`_@^X&C4hN18FR*215p*mL1T3dl1xzSrn2BxqzE(YGcpua+m-{|jP(lzF#v^PKR zAK=Y{?!#TN-oe_v&8HDvu@T{17h);u{Kc1NzYUkv>L%-)RZG(hB*y6W`hDjFDINt0 z1*45eA->(Wj-~OwiSOe%j0-pzH5{LQo8EcDp8hcqUEH@p@txT1?-71^;ds@PJ|hxX zx|~#+Y{f_6RYy8TYK|5YPgRH(T31JbMSP9}no16!7KX$$wCx9~^JI~Sa_~N&q53D_ zOb99I7D%e)=kcx&rAJ$P9FJ0bnXSst1=!hR8xzwz&#IhoS-DhU}!wp zAWbjMyN(?YCQbu-gZ1YwPQErMx5mUsaQ%H)YREi2L~g9r9gIC$@p^guT1m-Pu{9q) zWMb>rEllHNU%vw)#y$}o5emnR3+9MUi!B41)2 znn&{hsEZgo!7YB}KT)#cuAW`3(_BidKzFx+E`XB28P%JZo(x=DvlwOMVn^)B%JOo# z{T?(O==@_EX%LSHY8i3ZRIyu^jMLV?vqXN^@P5Hu;a?am`2S4c658$Pst621PE}W! z?OE8LVj>`Zy$_ah&dopMFsa{6;|FV~GOR0Bik$lH6MD&N_sI?P-7tOUl)Ao_!S+R% z&$L^$PQ5ZRh7SW{?L3k?zvdUmuOet@Xg^U3KB{za<-g)=voeylkoW!6^o}=tC`*dg z*3ym@%c;ELhN+%jXhDgxnbFX!_tB9E;jNxVhzl{7nfXpDEg@ai5ZDq_%R;r?9#-^g$uq?Ni#AN8Drc82f6@Z>~tJXF$P|b=L%1tDmEY& zbMc7!Gq=dC{tAdPSF>1jQ7WEyEBegMFo_+brdYi7A{fAUJ^xMKsAK&S?id?18Zo7w zy+8Pct-RDv`gE4@=xV*_5zsRPp=~nVcm&p?5Aldyr3@=<*G6lf?5_7o8{&$Po9-E2 zaqc(p68TXSLmSvj^uVL1tiYO!vN%&cr2w6x4EIDp!ffIdTFq>0U9YX9h^ji zsHz+Ru%dF}DbQMfyjGn{Ci-u}+xYm-x;rHTg$6PlfUyv_v^;Gxvbpk}G}RZv#ttB4 z1s%49zsz4>pj?_+Q&UwX zyG`z#Uo;90OK58)n^fvaMgk7qbbd;%42e*8-w-_mpE*p`Ltbqp#X5~1U$odl9LPYt zfOYe{sWTGg=4$A$n=n}oir?nu2s}j|0Dq0O&}yLaEGmbL^3LF!hTbip>)p_8%);}N zms|<3R!|?34ht7pS*_j9rWD4mWoOO0RgW7%#A*!6ZmhbC7$ETY`;LtCptwf+n9I-Q z(~Tqt(C>5F95>gGfUgZ9(bbGc!cm*7nzIu_ z1O3X0(8Z+Z$}&RV@3u>Qw&|g6vY3~?|1hwPN z%`B9M;ENB^w#skq>e7xkGB9Ry-R1^Q2jCg>NmsPolg0w;9Z&MTWeOG(i@X<9MlN@$ zumk1Z0$di<=J62P({czDW@ZkLkGIzk(Wl$no6F)R&%hi`R!P(RgsfKID$B%)CR^Iu zE{(w}#Fsi3dZ2M;5s}H6EU1kWZXoHW*9d8g%T+da;Z5?W|9R8Ni zxZvPe&K38idhNo7dNjIU=t_(?MoVhKEAz{(uK>4Pc52<>JpKd!L=2Ypd|6;x9VMj& zZL&A&+|PfV_&c~omsra#P9<9kH@4K7wr3gL1QE^Io!=`zX@2G7$I&q@03ib#rEjWN z|6jk$`DWSS8VENOCgpO@Z*_kCE3}3%{vXBG|HpjMi|2Q5`tOY|?{yA}h-uLk_I`@AUSDM`OxV=T%0YFQmN)*126F{kwk*@| z#YVgtPP8@U<=6BZk> z2n&Kg!|3OXHE+=y5$EQjUZ`n|RHz(O%ukRFCc|G)HMO_l3fQR9;1NH7wvX&1)u1we z96MR2YI9I#2mF(0^k|{Zj{gML#wrM;zP>;GSJ2*XHg_m1c%}`8WKJLO4v!0xIODwu zGw-OF@8E2fdfD>Xn}^9J&eci|Bpnmq0BhP2YmM#WjZ4Oh+VK+l1;Cy!8Ng2Xn>p@3 z;i2adffgS-ZZf$tR`cBc^v1IUpm^o#mE$pLJs$g`N4=w{%S<%CSGgvpSe8fd*f6FO zB*`XM)-rqy*`wxz@H@2r9!9%56W3os@$n3faL|0S_)mD9~6Ba#`G{Eo(?olmH^N5LljEIETl6yaNfntMb_n>N%fMD;&{(gxc;*Eg@WtPdH zuyxq~7??Ph|Hbs-G%{GSBGy_?l@8F@XN+a$i<4;9jPoSdO7$^4UGD3-Pw~)J(+HL^ zgO(L2U@t02O^rZ=GS#mKwvS~?25`Fqv%(_Ja%-WeDuObib% z8gKDdgPe=A9X{{Nri|l+Y-v|3;PB6^Eq*8|;rtCof#br z_(4v8+p}ofTOzRsE~fHtap)M`s{8a8gf7<8(^CTW_^_*PcMHJJhlNeTbPw}B339(X zOo-<~MaD&v0|#U=0eyV9p<$=qOSDo^UdC&EE+*jqPv2oNn=2oBPizaVv+feejd=(t(7;tX=d|%k0FDe&@j8{`z zE-+dpZ@-0-8mAHK`&C5S{D08Gn4&5>mxan``JTe0fQHk95twJV;QH$9A++5srq?Zr zv)UzMuQxXvC;Jk%*A(ux<%#IPSttFe^bf>VX2;MdNfUI&e_|-EIy>I^xiyO{w7h++ zPj4~7jjJtU^|iS{h)MFR%CiO^0PWuj;05w-NFQvAz39^8`|`^1-W1+Hh){eH`2i!| z7tmYI zXn6YPzv!SCyqPh=bCpC?cQH}XoRw_e&oDY^*8g{)JO6cl!K~*-iCrB#RKoT&CPyO0 z%rFCAU5=ZWE^pbr#3M>8-mN|nz&l!RnC zIENTaGDxE6rpTwNAGS%~TbvEqNq#v57PrpbhNohv58;M-y+(YjZUYzFK0rY+cQFVU z=U?^~`u}hFFXsE}{Qt`#`Ty7-f&TAv34g@%TztOW{B2k}wYRacWaD|@U4~*flv%G- zVbW2e$uIfss}mA_4mIW5zsAj1b>T3=KwJL7soWtAGi|iW1iCJQ)v|t`7 zG)r`6+yTk7|ug+%T3gPpvo8>5h7r})psRZxum`77MFF)|{+ID&x-tKsG zb%FVtK+z#_#tW;SR-@x0>1YIUx0y6M^%BeN=B;1rpCCPr-;rYQc7pfhn%Cu0r?|Lv zX2FvmvtQnPIKlLbW?k5gEL4dB1@8yTD;wCcL~*XDNIJ?uaT;Ohq}}y^(P_C}&s;j1 z^%)bQH>K+Ad)oQvJNPylzjw4 zSGmSn9Or%Md3YQ#Yisw6>gykXx2ZCwfl+?@xd4gED#&!ZMoMHrms|>$bfuwM{;U%f zocyN62t(>q)4THnf>G+PDJ~tr?edt9P9;SiP?vCtdm&)w{dz!-|1>cW^mGS|Z*7Y8 z8ByF7kUK1cO#6Eu9(akyNJ|k035<<-%mQ3oBf_wnu`xjzPMs7G>o5D5iM}{U1Iq{i zQdma1PwWXVH}IYuVVRdVc%wHMY-E z}5l%qm?S! zXZLawUQrH#s(dX(hmlp%5hq1meK+eAhI}{c%^tUjf|iL6gSL}gOZrRH6zgXAjR80} zZpHH$Oi|;Gjb$5$SE4@TB2QBC(p6Q})zx!kNd?}iI=ryCSgzmL&44G=7w_i!aQm~S z8INu4B6#*WD@O_S`I52w`HiyB?hqHVV3lfksyrM9E2E|3n&=m`-0@Q=x7lYbQ!RNT zyJ#0yZUC?Z7eLnlQNj#PAQBJ_*w}v${UaoRoPnOy;X(>jCEb#*FiQG zL@FY>a@uZFONZaAp#_Cf|EIx%+NblQM+ewvfmsg5#Z1$aug_|`8_#x@`IdQ$1=rdJ zw9BEcV6eI+G1qEd{I*A(TlEdGfME7xynX+a3%0wz+j&(i;4Z)3bUIx;Brt7ZvhCSx zb_p%KS@S);$oH{iN*o^)|E_50R3{Ml)K+D4nOv~#ZJU9s z0B<|$Hr{BYqlIlfiQ$gDSH#R;#SL5vw>=HBC=jiNNZw#?8fl@n=}4V88zdwq0BMPF zu;!Q>Rc!o^L{(>Bo9D{zY+6hdp-1vQ-aZ#(aM&(H)W+om#{+w9RIO*%)`A?8wMT%I z47%*iMBtk=o0QGX^kiW^+wQxrjM}e_=qpbAd2%ba-goH(QM|FX`|{0RvBP_iyK`ua zA%eMXcn3T)Gomlz#4=81m`nNp8tE2Y-x#F+k5iW?gvt^mM)+JEGBF$v3 zK!Tx=a2{Ts`#+8LoJ$`>T=)3*XCn(meYUZEmPbBNjg4Su?6H+_xs|*_COpk}k(2{x zyP5he72DJTIWF4d>SP-(k@BwixY=K4)4du%QMt&9`FYZ`n^-v8d;7Ckm&4%&neRsY ztAq_!7pGL>uT4^R3C=lWD?1;GYXL6-7lwtZcxyll$=EF~I6sUEUf$w+ihnFXtwH;Ik4;?nyb7yLy*! z)UGhk>}_j4bM@&SHa501-$+6Y1vu%XPi)-t=H``Q6;7UVZ)f^Xi5-0g4@k-2Ebwg=UOI;MU?U}0|O#o?rzTDTM`U>BjC*foN4D&Hd>R$r5FO7$A&GC zgau&sl#9eMtdTJku-K-*JI|8}9+6+npZz6OSo-7M1Ll~4vS8WZ8^?~XX}XU!CXL^P z_*rKfzw_mSOQK=KW3QkY#-h`^Sy{7yuM-p)YQ1}y0n)@bzBC0sY4jLKm`s>s=I;^~ zp(ynN(azV->sqh_xfWp#(Qt=j_=6*DC~V->da47sXjCxrK@ z0>zb)Cs8_ng7ZJ~50rDufEZ}VTzO5k#;?`}AgPpabefqtH1Q+G>}3-@bM)V;hn7O?U$no0B%v|m+*k%yC7AqsDU2ELldW;6=-d^zVG3UCC zssps^kxyNku!!-_SP8-C@RMcx@T@9pCF^8%Y{aZ;si2#Te6@$IU|jPkQLwGQfBxfv zveaF!W(zH6=ZdV=808*}2ys55f^fmDuWKpLy`Bis~<#vq6$g0{2VShsp_1&d=3lAh00=HLi1w zvJKc1{dix#Rok5*kET!d*BwVsXDk6(sd!53C9uXAg@Rr-5DA4(E|9EKFE(% z(9OO=4Ot`2Mj={MOir}uyrRqWvGhpp;azr;HiVEV8Hetx4zo+g9)Jb=#?rvUKRH=I z1)uz*6z$1rIP+Ns`?E9|!49DvRXAYbX`8;h@)~R4>3MX1`W$w0b6`1+Tg8~K)L3ZA)2eyL3{*crsaBXDPTxzVAfuDY?}`}bpiQ8%8p zc->q@EN6#%-NGUCsW&241I*g%LYCp7Du+i-6P7EF>zpUoD^~iQ*JF%LO>Ogwz-wa) zfuaOJGGsj@PfEH1h@!5qr%bWU<4sK;0GgM860u3Q9B^hd*l$m>Q1b6@#1L=0vd~RLA%*Ms)S`*{#cKx@4 z=i92EIV~ltfjm1SN!&rVgYEr?kK5W9W`A)5Iv+to(1c2yb5^QoWMC-B;4-3U*1_T$ z7>{sUkFOGgurekpObssY2v}2Pyx89xDb!W9l2^Nj23Qnm@Y>?}toK_pgbO56_Cb-q zt)+5*KOsTigc!5Q8JP9pB(QNFDoVD}ZqKIbyRD3k9qA~|J&Orc{sy@Q;|{CJ7gSr& zwuQB|H9Z4EM{_#^LoOhcg#iX|Zq;r`YXTKaON+V)O9UVJUPW~P6Hp1%2U~*D!ihzk zFt1_p0g!p=t4dLF98nJDdXCJ^b7ymtIje*Q&@B{IT#&hp@EZZr;<~*%Bn_=c z3E8I#mdPm472#Rp?<|Yc?0qjU=lzXCtWE#S54I}Pn-~}Hq};a+tP}Hx_D-*X^x);4 z$}0fZ1;CL{{WPE#l86G3_=Kn(F89cO^~PA$*;gH&aS0|!C8L+4dAMQZ_BJC)+jE0+ zvb<;@Xj!0PGkRiChP-=z<~PlG<07f4SU+~X=7#cw|cS|V&x72?2ft^rt_aE>jHKtpkOGgDAB>hAwI`> z6CJrPDN>xa(-8=mYk#+V!yzb&%4vw6jzSu)Qdk=8F zfLQKqdH5~LEfP}5zFav4;GvBCFRxvwHxxDrMaA+PovAtik$1-IM`qD-A)mtgNUrv< z7PzS+MHX*T460BCE}oN*3xZmh!@|z09n%C=lNHg2Bl?MKpz>gfA^z791tS7-m9;%; znAxqhxrr3y`0Lv00nex?i;A9S?)3{PNJ}f$uhwa5&H83ANaoE?Sj_;J_85raYE0#` zE62*p%4>u8QDK28c_rH15U0xRi&+Am$2ij;wew0Q5&dAJv5TSGa7?DsO^cp=%3i4HN^%n>UU;s8=H)8^E z#SdqvCpV{~euu0WIXt=mLjQd!{F^TKE5yBs|Nf67FyB;4)B%L1u=Nte2~UxNAdlip zeKD!?`!gC=oiol3$G>jU=nC-#ikfTc(&G_`htugV25>y`Lq4&R-sldZaRU0R#mY~O(#ot zeeD}4BT=Re_9Nv`T#Tvs+VJWRuGSf&+`I+&Bf<=!Xn}ec;cmyowVsvSGGm(#OUB&v zw-5j9ilGd{oP)EoMq(S(#?V=nogqg{tF^gpW1olU5qU&HvG~q7`SpMK^kEL#jWv>8 zfQ=8Z#5earqT!uV=KIr=MSLu5AJ6`=HqXRAP+zTU zOWr&@Z_|*AXu)&G;J@rGWK|iQT<3j;ivL(OMl^9sNJx<#U`+CE3zSFYFv$RJ^77Ku zzjwHwE^=#Y+X2+NWY30Cx*YF;a-qjI#fP#;oLd2QmN~ig&W`15{(qLK=;?6Sw{@*k znD+4^e8UB`pIMxHwy?D3EDX{jtE;onVgGgt$UEGmy7&f*>@O_3ydCpCGVg`qm{#mv z%!W%(Pu{IitA0G$bDPh{3~3%-2+QWTU}+v|np&HabgfLU|IIzb=;SpTQE2oflUG-+ zUSEj^E9h33gpr83-gvzkwjm6(#{e0GLn(jhalF*1jsV5%+q9-^qIASm}}1=6ITtzK>tYk#>QCg z)AeOp2-j0gMOkTy3Y%zD;ujNRSOtIX&?jXsZ6;X#n1_c8=0GeOLq>jA0#FKHW%Nnt zWow$z2TP2czRt`v{y`NO8d~{9M5&J?yMK6Cf{GB~nvK&@U$UMPSbvP5zqDxFXvoUO4(=E?w*7})RY$h`k zeiVLjUWc{tZM_An%Bwbasl$s2xQGnA@Gva&Kuy%6(DJ6t-Ow91d{BU}AI|SE;F~cs zvc>WjaPmOLWe*TzS|fRsh~zT*=w#ysoKT7;qf4tX{P(xFDq1e1PJNY8(E5TO5H;|Y zld!7TwY`1EO)EttH~i_D-HN5NE03G{#P*SCMU@t97gKyJUx`Xe+Q(z?R-yqN9>G#5 zKE7Iu&#u)1P*hcrUB58L8Zve{e|uX?3Ta1j;5ZziR^kslJh;?CtNU?hRwy|KM}|ny z%9Bzq-P;Jk-aSZ!3h>TMt?C`E^;d83?2a^;;*L4<#eZdZOHsSV zywFSuZ%S)l+WJT`#d*uhXDEV%gr%k3TZrsZ!W-Zd^)2%PudGyuifC%$tf9l>&Ii0N zFGf*P%z)vMD$MC=<2Bgg>_7eatccN^U}vuV^g&?XnoAB z6wZ2e?hfbkDbnEiQSyzl+`ThMy*Ni9rBj-3%2r1{9givQko|ad`6=dpm%#*uH~;p~GJ%+X z{IPXob!`$JDQUf*LgU2$McaEuMUi%E!#K`}ih_cI5(kl-lqeveAVH8IAUTNS*g%sT zCjqu0?ls)m?Y(aP4bf zTM#PM=9-`&$AiW>&`lW0H@thhfuR;ns@Pr^xbx&mKhU$jbH(u{!1UZ#=ck3tW zyx~7VKe`92^SeNP0F;}=cc3Nj$}(VVTM{8HohRcZfkq%+SGLDzsni}ib&7#{qbvoH zXeq-WEZ`0;C&*99ADWruW=6GM3KCD zAg5s3cGc~sr{X#ed6nFZTPoJX~XQOz(Jk!%dh<;s>0hzpBgLvN*S3VZONU?%|H`T)L6ob#sC0 z%0-BV_OJ~elBtoEZs^X$&BVF?4Rg*!?p6#@cR^EQg7)UrZ_^B3iqTwmSC$SEz4lrL z#9d-45maMNfK54|sRoe5|EyB`gN8spLDg_UitLB!?WJ4!jyztUIF=)4>b3pyQE??eKd<-(!2@Ol9@OACk z4VB^yYi$mq=5X#ZkT+pFmDEtRvp{41DE_IP8FQ#SV3!P(sUWFSomPBrOO44UUkYrJ zK7*wJWL7nu*w`r8cL$q8&}!lRTa}~(?yd!^KVZx&1@#P$!ZD{wd{ZCG&FD^ygYq-g zghUl+KGnF9nUz}rl@xe9oxTh1zg!(t>0AG4rWf-Nu*d)$HZzN2Cl3Imt^|a2XntV( zcwaxQY}n@cmAZR=MYS=f++4FDQAI-^#zgUxms#|==6Y~mMg|lY-#69je(vF-U#Uvo zdSzMVVtJ&N246TJd0edCe#khHunx>I-LfHuFZzGUecGQWj zjA^h`)6uEu9vK9+;;O&++-r-9IJO6;$*!14f)=JML1ZR99nB{Vt`K0oxi&Bt`x?7TPaB*d3w#Yk{WtOBtO93tZ zsY>HMyB|fslQ+U`2M2|D&OX3T?4XoExBWx_gT*<&{6=%B%xF(rdeV}aI* zTs&_0suph0q*79%mvEjoYCeih?g%cj3-U^wIH~)m9FyO?Jzt1>-CjdD zT{ek2^EPRs{IA6$(Y$nc-6CROYbZ8lulkG-Xp6C8f z)Hj7*e8 zx)*IvaW9z6%;vj-u__Xq&#PhsJ=HyB2bvTEg0ZeX5(&=CL`i4()x*H>Qv!rvtlfio z2{^9hPYIohq%_wVt{D7azb5AHd~Y(ryP=Z$5bc*L)i^OQ^6i^E7kTu@Yb276zE43R z>A>~=i&M4Csb=Y2aYjWxfmloq7q_U}g@8MbPG9vD>ElJecXo;h3Yzp+D(MW1j9Bf6 z^o{E`EB*0@W{oMc#B|#f-dGS4x_Xr=TjL35M8Ep*m8ft^$|p6MP2H+IVffX>O+Q~w zDVnk7I%X))i?%HdyL(VZjaN3?nZ&%X7nwi*J*QMumo3JvE;=?^d}mi$c>DqPNj;ak z9o3HE%XDs@C6m}$Iz4~&s_g66Be7u4VcqW1MG6Y0{mVoS>LE(8;n0T|dOC#b9+{1e zEp>NUx@W3l%li_V()C@Ny8LSzH8;mR=QqjhzU1utx`;d6c?NcdK$Y{^b8X)M8iMS8 z;LD&w#mj7{TwOZ_@Mpr)Fa=^9*?|q(SV>u&%tbirvEK8 zX|Vmm(`;c@-Jfh7$`0VX@^&{cs^;$ammDs=O)a)>C0ySGXGvYX9JS3A*M&jrQ4_3l zG33wVjoQuGpxP-p0i4`an)^`FH7s8ho;6o4X*f*l22?kwF!{EG&KY%3FnL z)i5+ov=R)&bMw<_HLV?eDTRfFSRO(1x&Tp+ zw2bn(uRf5m*U|Bf5n0kd;MkG;2zw}GKgv{0A$u`Hi}+|)t+!IFpNRoea)!v;VaIc` zh_^<^2ggFk+dPR|{QP!HAz+Qk({v=c!)~4J)RalRzx48Xu}ykE z%);(m4}UgGvQvJMy~yXW-(VF@6M{qrT0~bOA|PA{-B|s>3(2R;#G)^Jk|?%Q7*-CP zy?pK)cVxmrVZ=|SX@t;d*oka`UkX8=B1&rN9WR09*+Xj*cdW)XUPtzY!BXgh-bLRv zHkEe!k%N0?1Ki7{&VDrCL(W5{DTmy1XX()~wSziM@)i#Zwu7OHY(m&o`<3eSmR>C>g})XXmF!Dx|*+Z&4xR>nv?;1LeceeBP-PG36J!|Kx%wzCVJe26KoUt@o z7!1Z)d_8)2C4{GoR0^@9`(feKUZYJbfLRp&{kq+$Sc&y$R%tS=z+>xYpo^zNfudfm z0Y*{N)xNKlA3}-WY&<%u{`*_d@$I^Y%3m4CF_H0``D^;Di|NXoqtwBH(LtKzQO_qg zpe(}21dni@{JLqE~1A{+% zB7WWSSR1fOva+vZbgOj<6XSj|$*LfR2J!y`)?ZMGYT5BqoJGDG2%Bv)HkJg5D;FDRQX=<5;g>d-Rrr!=J zZM-qKkink4TF05+r&pn#TuXeF+ZcS zwF~yMl8$Q|)K;MRqNA^GymRHs(n%gYg_1wN zGElakh}qG*{*N=vIeD3DV&4kVfB}uH@8~_V<2Jr60am2Cdf2jhv&~D&1#>bbde$#% zlFnbMt6#qOwT4SYvOF!k2lh%sA$;C{fSQxWUfN zDdvX1ar$5U%WjJxonT3aJ2&_^T!*G=U*Kz9QVf(T8H*=M42P{vO_x??G$I_tqyPCs zoGl=(@+q3J0B_tf*}FV3elamEO8`w7V)4Rh)!O338c%`RGSQ8Evd6 zBEcw!!FU~AT^6WV_VwmS?1k*~yvM9#Z5_iTHV;OAKQl5qng$9R?U(Ixaek41eeH@- z{_$|`!Tay1v?RU}sUbK#2;rftbhCm+O+&SB7Ar(#VOb&+bRzyFzy1E#lk^@qs{)w2 zkv@AUYjuj*bozL;L@yJjhQh+l4KmgUma`YZQKZR1Yx_>D!XU?136=T(2oQfn$UX#ft z91(5xV;yx7WZ=>MIKY!t4R(V&l1GkU;;Tqa&F72l=lQ)Hj8N{2q$pa5Zmp+>U-HPh zsd|`y&MErRqw2_F*9J{vv&Q#6P*<=p7tPINN{=^wGxpjP3arZ(oSkUs#-7o|*f$`)g@q zei!Fx@poC(a09duQu|q6t_!~?`AJsO%o)En!O}mJ1f2h&Bp}L~O7a8$2A?cT)!+z` zc@Qs3;U>-wO2@>1?32-mg0Lz@b_BuT4TN%am@JKr@%t~M)~773SwETJ*2KbNW6{&= zqMmb?w5Zb4hF<)n{oP{%#^dVM^P%7Fv4nl60oKSBC9-iCNTSxuSNsadLbccJCP9<8 zP*V#Kw-}FMrJ9KwF+vRGH8(VQIxo8}MVROwfmvVS3VvyQxCZW^niQK2cXjQ0tuz|;0rm(nH@tCr6mScR z!&23j4`96A!PjVL#)rR49BfvBAAzo2zb>&o^+1IT$21)(h9BeK+VcQ?f?x0F#A-6A z{fh)U);#sXyE%kzed%dlKu5yP_WRHf`|MtwVSU>>)^wy%x%*xjr{tmga2r~}M%VjWw*HT&2-d%BdUZC{K%DAh%*VUn zN{pF7TY#LLnES4!kI((VF1o%F_Yx7_j`yvmRXAEMc6K}P_1+&`szaqrjkirRh!2`b z>G6io;~eSOD=4W4H@#}3MTN#O6~3j|+hDi(NhKTd^5%6nn(3^y9v1Kx z+vh@ECD4s8hgL^>Bp3|Vd-i^V|c<8PM`KlSGf+VD9m%TcgQPTTGku0AHHoK2wW47`TjZG+i(@pD9$55VBX>_I$7gaT_Y!xJ0zCK5QMwS?%mcHk* zj$hJ+sf19sP8hC(Da^h(Kli^SpP|=|!A_C;nSL@ysw!yACwFO zFAm`&`)Vcq%A`@{-Mf6c-p$~;KdLXz@aXR^#qS}=Uc*NA(?xkwnppKO^T05x@ZU}! zQhRpB6nSqCyAd}q3AR)_&yN5Uk2}s;wXUv8Ls#iBbo9LX5mv=T>d4aT2n`qkZwx(l zP#_;$=-QKHKtqIFjhidLX{HA=0(fEU28o4^0eOY=%~WY43{#hcj#lK1SKC}-ClE*TuMLZI?G3?9+m@DKS^+0{2q8u@z1t4y zvzq4WS7k!zbtzU7ks@=68~Pn4$Sy%XfkT&_D8bxPJnj!7OD?p}#&BEWKnR`qTB;7e zwXJTIQ#5R+u2xO&4uqYvQlTRa(cfR*ghg>%b${>MieQguqlWfYSkeSPLmAwJ(7@wMa!GB^cpRj-OQTO+@Sk!51lj=FPhc7ipY3slgOiZ}&FeI^NbM zt1h_8UD2(GvFbxkt-J%PtCseu_i?GW;n;fWP@xwu&fMUyog;EU)*O7bVDPTC9eR&9 zf-v_{G3DqwLWc0|+dhjy9^Pky#MF++(y5&VpdIy%dX@LoRT)eri7|ulYLY*W-PZHE zCz_g(Ko3FD|GiLsi*LJ@V63g(7hfR>#)9DS)i4VB$AShONj}CL5ij=jS7cH(Tu|3; z*owMIOiv^{HUYkSgyBeUFGCBAc+8{2VVs1++anarG&Pl_^K?ws-rOHy;O^)g8_$9! z+M^OepCl^Ub$g#8!kfUVi{?IuL|wezhiidJd_RmSwKweVjR8%r0^61o{bGiX%JrFY zgWVAx0y(*}%O?sw_X0Kb^$P)j@P>jlk2MwGT2`>Oc0#)KEhkGts*zV6D30OB;Q1!< z4y#0yM3QLgC$IMBZW=0*0~v@~9cC`!nuIulIFWl^SBAN@^)oBuT)Ki=2L}OCAU3CN zH8)Q)8=eOxg^M-jofQIMhn5*kF(v~!cwDO8aF4Lf1ArL7#dALLiV%%S!>OvbI&s;= z7nh1QtB`X}9^$yS;x3w&bhK;Kp2b~s(tg;yapDPgyI>Y~lqZmnAUOvX+NU2zu$TJv z&gRCARYy_PCVq&)BPW5|1a3T410)NaSI#R{AHM`uK3nK?q~Rq>23SdKIlHzT@8C zXAOXwEklIXkoQ_fvYHh|Hcb78t1i};gpDP9h_N@Z+nHku8~uG(M@Kg!{foH6&CLdt zy3EkAIyX1thf3{ls@^-SOG!bE0bvKEbb%r+>+Q@iOZvNYcEJG!kMA3Quh+uoNy~TJ z@uCyrFzbR#G3W@wR4ozF#lp4KtDKqyz%N}x*)7&!gRyA?r+ic-gp_LN9q)R6k7XzY z1YiuNxU}$*g0cwZ^&C|;+BKA6pQnH>$ZSrudgZlr3|2O_lh_KI)1Z>P3v z<>YW1ec3s$bM#Z;soNr^n+_Q$lp+YsYd8^l}Cw(ZJ%(!wv8b+ zmdZn4twsY>%aeu2@DA)qan#MDxyp@rP@mzG=Uv;p-hGzX3fE1J{r z-VFkOZ0E*9V$REB5}WQFO^R9q=W0A%4fem%0D8(r@tFJlt6%*s%*=Q_R~t4=Sbb&Q zxR+P?lzC-l2damyO8=m9KaW%trwe*;<2&iXJk{D*38D0$ZJrO%Cjz6$`3bnT-u=1l zmG66x^bZH(s}69>>+36HN!c|;{1Y`*6$_%*AZpD{MO8z?TvaxEvC-!|>%ay5$@Kan zATZP{vj~?#KBCUcefW@^)m$SL35LY?kqK5-#`Hb8LH<>v@pMY40J(I)Sl)MVH?=oR zsEV}-v5$`#6A-li>593d>hiRJL~I#nHol;`F44{L;X@ElKaqrWb+z^?XG34Uc#?3V12i8h1%B4 z?6In9cyxsKx>tSj=N1degOBM|tKxTO$Fmm&g}3t~DS=B&0|KqFWdy+tqmKav1^0Iz zU=>m~>ngxx)>qCemu_(FK#yGA++6au(`sN8;J3xCOAot8;`+_z2+A4O>Mp`S^^st`(on_&uZ%|l*zk`!}=TP zS303q83ug3vpK8WEUFPZH4Cp-C#$Nuk(`6}?OsDLi9}*G8P=*3Zz7?(-= z6C^4&t^XS&?Y@2tl0rXGQZqgsl8X85jx4K}bZcMqy4cG8(z7pcyLfMqVB;*Fvn&7d zqQR*Xln)u{F-yP@UT$)l{YtR@Evfp4JX`!f_-0*yw^ZF*{JWtLN`n8~so-SY|B!$C zZ;}iD0K1aYLagCp2+npo{(V9;Q!%OQ!1n2*d#v4L&Qi0`%#Rr2_=Eqabvph#IR1T^ z{@zAO@SbK-M{`Zr#(*JyN6hkd9bkCO(^!B$=;?iDvI`qkR<%Iq7PyM!jajSZXs3z} zWQ$GX`&y`;=K|x8bAvouP+3G`U8+8Aa3cHiJC*Xymi5yO#5nK$@9fVob@Zd4xa_`l zgfvl(?|lHKYn37hp%w`rMav;x*{A$LR69ugUDrh& zZY*pE*I^NrRl(PKRKy5X5vve5Sy;QNI%^2(8nr$MqdZ29rb7WAsA#0Up>q-$`!@hn z)Xd&ec5dTfzt^ppDPcX=K=gGrej@$xZ3`^z*+HI`^fU{Q=iSrxuy1Yvg}ba*IY-`9 zwOfv$j*$&b91gr3zHsx`ohAx)W$UN*#2!M__imEA*iR~Jx3&Xp;5g;T)DRUL!7UHV zi0uI|#xG*1*HNUx;~(D**jcd#`6>zM$O&`ZupN_0MqMK3A`d8@eR0Z%rMX!lf`bQ7 z{tpCdUoDuIoAclZ>0J+^29_Y2aL}RTJJ8ni3A?`X6gVro^p7G4;~@FizwRLr zquU*E<|3aucWS%I?+oem{=~s7bEzw66XWA!6GBrBfild(LfV;6F(PJ=*}N2ZKGtY^ zwkQ%iD=U$ITy~ItFoc*6jCkAjp{KBT`dvs{;Ij#(!gokG^PW@muc_C3o_&RVO%0Q#dVYRfaqdq2AC=|Goz-IACVW;oe&V17aHfxBq+w3$MN-_ zn+iJqJc9^js#@lQ8=+OPMJz)#iiqSI2aJ0{z#=(xjc^T3)_{q zIbyEpp%NqQL;;`QKLZ_;ANH>s-Xkkw11bu7rD$DzgKmMgt(jG>ss>=_WC`ziBJLi{ zNegaTbb8ID{((+QwdjgDC{6YAu!>GvF~~bYxS?qVBL2n^E5M5)xwmnem{jh8Y!hf% z%U*<{T2p{6IP!F%5zQ~t(A~f1b@It9gh>|AhI# z|NkR<_#y4U;;iX7t{GQq>;*h{a+q zop`(!h7IzFccOOiX0-0{A|)+PKgm7*&++k}1qu+jh#$v)>dAke=D+t)|K$?_`ZNDT zrYANo4&6w&tiUz;ypJuBSq)oEhvs$f2ZzC4>Y!GY?RL0`Rc#dYlxAFPY8 zx94eQv6!oU)Se|sGvVd7oN^!w7*9fM7m1Rze);94u|TBc(dL?keJrK-W_Wilej3gy zM{L~GFpdg)KgDl>l!FrQdarC(0R4Euldrl*aGor{^R9X|Q~1_898JZgXA5QBUXyr+S*3W)!!$j4e01>&-3;;P+I)x-#GJ@nc4>yO0x?f+eBEM7G~gChm@h2&VEqJe!1q)Cst1Q1$`9 zON7F_65My(ubq;|u-tF8l|7fPAcBxffYFRl-7{2jLyOetAAZFo z)|S*n_wsJ(0D&QJC_Qvi7+D0}#sPtY4;pVgKIl&JPk*!lxPia-LOuL-`#wsbHVv#O z-ECq1c(2I}kl|P}*fqrjY2_M1Zr>F(BPmBPDYlB-v~g=@Hw<ry~}wi|3Y!=H`r_rf0QYL-srFoRmXT0mo6L`bTij-lb# zpuXaakNoi?KpT9Pzo^hiH~$W+u*!|l--|a3pj*OKaqisWOsk_;B`wU&C9&NpRTZ!4 zs<5nvO60-A71CmW9RNwf7|TJlg{;+0*2ugt??lj?NMm7Hz8(!QOrHbqGFK6xs`dt8 zO?es|e&gce`hIiSgJFK7dsAUBFuKkE!cva%D7IyS0os3}?U-NY;s&P-Yy^QIB&^m` zx;`|%){e>IIOA0DNE?}r*CGMlkR^hHHGk|mR^Uby7d-e=zW@;`rH0rZ4pkGB$ zx}2+)`N;%HP2DDF)n2O?78Yw^VRO`cr~|??sRx-Fx#Rgei=dedM#x4)VOJK1(T@Ni zdH<;19CQj*QrfDDLt7nDlewB%8sb$hUZAuF9@dIG=$8WtqkveNg}FC=M`2ol-t*}p zz&K3#qW9}dQDa-EHlt%w1CEET?*L!2_`Vwi6Jww)MWBkr&6(}cy@+8(2TqAE#tzyW zi29Un(@7ekLAR`tknJAUUoe#Q87X?y+T401NdWq{$EU4vvl#VKchssT7xeeJT(~&k z^$6J^O7E?p?cue5v;ulJxF}U>AY13P;ay)&K$Sz=w_!?!22>=4WYhzxY{x6oKS@o` zFAFVJu%`4t3KzlsjKF7YCo9ST(C~CJA*}i0;O~AUH2lxHm*JR}V5r4c?!iU6Avt)g zH^Qw~BCjN3ehESrt2g{A1yjH2i0R&CF|4UIAD6YTXt}srD=2vT@{I&~dpfC6%Bb-3 zn0_0dc+A}PemXE=7xSD>oM}?)#-;M1P&@rIqltN^S!;NDe|rf2rx)PE z0e06b-LmtxsKHEcJuM44C1rEh;BH7%Or*~zHC@p~S6LbCrYnKyjmzR-(g$_u_B*au zg#PZ;_mPo|#dWA5?`+>spPucs6}Qd;Dg+ofy)p4h?#SydCS zT_kHdTPi--`!=?FmztXTJPKf28equJhQS+^OS+)?eJ1sIp-yZ)@c%I%bod#Tdv2Se zzkaruogrfoW8|95-K~K8yGo7q)a%P7eqYN-?!Dy`6R1~y=73M@Kd3hC9~SY zaWPQ+s4*+{mVY0e*{VeGn=SyIy{gBeybl+!l2}moak)L+<`)=MW;(aDJ~+Uw zPrM^8Gx5JfVKNH^o{O@mtG2{*B7ysejEH~@726F9bNc2Oi&vg_6WC90(jETh!6$&2 zoDWn;Rj3U01niJW*EI=x@0Fr?P9wyg&a15JquGYD(V9n%o)>`%uawO*+b_{iWVv!Z zOLcV&z7A@3zt?&8g{zal$cc}y@bsbAihL1NT>P!A>-tccMPh=1J>roV%_^|Xmm+6o zK9uh&`qzhmfwOaAW|BZVu55G1`t4~d#LC){dSZ_kPzA9X=Xq39YuB5l*>7M0Zf<2v zCyz(BdNX~e{+Y3{p|*C4hGvj5I^UkjpG<9WkRSNQf8EVZF!#kx%K|b!XE$ixOp+(l z6sO5{MJ=-r*`hua6BFsU<5{@yB-i?Ki^};N(fh#t4<)Uj@`eRTeHf<|j`C7mI42u) zSVb`(h<6Gc1mM*iZGDq*gH5D8DfO z);?}-3RJkH^Z-qtgA-&8px-ezP`xLp53QABB!CkMm_v)0cC{FdxGjCm?F}iH%a4LQN8-eZ>xs9QK~BglI6mBZzed7VbrO_B zeT(a`^z>trUz+dt@E3JFQc`*sq`}c&l0qZs6RPSG+t1&wrJ*74y@14y>IxNNDebsI zQX&~leOp`GSB#sDO@NP2jnI}->UkX65dREqV%`}b$Y6WkS}pINs*nLG%h4?A^_{u! z@q^&AoODtz-(^j10NQziyZh#9pkz?6_(6Sj8Q8M+_V$qu>=>lDf4G7frx7n|c@T`B zs0BH##mvlNE&e_-myVhIp;vu`w}0?;>g$o=;R9P)BV{BjgW0k9gDpmQhYfktmhPv3 zhO*tUR^*7GBG0wehW7kKxr^~4iJ&Ou`eBL48+P{pG;W8T7hB$O*m`|&37~o zG_$u){kriBta69~`789-xw%F#&ow1L9q-c~G;~|Q9m^lWbKuhnqJsKQ0mQ`nNQC2S z^BMOSb&0s}fX{YPY(`DmXG3F937o#B|X2{ zsaV*Vi8wz^tc2?u#QFM7pCJ-pt+x5aZ6W%6C+8Oy zzi&4y>&#PS1MD{tl99mqL**O)Cg=EqEZnHVQ~d6YefRB5K-WC)C-p}D<_A#!J7@CR z6?Mh4*MP4Gg8S&#Ok0{h@{sY{_&(8P-j{#MMD7B4pvPYfC&4pFWY7RotFsVPRgvn4d44!Rt1l6ar4%;5QW|r2BGzlT`3~FoVNDO*b+$`lw;$VpN_! zV!49#pZvTyr)preKrtjL7-FPf=i#&J7fi7*_1-y(0ly(+JN*5&z@74-I#)bvyAQ}7 z+#Frw2s4j($p#ftodzeQ+iQ=k2-9a7D?lOYAGnxc7x;+yEj-wV0JF%j!Td&y7zNzI zJ7Oaws|RFqmaHuX@^|1ohT0xx1L}tKx%F=!KwR{q`N(g^r+s&fga}=1zgV2xar8>n zJW=gZBWKQY0XRiQO#B>4qwmU+?|vE73kR#3jYWk1Vq- z!U!!*?Ez9WSC)*JhWm%VH;-hAPILSN+w#;1Mz-_5c?V^A*4acUCy!ug*q5cU6I#;F zP=U@JUx9y8_dK2GK*h2hrb!hQUW~<|CMs&MER1Z$=4ojqI6gta%OgPKLp1pAuOc{5 zz+8?N>2!J!yXu|H?2uoSM)BiWcIz*<_hJColr-d>5Xg15#vlrS2qM%0MaL3k{ivEg z*vluZKOa0%QrY&*@-E;o+YS{r3=J*q>>t;9~~yeppW@g&_A-wP^h#axIr%D^KK^eY+h48vL{?T*H#}Yd0bY zHlr+4npR(5;H7Bj5SpoKRxQZpCLkdS*h)e*q^03 zU@3S%ZatEA1JgykJzXzTHvf_z?VkH5u{r2x2P4=Y|7*PJzeHb%mZ1L%eR;mnHr(aV zFe|=H4A8l$|DP%p|K%0@m-x?7sNT&9w}~$j(KV;v+IDiRc0KG0ZN5pZ#U@iox9g23ZE|MDpJ;V2uS;b?Yym-l>rOa8A)>)#eV@#{{TYtz*gmVNMmu(=Am z5z1C<(7m_o+tb5WdxG0_FCCv5Z?ibdTT}cL)BVCb0rGYl$#DYJ1o6GGY(Pu%0hVzm zWGNPGwsa$RyEgC7K+z~Um@T3W6x3B3<0vBmB=dx$--3Z_8?%{FFl*7SMJwh5xE@ro zAFq`a=lN%s%!~(Q#)OPM;}V$Py-l~Zj*ihveD&?-8Qu$>OBAX@ZJA#9GwG9`xb<{} z&p>DPc@MU0P8}gYrMA?b>Sq!7Bx<_173$8tHylNIajAFK&w{?+e^sZz{BuCTcVMpt zLe(w0c-;}pc5c+64XOPqjf&n283 zoyKbMRV>~l$IL+q$aMy|i|A`KRxiPzyrc+~mhP!w)T=rJO#9xhqVV|O_pdx(5a~H} zXnJ9%{8T@YNzlj-j+j%oL3!;t&mn%%Cm98b+OGKq4y(-Qjzm63XV7w_eKX4h2ZE(2 z&_*nBNm=E@RCV9oaK1})j}-(phjuONs}I?mpeAN4R_K1t5nz1Ht*?0X&Vf=(Hri+3 z1ozcYTomcE@HqOBEAdu+o(^HO(x~hl=dnS`8iHF$(hcNpXxcSJAgruf8Se1w`6mFuJ6w^rw#cBAc(eLh5W@< z|LS(N1J>9T4a7?OF4g{EdJCLBeqV4CpjE~T(`nt=k}{`Z91y6YT{_n}(h6i|7`!%> zOf}YqLSNCx3O&?V(oTsTK4&`(LCUf+#h^x?mSLVsmTCN=L()Aj^n>>pGCUps{Zbh_}r#W`G zJn%8A%3xPVJMVEFke5Z@U$;T=a#ia|EhA;zBswiCB_9n`dRYL48VAsydKup104`r7A;?o z8i1PbVhzLh$ss33ESR-!muPZ+7otzm=7Q;O;TLjy&QdVBx zMBX2j+WptS4@^zO-ZIbd^V{n(IGFSzzyVA2&UWmVlcFh)jXEE)VStZ2Dd0=nP$|+X1qDqSO3SIweRWD8f^t1-Y z%{XCR8pZdoN=i6(PE@B{pY#DF^P-{!8#(j3%}P(RUS7##te}{UJ=%<7k^Z~a7-$IF z|DC8?`^Vb^)GCF6A$vb{?<5U%7Qu{615y=KuAl?QMh2Mpx6g0Fv<~kQf`DO) zI371RQn)g_Sib8U-sdbwELKpSD;%l!F?I(%?v2AU<<(-ohd(qKDaDi(-vaVsHJ;)g zlV45hcSsbyo$Ng~z=k_#cbM9E${T%%Mqtiq`3fybMr#|vDn`RYfIl{c5gy7Fo zTkTiM;uc%Fz7%@kVlxAlPlq*6I8&6IkQ*(ze%PUYn^w@dk$c%0?y@o@QSQ3=ar*Q; z-*6$?CkD6gkLL1f9cW$&M%tL010Jf}ogE6~(UJbf*suy{P;kyZOsDt$Q-oP?y*`y0 zC@q+%4=OVPTa;4*)r9~af5q0h65*hP!xVu3czLbQ*G)Ew0%RX>5BeKHliRc`h&@rZ zU-UNJ*!zkE`uKN9UG)Wca+qGRJRal2#)ij!Y#S06O%nEGZW2EG#4h z;&Y%_j&SJt7%~xKVQT5#duaS>zas{-5folr0R>K|Aiq_{sv|wt19Ib~Iy)b&Zk^FE zVfx$@xX|Ig_Vq6!BMK9f=tnDSbKN=c1CY-6f^3=)H{vUX8ZT;(04cqF%EWL`++CO4 zq7?^;$(v_F3R+h(p@U3Qu_> zHhu))NggK=nsfc&=x2+vU@`oNk-pkUw|8$3e+aGkZO`53RZtE0iPh>pnT6CHAxB@G zmr`sy+BFVc9LGc{jVe8Q#~Mtd%<#ia+MZNyDre>_>&#G_VnY}pQtY#CwSQ8weQxfT z{;vsg{}E&9ND_gAPSMGHCvv!3vHe>LYYMT{Ln-3rp5O2PLGrZWw`a+0iEreuUV8AB zP3?D?-`RT*5iIg(@q6c&YqWZJ_gHkqL#l4x{76EzOPHuZ#8$LUtaq#zkH(wbIcb7E zzT+U1|Kb0~`Tzg;q5p@U`Gr6HzqyEBsMshmA|5`5|x{xNp(wm>(me@xDZaH{USeuazY{cOko48L5tKI?=s|Ko$9FPgGhWVf^LGmzySS^P5%o z8HK&kW9@A>#)*i`JWRD($b5?Gw3t+op7ucgHIMB26Xm8zBh`GACzC3fRpGaskV zoH=vj)?Gf&v2c;(>BpmP3|^}xQQv;>YgC!ncL9OYvthl)e})1gPrU?(7r9M(y7ZK* zC@Cpv+*q8S=iYVP%i|nmW?>GL1kAw6MTAnkKw5%74f{DFRm`{7vqtL&wttuyR(px_ zNPO@9YpVCMbV`By*ZyHl&Y+<#cJ#G3aY)VW{KSNJ7Ck5L)J7ReRGv|d52v8GG!HMJ zuN-vI@G=}TFt#h&?ev6KhjS>F)oVNT+z9)lo&`^BOieB4qa{W zN?KZ4m-%^lVck=qas*N4$kl<8JR2Al{pAt_`l~Ws!ZWbjla!4PgfTt+u`@Mh@I=uP z8+Ly;MCCYdwosad=|1uA9>Vtb3{S4<&oA0VkVnge^jpaq8jrWQjW-$mi@c-SdLJan z=ElpXtB2K%)<#DMnsv_)S9&AAsjw#8aao_5Ht|kCCBNAlf0-D9cgfP^-ZGQ);q;@r zdKGmr`z9dZcI#OpRjt#nXQk_JG$CH?uX9Y7baaBm_|vl)sRpL&KZx4*4b_ujk<=H}pVa&^A8 z-<91qZ6I0#2EREv=d%XT2ge2$LGHpJobja<_2S|b8wK*F1Vv<+1kv-i8K)XB!^0KK zR}I;59uI{fQ0))jLNm2IdU6hf-7G9A308UJ!3E!T9Ou1zZq2sbjuLEOclXPWZ`%m3 zKT=eoYG=>1Dyk(s6jvS{=v=9&V)OTXEthb|n^A(XwXd(S%JG#lN{<-&!{2Ta{;e~0 zH`?y4CE(=-1gU_{N;R$@7fv`THJa-3mRmuTS5MUC-EcTac;$2ID{^P0r#fITHCs4u zvKlntuIo(ZKy?#pz8RqIPn9d%)HuJmI69&on6t?d{a8-N*f@FDXma$_L-x#OS=mw{ zO{HoYA&m8I#s?hI1Y&JM0Lt`LMy=evx3~oHgaP&nHHkZS`oK|CTlH>n^b}E>DtMTI zZ{Iqq<#UsQ|4@ZVwK` zA0TjR9C#4mN{cM?)0BZtv5#HOEv)xVxpG;A6rRwi4Num`q-1hfEX=LA#Xe{}>6Q!a|i^Nzr_P zj3i_v-zB6brxSs`pn>0RTn^+oz0KOCuU%wZrER&=o1xfJ3L5Il4u1m$#Tw7?QjXyS zAZrR3U3Ys+K5N%4Rp{wmb<={Vc=f#J^7!}n_r*oaw!SQyo0?vuru5iAhNUA7gamHw zdi4?a5ORm}8zgUm=DK1FoN{iLGGDa4zSk#oJW+vLe`3B|a00Ec$1%&HE^#Zu$=zNV zGkys8my#a))k9Th66ha)F9>+ps-A{a0kw@i5z$q_r*|gWptc+0OzM3WD-6uc z=_}!fn~!8qo^66diA}Vd{bci4p9WWKOw1dzc@=3Xrf>om2ZwG=t~*yPB8oH-+N{08 z7O5wQV|R&+^?x50E9SGEpDremdbVPE)eYPhplQR?L)_es+b z7+Kj&g>$YxRL4BQ`tJJoPnl7hq9%$w@_EWF!HGV&BNhEPIyYNGohn!E`H%-U0&!qH zkcZaw8)TCat)_l{e(gR6Vjd%FGK!Iqz_uL?4JWB0(epy;TeRqfnJd(k*J;UY-9v&B;nhp=|)Q+thELq80v-y=h3nWZ6=uduaYWnaDb1) z8PY`36-9S?su}5`F|hp9La$3Wa)0{{oxzs>gkJ+h&v5va z0Os$x+37sdLa~Py=R{$`1S?srId+Uh&zv}Nk z%ZOebg=+PP;-DA$wQsi96$+ZX&(0V`yZJ=u_kzg|J-(?_i5y*l0*@nmvrPs~O;2%#BY4kmx-X*ouJE}B?%P!PR4%pFQ z@Oo-bbv>KWG;VMGa^pv7sHxwFo?MoOzN~DTdl|Q#(A6rAm}gDRx-R0t?9CWp9JE|S zL?B7w^>|^!fOi2^o<{?p615aZhx@7?4|bt+%PiGSmJcd_Q~t$&XxB~*7j#XqHuu$d z0?J0et(=_@y{-a&D6Y<(nVGr2r$_2MD1eF~)<0-(IJnK7aSHz za(-^+2>85!R(3V{$%*l+A%6-B3SQqFc;3`L0K|N_K)<9$4Q}0xLEKRYP;Yxab?N2R z$n*Sm5`U?(^NB~h17RU`0-G>z9yup~aJM*33)Q42JZ8<(KUf*|4deM=!mO<~cjl1m zXZ#JTowkAr286$+KC^bc)oO9gVGX{Xd+m1jd%7yZHGJzGZ||Civl2nU!GU;--B0c$ zb~&qq@Tf>`u9XXPcb69Dvl9`^IhPqOx zfOEbzqrfpB^#a2Zm0J(o$>~Y`Ecpixd@2BXfRjo~OXCQM<_?4`#NCh|3Sb__)43s9gVRh*M}uB19NHr7-i_ekYb%@)-KVz+ZtdVm4Oe!l&n z_U*?w(vTwEzWN!Zg3e2c3psuLG)p3{nF{?m@1{0voXE}LoSB>^C7;=K-LB~bNdL*5 zl-+O)V$;dM)U=Xkv|AshQ{Vq8N)I-qm#;c-lNuacf_giw{ay^;W{wz#sP;=IE zJ)0-Kw&RWom@<#+3-;THnYF%uza8w_4^C$8`n+Q}hk`*!p6tngTq}x5s2&o$o7tbg?JzkX4HN`Nb0y0o&4}ME4zpN2O%`Rc(UjlBR#!*;t4PZ%N{|f zE31Pd+XMf~-FpE~pDi>s7HO&t$5!sg1VE;1rE6*A{aTd@ z&S+k$v3M-ElNJzt1Y$_bO2R4DH#a%0H#t1kUxF8Sh*Pl}EIEIG`;%#7lO{WYgfrE83A$CqMy^L?}Bi>_4l zw~Zgcng+~V7@O5#MS2C2f}+y0>@?z12w`07;MO_RR4~kBvA*%zd~+OpwBQ5Y-cFb- z?*@K809xQZb8vJ-tG`Tj@n;DLmU}NP4P3BJ_O^;i(+@!Z>o~uKzj2m3q-6MxW?;t^ z6PWI4k*t$K5@2x=97IH~*j;q65wyGm)~b{5ZmjA}S52TC7q6yfj)?j6&zrlIrA|yo zMuNdzq57r@;)W??pd^xv*gu6w_6;W8=S@_cLKge-^SAE{6@bN(oRf%BdlpueEB1xMNa*HM=??L@+}q|b4PlEXm1BQ+Ca^PSD@?7Teb zPJq=s`ujMr{_Zcc_K!I`I^nWJfY3u`>A(AQnh-Y~{5{;kkN;2iq5r=ZND!2~1Jlye zvlB6hQrkZTm1GPgKL7icvR{5%@`eJyUgKuGPP4|M^Q>XT&g1eLyFP|F1Vz;GwO~@j!8zHs4hwV zEjI^G#PVKnnQsI}_G+Vi3K9f_kipZOzmsy% zmu&$|Z2n)JBSeLBb;y}%Q;>J3|65t?XkziS@hf_Txj&jhSLAv zGr_w5B>E-mcY-kOT$!#{A7W+${_ z2%MZ82L%O5px>Fye^Z#3rX<44KYH;P7N^B*JzHb%pe1zzm8aFOoOET;b5NjGAXWQl zWQ|kbY5C4nMQ(&8Eh8i24M%Er80GDUABqm%^~!b0F(39HpDM<7FH1Ow2qQ2dgcCH8O)w0hJjDMaNZB9P7yhy>> z09vB}Og_m^ZHnEwaSjt7X2_vqwm)E0&RmsM`|)zUH?1_80nU6xfiz|@hO9H!MQ*3Tc2>Ef-&nbp0-+p<@el(wMfA=$k z`7(H3tJcp#F2G0*if4=^`GDP8laq0804gZ`Dw^0D%pb!XH&r0r2Hp?T)G2P8WZ(E% z+cSH%p#mbWo$-YnUPsWtWwG4rWbna>KolaCBJVa$%LVHI_;ttZn`Y%TF#Bg=T54-2 zz?i$D%PIX)(fcztI?o;os-$Ne=cy%>+j(kjVfy{Qt*!S2$G zgd8NdH-0eG(LopR(5~9!s?hU)y=ID|g|v|uYH2=MU40_p7x)wq`&xb52?=9!H|yOK zffExv3(vy!Y2FJtELRm54}6Nz4O=|k7!eQZs(Bq42y7r)@p}r;S+qVZ`UCY-jPwy)gfHnMQ5Ij}=dta8j7yEwI#CpO= zqiGXmc2K`K#c-wTqqE1N!%M}6VDJS)z=O{`*9Crr08KXXqFG>=bo3CtTaGp8y;$E_ z`1kCf|7Ui{-2?yHijl*+UHMeI3|Qf|`<+%7(A)!UZEa7_8X@r6xkQ{ z+j+V#@s~4bOa0G{u+8TGdOuh3&_Dd>OZ9|UqPivMhH7z+x8*@Sd- zIZBA6O8YS~fG!;$AJ;hVu$#Av^j(ss2SwNJ5bUp%@zBU7GB7f-b8)!?CdpJcq(HkX zA?kM0fm<%VYFiq*!EEi>;;+UnPK(+YxS^pb`MVHefq~jNqDJzMuR=JX+9k5{=|Gp~}_Nj;@Rb|B*!e9iS~C4zV5v1WSV?*)!MivFL}L$@0tNO&u5v zloRBMkl;(C?TA}T-ZC{Z%YEokqXb=n;u=1mo4~p)LGjPKO+kN69EzfU`Ikf;SaLG3 zcbA6L4nU8+rMMhJEgWlB%zqBcc6-vcgpiVz#M=WY1pUHk1!sWVkMbABdC|5`PAxBD z-nIVwMwN4^1v*cxY}Py2VirLbh=BpJKu7l)&mE0b$Y6moCMJev7UYJo-x$G9@RHv> zURlLA4*uc&L#itLqpbw!vX`90O*H<9qF_qkp1up<|Dq#F;ih-pF%n3m;Pl+0N6ETooH-o zqo7Q~^9J5hNx43gk&;+Aoa#Rpg8sSE%vp$w^-QXN!;L4~Or_bbzR8EMWL=vnmarg? zGBHddidms6z0Geud|+b)q~gUYU(;iB^jSCkH(-28knI>)8}rSt5ZKv+qAc#?#YzD7-rkXE!V+p<{Q5Hul@1S| zxn1@A0l1A@LL*ctG5g{1ffc*u;To8#@f`Ytpo;+V`Yu}CK?B7;Xo|Uu8a#US?|76i zzx__kG-_NbOUmR??US7puNdD2d=i(Sjx5kRZHjZ$tk@YR!Xsf|%~S0wOcSdEc+NX$ z(S!|5I7-MuyUw8{SzONVsJ&(4;6zoI@YoGf{5Q(YwgX_PAPtvl9TCxsT*v^1{XJ1> zcKYCNI>2L*E76zHGx&!vFb%H2frHzvP-RPdOI>Wu}0nqyt@{qVXU6 zC;xa8KLT(^H><+M8J3q4RQz;2%7h_BYrF6Mfuq6LDO1OI9%4O}S?z+@-O94_zW$Nc zB8OgHQIfy6AojlNhFUtn*mMVsBRcvc0y2&Rv++tftbedI@Qjod2SA0GHGg|B zjiG+Iwdd^9lrDtv>n(w`?--a) zfS04=b_WGF=+2wq6x~wJvBp`Ix^~Wg{(0w34+Z6;fA8PvXaDx!d^`95?Y?J3iq!Tb zv^?%#n(T2L0$@GGC4ajv79ao%2Aa~1p;GgaLf6qVp^Ws*-YYk~ivfIO9X&V1t~cB5 z|KRd+1hs>;Ap54j1iqELGjMTXTx)-Q2d)T%EfYK^#v{q2iL9>X9kF*VxZ!{yex(4A z!=!i~Xd>E6by`pd&9RSxRyV_IIM1C%pOSKCTDMxui;J6)_LY{7+WP9^cz4!;aQ>@T zj1=sQiusKLjfJ*Nm#ncWx{#NGni?&ANp0hgi~DxPjRl(QIp-*yYA*(Q>cZhr-ZD#5 zeZ!o@Db|p4$h95PIu%n6_x|cic?Zj3L&3FXG%GyA{;VW z78ca}KxBp&Y-&2+e9bAtD@s>>-hE1l{Po26c=kt6ETCq$Ye-cL&jSrYc1cjR3In#H zyu3_tZq?i)Oo-h29f*sXQ%6^K4)j=?8QBOI83k#v!5p9Wi`cVfJAg_%Iyy>Ql55r3 zl@3EJS(dS`EYi@jJ1FvyI9VR&!j)}krPFH!qqiHuUMpEAYNmMwT2jZw$GVQ%yU&vo z;*+C$w*&04=F~|iSm30S=^i^lwOK}nv0p2kTMi)LoTJgfM8g2Bp*wL9v{@F(G5?E; zZRf_}ss!Czv-0jDPGnG!2>CoSdf%aRP=I)$T`CU=^y~%Y76 z0-L@i;@5V97F{~<$-qfMLPLUasy*6-B%Udb&7PKq`nqvhrF>QQxEU<>fs*Q02x#e* z=+gCTN8!0IK?x$q!|s+6-h@d`VwRfs=A0G~!lOsk;U&CjtGYJme10dO@;xgn8zAM} z0S*rNelUy0f*|>r5X*9=z?P(&)2mTwRqFue0K2Nv$floXvlbKGSriP<=uWRP6EnPC zX*huRnN}KbbJ;jRJVEU$yak_4W3@0Y#p+|$ND#(3K+R1-}8K(_tKY#v{ zPDEF35AIbz=e1;ba5{cCKzk8D&PtiQu zDn%^R6YrvIyqxO@5O!g4x%&^k;^QIt&;got{=0N$&I@TMEPQvRy}H`fkmLy<%`!0C zk_DYzH;NEh8OF5lCM!)}o1M=CEdsKwU3m*E7NCyjqbedhU0&!pAqbib_f!6$-$+q9 z)LaeVh(~eTX%ROCh~UtjO3fLMs1{bYqk~v;q33%o_mR$MC0pxQilwZSEYr=xD~9;( zyYfC?zKFh$R%>K0nR(>Lb)XHx(< z5g#q%C|&4-2?hGa@h@HN8Ozu?{4APx6AA2})Np)1D#5@KOh7W+FaDUSFKE{ z4K!&N6KfNkc8hc2a8Rn{@6oto4i5DX6mmLDXxA%7Kme-(xG&W!>c9{!8e1i^7A_}C4VaB`A#;dKJyFg*x4-bQQl6M@yofaNprKcy zq=bYOXb6d3`18JICO^nie{2k$h^9R_F}uWU=+`_^(cc=3v(M_)zI0k(5fv46B-!>L zjmNTucoYOvH)JG<+G&CqGcEBnzmoh~I~LL<{W(25fypK+!%fM-q#cGr@h`e8w#Y zlPM{x{nG*B4kH!5}A3asqYEnZQb|B}jUL+3>Pt!NdOP27w(V6+bY}q%r zwgNLtg3B|iW7XeVTr(lwgcIt%k7cu76@inPf@w`7hmtM(($Y9oBdRpO->~a0&JNW# z#MRr34j1bL^XJjn!LPQ4T?hP2vz4YMIvPhObASGnzoz3f0?G?em1v0ix@g#)TF0fL zwinqLjqf;|mRN7&xQiu35b*BT>}XrwN3m6b0RbN84UNI=;p(BbydYaudOJAI9P?xo z#eE<#1r=^fO9G3oa1Cz?k>aWP(2^)`c5^an30QM>PpTtgP}aGk4G{A4jWNg6y@CFrr7adAHZ~4CazK0HWO@=2#r!mzsd$}S zr;A~pZPcY|jjbGHq8un#YGSn+rDeo$Ir8IUIO4kEAkRNp#YVvM>evJt=<5ZEbpY zK$f^UQ{@fB)YJF7NcFhAdcKi7lI%!T=5=m$*?3m~+%rW+OiyRhGY~g7NUq2-%b2Nh zQ*xcu3o9$5%b@^AECJ%V9N~B`Pd5_?e`-Pvy6L#G)ZkFwEfW`Q^Ww2@SpSW@p1LGk z^E~+l-{!M{9Sa0KfiR3^h!`+vIvzhJU`Fn&`76gY{v`r&@Kb0QTq+gBDcdYQ+!MnPt3%#(SLB%(|~9~A>< zp3R3XZysNMYB-tE7NJxDQi1ihrLGqtER2k}vGK{e{UDMVK1d4+4rD<*wVbh4T5E2m z%AYGS9(T7^5jKzk1|(9B?U_u$VBkZwzPKNf%zNgIS_E=|aGzx>oAF(wc=PA+uU>ru zl=n%mkD%s7L-SK&?#Tja_myf22E6J{K=aF8dWdwb3UbqUOt*|#%|5HHUY|1UWO{l#p+LarxwqquvE@#S{(1lx zmnzADtOKYD_;Qg?C_Rp6i85xTf4s+y5GsJq>p?s9jS#tDk`r%SA=+0SzsssqX~*s7 z`~F}#b**ZT1FF|rt_R1CfUy#kFb zzZ~P8Eyrut&SYwncK4&|6Jf!BSbRSkqGcpk&UoEZs3t_-h;aaAoCbKvx_0p(Ke(=a|n| zCWEK&E(ivqCyUSek9h){i@A=b$j(T)<6>fpyyO%4?GLV65CJ|?#{sHEb4lN$w0e14 z$Z0}cpYgEsZ_ah@S^Zp>4%Aot!lG8A=y@a(;BTGGSA<8`jIYG5Ma`q2DF8X(pwAz)Lk`>v6rnr}vhfU0-jB z3=xEHIRPd}Tpa(hTCPbjF`umpV;ySOJRm-^Wm1sjwFvuDQC-of@yUv=L7Tej*bN*z zW>CSyX>aiD%AdRtjkqtCZyh=QyC#ZAB~t#gnOUE?wh#a(D|xTE;W`oI^ZcZ#8a@W0;kW4u{C-y{wwk`s5A8gDGJK+2c`=iGh6;_N3_X zVyy<}8QsP4xG;!olTN02E6fY*CcI5B>v?`A%N}@^HFQQs`w{bK*$FkyzeVD07y&e8 z!|x1gUSP6QS~dme{$iCl#c`OQb?uXCg5H3pA-|<^iA>pnAo@SyCXY z&!fEuBzX;&osSiFg&YhpPAH?=7Vq1BYj~YfQwO_C#qhZZ55oi0ZupeD0;eTCF|j{t}Cn?j?*D5 zY2--WWEKB#WE<&^eLq^dCESbD^8v*Knp-X6ayZ2y(BL_v4%VVr1Qo8+siA z9d<&S+Kgpiq^)BCDS(&2nYgKotlh?D4~aWgf%>WTjz-3`@fz84e$QSC+bG!=Xa0+> z9KeJ+Q?vDnq8CK*lu)88RBc^^HS4L3-q8nycIMuCg5`_2d|sz;uRg_~J8%kFxr5!^ zRmSpS+$6J+Zctnv1b{#{Y$)KHl{+dwQ>%4&PB##(NXJ=EI7%RpJYMzXv8q+wv4>m; zhGA+UU07FeaL}ZXlxYzSs53yBTa^6v~H!g_;8SabEqHjxlM5Yxm6*~3I&?=3pg zi)VI)J_ZK|gS^ub1XcsrE*0Wyaphga08UO=X9;YGZ)ot!IxT=V?d(|85xHK@K6;vA zD7uQ~&OaHpGg&5oQH%hUZnil0ut{2f`q8$ysAN!>Z`?$)zZAvK(+Zia_Ka`oQdwJG zo!vjdDbSUH#0&k=B)Q5%KCKgK)q8EF4*nD~q*RT+JeihyK!)Bd{rYXd-p-wHmP{)!GwltT{ajDyY2~uhU0#R>BOF@nG^V=c*h-xr_OnO(-Kfz+`h< zUzfC%;7q?)Sfb*%NYjf77>Q86Fo zsbw<_S1A$;nX|#^IUht9{YpB9n?=&U1C>hpa+GB)mf!J@=YPh2O2J#^B4!-1rCXn zxs9-p;rQd*aJ^#ZhYoFL&W?|RfR2Vh$n@86S(0y_lJ-?xIdXu?+1)Rww7ep>J%>Dw zV)KgDP94_6K^2pRs&35K6;dq_l$Vw5ZDHUN(fmC=J`P;=O+D3_a_oP+8NztvtV%;E zEPOIJcqaQsqNfOF?}4vBE8^$R_SVS~TOT)iO1;jFmxS*0&;q7Y1xSGWLvl^z>BPnv z@YLgvwCl4hOibn%Rw*Xmk$(&icx&GyI*NVY?gB#>@x97r;FnFl`CLj;yc^DIDh?(N zruwRposB}TOusV_D1iYFG|uo~t;FUp!()U(A_%O**QqTL!}(W6Y%U5B*~mHMHc~LnLoVW z0yY@eJ6z%Qpijz(+(Ijrua$v=#oM>u4xde!|DHTFw6Cs1pfI>$B1kEtz1Xc*hpPOY z)B2=V6BKy+n>?-2zQ@P*g4~CYnP6bD!U__!IF4zdF;@F>c;bIoXY;6+s7Rs70S}!ongyx`>G*6$2- zx^gzREcJ9gC@=!3oZ{XX+?H7rp4ipFVG;v!*ur!BbQG_nP8}{3P%+(KiB3!Chj0{Q4a7tx@7&TE9!g!{K`V=xZ z#d=^4r;c0&|K|JSt@HIBh_Zkv7f!fP>d%JP_3gof)aK``CZD9bO0SvK?4)upB^4Um z<08K&*q#1XE)EM1e}RkbbO7k@j|Zq?yr%4ZLdw_;eJL%>EhA zZ2*nWGO%J?A*uo~52yLy`D70tD(TTp%Rth`s3ecH2$a?~L~=^POD&7NvT!~tP)iGa zRTb{*QuS0)L?u00Yh^;Drl$I3e`m?;9-w=X49_g@pmY5s8FkvzocJAzDmnXAe8X~z zug$2fJrS;-B{FAva%nmF7qMmK$ zQ+X%mKru*1r{QqqqsBT&qEl^YYd(`b@v8rfO3lmLT3uXM8*6MHr{xrEWO=E!eA4YI zEqJk>C@28cD$3lUEiYrI%UX#{)J${kPnBf?!b2kSyy3Frl(3(md7J=r?aIY{98af@ zr~KxvKkHEAc?13jFp}LhvlF!C%^sUFmqcfqTW>9Ht7SReB9fneeXz}D)T|cd<#hre zG5wYDX5owH&)2s$@Y+`bRGc%w;DsZsF#`RdlFITI=ZBtHm_3)ZA1z=`4kw{s2yujV;J!;4 zCYxOT+`N-99&U4ZEA&E4QpV+V^LsqP>D9e}T$M(Cl^Mc%;`(s0j2NW$Iy0!G$}-o; z)*lPNx~c%QtA+ydYSJa;h4qPU74(OB-T4TVDLtOLa*Ka&Ruj_S$kLEJKzc*7bBdr` zzi(p?gZ74A7Z46BYLh6E;qeLH(ttRfVAKjUvX4lat3WVwxXb)l>)ovp7HYGob zOa2Ly3YQIsQDj4}W(E}1VA&fSurp>9@~$o&Q9i>Ed@~NM(4aq$rNUyhz+<58LV66J z5h=s27IR&qQYJ`Nxs=}hesloI7ZXT|C>1W_`i4c4?s3wT_z6k@k(-}?ux^#ZNZ(^) z!gX^;%`;$01Ks+fI}G#%Ni{aVvSlO_W*C<0t>RnqB`W(pI9fE!Yx&(q)}Y-ei516n zvN>9vXh+0IPi+O3SgV;wS`B7OVC5%WlFYsXb4{cz_XsGf&1Upne*%(#4uIXzSta)M z_01Kp^-n@Fe0%3;JJas#zDCdFA}3lFW#*bgb!wwHf? z=#2rX9Uc%lZ*-|z@_SwWPP!`&TtR6rHop;?+A0lQC`(6xx1EfG;(!2X2?`Esg*dyr zhha#Qex}+E2i<5eR99D)m%WfXi{Gp6?X^M}-|8~lEPVli8>8!S4cl6ElcH)Z^b{Yp z7~ds(UgvrCcH?ao6-lez(b>(c2pIHHz?%KIzBUj;N>b+Wqrj&d`zR&kY(_Np-A2^m9*kA$ajy|Cl!!GkT|!k zhx+l!349M5+xLTTpBorbsekrQE}SEdwvsfw3~%Cf4UIs*14C8i*i#XO)q2Jit=^ul zuC6{VHa4a#PnAul%ovML$2sBQA0^=l2Pyf(R}RhYK+69bWNM|EAn91}C3Ch(YO=#60&q?hU->^*kMvW;ssHypy7%0VJaEk;B)VoQ1`;M(K)% z&!0c98%2HnlD;g2Ne1Yld_8WRFIyWM9V>?k3Dam581#76)D>TckF@Q>$#`0hW^oxe zhi`zXY9ahJ`NwAv2znHrv`Jr4IpV;maR|_SvWwuFGw=W{j8ih+3XckgD~}1ZD|L1~mCJXS zXYT;xme1z)SB670q|*$*20^`vjP&&2;tH4$V}{RYaiNcd zCE`;`{uSAd=f$Lq6n2VYzBKbSt|j$JXoM8>CkWBm>%WPYbV*7oBN}%m*<gG?pUEn1-?Cs7{27=;C(-yb8caa^s9@D8g zs-|c@Ej}gPC4>~_0G^r!*r3()Vcy#Atw{*BheeBMHqq451Mu+j+L{n2%~#N#-UKFn z_!9n+ac*bHx@GeigNp5mynQ%E@KOa8+VqV;cg@Q|0zF%gnb(cw(u| zhIK&BEZ%@wZ2ZOjdj;2?f$aA+z1}t+j>H2u(nz7mMC%K&Kq9cu9Sm=*bF>zZ|0J6M@ducUbx1qiO@lxT$?+Vzw&su()ub6<} zQwPchc&Q!6J7j_uOO{if^T2fj*H+HBKE1*><#l;WEChH3+{bb;^+`Zv4SE&=>Jj0z zgqcGJia?_P4RE`Dd9N&uyvbVDcvnpP9Jl5}hg3`7gBP)}02WcK9@kwTZa7IVrW8H@ zVmb!}z$LtaEmfPZyr;J#@wM~8%)Gp)!86M}?=-+W4bsK3y!TK5MzjlQmD#S&2t|vv zS>Zu@&O5M#_`o!`Myf~Xj~GJcH`6yb_~W!OnGAjI`bcq(uaI-8$;lNB&l!^WE~UT1 z0j|^Gy)X<$x@|R}Ef*Mrcm|!0w_QYt{<@J)H@JEwQwlyMVPX*TNiH z)oO5^Lgh_7wzspN^DISuBDjJXG7CAav3|-mk-#PR2FB#3X!A^487M0vGVnk?iAiKq z<^&&<*N$WN%bdO)R+hbIAIcx3n_&4ph>&V9XD-1fd*w!dk_k{8^$y3~P3VI3Im{yy zbfKiD=E;T`1W!Wi<8ln6{wm1yKJf~xf%JrLwk`%`>=aoI<%-V2L*X3aWjO(Di=N2o znFM3mi|Ogi!tby+-XoVcyt%6 zz0?j)t2;L)%pC&TJnq;#GpZ>JPu&h1};=Sug$t+ z=-gUCFJN@)sLx4B*BCgpvp{d+qbaJ$%)YD3uGhc|B|3k#*RV?PoN+UL0LN&w(9p_M zfrf)gn{b+pKO1oz3KpAwKxqQXV9pKoO(FmU%=z0EC%AXz8vtp`xi`)t%VnEHFJ!e` zp__8-5gy&+0oJ?CIl}`|93nD^m{6e)Ip>=!F4iVB##d=}5Jt4^PV>f$H&5tsdZzin zurxcH2;6>VOzwi7d?2u4$yqh4%vGCI6>xroGgz2Tli#~iH4@lRYkyXIG+$Z!SG+ak zdkzqlK=iZY3x-aD?}3HG7lC7CoH2k6!w~>LSY4%#`$oBHQc;BCDnx)gefZdevCr@h zO1qRg*pM}LofV`OTfkBm)#h(Cdk8UZHJfta6UOGcXH*r;wBJl@!Pm5s?TcV-LI_ThNL7D zcfT|Z*NW)wZueAmMa75gIp5j|BY@{vzgO?T5is0y?=Q42zUL1&zSMws`m`{`y9Q$4 zE?=o|LmP-(9tL~W&{EoP)g%h0hH_V@sUD~R4)qTKW3`%vrqZ6+rf;ELFbLl0t9Tm* z-Gqh=hkrVYjs{E$u+mpJF_p3iBHL>X!c^5k*>UQS7rU@AK`s_K5fF$uJ2`ykcQOD- z3&T|~Ad;c)0s!a5XBDUi2qO?m9gOtKV)~$+5bP~WWcVqqoZPU0WT>r~Y8nH=jP$A> zri*PcoE^i7yJW|{FAkDS5aflo>@`g2x>`Aa{g&^ORw9Wgfkwgt@I9wW^XzA={jL7a z1=(R@&XTyl9y1*e75QuR%)F^$?X8VSv8|Y^4>iW_`~p_T;*L9(q8Q{P?x{gZybBNs z+^g={*WU?64f_XX^65|-PD~iB4k*wdLQ}2s0+9nUh@+8oxuc^anx6Asl+^)?9C#3c zGEzjWCB0^2!wZi1>6Z<+(hOHZ?L zflOHhgQ7D*SH((RtB->4BiKA5WR2lG{a7Rd)oZ^Lpp`9W`r2>evQb3Agu^(GfekIC z%14p?1fW|i0i(dCsLcBXOjexQ{HhpATr6vu>T8@OBCNMf@wEK&-hGQwL%W z2pBU1Nw~P!EOjP(x$Kvm*X=O%6Ht%>?`nv0&PGrGO2 z)&h6zhzL>NT(NYOV zsv~TWeQ^@^POb<_I*#MKa;{S9Y-1CO(&B0G7!e!b!R|?l=5XSk1%QjI-6PSJ68k;! z;{Iw+^Ob#Ol4NXUVbDb3TY)>Sx4&cuQRNQaV^By%k(9pxX0S6{BZ@<0E*;*I>YrDo zE>C=C@vnZm*go+KbiE!?ud?RVnZw7hYE{KDl&GK-LX^&JD`4R6##2QrJ#DDgb!~Ve zuq9dW=vY{M^ou82U{N4}?_yQox19I>h40Gl@>MVhlBg8SQEQXYQS}y?bNX}om%}p- z4Nt&yeguG!fy7u2M~e%&C|o9n6Kr9Fxr!Myhv8LgmzpIdYU);ytZi}b!0rJPs?~Un zCgW}SgCs?A`J8rT*#u8$Fqm$$rx(QTF}><5`?FMNYrud{+j=LkHjJhhRD&2E8G&PN zoB4Un*(9q7+^Ko+tVP|yqK@`HXkgRwmf9=_)W!m)MJ1dO$x9qSTMV(8k8xHi#J`~j zmhXf@j9pZR1aS`lg@N?6yjvNX@9gOcl%w5s^MCM#lRpdEYEr-%Vq@hh(8z!2>e)^U zLNB=VD!urX;YjPuRcLqO>)9;ZyG$Job#>POd(MF=dd*i2Z{bk+g?~J-R59e>7D4GQ ze@!eXZ@;gAF4EQv)JEBILCWFEa7soBdLhKG(xi9?m|x5}`bioED=#My--;+fai!d0 zta~#!P)6x^T}cBK+mVi*U2b)CPUk_+^-d=PZv00Ji*)`VpbDnNlvbefv!>fpRe$#O zI6;)x54WqcGbF&DfurS4)XE{p+(s#IsNs#A+3XA#I(ln)T@Sn3()-(QUa3dIcJC1% z-1CuzS(>eArfB-%@IWd+a+Hyn{%{PgI z$h66u-_RNCw319Yk(dB^ijuawxw*8$)HhL);~{t|K!4ybxyFJNSex1?jq%vKH*D{`aUxe*E~><` zvT3whDFj%^0hxH^c zPXc@!j?7SA_2p(Ul&N}x@qa0?V<#jM1E09Cz`H!7*;Hiyxl%$k5boufTi1FVSkS%V zX1|7(jH+sRxbpWt>RXzce0WNf{1Ve*!~Zxn3dfMd@!lQBfnwWK=(;)vZfR{k1iDU9 zQZ|7iL30OK*h~Tb$)W3N`g^P>k3+_R)n{L`3JM$#kcv4QUVpqzm3%3Pz4PUwlsYe{ z$8yfoW@S$^y^(H!SxUN(+wD4t_MK=6Hr|309pB|c5c#6R717&#EDZS0PPwSX^NR8rY^|`?tKO8xY)RW)#&wy%|LoCH&tfA zYxx@m`-YiL9o>+mQ<*hYTf;3T);62%}Z@t*;|LuexHYsP@RpU00MMsq5E@xvCzrR+p8V=`@gh6+HC z7RP0p5N^BIxA^X*_Vz}qbyjYWxH8C}EB_Mr;1o3@)6i7^9v`2PWjy<+${kqIx!)y_ zkgWFID7w0+=%~>~ykl<>T|0{{B-z=q6j9w;-gumPwXaoFNCutvdM?!y0^+~Jj5H!C zDanoil%kfZlM1H0^yM_<@Av@?qqzq}zFqN<|BID9e?#@i^pV2H7>?6d0YPZdgq?RO zQr*nuTLrj5)XNi9`7aybl9UQzTY&8TD^3p2C9 z`DZib_lC7nmDr7j(WIk`@XBlTz6nYP@&Z2S?S1h<)D=)e0+k=Nry!))I5-E|2aQ?I zo;Zga8u7X}@+6lP?q^%Z?X4qzYHu z;}f)55n#JsY-)sx7`!_AQ;UCm=Qb@&|CJ)Y!eD|6D7h%#CF!E7p7+M@Nd??*SGwi$A>&UTD@I&uS1xqA!rQU(1#C01;d za3aUhOqQ7M_G5$Y<$ygqp1O|!(_3oHc`K^hQee9t;^G@!exX}WID&-{%I|s_Ayt*V zE_AbRd9r12>$-^ovOQ+qBfKJ1sWaIJnOBy?za-b|o#|3ve%4dHhX>oPKb!4xP1@-# zO<391-Wx8}+pDKUmk`H&`}WyJ{TboQxmS7auf3m!xwN051Y8dp<2!n{x3^iZc8#J1 z$0+L|#>RcO0&IvYsey%z-*zc+X>~(fXV?L&R<>^8+)A5|nK{93`l8b}G@PvA+A1QU z>88dZ=|i)Yhkwh$@rC(UT_NVf??;Aash&jUyu^JU{?bra|56|7vbxX|uO=mB_J+(C z#DkA!TPyWC66&Gk!vkn#b62LXY-il*B4Psi+>))lcnKui?`Px~$D6XsEN3Dre?;v3 z=6OiWE#^k4YURcAu)V!IR#5gYJ?4hy6J4x>1YHc3k_lBS!R$lg04`$diBxri_r!a0 zOdIx1E%sUHVzI!;rlh4P)l2#Lv#?JCti%?_XQYeZp;)CEyvi#a!*aGe$|39G(l-H` zlbLk-E@V$JS1=gyeidvblf`?N$% ze7y<87ncf9zL7NF9N!)ts}WtPJ=Fs`V?GJ?J|8 z%b*W`on1J>A6)WQBu+j;J==n&!mCgxk2s5T1ZpbI?e^mDon`)9A3x8e7DGabZu``M zK_P*DL9%`A^bGnr*|Z#7Y@i2KR8-XU0_W@3Rl0JNNwUL_lRx#JJ*(l>`22P)AOPKt zLJD6^Q870o?Zg^A~oSf;QrYgGrTaP!$!@NF+-CGjJP$8(&OgP56tAi%q zW`p_XWzF)-ewDv^oF%kEnRRNV1aKF7+5aWQ(MTjqth4qqed}5av`^lqKBz=a=co&KI2Y@MsPZfb%e~zB7c=f-Yz}r9lNdC7+|Jwr#%x~l5{}d#-J@U?- zmrf6XZTCO^r~jAd{&ODi^Z&f8N18I|q6yP_hf-H*#imlXm3)*}m%#N$U`uXxK5oRb z_?}Dx9d&g{d`L}Eqw~s<)mMjF9k(OQ`qA>%@6wl#bB?n9>+E+hym50=iV21EC6#c} zWXz*Uof{U(&NIvM--~P7kH5(R)8RKLpc5w1ouP2@VLH;d=&}KNLeIoBG&=m8w>>8@cw`Ze$c=9tSz}E|oA%^DLc-{B? zt?zlBwchovcfB9~T%jqQ9o*`!T}-OpEZn19cjoZAC&s1-IIvwke(}Pg zJR&?icU$2<*YxSe^_hdF+}vEqyfaphRs1AG$Xlx+%#FZ z?IT5}AoogJT|FCFVz=iNAJS_*uYj2Z-p=1Ee%A%Y`vpp?J~w-`m)DSzqO|5nlYz93 z2|*e-8x6GWWYGus1jPjv=pfcI%uZUblE&X#vjtH%mqXDpc6PD1Mr;DIqo&f;y2 zwf4IAjp!OgKboL2bD(&kEiccfk68v=VRlYdQ5)3$#p4Ywg`n-sm{#*<8Q)k8NV`o- zGu$sDUz8%vNXr6|dW1xYyMJ1385z4e(G9{tzW>R}Zf4aYYDv4FE|84F^~tmtIE$WW zd9==X1yOPx@JC7fzjuN{aPI}Qu%F~ExZ*;2eTlP(;MygB^68dS)O%+@SccNlkofrX z-uG5H?2_1lkua#NekF?1tmG2SXyga z7qB#cZmG952<+(JCA$g%n|0mik9QL(g+;=!<#5_*V~!(k;tOFU zWEHE-BISQ~CFeSq`-LgMY-(1%=+Pz9?6z|iwF{xY>rG_h8)yEQ$79!%wqL)fWm8a5 zScpwcZsJWMB8YOeH_FNCIyQ#G8R^r0CMU)yXlhDiJk+fHL*_d)tcH(|AL_^p_MTQI z_e3hFb$;S7KO@!c{nC+HImHADJ3nw}*4HPWr>=gNaw4kg=aUn#Y(JfAEG#TUtMT5Y zd-FzGNg~@TfnwzmIBA%e7`E=CWFmuu!`0vYMAS;k$Y@JxKIPJRr7a_-5hnFwuLM0& zm{%2ZC)%iZv~U}4(9rWzMg2)NSicot=T{*i7KUd45}-f5EPYQ-_64Fizc4OkDIz2q zY@oSW$=SI`n);fJq@*9gyRPESwLBbmJwo2PfG!w(eJ_948ou%&)T840^_Mcj}mOv1g^{=)_OD;(CAc*_ zTt=%Uo4Nbx0{VJqgQUYYKj&0qYs9)!^zlX}euIJ}aI(sutWQ=&jG2Skc-vPvGu)A) zDwK$8vNm8ZovTjVRqzx4AyM7&pF*&08u}UUt@cL1Qp+LT88hw;IRzylTRR?|O-BlW zw5!XjrrB4IUicHUy{k%JyNOa8R+Z`9;X`c3WOTP9IOs#C8dr9Pv!)-0_MTaO{)$SMgD!}It3+6*?cgi&dizkjxMZSmZIokel+K?3{vctIM)^qr z0>iYa`wn=`_amah@ItNf*ukp;t|iE-6`R)o*(AJ493mH|5E4iw_P= z$NQO%K+e*zR&&d8h0ZYgIXC+Ylri44%4(S?bm<;pJ{^}4ZCcuVNu1+-Ud>QPTMh-P zX?ewWJ;VL2U9I<=2s-%B3v;}$IpRBE@{yTI89DL6;g2XDHpVWFMujYR78jX!F?QI( zR#{!@A74I|=^g4@)*U`vI_#&Z^*4k4vwnDBC>=m&rDqj_tBb1ZsA$N&O0c^s0HLl% z=1ny?7NSy$Y?^DK`()MtSu$pio~Nve$+pqe?fpV;33I(7eua^d(dOyfvth}P?yPwY z#Ib?Lzz8Rs-E38{EiW6Yw}ib2%30B!_H4qJmG z;okUt;3xO3wv8*YxC{Ywgb<(>3-fVnztNJGll$yC`K$fBTMG)aty`$Ga@*OjDGGg> z#oTurK~C0IeN8?M7|_+7AMj?0VnnXGY000fjaKrY1-JKIJew}`()M)7)V;Xz-qpS= z7W-zJg^Hbp4VNI%w@ty83HfD&D~t;ZL=#nL%s3@~j{naAmegU$YKm8<;vq3PJuM@g zFAN#5K&>=t=>!{+Pq{y@CQYui9vdf5$i3UgX!DsBWH~enE#L`B%Cj zH5NDDxM>+0rulc|k~&gIx)QgE9$XY^=T;@9oawS;3nMi+Zm|7~%r6J^Lt~xwu{&!gU-#^dvluWWS#2}XdmyWx@=!e+{#?_xP9pS zSVxe_URS_XaM}r#jci?JZ>&Q8+lbLEnG|UA4yiRjhL4K+qLio1F=P z)8(a+i*sn;@`^cGq##hfJ|-uf9{Bw&q1&if5OrEfo(G}Xpv4*6rDRa{uIFaWIrERn z)m=Z?Vhgo1d73)crBfs5vRdKX3W-Zg8#XLc83ZQea}VzL$cpp1Vj9MKw&uE`cnY+X zb#l&G%&zskry`sF9{re@iIMTd?Su^7K^!KwwzhKXeD+(tTdRrbJMj^n<}OTEBD;Hq z$WHpeZVMIgshEEIwrPC=@|WqEZk;L25euLm4E@5ztD}Wo(ze;1s6{hOLm5&xmdq>t zm1Jb}TK%xpYHGwbwOVukJk+Xj``Ot3i4MDs>_!fvT+Fshsa?I@>e+5+XrBgFv*q`| zZE8_m(wFqg@`{fq>Qp9+h6dghn;0H0Zn!wYpgzH*r7-u`a@K_*$?iKa;dW+beYvP^ zNT;wTX31w8R5(#AP{JOla&fgLL~xsNRJ>6ooP)#9bi{)WK~;iYT|C`9ZT7|ZmNFCd zf8N`-$R%H|uBdVe!;@hmOH8OIzWW2XC0PLPDJ-XwlE^`lD|jfExSk4z0Ya+vl_#*1 z7@0#J35p}3ry%`GNhEq#DwEfo`?zn<#b9>HlcQ?X0YWOMt8;1;>&^)4Lm12@$B zK{ySBEI%upUG4O)h3OFC;}b8h-KU)Y{r)*+wKD_r`LQd9i3ww<=lf6EVj4R?1NU$f zZ7ehC)t*P+C1;V|S(J_^+t{@$g?m(TG?f<>ZSA{61cw8s!`$S2Q$rIWgO{+?Xo|lj z>p`88x^iTEn1;H#Xuw)}R&qi@bVPWZ1Qakwd_d8GA<5g@1wBQDim&zRC^0PSr#RDG=E6>)h!d;DW)#>(+8wb>Ppx{CC(#` z>0f$53&fbUQdGo#!@0^ep7Qplf<;l0?HrzhRKUGDHfrr$3??_1tnWNGcis+o+?7zL z&H+aY3w1U3V_)c) zP$EI&@hR&0n&>PxL1PeyJP#>pjkdt~(FlH+8gX>Oq(y7a{Y21FpHMolF7EyD(NMJ0 zr0vF2W&r)%z*~8Vm}~Sg=Fh09wSv^f?M_aVcI*<)`xOu27b1e&V)-%CgtB~f!oPHKj{q+3Pt+g!CN{#xH*XMGmZty#JAC+nKy+ZXm+N+Q=0=`>3T>EYqK?1VWd z?I!Xjj7y(YY>L?TVyjVcn!_?16E!E3+6OX;oa!DsA4V^QnwYH-q8$CAY$(fVVK91w zxAt(jQ4N09eFR(1VVZ3~-6uNx+OzUR%W3GStWPk8}$UEi!355HcQn`-$ig@MuLMtFN$YaW(;cfEITvVHm&^ZT*i za!4A$=W}*AUlnDdrM-^H&&fL+-VB({4?{;sU0b^2y?!9~rIi`PVV&p?#FpzKxh<#M z(6p zD&&ul!pGgM%S>$&yJB@#6XcJMUWNN#gF4kTHJ|px=Xisy=+pO1kEZOlS3nUP6M=Jc zZ)0?#Vk2|Px8n-QiO8cXi#$Hxt2I+5bp?}T^R~Dm5NC}CefH;EyAOFUUnuYtf z4+alvIxFH6<0e3K+wzM-BYXnVZoLSz%$&|Rb}t@6*&_k#+df=IRIujTN`}Te)8vov zqJHvZXb;^=x(xb|-l|#G4@>DfjWM2|6E;0Y!}hIo9M2A>@Vv`TtL(JFD?R|s-+nrZ z%7oEW)MI_yB&H6RSSM-dB|(G|Cl|M+-n&*KN+Ii76oiVi;x#nPg}TG%dF$-kyI~Is z@5bryIIc<2frCDiG^@(awB5zxQ*5lh@PHmoJ7)+`DJjTCMYkbSfHgc)c7gydbHj0}6ni_dhNSRzFe3 zripMc!dt=f5OzG{%G8ksEy4W2u9)oPw5J;gsE-g}bH4F~&9Uk09<=Fp$Y;vn9omw5Gf-`uqCYez>*n+zO|bs=*FG zvzR7Z>tJUm4VrQKn;}J6p~3Mkq~Qmxo>4 z%qM?=&@KJ`0q9+VCb#3>&~E}QY1@^&;CUVF&$W?bOLA| zY}P*PeSP~nxVOdGxGamTJNw9w&)`xrr%SuEwP3JmsRko%1qe$#+u5Io#(mv1Fz4DY zZcC^WkEGz|GwoKO%Fq*^j&)3QI=TVcUZB3@RxeSYq7yn8 z<)P8BzBrlG%%mU2?Px58;PGQ(Er9x}sb!_+iMS!N@{P<5lG3xEsrrN5@hJ!sCuLW> z!NACf9z(60#_Ll6-%qL5QdAZMklZ2?R z_uqzWkgA5HFYC!E9+mBf)8iKg3m?5S%L~hL^MhVh**4=^*hAP|7raNJRkY^Z8lVT1 zRq?a)RWB`hP^iddG(Z2rDzo6Yr%@2Zx!4`Pki}ac>n&5L2E7fnC%qu;Ob&-F=CR+R z^rYY^pl|?yC|8}6l0C^$*?|=f=S!ParT%oov!3wuA#lz3&(}M8r*P)~{1@~;!}9<6 z^ZyR5|Hsu5><0f6qW{15t^daV|Ct8-?|tvT{`UXTw?Y>~i2r>3-@e-a{u}?j{{Q;1 z^qv=9vIo!>BU3sG^;$_pO-FV~_?f8F+i(D(XIY1dMn)a*=$IoAhzLq9wjKp}c^W(u z;%GHBUH}~dyusDv&Ye4Wq4M+d1^^YE#S1m-=t3bB;B=N_nPXs-3h)RzQLjPTCmN8i zIWV@l4u(%wnu_G#5{|L;M#UWL#Q;)m>B`JcPsvPr@|}*C7Lfg?0dS}WXJq5$0?dU2 z_K1jxyLx?pa{oK|={cxz z-q{SwJ8u07HX`Q8j>G%BkTQ|6W~C*6G~&ZB<#h604}=f((V?69h{F?lo`SZZpxR;iM?;Wvy3wdSvS?S`no^6#^Xu__9+?-UA(rYuUi9+I$J34vhujM6P zz0$Siz+=q1F|3f6nW>7g}`#9Zsm>Pa4~k@zt3gS+y# zEC$d2lW1V09=V{soVH9^1#}AA#W@PwY9f{9tF%VW4O2$wbGIE z5W-A=@w|PQYAJ9-PDP>d?D&iG-Xb&^=yr}zKon*|U`Q){&%SS@$kqnuj-LSP>J;QZ zf3JCmePDROA_Zgv6GxA#$E+IMMw$FpcLFa=Ly8__7W8G(TQ&|N+TR`0I?UK^>@Y$8?4m9 z>?at&D=LnZ-j-Sb!VKdDZ!~?kc1TpzG2+bb#ur_aUVp6VTL+oV@XB_s0VQZc62IqC zV-vmQq-{k>NiKj`Oz=cGc{tL9nV48_-@SAD{ISxcV}>+Y>+#Y8DYjdua&o#=_aw7^ zI`70bJ?LC`+5?)dyzy=Z{F+-*kDA}U zDo9RDys{c<2t3v27qfD*`gquv7KmtQY((i{tc-^MTZzd^OG|F7sf=(Gz=BBhii6)w z-k)Wp+grL?)5JtF3=NDH<}Xf=WuK3Jf>zoMl1k7VhM9?(T1bSMp@o!(?Fp2QOar+S zt7|JQZ7nCqlez0C5VA&Y5$az+16JH9FQO807(-&JM;`D>UBVQ!5+F0{crDQuLXR@g z)4MDu0zd6Wza^R8-k}rAy!@1ur@UM>AQWCwA>Pp~hjdbiTd1(2;`d7(u$Dv2OChjy zZ*P7)90)kKv9U>S(g}bmsw##@CV(ti_*{78S+=!&s_r}eX(Iek17hSOyAn=>w*q~h z-?QX477<297Fv^*uYWLRew{{PA|dHAgd!48ov+iSAq!QXS+Lxzwf1%j8Isr+x@SK~ z0*SsP*~(~nEdRLg6Joe3Dd{;2H6$q^5z1+_Qs8}KmOM$IF6jx2pdb~G9`x79>x|BkNqvkR?uO!+EHKKqK)=nd9 zPAZq~?gM(UV+cr0wQdH6@LF%fnTIS>h2jM)U)b9+Pdjabf>%yXnQsq!f9liyLJM>V z==A%&9htTb9B0NRK^*d`EQ_>#kyL1WFu_cxq*Iourn|NE?L%c|X1`H|=FQV%e6?HA zS(%y79mguB0{4C3j0Q$d*Ot6=OiZ%i`rZK)+@DyXrb+MoVYEP3TUYYgz);^~q0H_d zuiP?)v)(18d+|lvm7~FzYtRCAf{urgg(3BTTmh8(qX5%8E-4S~>$8ortgJxD zS@RUB<*{@UbtJqE7zW)`!FZ9Zy2R?AeBtHH(eWQ!T(SE=0lPk37D$s+-qo~mGs|bu z2%`&6C7Z};Iz63{sy|2j;wnb`{`%Srm$4U9-a}Q0BSAZB7s|6kk9(srHRS6?vgGv6 zH1R4U-rr8e;<7|VBwiT2EXcwWy+i4Wf>J6o(&e{9Q89;%O#AbdImx<|Sux>Z5#&A1x~pFw z?k?Ok{e-b?-Y4U=u16a_fDe_cl+*5H3|9<+JdpyAsgS`q9{nrDGlCBelKuE06VmGw zl0v50Ot$F()El4Ilc9XAp`j*UiC_HTyl{AV>om_sI$cZh`(rO9C0605?2AK!MAK0P zaQ$@0&z`n)JEDC&=KL>XrlV7GsU2>FFPx2Xd06bmxAN)I%86z(2}M&Wu_*D zn59avcc9vL@7>~ao8hzKkA~Px!nLCA+j*>(u}{C>lubZrSGgCo?bRU8ZV_k@)y_5y zXSQ9u_a}{E$7Z<~8~O!z;9{o~ZErpw%t>-IQ(+Aa%M*Ftd)N?AvMU8p8w-ohzDHda z4f&f>E758iLo6Bnnw5z2iG^{4*P`m`02fI5@r|b1X(K-jO0CH*k_U$~=A}9-*+B49D)ZyVDqRk}zq)_u&!!a=<`&vkP&D^0%Ucd^+ZzvxgH1ibBU#M`w+?iw z;BY#|O=M+ZAsr~lU|^<2DL!Qtrl+?$jXVV`j3(*3;xTa(DDeCT2NbcVZcKGaKYWt4 zt*s#wuyXAG5+26X2Z1Hei=;el1ZP5$53Vd^4*Os26 z6#$|-aU(SsnF;)92~YMtCnhX)6r;>gOtt8M1oBW9dJ z(dp#gLhil^5SRTr7$IauCsRHhIs39&h=e!a7xSRK92D&E&58`{$B`ZBNynjttIi`RqKJpl>rZ&CWm~ z(L#0&DF_8!dP(_EgWCY-MK1KV8Qwz!(FX>rw&u zXVzwj&)@}pLSr~kTG~HTxSyTfaZ~u>ZXO4PuvCUjSA0d{sBAue><>zu`;-&jzR{pk z5XNrN;sHxVb&c}J+`0*cA-Vz7q5QmK)h>s^w%D0_&B8ldROlMA^-N*FXY1CkHOwri z)@vmL#ri-8APJ4{Lu9-zD@Img{Bo?^w-_Uq+hxTiRz~voiW-X*!y}lOna_FcS3diV zgU9uZr9=5tAgIdeL_Q0S|U0l>?Xi}_uc>;7D;NUbd z^WhBb7!$_c=|Vw!*QUwj{{VVq?y;?oHJs7-J4s&ET}UBJ=biX;C)00KG@kb^T0|PK zalmYMaJ=3UK6JJ7iXHg3G;yOYP*R$$KP zSxm36A+0XqYaSmkj$7+WUZlI%jne<|V{{!X`*vm~z@&Ob_`0;pKX-ldX3)*+-T(`v zg{0$UI(#N9R7c0MukRT<$mCzcdOtMS2`NyJlZ#qPxEYjym_72~Awv)JCzL>8Fd}pL ze)h5r!Pj^2h{>CIO2h+NEcFIc4z1c3%`$CCyu2NM*4D{RLbKdI3R2~`}KZYSXJ>7VHvdXI}@eb^A#{R z16%!$=ajET8YB2|*p!r6(76Go7~txgW_~=e#Z8ppc6nEqkm`NTPve&m zxc{5&jah7zG_$0I^gYV+VH_jblq;NoN(*f!ocZ3tXw_+oIBm{osHu(C;Tl&&1SO`l z@@sgoP_s-XTQKv;)&)K6H2|apxfGa=va(U5Y;o4iI*1$j9&(kkq9O#HugUIR8v?C= z`wW;^;^IdFUO}OcmRfTF8nnD#!pUVML-FPuo-OyA^yKzYejkb4n?%Wv-b4lZ5M_u9 zk6A$Lz~EWU3^uEphieh8?o6k2Nj?nBCpU%NDexaj(8u!FzCOsvWainVPeaz+aUSk0 zn*;npE+c$ma#Fn59+16oK;0`eSsQ%)aDG+DNEF*nIMPyY{ImbIUO_ zw0DEP-MMprL0Uk+FH^m7)`y+Y4}~N2Q4gHr^WWZ{!voqdEH3$fSjLs~o(oTtoO5%c zz-;DCwQ{Q0{y9plz2#W~KP}Dfo2pxmqkfrnc=YC-v-f3n6g4~Y4Z`wTiO-l%6iz8- zS_|WA3#K!DUcqic)%>^#O-=Ojy32nY(Qu2fqvRlNd3pKbbbmKWiq_FKDqTC$oL|a9 z_M2m%?sc#{3&47)3FbTH-GQP0+0CE)#WG2f0c{5SESbj~U(wOgDYuRvSg)SgA0aYI zCMH}>pVG00pumvgVBZF0+$0E^7fd-X&tbD69>=789x7e6QC_eNl71oN*sLE-$i2S>HJN+CHP0_IpVf^%#JX|=&Nc<)n8Xl;COKEyAxahdP#;o3}7AGAka=-3YENmt1e00Ov zCiVQvJ)wf882rHOdDO}4+QZIHsm#mUS@qv)PnW?i13OuHcw*m>cUYE}=VTm8J~Y-h zG1*>M8(2fEHah=)7R&yPdhK;NcY*k*`KvoGOlef?d^$l4kXxHO2yrO$M2Z7&?_ zvKg3%CFNZB$8>7;psi?c7(pPk*j0Hr+}ax`^9~Hir#htqmc6AgvjIWF0n6=6q~-A( zPq@BNz4*@mr9SsSUd1TA3BX|P%d=;6H0#NJq*)_?OCQ#BKX2pK>l9F!+zjLkliqJ{ zATBB@8po0g?si9mFAqT0M#;jHAuBHowdRB~A^n zyN{6td1MICYC3CeQUU|loGP1RgQp}EETb{kByOFY~q@96cx!5 zziZ0LX^-D~b^84E$NTm>PJ_~g{{8q7*JDb2iyob>EY--0z@0Xc^LOdKaDI7edF>=e zvw{oGtr-EmiC+6)Hl!NXAlDrEt>tkaKH}!?bjc!-sRxPw@_4MXMRTCBwVC#oMp>!IV#jD)&Q3wot;TGRV`Cgf$9 z9irML*6E|UVOsFwG?xF@#|eSwIr}Jd2j>nn{v*ulZ0*lGjA?%rE!tKMQ7M4cD{S+| z-Q=)n5a$!vz}s?O+x%rmV(J9#OvMafoIVuO@^JwpgP1s~cXWYFFc|JR{)X$!(U72+g9UUZ9d6CN`lnX`3s^(^ zrRB-2RwK@3X%uS`!UD=~OUe1Q;c~AeLf>yroWvq~dW3UOu1lZw8)<0))%b?z*GUxi zD}}(@Wuw`WDhF<5$wgVBe(gETL@_Ew56sxWJCW*Lu)T$uw%>9hkBp29)m|VoRXoie z4mp+sg3AObpBsGwumlW%dK-$dwEaPudCukRJz882o1p#%KE>sm!--Dbp2;MxAO&Da zh7uNNN)sEw>D7^IUKeHMbdw;<5&V`3RUY3%nSN+pTO1gTEl0O578h~q6zfsdB;)w` zHgz8y=HNYNY}MNod8j>qNj1^C6lURLCB!8LDb0JgVeZ^@cG+5v2DHes(!%U<-*a!z zi+3~VPfrY)cPUz;d z&*Q#0ex_{mBKB$71q&)Zj<>=7KNPw6bW2w}LX{&U{Vza69ZdFPGLEtpJjvCa>e3Iltje43~;C zIyEG(*#7iR5w+EX!~`_M z&lh^H1ITse{3!UR`aEneuLFXgl^I}!0VQh=BCVqN7fF!Gz;>~%3+a>}D?kb`leu~6 z*??E@wkUlGX=s44OjS2jr_GM8tl63)xAc<|UJ`QonSCUx8jnj!U5*yyGi&=!;Sj%u znthrtS3`7hd_qaD>L)K0B>_D^`!pLnNBW?|5Iemwu3DhVF4jU65c~1z9w@?zc}wcj zz0{pWhd^KI+AH6@8P_{CHycLCa;`t}5_oN%dZeWMz)hS&-Qx$`(g;xvNop##!?U?1 zY3Tr(XsNDCYb2|KhGUcY{e}HeiHL*jEC!Cp{@T!eErjsKXw6zcwX8U?yLzSdOPyxA z8_tg=W_qW+yu!#}2@^2qQbo0FXU*CqduUTZQE_=o-wLs^-Qk#}z_PwbJ77bTVW zJ!2z7Gti$F4OW!MhOl{1=l9sR6Q?hJv$ck=fOcwGtUc6+IIQ`^e*GE3Lc3aOH+0xx z9dgoZ$fQE{b!nmwsRa6_%D$whJA>(kjcHZ+G_%4S9az!SCV)yefmj}Em@r>GcnIFk zoX9SXO3UM})8_l@nzBJpQ##3E3z&donLr1yG5-Z}y+~F~0f{)+qTFV~fxWk(!5`V( z(g7xwj#!LNtyAP5^6FB<-xkmbfK4cbDhj-+Ox4w!eq~Uk_>U;i!r`qoP&~*{E3!06 zNYvTd%HlWCHz<7V1#Hzni0$X{SWOp3H#t!dOOpHfs~9~i@BI8t zp>|~};*sHA(5_!c90Fc`krXcM+&+g82g){ zO2LgWRWp-p6-C9x`NbN*nRKJ1s&{*i;jt_j3nT};$d+u8ogZ7Wb-8D@^F8ik>M`Jy zRJD)>C?N?urC8V0NH@5dV&APMm219C{veldu-P5mN=bUNu42Q2Pgy3)1- z=Naz2XYlH=qT{U!xi9drZvm1hP)T#$l_O`*3-t-njU8h7GF?m0% zT{&xj$Bqd_y%1tU^asAl%hLb?i3OM3TU>;}-l;Hff@i+R6{iQcxbB1sKiib@k^@9OH?)hE@pcEsK>F2gbj|3C3 zJ|Up=FPK0k2;H-KM6qUmjC8f|N zK^%N|dh}OQt!`cRvG3v+S}Zm*`UOKw79ydm9u4NQCd`_TJ^CB@Q3$sT;FKx zxczpU&;v-pXX-9r=nf{wJwfsw3y?(Gjr|S*^BzH`-41j+(K{cP5yvWQ&)^1-puj&; zUao?U`PeTqrcX|;;G#c~I#TYiYJ$bE#Uw(0m$+wqp;G0oK?i_nYt4lY!KKGnIu+lS zEH1R9s(&B6zmE>2W-bN*s7=uzp^}!g{X(~G8Cnn`!E9bd^zMNEnnb9gJL*)<^0Yvb z+frEEfv}2JflFfn z0dg47Hp2OlJQ5MeG~}6P^*VZP2{6$cpwCIeB`gMA_#>YO~sXU*k_L|v*iy2h@M z9=nP6Kpfa{;`v7^s*YR4arjwmmWK{oi-`vOOjpNwavw>FmU#f`Ga%QYZZ<_LG{sfB z`tghc)C|o1!H{>KjPDgCUQ5F9$YARIXyv6S5sqQYu{SonuCrGJ``_t+0D5oWU5F8OxsY z^bhwxOpua$HS?!S7|1;j@d-4vwJlBi4BkyA0N#G>;grO#W~fEE!DW7 zO$YlAiw5Ufe-3Ho&|0oc0fiRe zdkWcN?+1OO6Ojju5b69_i<_(UL5`F{)KDPFSP+S$C%fyrL{J@$93a6Ssc`jX=?BZQ zg&+diM3)dcIc;rSgG{ls@W@d0!<~uCZsawP+toR|uvS7adU?*l1s=SoI}_umC{m(z z|0uhzT}zt!Gh0kc9GJCY822)Ki~5qa;+iOl@l@30MhAX>01yY52hLZI?b1HI@#Vxs zk7k|DkcnSD6!5c!?W}?08MVuUQA4l$>!*XC&`K=X3NEqK$K1jWd*8nK8b%b_K_y%p z7W{TA2TJ_(IZ}NURfw8a%A5Me>JrwxxA-z64jOuH=~ zQ9bo-H2|STGTdz-NRYOcA|$RrSNtPuzN)46aXW`&A~+}d=sYLVYE4%J`s}{EehqK$ z?eAS(crR8<`rvVHYw;a}pz(4GmZj&U-+6hz<$W6sG7&RiYr1h*(iAX;0U9 z$K1y(_x7&x4~I2|3i^?yDTU($mJ>X%4@LAfd%Vq+E2$tyOR=+A<$>vR%E&O_6PhEIKl5yx<)JqvQ9*qrGg~<&*-3pl8`d2dMrf>bO$=LvT6a z2@O?o$A6{Yp6CG0-E+5D zhJnA-niG}h|BhC@_&NVy(h&479q`4U{Qdf`To~{FrG)t3-}yhtfdAKArGN8}{(t`1 z|6DBezblyjKf2ogeC&Td_3!oT{a?Px|NUbx{?Fe}{rCO2hjdPNd#0IN1)I?J`=KEohW7Ip_F zoo?p=J~liYT-@vf1*`#4R)6;0fX!;94G@6IY*=t|4V;b7-T0ED9Z7fuQRxlMP6O2m zSeqwF8qeuityfVCfaVlQ*jZ~_8|y&Ex9N%)_;HpJzFNmLk2SZ^(-O1h84&$J{m)(H zq?4aMI7NmJR7wt=RKI}3-2}-2e@67)20aT)z|n$|auW$J;uS>i4Ms2Djy3MtN=Iu7 z4=KCMkb^l6Idpq7jSddKyE*V=KajGJic89;HLLnQMaI)mq{R8Sr2E4Vwu`zM4Eq3! zw;}!TbFk4b3ju%1{P`4~N=irae{N|~a!7VmS`SZDgo@&Cdb-3q8-6gZ9ZuVMSNC&m zb3tENGy3Y6DG3RQ&G9lIw7#YB;9ge9?K}7LAi&P5`X~y{R%_3zt4+UKe)OV4&vDmo z{d0BGVy9o{$k+KnGw3aaPe5=Jb)Wa+dT0-JZjq`{!1&^j^`dG4)fSdMI6*?4fS%$`X zUY6H|y-7^i2P`+0+(}U(xAh|rRLfUTtN_k{XVeU!LNnD7;6+t9HBthW11WbVZaf+& zSmCuqGSCHSEB#f#Oj z0s$bKxDex;iPZuD$}r#=?)s$K*_UMJ{JK!A4WEzIP(5GXcthlf$pfmg5C|piLS6lG z`8Wt@M&{QH$i?n&$a*6I0~5@CfB|=0TnsBSwFTg%0NIE1-Cd`s6ENTg15psCP&F2# zFM3x|Sw1`eE~o;-jLu(6cdMU}BFasBU6&T830X7>GBbZTAA0C#z;1ee4fXG$oL144 zw^=z7ln;xJ9-Z$mgSeFyns9-{GOd>4h-!9tkKa9pN9mRJ*=%Vw$Uy)QJ|N0jjaL~; zsPn1GfXZHquDJZo@MwRN5>c={ViN%fX9lytfCuvw4-C7?tW-tlroMg!^G{JGdU_sN z4aqN@i`LXPdo_XqLripGb}=F*26!uol%4gwl5vQK##N&Ia;kY+Jr(!bXp37Py*<$b zlmCgt;k&EDwt$O++9}@=MZyL*wA$1T}zA1rqQbfL|0?jq`#~b&`uL`_n3gj ze)kaugCi4Py-NB3s8}`BxD$oVf=gJI>W_H~u=0Mpz;^@t2h{)-X{d=20*T0biZ z-BDDz4LTa8v^eerFN^hc<+<;YF4^Q(%@R1#*|bc!P$Q%c9d)L@EEceHfzR-#`CFI(dA2!0V=E z0eooa!dhKOpSdaI1+%}y&L|(U%w#i*YLGHNn`*iu`(&4oBgh(>@x+ah?R!t4=~x7 zGi&xlGSwEjz@xB6Ci4|y$woG}d{Zn0T-S~gdv z$CJ%)c8s!_B*1P320iqv$XzIWmm~fHD-qZdA0iBJ5S} z`{&;i6u`~;dK2sDoY zl!9N^agx#0WgEOF^U+KGt*Nbdd@GEV)Z|LXhQAh6Y;jXp$T{6E_-H*quY-_Ds3F0N z8Y!;c7(Dsr&LFO(rC6RAH5#Y(lrY zi*!SuPouF%+ut`=vtdemiqiIjl{itj=&6^dO3$!^8XsrNDroQTm%V3xM}LfvF(Z|H zcT;_@AxOmAoe#%L(B@=fc4TE6u!GlxZyX?SH!1I^?Ah7dXxD&h*UnWB z^KI)gw=&^tBH_ONL9NGvgZ+70+UamYDJ zQlfxlB43hJpfS^PP!ytJWa+<%-z4x4Z-mCZOy{Y-8x@x+6 zYVTRS_P18A{+7kBWhIVZtv|o5l++$KY-duJn)DBuaXG(f)?YHGWlcTbV<-*4rhXZ1p!y_M3DYFZJtelv?XSK(2xjSf$a zoh>I4BmMn{4u0ARv}7|QmPTqrZd#VqE=^2>{nQ}}rA?>r4 zV>~vFYG7|#rGuF+t(`kz!L#4LcY=vB&wNPQm)=F*V8fIm6`tYQxrMM<(jLNvoOS}% z%-+)1`Q&K}t+pYu&*?ZDDs%Q4O26gq?>SWrDPslcD&PXGpVT$m%q%VCofn7iSob7h z@vg6J)JQl`qljn;#2p28?vY(TC7I)2Hy{0 zG32?U;6Q9|jY5=i2Y5;M7qyi%{Q8SPeC~3+2kN zRDtn{fq?-5-OT-?nYn3K=NTKj^FJ(}s$rTE3&E99ee#amQoJ8A@U*d zI8M|ni*9F1%dQVld~yD`-0cl3of)E<$b_b*UL6LDv_E3A&!aYqn`8y4`cY>mtla`H z*EA3le%}5sFEQ~RHGjcZ!iw7L3cjlK&TUuT02*_9G4C+$oyU4PLs|&CP59qt%!<}v zFzJMN5+o^a2e4xOb`E9LCt|-S|HRCQ{pvsO(ZbB1NIVVAnK_aJlYry1d>HK{k*I8K8?>H~4EwG$tSoj!D9M~e|%QfwsZ!VUl1s0ek-R36> z)pJakF6MWp)faZA#gv21w}W>ClNBvU36=QEmf-*)zU~DliEvNH7kvc39FT!SJUcVH ztgekZ@)86E!iYO2#$virQGmwVcTb<0e`c>-GS9Yy6&no-38lU_!esAR-`;&pH!LPj zGV|)~sQRX;25rzz=G|qxEz;n~^1j?>L8D_lZ#Cq46nFU*Uh^%^zIVt{_Dn2(A-UW| z@&gf5==dwiA@5Z$b18&?L6(b~%YE^}urSr6ui})Tmwcq;f=9Y*d4B`AOEF$6ATI1_ zuO1$itrZu)yU@Ts>E!Ah)YjbYEsROZq2KoO^zNOj1`!wQnck$fpqHR_0}-(uo#XJ{ z%vJX-J^HA-x(jNQ#F6i`mgh~h>MFjMDdSq3R~_>qGN(GNefSa2koIkPK^?clrm~NX zbRSKb*N#E@&O~Vzf4Nu>JuEVY}V)kQkF?6x*V^Bky zf-N(jV^C#@m*T7YoiR8Q22TDT{U|u@M~hj%&mB|~BC~$qt0fn%^r-Xxx5}*auTUp8AYJ- z(NCVN?2=nYhY|@YHyPR28fRu^JM>xOy`K?=q-Ex&L=Br4aN*(}i&t7CeYm>(wq7!6 zv6(B0E>8uDF2GB@=;WGel<{sBA7{>rdCm4V_3MGBn%qQV!y1pzK2D=AMdbv4{`_{k z-+2tNw^ZWnyBF2YrX9P9+Sw58(tJlmDYlYq0EZ@*rI;t!a>fOAew2K^rvBrgEf->0n4FZ-cF~TJQe54`wb6e)I}2@Zu7r(-BR77(3n36D zpuhS#ImF@s)@Bw&(A~V(KgjDmY_D#XegUf6P{cSO4=Yl0n?=VBn^>E0@1o z;}2iS?{;h|V`O;GqRHNRFVF{G!8*-GoUJK5Y?A25qy-Y8_`6ENF5$U8DetMg*e1>&u8lUsMP?gF2=bsfCFM1wt!-aIMO zwe1G$O=iE%;&1Ol^!j01JR3Qy{JO5uT(pu|&V~Vv;&i3Qze!9lTyF zd@v>-1S=zU+6G1!90xxMhCK=-CRx;(1oZ~?R@xoUlLYkFq5bR`zT{O*>7&i zjt{?)Y`5ckr2$7>kytW&imKZCH21Hfn;Bwn=o=dwG@rE4Ud+U)pX>dYXs@(azrT$wNH*dzkgA%Wc=_(r%!jEg2D~$L zPcWm`*bn7FsS;^wnff1{z0qlr^|Q+RloApaj4b7869AI+={PO<8~O-24vyrF&BQA7 zFq5HImwa~RHP7hgLmEYdz#Iqu&Ol_hyIW-r?QT}V(r8_2>Bf(UO5)}}bnWX|${F7< z4GADpI$}(ILcZ*7KZaB@YaactCJ|M6M4N9Z{^HTSyh(YVYg+H6XiIw4ffVchjDf-}d4Fg(T=A(|}t& zsLz?Aii*9WHIez6i_5l?D^%F!$fd@faIT%j}+-sSZ>kpPpSN9fmA0~-HFCR29M#{!VU zX#_;vqxh5TF!t_-8lCIbw?>kz0aGv)QMu7jur@V1%KcPWPEgVTrCNst386?C&XuTw5#`yDXsJidQ+Lr zQ{;({xS&T6SiSZxZ5=El?2%(+?cbi)YNvOUWr-qbn4@iCkI~mR!XY99hf`y?)Xf2C5u;ZLPG7`5R?Y-iaK_WGq&#nCI~m@o-@WdD&vTft z_lItP7hs^~Q`5&i>0DV8SluaMH^K6MbT zObj?rOtrEV`epAeM1|1G3>1%;@D025u zId-A#Y-AeoRqWAm)k0m#=g+2P)puy|t2H==4^?fBuC8DG+AJ?rNG&U^FzxcV>WoR$ zkPoXWtr!oxIbA!LGpbn0H(ej*eQYx)8yeHd`l6n1#{3JWjSrtECW$n)C5;5XbL z%KG;6XKd`h*PmD+r_U_@(rG$ z^l3r#7RmeXOFz8tJ-9OWp^c04m?^|>?-)@j4QeSe!?2qicYkHQKYdCiX%E#P{Drt- zTz{hOr*E?K>Tc+$B?Wi%TQKK-dbBU}zE~j#0%M&Z|4`QQsD0{uc z(p?OETFFlIaM?U~$iCu?23VEMkwD`Ong`y(*_#A~Cno4f3)kRxW!w&1d@TB0Krk68 zjurLHv-*DPHLYktmb;xj=({1RxC|*7C$w(Z_fKR_>=`qhoV{%wRVwFOe9@~Pt_Z~2 zMFJH{qrJu+CT)OIzRC#8Y8HAC@PeJgn z1=~`|1YeWdvG~g}C3BTzKJ79;JV@OU&+8t~=?-P_9-Qf?eDrS7>AB!XKo&jeIR}=o z^f=T8`@4$KGR%zRovSO5(bUw57*%@Pk+TYJq;KGg?)Yzqm71G%%RkQj^z-)F^6?RO zuMCj->>nW_RZ0%sVVS9~cVH&zq|0qVPYZA4=J*hd$7Tn>QHlSd`*_K*15}SYH3=NIYAnLoXn0p!A z_Lb`N)bBa1aHZ$E%&4<&A;xtF-tmcoKmjScQ0xjOfXgry@Z*;N$vTuVwK--d+5lC_ z$M<+|YoN5W^rc-fJ71A=kMif_RSal9>}3H+%uFe%zsU2D!`O%T*+zLi%w%PI(gO_) zL<48d+;(I*1D5J2W1&#hXsP*mP_3&o_@4VavtGClnSfm90BiJ9ThdwsC#vA0ErX;v zE#g$bMs((`s;I`JRHM==L1b~@_*V(g{emX=il4?1Mk6n^SnEzx!;}Ixw-u0M4xKtr z)^i{B*t~l;X|FwgtcSgFT(hCE0qj956>#`m3@DB%uaEXR51E+SI$hrX8vcmrxpcsf z7MDBMI-*x!YJRm&v+TLKf6Tx{fyyz8hkC#(R*bi%@0lN(o%{SDibq=Og3KmTx^HY| zf=ym3DE4p@^VRzB4ZTTk=$F-red7LmW*xklo$jksuO#)sVa7$sF?Dby7b?)z2xI>d z9ZxG9x1G{LR*P{J{c=|M3U}{8OQdnFnz8`L{i_#awmdvMX%)++PmPs*3$c*zl~G9m zJ1gs~!ogg7%rIMq54Qsw^M+Ggv2p9nz0Ax^LR0I`Hl=|mm$j8;-FMbRtLnOm*QyQO& zN&VmRDiRKSXq1%jP*K3RczW92xhZVj^87(89e>(_tr0{fF(;#`SbUaDic;xt_$j(!r&q@Fe z?b$PS?RWI~QR~(B9dVSKZ6*XyxGi33T~wZhGkbb!*qewcWcP^L*+&Je&iaPAnfxMH zPBWeQsZi;CA%-h$VT}OH3KozE!6Uf&H>)Z4x|8Y}8cp7lPZ}+`ei(Bfro`nwXtid3 z9=pU|*6P!HBChMIneFaLQ3aTL2vVb)o>5XodnQE0xhu71Owr|z6PNtL z{P|QieQvyU<>zM+z@JCkATD{#$K@inLj3`0d4LjkLhn5Jvi9Izjoa1NnzykPqprEU zR_l2B9aK>dxebfTgQ!vrURU6*O%Htb8|Sg}p-JmCjPD|Y-NmU=t##m^{iIX&(MHRK z)Ajbkg28Gc*tV8kI{xfQ4h+0&I*_^e{8^q*;-h%T?1s)s8T4-Sv3T!1N z_$8G>9Ah4S4jk!eU38)})ap%o#;WyB{|eqS8}DLOvmDu% zQZ3P;I9E+Nud!R|1eY!!vs$T}Kh*rD9AJn43S<52;6e|y``je5NJ#7A!+8==?&G%8 z>hkW(#$!N-4Grmjl^wK05NZ9J#U{5=S~BnixSX4LL**3AJ*T=^uWHCneX~TbK2+wf zx%V*`X6y$YI*{mEX*%lv-N6$&k+an@Cwm{d7Ae9Gr|dK9yNGEFV$))J#T0d95O1u9 z9C>!YCahl8v6mLmd~|bd-t|7GA*xDwMBPrf?ihHj^thF0{t+7~Cywisj^<-C+!f>L z2y-!sBW=uTUr`cAzTM_%=kca?jeLFsbdhbi*w`P&0cPkyPi>wb_qE_oRoMBl3+i9a z;4&F;*NeI$%Mo)P@jQ>kti|&yTl38Ljm`@X;qJU~FfpD1CsoxAa^0SjFF%8=B=7pn z)Y<_(qam%>TKA<7pFHH5!NvDu!~GedLpV#T#P~4rz(_1->fX7 z152A9O7iaP4LMh8Q3M8^vX~T6sZKQa_W753;A<8E^TFgt44Z1oEykog-S_6%0i6EG z@s}9-ENC0%hLl(ErKD>NF;m}j?nr2ME>@C^&Df8 z&^ersE)@+&2BDFwLun<6*H!6K1Sjy7iyIPg03qxl6e-$;*hC>@)5NyC`WzGCTKJ67#Nn(^%kg$ZeSUGvwC=>vn*ls_L5JBB5aUmKz{wfqrL+Q&eoL) z1}9Qf4f+~+6H_=PZW-1Rg05gry+XyXe1~VX50xO+JsYlZf{zdAOouQ^?QxFCEM-F_ zq&zqcX+}!ba^IuEi&rbl#ZJFD+FVYC1E}_)o9T$Q^UHNMUQWV9#w20SwWX=7Jdc*e~2}syFNJVgv`uW&fyFUz^7oE3{qto#Jc1ECH>b%?X=5nQbR*`X81#<>USDLSmGLHVuoZy~q*A1SO|3e1BDEtGX3|GUMb_;|v4i)bGYQqQ)N@jk8`9 zq@*+i>g-6VXRHNROk%kWtGbP|yN&vZ+4?%vV4BsDnbqSh)X^=}V@TJLNY~@E3%Oa7 zx7SLC)Nq8kusAEXi>=AwFCHAAy~)!I`j0nE;naK-|;bVAb9v z)rN~`BXD6Gys%AL7+eO1xXZu6%)d#@4;SJ`JgnN_u43J!u7XQdA+S6)*gZDMJ>X&< z2>ksG{{2n*eRu#&Hz+!&Au+A7Lr)1PoY78Pue}^J#zVpRR*p4PdC_?H9uId=BgtK2q(DwJ}Zs%|zo9(WkOl#(w0(Q|p zlrg^d)O<+Qd~t_-Xor07a`+H)_&zN0p)T>oa`Yj0^u-4dKJ*CRI}|=d6uuAAe5lfV zF^zo4jMT~?M+0iM@r6XE0eo}bvf^GW;@-l|UOdg-(g9u^0p8+ghtm6SxqZYP`VA)f zO(J@@AU)zi?nW8sN%bxBa@zE%ePGJvI}-0rP!*r^WKLsk+$6(mABL2H95O?YLeWnvpOV&P1EOoQJElWwMbK2O z&F1e~#$q3SP7@oR#{_u8@lr_&AEL8U$I;<-z3QlH*5|@p)~P$5Yol4*9BKyp*Gc2u zpBWIINU?)M{sW{O!yyv`QZP^X`!t&H4D}4H4-{-apCY+2Vwe%I-a9kc9eT2huif!n&G<_g{-j4bu&7196hb$ zeamL4@;e@&aI`4TCM#;*&neZMRM+~^u|M;P&x3tjLJU%cYm^}nxqS3I#rslzZT8Hx z4{*Eul6c)ec};9@zRNHN7<4m^2DS(w`S^ezB#7=2NSx{Ii`N^erQeDq7;-VmRlo;& z)HfC?7jniCPZBc54dh!ecysDRT$+&pNrv`><3MHdbzQ0cSbw9?wh-Jddy=L5QFhuz zdLE{6QWzT*F{KhxV)0mb<=&b14d_`Yd#n7eOmuR>CLATywO-ErdM;wP5HnUpMg&Mwv zT3P8D7U^2y_8K1HjDg%;a(lxi!ULm#t48g`$j%>GVFlx2@`*fGpBEYThi++sFaI)J zhmyR3yoYmjT5GhMKG-q;HC)~yum7XrYNz}}YjjNbx}%_+zN76D!a5GFDUxWH>37tb$9i+(QA@AhhrhCX;jg;~{WdGl->X0iB zDQS7gW{DJCIsD6)AcgNYUz*4N}V zbh`T`VYeT)+gmkm_fIDbxvG`F@5OlETcAFp&m4a>ZKhLC($uAD)wF6o*yER~$C`r& zIQn(}{a@PqYaY?x8q-$18O~6D5F8N_ps*8S6Vf zR5cRI^T;qI=&B^Jcu`Eh_jCnLx4fPP^y*Oq8~4N49K%-xYsm}0l;qr>yhA&zNIUMfDUhZlkn#{$TYz72nS<`*F=%WW~NVm{7hK5B0cBQqZ{FrN@LAA4awiMKecwK&eVI9Ry| zqgxz#zBnPfICjti8RbTLBD;2*Awz`~!GPxYXay{2{rXOHeNb#nP?Yt`Kg`E%W+yFd zEqC%_kVLn7g*ER^$j#LX>u!CxF|ZhH;kvc@_v}(vtA&Z7({~*wJ=elL{%K{(H=Ay6 z&6h=^? zbIz9|=%oAeEx~F|Fyj}Gxf#!jU zbQmt^M4Ieen`~&B?Atl7t|Yzl_D+Iy7L)0IWaRuveV#tT>+dLjYDaKtbHI6JqL7>lSz!&I(&4-1Om$@s?2YvP=0eE5C1@LZ2J^=OU;o zdGc^j^o?l8m%R6`+RWZL;(;H4X3B)?KqTEV97AqK$vOI)sG>Q@iQa&O$a-9pl)y@Y zdh+}gJrzV)B~&GxAZD~--7v$MyX5UkZQ`ujnksFdz+AEO z9^14@w%X#`$aH={`Nj{H1mOan(I)6sjXQW#It$xh*5-!_BmJ8k(gD}{{=M(KM2X~{ z#MM?3SM~6`8p$k|J$@J|g)Jr8SHj3aR{ZlKJAUVDL39dg!xhys&Le7_y%DPx3vtPT z=w~sZjCc4mbZI%d6}rkbNurn8N6!co$t?A|lisx$?+B2u-G>@jTGGzevaNZI2gEYG zrHLJT)0b|}ojdw_5f~fd#yP4SwO`Z7t}ITJh|8Yc{UR`w=KETc=w#_fl#XPCwSi^X zQHCa^uu)KwwJL)liNg?gMps3D&Fkl&a;PEPXA^K2(B|1}_(tF3 zafR6`aMyoABjJf}>J6f)=2dT_#(llFgva+o*7QxAW6#FAe`%0I2`aOmMH`foK*rLD zD85oB4B{g>KyMZunm$)1*RItO5oFhU)2E$bFQheK?MJs3Aha6JKHg&!%0@DX3JKq! z*xwXS<%$E+|6CodWqs~BYk!DAt4>F@HCE*s*D1m| zbu}(~)g*U4$ik|{BI>FfLxvOeX7xO(6WffkE3A7|Y3(V)8Vj!N?|XfsJmV9t9X)0p z6VkXwwO{)%qF-!s^_AYP;YNFtRXtS;(kzykrUW)MjBmIF7Jt)G7}?{6g_Agf3|nYX zKZh`;^(&O_M{*8jl|9D*7wxcHw5Xkgn$TiKR4WN=hU&r_w>+#RGiJxL-zzc9Fgwp0 zGrVs}$T~YpI>xIC)fXTX}0V zu=C7ce`RhfTDNwX__x(6>N1>I*-G^f+){HA{rcF7=h8X*Pd~uTpUUO zf6lA&l9+eG3lnz|dzIfCI?9L3^TU6ChI-$*FUPKh0f)z>Wr1qYOI|?TzGVS-4(XLA znQ=n7f2S8oV03EM!RH&X8fejkH;@4g9@3IJOsk4oH#+y639aVx#eF}tBic9X4KwnM zZj>EA#*Jgf$ zB9X|PmUsiK=6$8go*2MI$Ix$b^CKLpHchJ#s{3nV`@fj>H!b$p??eZP^-2q?1;%j% zWvt$09G1_hEU%J9i(B+waW_*gQ?`y>;dWEVCisxYYr#IUvAyTl>Z*QUUtl_A;QPKx ztBWkDzDs)wZ)C;7N9-oW&sBA5Z*1MrPbE?9TGPx$SEgr&sj#lTAmH2D^_MlC{JL0}^}?`w1w6 z0;R0fz4SpZmU_ z@AZ4}oELL(X1Gp%_TFo+wRZgaAbNV@5c*fh0|+|gA>^JWyoC@$NFbyTG6*?@0zwIS2cd$zhfqUkAhZxV z2t9-W!U$o4Fhf`%tdI{7HV8X}1HuX6f^b84AiNMh2tPysA_x(x6OOQaz}XmSDg5Xa zz(};}^*~XS+*uL&mFd{1T*)dXuzF8tGWi+*`*(yIYI&Sbp)Eb87ZzzQ@7SkhdcV5J z?}f2J;eno#p(h19=q>^1VVT94T{6k?jH&F>yeg9SCN>xSH+n=0QquE7oxBLgy;9yw zd{q^T`TeJ-vw03bb3ZET)hd5lj{6*d1U&vu`6)bUGibjeX-cW6)00lu_LZ$=XtHc( z=C?8l%5Bk`bvF-C^@zoD;Z#(4Zov9Q<5$06rz3lLSqEp34YTJa-zv=s2R2=#0c#iX z84(>R@yHQ5n|sF>vvE_3V2VqDKK%VIw+G6`5e^Rm*YQ518?&Ar9FAUy$cbXJivnpI zfWkyff!mF&aMlvG`_IJlZ^?-XU+l5o;ab&=t*Ocen9i=@%SEdWbG|ysPquA;Hu;`t zH8_wpZ=Mg=oZE@JLs?#T}GKq5Vd9-WfuBRj3qsZi+X zw^nB9CRq~ZZogk)Z7PYdK$^42$bJA*XEkA&dy0!Rn)IYj+A`pJsf}9FiH2hO8Ex)V zoUW!uidF+z?1#ZcLd8T4TDY(9!_?BV~9{G=Wp-`>d=^> zHx=1$^Px38Z)s7~Go#Voy1w$TC(_mP$olUEd`fsBCvJ#JHbp z;bx|IPC(q#1rg9R3CKcl8kcy9>vCSu&e6sdM}A9{`ph(Vtc}iR+ertEJQ0T*;9XWlsr|zm%K!}n?KI{;?_VLb^_%1(2eVE`0m+z zX<;P)vV9(CSQm8HV@<_sw!K4WG#lirk?dQYOjJ5};+E1~89d+p1Kr#U(?4n4-K4Uz zp}tYNioe#EKW#977>g)*%uxN}cNk_+(o%}*`_C+*1k~fB%gUPYtG}!!EjLF;RbwpR*(~b&= zxHbVboj|bV&7fE$@5Q88q>b*zcW;r`v}yR`89ltuWvViy-Drj5y#pV_-JytQ1(i}b zgePlXOi$(NVC8@CG;A{;4}dz6u=pS8q7z!WuB~oW-Q>ac2#WxpgNOl`!cx$!W7)aF zupq{D<(!;E^}dutifhdEi2jAvGUwWAMyxmsi&prDZg^OaG`y;Y#6c5EH+7wu38>6D zaBS}4{9rtwiYB)MzNh;8TuZKA2q@i{%Y;hyoo*haW~Tyd8u6Fkbl(Tb2$`S zKWt)UE!2C4q?GYnSSeOo#^A!qKcj#>oYXNvll^Q@V1d3N(WURgp~MoA*;br%_{|v} zM#vr3#1@=3ak#TU+!ml5XYKDqi*b543S0vpT!=iAR4j#GH(CM0TQ~sQ2Tootc41*G z7guL>L*`Q0T9_16Y(FhBF>3v{D2nRZd4uT_3Rp%0a%( z|GSufzoh8H@BVE)-)J373SLr;{_xXB7}jw(C$3oYhTzfSz}rxuQfKUcY=(aH;D}29 zpa1&b_W!$Z*?x4zj~_p4d-7gu$*HL+DQPj-oShv;L_}TIt!N&T__Rh}AY_w2jX2&Dv0S_C3YC0xX)~G#kf$r3~s1 zO3KPMsvHKGMF4N=Coh|QFHO(m@ z1`tYM~n|(b!gNAp1Km720uySFR0X_G*Mp0eQ)OaV9RrVFsSU{J% zmYX`11dI#~+k*qVYH4sJambZh)I$Vgaf6;i|D;0W(@=JBzgr|Afw3o@{oKw|{0Nt0 zl48g|O$N#MM|jA)Vr^}$*{Z6F^gzxQfqkBz;O5hQ?`CxMHL%UZo` zvBB-wP^jF^t(G_UWFle51LY)p&MfxOudIdZF|Om6#W9Ac^&i&@xXkl&&MYV)MW?5v;btNx9xUR2X9(>Q+m^lXyReJsJ)_+kdWZ!U}vyCI`NvtMnj8s;pE~1iqUfkr^IhAv6ISgZ~K&5;3gK< z<@xbPK7hRZZnxF(BlJ)He|W+=E%e*RwWE&yOopF6{q7(8C?%CPEF&!)J?K$VTB=;FKf5TueWq{i=XHHyV)PNk&dqIB zp$49x%S>9ccKpfQ6P>SRYa6R;Yf9-Y(e8I_NgNzS5pk#wFfmNtKdHi*G@@Odh4Mxp zjeZZ`?;h?YB`F!Zz+6S}2?+_`yq16E-NJu^Ce;F7?Z=PG>aKN#l9c{&@2JMd*}i@H zgn8vXI5_x@;CkgUPN8o6|bv4zZ;9*|vWht>NuRHoC{N~M@c-5#- zKxmAZYfMa3)YQbZWQFC5;?K$YaOZ>QaTz>3JQ--zy;@d2A8yTXj%8y3wxu&nDl+3O zpFO|8lPn=%s=>_#;VgsWF$eP{(_Wi3$6rkf0*D=64+B z%gw|~x?XC{AS&^S$4Ev-#r1)x1Wp1P5TlavnoH`;!MrQxc&FYqh`y-eISx)92ix8y zOU3C2Ha4~g-+oI5dNVLEh+;TXo{9zrUcKX9Jl?8W*Zhn;l+-bCb#tq)kCjuG4etQH z20Q;?-2PIpK4w0dDDA;f@)ry1*)(iM0s5HeuCcEVyLfVHGM;H#p05i(>P9R~1o+7c znDW)NW_5J%rqIW$m%ioXd6E(FcY#$3zviuRQGr0u$WZF9xV@vKfO1kU7mHDtg=O!_ zb%;=2vNrNrXeE2MjAcugqZ%t09j~@4RNA+W`~EA1*h3)+3H))PcKP#8fR#1t;)2(H z{?8<&{`)qg|NUi`LRo2P>Ko#a*_Q1u*5eq+9RwIc!HYHzQczqn%&9yXBJQoqucQqO zGEyH9aap`x$7f>t18X3sgB4R{S@|O1B+j|8F0RbW2M^|br*#T38%jc)Bz>6K*#|`^ zl=Gtb3+x8I6Ww5lKO!#%6d0~PY(96xH#ZmTuBQ0NYlylU_Hq601e+>V&qodRDUpQ; z-Deq|lQ(e{8`09X$IDKPQ(@LP`c8xpCY-4Ydd^<{yr>xn(!?YA#3Jv7uub*8FG*nfI}C268?We*DAMq+C&w;X@P^3ZiT zF2va!&Wu%&uJ+DIq|dO9+7dYDw_9qT>EmI8GnHwjh*)Bozdn67Qx0#iiHb8)9c}h) zW3>ZRRxUdEH7Gh;6~y_3s5 zv4(?0(reA~;_Blv01}@W?e(uAh?)(PyqjLV|@3t9()=V&=O2crf}t@dlr?VFefA3#wC+rI1WS;>637|Js& zdr}cuzprxn4t-4{L<4#w$T~X@5t=Tp?j&x)e%l8jmaK*z6GF)&;F$$)<+tHYPW|A?dk@USz88?Ve+ zT53>ZWeF>OaeekXf-$~WOtnH08&#ZrlrHOLe`zOh#;V)EVyB{HKvR@Qi%TwWxmqA_ z-tUbxP_NL>kvbpoBlUjbrhl0RW>Bok;e~VO(~%Yp>}&qn!ZdYR)gpU8c-wvVt(=b~ zaNgjyN)7?js&0qZ%r$NZ`ONq@YASChyu`HWPpjMc%CDS(fKs_vdfB5<-u*Jo`jbJ zXp7Y_ulE;htLpoYxT?Q7xyjxnZM!XJs3~T<)`d~l&&#g=ju@8%~vZ0e3`MM$2bn}4G-aU;z1?ghdoI>wr3)krcP{4|UXBPU z;*tri%;UQLnyD%0wv{&8QCt=NBdHsdeu2oT6tWR4vQwCgNJNnW`*&*@J{jArgw%Nix&tqJyj^$1E6m-0OKE#v^fE)0GF$#0o(d#g zPNL;e<=B%BHB_Y5+W||@v;-+gSkof;iy&n!uGYjv`1Q`io9mJ3#|RRvzPPcm_^pbv zVmd;ZTi7aCEDZgEu!SBbwm{WI$9Mi36Bga-g-1M_)1&Ge{>z{43xY2~R(SSIIK63+ zJjnpCv$AH{(1uxy6*%pS7#xP4lQn)>rh~;trSR^oBjT@tr1zhfeBuz^_}Uzy+R48o4iad$Z!Ik_q>IP!9H$(x1p zv=oL)@rROz=LZ0Z93N;+P9O(c(8YUfdv~q%@xF`&{j* z5FzSjJ=Kq}jvT34?FqxmP-}Y}{?Z$hN#l1?)p85qymyvP^r|1n8buNy2 zX(B*w=3rF3R1cS*54b*Ct2-7B8L@Gi79(D0_7R!RiHM5o@VXX0#aw8#qrW3mS?JwZ z_o@V5_#6&q2%r@k|BA#yw*3s)7m-h~z`X;p_-On76Kk{stu)V~p0Jsd-0!G^y9%Py zL`!}aCt?XTA5+&*A#f(-?&Ch_RaF=$+2_t& z5;t(0y5o(|{rvgGePtEvuzaJ|1x>iq> zelLJsL6PM+C+oe?B%;FqN8b^#v$IUnu2X_mf$YZI{^A^;*8?;MCrD z^(40)?Vg`3ni}JlCoy{L!%W!BT!#UF{a)vFb2!#&e?MJ`zri-{XcmYaR6NYD(&Mos zuFnwz9iI~wrLX;|E>>Dak;s$Pq)xxb5i*UBI*B4Tgz<)Z@Ak}I*H1k;~O)BSA(;-|{5lZnO5ECh{oi>TApmWh4Xh4@x! zYHF!il|57YuMUEJ#IaUKYil#&%uX-3>D~ZIls4@9T%HP2Tl@HBAtyl$?>xOrP! zP@~}NR_aO5suknJ-Sha2I3DZ=Z&{qBn7QvPS&iCG4-RXY)&={yt%WtKV6lb~db;~_ z03=O@c0Lh@@uWs8{J`lGP`X(OZxyh*^5yQZ#I?D5LgtD!7V-;stE8;t5ZKfEX*N8- zQJGst$)9;iYt``yFm2Kn#Q(Azo~g8Nv;4TzWHj-??caV;=7fv<%nUw_&yLhT2@4a&Xq>XiuRGEWhks? zh}s|~4hDY%;FF5&?AW6sgDg7siX(hJ>GW!noAz9~38D4Ta9 zZm89R)GgQWbsq7_Wm{UycuoDN`Z zF{wDQMb$HzE&-heZ85$dq$ILBJR{NyZ4qOY_&h>3C{=%fQ1%G;dbX`mi%DC)RsRSj zt-p6u!z+>Sk3o-Z6jBOj(AVcp>#Bz3BVOs5yz4`A)&D6Qn0QKaNajCZ1+OO>;Tv^MOc|R^DkHW$3kgc5~7!w`)jTIueA7SSXb+ zbw~Q&o|&Fpr0DObGLSLwGH5XMG5aoVFag;46IVPHdM2hz-xz*?HMaU2lX!Mz_E3z0 z`B1#zrf=Ys;p9%C{X)z`y`9NYr?u%ed!fhGm8@uv1|BWSl?Z7pWDe!ibSuS*f~Yb^<3GsmVrIv9_$E-lp&_M+xT zgN$wNFC7}K1!{@_ULL6Fr1`5w;3lncAM6R^Po6-~i@jihT%}qut5_M4>DftR^8DDE zEKcjAIp`4b8-o_d&A$(xb5{}m)v;Gl=)#s zT1>sDFU>f!v)ArGQ8ilb@h{vnvk11f`prE-aM`FMrn*`b$}D!Pk&{Y5;Y1YlIUT}R zgb}fae$obQsI;1EJ=_=Z&w!qq{?^yE46}&HZ^{HqCUA5yIFHMi#DBK4^dnS?p@8FR ziqatB?Lp^ZRF?CPYDbZ^>THG$%8d!)^YdMnpg^41eZu^g;rJo(L1js!F56aO2tFs| zP4OW|v-ZOT$mEJlqdSmozqNYF;eC$G9Dh>+az0{@k8uT*(gf%G$G+CspSD!!TZ2`Rh`%^0zn-{*Jr{LyD#x;D zfCT9;GNTTIHvg@jAs(GdRL?*i^TN4GR@}x!=%i(eco093A$y}Om{IFusn9@YD=F1^ zPA+1lRb!-3<8_3Nw*Ky{o2D`+bDYX9e;ZMobo0rG&&_&fw$VkXq@fZ`8`swa+L#095<|Mn-;zTO3VD!8} z-P8LDLo!+$+ttZ+0`=<#<_0@mdVagSK-^@0EA>L9 z3gG(a^tvJAv1j%3A&sF;&2r^L?r-y#@no-FQB7;eAY}Fy{s@hQaVPR$@Hi|%JvRb# z--#aAlw|ME)rP1Rirj3+k!LXaq+W&GLEVtx7cNhTbB*UDxo&&RUoz27eM7}jfK zyElm8M`}APudv;EySBBDxC~R&U9hJYxd4t?XWHCvpQoYhVbxFt+ET&d=nPK(CUeYl z!$TEBSG@rp+l@u{&)Tab(OXX1*PR=oTv_JX~L@fKuKst)w zij9BH>U%2UhU2t#ZUn?8V}@H=oBmK zy(Z3X4eyn7nx+Cr2@rqp-n+WGwg~sjzQdmhD%Wn`osIbc3Cvfr@|823w@kINl5s6Q zE^AIroNBPs9}v<_kL+FT36)RdY%vLNR{ggHQ1SE}Tw1x=GQN>hAf6F%cZ7~rXl)r( zIcn|oQ{$+$JEAx%y~OZ|fY*^Om0GhW$#_3#B8d%K)Sm*@m>1WNRn;h&1k=4qY#*FA z(i;?&@u<$@PI{&A_^@lxg^b4LG>B)Br4-KfwmjrPk#kjo)iY~?*;+k{_$CAS_U=c! zAmDAzoFQ85(IvsKU0X-yueN2a`f+W{m=HCsbZ29TNl14v!)<->u`=eDksiUBD6%r$ z%nskW9|s_sO?xR8v>Kdq>scXORp(RTN*{8rn@Am@opuww^6S@MPTN zrF*muvRCNOO$-r^ZjO{k?&($IxtW-nUS{D-CE_-d$l{9anel&MkLaN#qb02T850aP z=eYZD)e;jvmZOjNs@N5N3j_caF}=<8!T682w!kEN(Ze5aKi^)T2|9mHW3jhRQo*=F zdV-Y9jXw=`qK=1kt+n0J!J+z0+byyU9grv+eiNmGO6=3b%K)hcMiX!%(N+Mj(~ji16`;hv({N z28I^4{$?uN)W!meQ#IwiIM(-#_iC{b-;Yk_%3R4MMoK9ee7ARIzfeX1s#;=Msw2T) zxZ!>0JosGdg?^V;8TGNXI)RNVLin(LiXBUn!krZ@IF6TA6YZ%|`Na&AWhhoLl?)+? zveXXPrUQ2uE?Upr_UC_htr5FnG}j-_3{;kE5xU;UzAwox^BEhsboi8X7a^M=(14f0 zt*Hz~t3jjnr)e^-#c1GsmmH}*T~|Js$@mNP;msoaTZFv#cCcK1X@X$8?^sN2l^!L?CJ~WZQN{RT4rjxmMa;#{x(}B#ck;|+RCQ=sf1T4w?Bh6C&C$u^`Mx< zKfcBP72;yp>#}v;jvmT6$?tw9P&XoqmLqX2`A!sBp~syfDk)lh+zfk!rx zPP0HYU{itE9XZC|=q~_C*UauZ1j&IKymYw)fT=J)zu%3an9*EiHvL*B7KsT0*$WXZ zi@xXw&ua^#ioCzfmEzpnE&j&h<9hydVV2n-C&zekZL>JlYNF{b=2ow#b(JX%B-SY0p7Aej z5lB1fb&pOF48NUt+9`B#(qFAsTNq(Vh^xxcvphm`T@;v4ly(b7)?fzrPi1dYf});| z_})qJN#4%b%+^>kG|@kQ!qsG5mIna37yS={mFs1Ds;M3;WM$kui94>;D!`B2`hGXr zVE{G0u1O$nN~uh$goO-5e2F=ePvP>QCw3GI&guXPr(WFm^9+~2y*gW}Rp`qEZf@o* zBZ`YvRC5VFKqV-!&70=nF^;jJtOc$xiiA_~1G1}k>fUl!BzQ<|vP++Pcpf074d}RnqMmeh6<8VCNKf?-hb{}t+XiwU1oLQH?M?MjP8bWwb`<_dPU5(bRj*li*MaMk-tB6F zK-<}_h>1qg=3$q>MIO@SCGehguV=KboD_LuJtDqq&4T@Ji_Dp>_iow3k_$M znCtBAhIBt*)K|c z$yV|sw`TomQx+#mbg0h}Vl@6F!kt<396t8JW(kmi(J!h@4SfJq@iHeLPfkq8>xo#L zydcb}KLpZNrfCp{(rV4v?|1rAQq9JqUgaCe5S;0{PZk>%m;}@;Vqk$UL`8PQsA;g$ zeg$`Ug&-CRqfB9(*# zYc{*QvY#fMjpkBst?5(85@P5MI`96V$;hO0M;>PSoOsb=B>bsx3 zuGdf#1}G~H2U2{Q%(k!2RV^m2iLTZe3If53x(EK0r7H%Cv`B#yv%ql`JY;`N=w5b-q)Q1lhYgpP^swTood^W2)?z+Z{}K zKI|+E3B1~#ipVCc*lRn6kWZD$3AlVhUC@vA2k|KMsUjyqSdicgA-= z5c~1{`zX}VxIWE1q7=m5cw1qT&v)RBje8l;D7t!3hrKx3>)YYkZ~BU@!3hPIbk(k#mDD}nLnk?yXmcHXwAXiEV8n)yE1FgGhXx8kxBMYX%<9po>Epv zxprH~^$cP=M5XKT&)++jImUQG!TF-R)%Hu!d%7?uo^m70t!XaxAF#R(;B(60>BP>R zjrpYi0iJ1hXL!qH%CtgkAqQY5m2mGcA8EaBxa1vnxZu@dbDdFWOo1?bfdmfsd)DFn ziB2{;GOi*lQ-;S0o>k^?Z5cQ-x476*O`aQ=#7hBhFNInPp2ZA~ataBNR#sDb>vhL$ zh_M7sQ2F;WF2yUaDW7!i&2u`;wT}OgQpytaTRGey=BhJn^B*!mega@&Nw-Oe;In9Q z{6)u`I7`;KrNP`b-`lgLASIKt!EPRxd0v9Z`;83+3=-(acJKN@COcXs=n>#A=(hPT zt+1ylq-azPRgW-*M;ILJcwc(A7#TF1@SW8Sa1gT87E^O6u-y1*yDLU>W{8fYL%SvI zt}kIbbH_8S%nYonzz@)20eUOubyGb2JUEr4RZn| z2PIs7|EAonZhY|In^xuW;AVp~n-;0}{6C@RL4Va*U^90}e^+ogF+9t()fMw_q1^7; z$teYV(a(wayv`fyQzB2aU^NU2>wD1WXKve+uZcsWqP76dcZdN-#Di#zTo*DdCT>-3 z6%Cab`>bOJE}q1Iwfl=pa$)aS0}7)&Z~SF7k}I%1;>?2yZk1AAXL6TE_^Y&9)6Hm9 zsz)vr`YoR#i&q}%;JK!H=rr<%zNXhNFUBC$YCHeM%*%WDc)Z1pX(QG?cr#I_!%Iu{ zNuIHR|5KpXogCkFm$Fy}!(i~mI@cY6Yb<}0Wu9M0jDGahT^H>wW|UB!M_m5eQ{}zh zK=z`pC=ulI^S=)wLBkW+%45*tEYw)sa=Tkr|L_ET2JsW$#Bb^Bqt%Ws296stVzXrX zGQG-v@4V>k?XBs<4hjqyulzGkki;A0tuSP5n+2ej0keZ&%k#P=mDi?JPuHghu+gIZ z=I3=AmiSeEK(5a(LTYh{CpV%@N1E?{?IjPTxRnveo-as6y=k)BdQjP$6Y2w-7TQI> ze5=7&o3qSf(Eht4;Cyl}o=iO8$h>L%Pu7uSQax%hGR!G#oBvP>UP$%=xGnP98T~dW zT?4olRt#RBDw4eyqNH3|3}7^IzwgmaD?6Bfzl0rZ?O`3}! zHeyAjgPF$4ry$oNqLhhx^6KhfJWqd46+kqUH)+1&awX=eTP4G$O>#|<4nRgUVe@Kl z{s^aUTJZ$e)S`ByDE1>v%xb-;D88%|0h?IYSISU})A{<+TL1Y?J0Bh1c93khCi!{g z-1%B6$Z+Wq`tRjisoAdKmtp-KxbkYs(oY+ib=v)pAb52LDqAo`DaHKjDR9LD!oEzG zL=BDh@GMR{vI-Fzev75vy&Gt6zT$){mQJmInKxy&kDFy4KbSJ|ycSS#eYi!1+MeQ| zyD-aR-)+9mU#I8Dyw>qZ)*H?2+jhDXv$i8h)jB01A%TJLcHlZ%WYz4{I>C9Bm{F5% zrq1aWBt%{A?U$;7&mD&PiycV@GoWnmcudH8AEo%sy&{d6hzJr8aD8`oys?9O_1Z14 z!{0L!TQLuwUX=wpTBfG3e*e1Y_O~I`aHu&r~Ji> zYpjX&!<=l~h2L8YEPd|3er)c3T)srcZ!OzG%79m_12(2}E#@#da6`bN(^aVhgTYor zOuu?JoVgn`!@f<~Bdx8MTHIf3=}p_uDxN+eZ395R&^kjNDC3C*DIHvS4g{r_mSPb< ziBX>7Fhy$f|0QHKB23?q8j;>t_Sm+n^?@6s3>`oQiog!4kn*8awZ~aye4Jd^f4TO4 zI~IeZcqKIqL~vi#|D^MyK*nYFF2^a(!U*bARAiSFOC8$R+uQDQKYqF13<~&M4)cIA z(x{>&XME8BNe#{Z6%EzLOm&y4)BMbP58$_u_6Pe4uaWs_%nu}!g zs*fH9tN%=*H|YHJ@qV-7T28B6>AY;A!CrETlTf$78ETnjtVSC#F)``N=O!%m5E@&k z)*qpFuATq==o=7KbCEITI@91ZWDodvh3KW_`+NeuQ)OAgv?BC6+%Oj~6qXyeG15ZK6fb?MMhK*L!INy{^!?xaol6->+W1601N5 znF9}$NW0th9VqGBHeHe=1e_cQ41qFLo5y{1KtOY=@9NJW(Igt_e3zhUzy&;r$#ak> zCM_o|lqb=FYT}qvQ%6<7qoCEEMLRi;SEkl#<40PROC~&O zN264rVp*3XA22@vqbGD^*7?A(XP_wHPM*$fGY%g-+6Be+eE-A0AiuJd8NOAx1ozK= zE)VNHwQA9c!{nX@_U6`l&PW9%S#w5J4MtEEHWN)`VY-G*t`PaX=)Dqbm>>EHd_0(H5^- z&GR?6H*b2UFAsM96aV@XrtNdt7B_AahlyvFW+4p{0LkhWGWuIv0H#qbV#4s|!K3`N z>bZfSQ2ar@y`{w^{||&2e7?K!iHVyv615WV+x^?-)pcqPV4xW#C_FtK75iU9r4x<7 zdxG&O!-VLzT_Qo{xgUK+O2Asls||02caVF?S|`Xua2Iz7oktg6TzDd1kMEG1iGq&- znEer71#cZP>AN&9r)3B$-Wwq+aJ$%+aSt&fD0R3q)#B5X(dvVhBB!hKGkP`&8Ld_F zr3bZB)W4rL?2ichOkSR-Sgw@)Dt^D7%BiCUdc3I~H|luokDi8>ER>X#RJ#W}igGJh zvmUQ;J9(KZ;&mgPlWKE-_87@`>;n2v&F?U5d4;G`-e;-Z3ryJ>0lHU{iw`us4DhB(xGbTYGQEeHzeg{{4Gw-+-VCYJ%wL z{)ZuUq>_rnFN1q=Av^}TA|*007}yj?*3#7c?Ya$8YB_sDaWYq$!1x@sf6xdxm+U(q z3GO7jB9@I@DIILhcwLnpH2Xn+RSxDZ;sreAJx(38~9z)177dIlB?$9qkuIn!9Na)AO`gGXLsG zE1i}m==fjV-N8!qjO_1LflH1W4vYHCOw^G?tyFkM?=A6*H=L#;FYnNq)3_{PB7y*D zfQvRH%8;UKF;{(Xo3b$e^d_Fl2y^;p+tHv zi}mdEA$M?Xn+=CvT6Lj7ju)}>neplesG+@E-m_XL(3^?eg`e5!Dho?Hy%MidX0e)m z7uMMsm0hB@5SwZ%w%q zxVN@`U8>rQ&SR(1Xn~~O!SZuIt!Wm++v;#EUaNnt(&gvLtIFgs6$P2iiD~Sxtt*T)rK3<8Fhk$Z78W(&x+^jJT)E3-$kASHg)D+D%(=@f zC~nU{`#>8aOt>kN-0v#s+)OSNAaik1(@Rt2i+MR%3BDl<*MOiSc+r0F2#Gb_NGkX9 zjRtv!Y^^>x^gg-gcN&*fG)jRE3@4B)^0~O6Ki{udI8T%ITw?2+>`q<^EYC|rhUA5Y zhMG#@7NpK&%}LuyMRPullqhyV2rm zs|PM(bpso&DeR)>p!@86sC3dA2BB5d8#xvW0IGo>Q4{BPEqlNy!u^#!>~Y}CtaLB}(> zam@C5!UBNU@#@OEH!!e3F6keJ z1BAC1B45X~@_j}Nn^UZ0Uqqj;y%W)}wjLm*efY4Av9hlbHJfx%)(T`-Rz_-&4&jMi zUtd4|88GZQdcW>Hk^`iT-|Pd!U)m)p-5MD3I!yCrorK~gl14~D3LTNYgU?_X{tQQ# z+kw;7F{}RG&U9$|i%nSVTJhfk!{V*c^%b>Zmd`}X#r_6&F=9rPivh}-hV7msU)$Rv z`Lg_vb_@0$MEJga#WEgxs{wi;|8!oZ*cWbNZ2Rz(>$G}$9pR%0Sqy=K(jp@VJv`&* z4KEp_iMflRZhd#4N(TzviA*|wax8hy%JC?{WHCnhTLDHiPB%P?OGG4M@g)d!wXu62 zolui@b=+TIKYEIJb(O%7xHFkkuMCYTZD8DrvR|x8vu+vt`LlB~%dM z*Yv6vvpIVXaKIHbOzt6E$g@P$naaw_s2ldmv{nKshK>eHK3Y&$R9DB0L8*dHfllJ5 zSIESSpW)epGee2ku5gt|E-2uvQEUl z4d5)*tl)M1yw6@BSGmw6v}6wr4IRvBO9Oq*8@@-2qgfqw^>s*{sa)@G-iC;2?806| zU0xH#)3j0G2?0=MwV|U(Ycw^8PqxV4VSc^5a88DafuRO1kSl6XcL)`Elgj46h@33_ zJx%Z?8EE1j+^e&B1+Y4PTk5)QL@l~(x9&`d`1@mGUJa*llL$BMFPvvh2w{eE^lO-# zQ**;1gb%jP=Ev8HK(M?+-)s1amiz*4V{4m?^3_~YliUBh5-&i{;cZW1ZVFltqxiG_ z>6`mTjaF9hTKMNBB{3%pa}|ZSp1H!F;bZ`ci}G)S^rt6|;Yf+e?{00b8E-|(p*g2K z(#a|)^VGI{s6x>+bJ_$7>assR;~?_x-8=d4h+;0e<7OXKnKFWir7U=E?>Y!9{JkzO zuI^+6k*;+mh^ZQhII#1{R-i7s>rr6wQZSoP8bex_&fa=u#Yhx@7|ectXGe|;*sa$; zJ%kaK_3kwzKx<6ITVXUW^k}IvzHc9Is`Jf0`ucv?|NTanjhj0W`Z>L001q!lL08Ae zSB6Z@ttXamx1@P?Xf=C8DIry%8G%simCxYCdhE_}wU^}iu&g?V49f-WBdj=EgE7PQ zYSHI!?>}Z~0fC3`9B&D59C8de863L&8pQPABMmqy*wy)pe7xZcLf1YrG_=OxoTZ-# z?JKZ*H4Hr_sA#B1Xg8~3fTLL8jN!hp?|h3Xd009R$X~y*!L=^p|6oYd z{_iFo1A~R@2I0Sxkzds*!12x$NBy5S2i0mfe4rHo2j?_~0}bsSSO$SCOgJY#McBN& z`AsA9L!B1qThI-C!za|@Hkfnx{lBqM-Y0zzznTrkv+(&`Jz9T=_Nyc5Y-J(5C3QYM zQ-7Mt;|M++6bW6!I^9@?I6~r3j@~iwu z;cNu;!lm-qsL=mLMbW?qi?-tV-~8);Pl+VRQSkqM{us#^ymEo9g_^*=24`i>>OoQg z;r?yj?EcqhZ{xV?fmt)~l&~jLSVs?1zWRO@e6H08PRIUlTj!L$8Z*NwJ`wu#syTgTp>D zDZO=id<|Cgek`=Fy9Z}s4H^EpmlXJYNn9Ukz{`5v369>P)zGjAfBK{tO+gEv=k(O2 zChz%ESrtLp&+Ng|wwouGqNwK{`=OBF0 zq$f#>rY8x11LL^VlZykNv0ZS*%EWh0U&untu6lKd$Cs0^olUAVKY^hD(oNSIoBNT- zNmnNjt_xl`c>iWNWE)uE`0x+iv`}Ir!>itt+(xs*4K{m;)_2rltfM}5WUSE?$O!e2 z6o&tvZokxj&+C&(B=mbRkI6qz0Nf5n|5@hXfnJ;@UaJ4MZa@A%*4{FzsxJB$J|-q8 z3P>v|9X@oI2uSClK}EV-a)XqBl*Ezl?v6t!NVjw=U5D;`*8zS0&%Gb+xZ@pne_@Qn z-fPb_*IYAxbI!GE+|i55@8+Ifb|)FU27PT3pU>qMu240k^H`B)GIC1v`uTsF?aoK0 z%J_n)*JnBZ9d>|q&W>xB6DJBoGl#>f%|$Py zV1^5985pdYSEyQtR#EbEy}82w`VV%Y>f3Li%!^?jV+o==viZZGEcq+{k;0k#7bDOL zRp0l!*jBDpSAKs0PYiqf&$NTmD4&3KPS|_tDt^-xuEdE?4HqxsC;5n;yXzg*Ra1vy z3|X@uQx{BmjP?Bunta9^1<;3of6Nm&r+}R110YQ7FI{Kb1K=V4Id-xojTds)E~q^a zH4Pv>``wGMLYP*aE!-y}%1lnye`lHO%(7cW9-QvV%gRz| z6e`c#(q$^;eA2i9x=-Qq4*+K%KE;b6gKB3~*-rs1gTx}jyK&W6N>qP{BG3$xJFdcL zSLj(m1^xc?sX5{{-1&5q=iqGpQMeoKy@Tkipb(`Zz@G|5EEQEso0|20;!7 zLQOx`=N+9SAtGW;arydYnYa$8vmSTgr^^ZGKTi?IO?rOq^7o@at^B8e5Ju1Rvi9Uw z1(xegz-WMrzb1t1_kLwbrXsBt`7>tYmZc7@TF1tQhF3yD6VChZw>n>LoANmlUh@p~ zI60nu33-yNip|Wf6JI4{9iv)ESsBN#y!yXND2-N$mzeFGlNbbrevhkoa;n;&K+<%HStjJ4LmZHIiuQJ-Q|OWBoWUiy6nVkBG8~rh>+@HB4_>QUS4J* z?G%zVPvtRl)9eZhhq^ZecJ8$+|J}2}TuMz`#C#9#JyNeRrfOQ$rct5J$n{&YN)R~B>@C@C>YRb@LwfDT=5{l4%!PoWzAF) zmwlpm*%z;b5?*Zn^K#r!dmZ$;%V=kFLlQ~Cur}#)JK4UK1`GhS?Y~Nh3OtV$_$gG} zA>N4drdVVP-8U+=%4a8_PNus1i&l;Ob#ITKLKTe9&n0F5KYMU963A{;6n^g{Mqu1+ zo{}8|qHA!CZ}@WbbvVo7+TdC&!z5HS&ET}T^88RORl!1D`{{}(czw-=^nZs52g9%# zDFmL@SVFb$#!JokT!5r`iU@IAJMx+>Ld+VX8r7ybOUA#yH25h$KS6HRqD;Fq8=9Ix zx(8*&X))dTGIBV}kz(Z)82rhE>?en(xPMNKfb9L`a-EEp=k9`;*v;3B#$ZSMKtP zuMVo@+=0miu4-&-Y_NCo=fNz6`1_EL&+Xk~hTAu9@;Lt91gSu9rUOklgOT>Qricia zVh9!QDFi^PuGrec-v{CcO7q>Z1of#Mm+Owdp^0}swuS>o|M4-c=Fwzi_T1`QEc z4#peW2Ad?l&XV41o@>`A-pp=0HB^LXEY2!xPr9oI-%L(I`*D%TH?vE#TPw!Mb-mM# zt+fHGZct5a+__|XDYgIUsX)kI7da8Qud3es@lM68CbfTUM`4@rh5 zHN{^}g9ITKDIaPa3LQQ1ON4F#f7JcLi%2z96?5b~yl`?ItzM=jj=uz^WeW_kh18$eD+@JCK!OfeOS+}Z| z6%R*i1BnUcV+S}uKWB1T?&Gpck0-zPqg}0Q*MJoa$qXm|uH{;x*rHjI7yS77vRXe}PR?O#&RJ0KGRZ$TUpGU-g*CXn(&zs4P>_ znV)pDPM7?$+hZ_b2QtuZjjv_lDL4p29B$^a`?FdiV`@wHb8pYUVbe`l?vKc;E)#WO zs;ZGpsh?R`S&!FyroSaj?~#;CM({$TsbR;Se#&3X#`DQ|pG-8(mo3jOtJ$p}%{P`j zd!^=d$;U4H`ub@1tc!$5n|eKmP88~^0lpG=;kvio+o9Lb0BSGj7O+R#cIJFs@(-R? zB*U_5Voj$HZ5F~-H=iWD-3mY5+t?Ze8TZa;ZaED}6e)!((_aUmxy9(6_o?_8!#{w= zvb>mg^-XqVi$kf}T_7o%t-h4rag-sfq?88Q#K*QFkAnKtF3n=+!S6U(Y`JPBK~Whg zDYK|8t{0Te-7;}!bl+bO{+ho39Tv3qOGyYwe}v~B*IPw9^x>h??5Sq7CvEK;Eg&y&kF&e-4I2mOqG`z!IMyW4vkPzWBC z5shr-m`89iMByUSf-UT(5hr6q9RjQ#|Y*t&P)gaBXxVRWfr|~6f+H#4zS}pN| z?5+SvWdcYnT>RETQN%#ZD4pi%BxouJl@H9O`md}2-b1TWqe*W8i^3;( z<#Rt?{q8z3er9p`ab8b0(5+r(cRmhp_F#L`fzwuA__TmA`%#7M>1?} zTCbHVI4pNt*WXUPs&lsem0R?iH)f+yG_2Xk3v=ng145*6gO`YDS2+hK=Ll0;z^s?o z{E+SP$}H-r@q`C2g92bjbWv=YmFB6hmIxmcL3&c3{ryJ^j((w#SQQ+_L*YAZ2KMh+ zV1L8c_<>a5+I&L0HZ%9!_#ZQGeg>|tXGP0VAjzqlfB}L*tF*6<;{g#~4v)p$muP;| zH>4Nz2xE~B)dt3lhTs7`*64LToFy|eE6t(NjBOcn8#o-j(`|ODkg&8;nq|YL*XzJ_ z96_WDIGE%?Bc}82v7&^8F{qyrLr}XBU3Yc5bsqNOGv{Cf9To?iaaSU2Qc}J*GsV|! z=(3b@^o`%g=cM2e;O=vMBe7sd-ydILSeX*4Whk?La^r_9sgMc^4eP$M2RHs=b#gJe} zpxke)UTZ5HN26BxBXRh_JgutL=9msx2XieXSO-8EHzt}`*UKQ{#uvo_9#W^Qg3J)o zX4bt4V7JzL2h(^UV9w2jSBd_H8t*PMr>lkdKS@eY2QzrhaH`i9N?oJhnfyG9O{;c? zUroNqTM|PxHe41C*KK>xDbfxC(Z9>{kQNww{!DlcOG7XKDmcWnq|07L>26~eNCemE zN+AbQVzesX7mP=R7eO{Q#h$ni^a9v#g6Es<*6)F6$Q8N_qN8Eaenl7rXE=g^)(cK^XlxUc@+CweQJDeUCrRB4u@ zRa9=5XKFAnA|y0m1pWNJm4=9j@Btm~7f;WwrXrmG8-CpLn6~nW%rsQ5GN0$>{CzRt>p+ZdwjI#HWIu@4uU zbb)FEkeTrwy8`e$c_;|*CjMVwO4^M0)W_XBX3-3knJ(6F+01DbnN4ae+R%L6v>NOt%A~UHs0n3vUGj9IbGHC8MYuyVt z!|T+lB6##@tiYKDqLY+NC=(Q{pKI5e$N&eDbr3#Dg28^p3H;AW>^GsG!6dd#sO51~ zPIER|^xb*;_P020yAb^);QU=cYu^8$$3GVar+}k_wxfX<49|`Z0=8mohh_hxqZmg_ z$q44m{0b^YK-TG90RB8y41r>CzoR7=Txc5;H;qZk6A={D-COGSQ8051_r}!_y>rceo>qz!3)9pqT^cWbrUs`s0Cd@9i?R zE&;)<>T2f>aRqq=h2yPWVtjyH{rs3-Mw?=&b648?DCJ*JkKsnsJnia}TySbiseJQKe#1D~q$)7ZUkjMcxeNGV*p~C#1-SO27!n_5dGrQlum64I)kQsx?@p9Xn z)%#W--Cv4@=4tEdPMxmcT7qCT`5M@#;))Lw|A3tb_ZkDb1n4QcM1)`27*C3+i9e_S zI=bTWf5D06kXu_D9~kp-^w0bs+y9DI4S^6lyd`S9Qrd{klma7NFwXgJ^`Qb z!GVFz)7@JK01`dF=kG0PV0~^L@PuqBeY|!B{x%Otq*>@skF@`8P#}|ltBm@`A!Pqc zoNa%j5q+d&kie&ge`9a^yLB7r{mf5f&dX^xY+mt%Z5*uG#>wpIsO^? z2Dmlo@2x0u_(T z&SoF@zlGg}j&)`f`|gfNh9jh_Vjr{8{`!F|H&mp7Bk-76FzFKTLwX{$Pn~Yo*zrrf z?^Ao-rCwy{s8d}xR-2GVFXU_y(w0@QuU5;lwR8W`1B$ZgR5x~U)~tK*0`uZe*REdX z*we}k*^n(GN+6KlH9!|slDil6STfBSWWG0fOlDt=cra5}ZD z9bYX~C6F~IQa3X8#a`C;Jy`B!L4G8 zeGR9V1a95h1Yx(FG&om`q3OBxrj?i7gc3U7@s|F6{mrhH))uin93{pOvv*9nJdVrXq`4Z7AF{BU{(1{`({QNULMr@YWm zQ)X*az--QuT7D7_^s>zq26g9y!#bRu$E~mQs+b>5Oy3>d4I~;vIxnYIe7&>D=eT!i zLZQFOqDn<_Zz2V9mlGBrb_}nU<$v|cB0iCfI`5a@wQGCdd=kkp)m?pC@8PHLQ^~M~ z%2F0vuT>XKj9z~!LGhGTvNn#QYiDi-GL!1O>whx43yv~w*^1TJT+haT(1bZy8qID4 zUzkWpqOjV!#Z2I@Kz8#s%6xD;>Lus(WwZUG#vHZMwZo|-Xwb}<%#8D(u<$Y6(Vo)M za+t7eaLtmN>pa=3cLhH>8;&M$s7NJoTC$x;)SqLEdq|4;-PYqRP*_kL+-F!36DvF$ z&YAv(w6Ul%J@enqTv$J>bLNAqb5dK~?rzwa_nS%O_B)~MszXHt(Cuker}Rw~$W~M$ z3#cql*fq}*ATJ&+sw_SJ3Z(l|nCLYcRg8*~bXMgC#Tck4y{7XU4XM?xupLhreCU0N ziQ80#dT%M-#NM<=O(i0NEmLWBK8c2&m_>gqdQQJpd&gS;Iv?j%ZDeRUSGp>yjg-&U z!v&Y&E5|fki`2|hF!)Pk<0w%1uonW+;CNbB*&ngIcE>XjDt42pP^I%=XH2D+&B}OD zR#p)dBKoYgz4c&W%gnQ|tL894x+_=Eue80T=k0#bCAFu!V-{^fY)XjbHHo&?DTXz3 zl$Gr`xvIm}o68eS2)`VLrT{M6)dTyEyvF6Q@R8ysv4knDM@zZ}R1aLflp`w`AEFs{ zb86)SjDPx;pD$C9WoYYCw@eCr9`B*y;9ba&ntG{vDp5G;{whU8v;pO#v=Ly$TpCW(;`KuKyHSzRdAZqtoSF14qR3fVirwI^}wX-E!!! z2AxdjeXM&|pv;|b)jQ&;FTm}jm}YSbNZEth+#v2Vd^kj)R@IU`=Z-v&7m`> zKCi>mICAe!mn>Ew0RqhHAyl(Se=KJDzf>%C$8{`+cJXT9#S zYIx1*gz?~0;idi>21xDT>_w;F@&pzOJrqKtBCLedfr-sV`b*W7u)#1}(SGLOwOD zn$j<2M{oLy;pvGObR^SP`XPi@0!|kZck6sh!p467lp9)a-hkxwx~dO)M*XU|$4gk| zrygjkZ))l&hP{&PjthyZT?xr^uw6Fp-V(O6i5X8v^$l-{r4(u(#S4qDP%!W@(KLTyw=+@U)cCY_LV#u1)1!2+k5IU#GifX)$6L=UOF1J6&~l( zj?D4*&u`JwIthx1v0Cn!PL(2AR<>&S-dFq?3gkOm-EU!%?ZK$W*gKjTSK?LO=ot8v zj);R8DZ91K7RS`1^*X9dhJLxwclJ|!#lt~lYt%08`H#9BOy^Y;F{Gu&K_~J$iv6xN zJS2|#sUb-(UmFa!nn4pJ_$L1{ZhQE{TYRF2-RnVy%#k=WkW7Yx-6U_yE|l~gbr}d` zG^b;Y8@o%Vyx$#F!Iy%?G~`hYw|_Df1)S}Vc;tT`yY4v4xk?rF>-3N=hGGm_oY_lB zdrB|?6$o11%_s{rlkYVx{O~ojt}an+)oFddr8mxL{V=p;`q4e0f4;r{$Br8opJ5uu zN(GdKPx)a{l3gJEIMNFaQ9T6aij<97oh&1EkgHh!NZ`bg>S62lXXD?d0&jax-J3ib zUr!5DKJ*0*Y1z^05h=!v+e^br?*gLr_ZPjeh+mj&D&y&;6>i>oJG!~)+cGT-_HPU~ zP8wZvd#eu@)HB{cyGWtS#ld-l&bE|Qn%hzfW$>P0ynn|!NopA4sK9kd8ycKnCUqHx z7KRpnyWdUyI%sKYnLeyI!9tYXEFs@MrFf>!{!KR3uvjTP7B!wm-;>)E%c-XMHA`bJ zwnXk(naQ>f5FyMxO2gKnFAep{GxK;76{wn|CMlQo)t_8x(mMQr4KVq}VYK`#;SRh$ z%k@U8Nos$BT$0j6Y_v69+&o`>^YnP;FKl(Va&ScAoVq(#hpVt{mC0Wah!ZE(SSx;5D)S|8%x+ottnL(#K z@n(S)MS24&Z_Jd4IoX_iMx+sR|L7ms)CYj2dQRaC_2H(5HLSJg-0EmeDG&=&ipnT6P!=e@o7q9U7_x)4=sa14qc1N zJ^DN<+!A88QmcK7dvP3&F2kdF?a>xASI`FgEBh2MKk~yv%FB~sqCad{;Bn-6av7dF z1}cz>jax}cj0SN=v3s>fz4YE{6V2W#A8mH?dpSogHB{)Tz5TUT$kH4Yx)Ffsb<`6} z1t}scV@>-O<1A$QmGTegcZ^&{`_l9V_2zi8>2f7iitL>OS=+A`?IWZQ$Bw$jg{O7W zdZdSJA7yLs?QU^twC&=dCbjzJS-mHEJs{lq`&Pd12ErOptP5cpo(f#OLDdpAk{Ncc zi=S~+xCz2ygE?kPkER)(Rqf49Ij+Tax}ujdH4xzHZ)h=X`vvUF=wmdM*N6H|qawnJ z?9SfWDItJ0Tpi*SrLEycj__bHP|{aKyBSK8bhUS~46#m6rWLU1C=K@da~ePttqk;3 zIa@t>HEKwg=>bruK(GEnggj@js61KMibCYb=+W^%n&gOvEM~GrzLQ|igK}vVV;Q)U zvG%KXJ8Y!bdyf$Yc}9d*^s&W$PAbUQQ&bXbOkIj?E?L3Qyjvc9)sUjRV(w%X9}6Nb zo@(;Vg2T@K(5{u~>T@F10)HrERTi21AlbGbT9OS5QGT?Ob4{h{wJPECR}F;|h6T~7 zu(fNoBdB+DS*OA;s^niQM~N}H9LSp@&zz1Mf}~JPg@uz&d)g=7sgsuSGNZ7GCxAPy z8Bai5_DI}&SAa2MyLm(YCgRKJ^?iKqOa={!cAWd*`qw$bB-Mxeo$3NJMhxNQP4u}WRbF& zj8m;x{XP~UY@WKN!ZOLc@8GCe|BG*Qc&2dQV#1SgW42`)*w9sZEmkM-hq?;SNN^pV zS8h4dEHBlj%Rp2(uO`Q6j^{1C{R$atpu~=+e98)4w4^H*ZCMd~9fNCO3dsgj`wM$Z z!3D%TMmCK#qgQ{EXNSi6(0)D)r=4Iv!Jf9J?r)9XSRSVz!C`Yuv2WUV-O~TSHipAd zvmBqh52}xuJ2yaC1Mn_T*siw)g~uw4X}P?+ysS+^w)bc->-fAGtTq z{I+lX$%!YU5@JdI%h^b2o9dhA2+<#=x%9j&_}jp{;ksIO7aWCFmAEOiGHe4KKgZU1 zO-a0hy1uhkLUH)3$ZxV{zB9w6W&%HA7{0+NXB)%VZfk z&JD9Sy*)H@+Lxz`w${CxU)FQ`I{OAFsdHv($rrCRS*zh?#AM~?%4P{MW_^B_lB$7h;tvtMJgA~(Dl=NB@B^g1WiHc+Z+{S|T6q}8=3K9K1U@0ue*_i9 zIMcK{j?M8Mq{}JpsnMk8@WlzBMe8ORTC{q2>?Ar`Tm45M{hh8_LYK=YOwS!^CwPF~ z)BCaNT-TAviEDRi+rT9ifT=BldE(f#8VGmF8hsC%>ZlAB7x7gp!-x0AEud4ucd8C^ zQEhO2wd|QqA8l{5G3 z)+ndhcNMohr_`MUk)@|2_GYHpbFBy~)|ws_#SEFs3Qmmu15DBp zK6XkOwFe7p$o&cmR#P9*uI5ZRO6%UvcKe0Z(n}_ZSVS#t3?ScZOEgxyx8L*IZ*pHI zrOMJ-A>qUDGDEgn?ZHqB6LlChW%pWX#sd!uUk<0Sk;B%9xRS$Y)5#j4z)9xxi!`dV zK}({qTF&hS`;N)G7hfz?ng4TX0mwujET}Jzs&Jh z?BXHaC6}6)7`xS88XD4UhEXqchrDWSSB6TY`2w^`j&sNZ?0@7sm$DB0c0U>)dii?E zgrs^TcCoqZIbbVH-y)JCATJxRg5dC4f8z)wMvFUm5G0roB>HTJ2b5U)tncg+@i?sy zeB!YiV_Y<(h<+_0#_?h1M>SLGu(I0tH~L_QD9Jn_Tmt`VN)8}B0CUAnwV&GWQaUMT zAi6ju=B;vr<8;Cyd^t0#rT)53195HV1B&yiL^|=U9~+k9;%bD;Bd9RNTPbpj zZ_MVWQ5ln(GRoWAerCyfaIIKL>aa|O%q=|Z$)gWC$DTjDSCQ^XH^A#Ur2^uh{Sg-9 z5A}TGqSYPQN>|3j!Z)$bQCCPca?WR(G?r!hx2n~>+jF(+4r2#?fORd&XG=rgan2#J z5vZw^?Y=RLN>7{2^z38j=~B{x(~YJ%lm@5W%48Y_1~$y~FvuZFD~~#~3V?zzSp4ii z3HU=H@;|3KwI0Gj9w>Rz<;j!9U%?ORTA8+V?bfi#uzj*~vRA8Ypw0BR{W^Rng>HaW zerM5YwYPcbb%Ld~cueh^1Uq`_$i!>O#MinwZoAdj6>6xR6D6JwZ`zX>i#w>SJy1*;p3i!8;o$%pZew zI5vL1Lk?XOR({2n_nF2b=84dfp$roPo;cG7N){e@Wdijv#kQub%732IxnH>nLMS=2 zLbNUi7A`nal&=((9h;E~t*v72+Fr*B3lG07?&vr=qBcCN$Od9%w)}YxX*x#zn|E(o z*x7Zowy?E5HO$Sso#W;m#pDhfqcn?wHb)`_{w4kixrMD0@Ol#>{YnB3_g<-&WInw0 zmMUamqr)vl0e(Vl*zfk z(@8NkZS?#jA|~Dj^!*d)isPVCZ)20uP|#^zYvHVxiT_YYlu}UWm)+lblPi4I63?*5 zosS*{kbwz^pvQ@Xh63#v(QGJY?{G5gFO89{vs`!89Du{`hJw&I7esYd8#hUF_Tht+ z*gmVZM+5W;#2mihlS%h8%IvqPlYV|-JFlVy&*MW)^K&@r>*Se}k&wZW-u!PI$ zVN$BC0(G_hl>YC=f3CoO5r9o8pEt1>?4-nuCtlru{(%n+kwE zfp?Q^PNf8XkkR$rhP+=_w_Ma3lY0Pt4`X6p3(|79_V2wOFH}Imr|u`(Sm>`HpTF+q z1Uz{nCk2dywtyecAKex(lUsxLpvw|r{I>Z`e)kRZ!iZ@=bKgsr&t|IW+e zndHEckLZWdRJ{@Wnhoe+62g`KqXAjYU!U-gCSYaa1uJncR$<#%PtI&>?P3q>pk4HD z&>djToG(-^9p)d_mEmt+sD=M>ZGr>Uz=bYjgQ*gsmtOTg7;B+_S3+c>9sTPc7OMgl zqs?RcMb?-uFcbs?@hp#bSkpA=R-#Lp2M~53J4{Z-LZ{u-$pekrtptqV$%7A7tq)U> zLfe4%F0!QO*c(H^ zFa7)CK$OtTbCe?2^7Z~p)$MP)3v6N;RhxV;>Xaou$}6y9p}(uLwXEp*Wse2+OLb-j z2Xd<1bfggRPW`c=qcba5AVJQeNu;H&PR8vSratj=`hyum;CjLpaN38%{(?_nc;2yD zD-1?=AI+(`&wsYdaZ>>4=kDAeG063}uz#t?y%6_72~z_OQ-yK`>bI0ISh?(fmwHPE zQia>@jx|&rU#6)T8}CRliH{rmx!Ecr2k-o?U64yJb>d*hHe_yRE%EyG@HZEp7$RcB zHF#1_w4b!7c-KSmXW$HtolSE#${^9_0&XtYKINSz4A0cXB-<1y!vqn6oL2#cxwKLx z6Y(|lOLSze3?rcZ#Wl^ zI~mNlYta{*zSx++Uw5Z#dC+VuQ%3zDyP%{5pX*(7c0`%Mnx=%rsrBsM+XQ06rVYMa z;EdBfrR>hLC5khrI*7rll@NQ~Tk`o4=bP0P$8GtY!?}P-Jge=jej|v9vB-uFpo@EB z945yL=I(kv0UVn2KXhoPW{3bygEnc;9#F?W+$qbq^N7*Wqqk|8Im= z+q4=PT|Z}eyv~tH{iM;L87}_i+9D zNpuX6XZTELaoA+;eYBO{ZqCMs!!1p8Ug^wEIC~c_CwXH*t0NV?x#?FJGptuvBF;L{ zOMeaszw!Ek`$rcu!8K){&>rm^4AL^QvJO1;zFXPJ15O`?BN}dd-B~GOm|Iv~_*!eX zV(>U%vRo%kE`x}V)2Z(3??Y6Q*IiDh?U`6-ZxC-pu$LYt@STk$$Yp>9bealwY?y** zin`QVW+hjSOR9JOI!gujhBq`X!@k})*8<&Sa%%%X|GV1XzO~)e+NrTKUU{@u)8AW< zy1r_^ZC)?qxcx~0nWQb%)WDYCUl5W;m#J4=)?K@q87`NC9S#7q5hR6l!_BsWr`}6? z8|0fKK#dDuXi4a8Z1Tk1Bp7`dxt6Qe`J+Ju#rJG%WZ%Kl`o`+d3hmm7abpwXT)i^J zh`#=Y;rO1mdJX09@Nh-AVhbJJ^4JR+n!!Zf!+mfLHXNR-(bC!RApFv4&Dn<0ne{NX zd$<YdRcY zp6Xks&)@lozTf}-EKIW#LxI}Hnb1($jdUmv;qE-%TWfs-voANjzn|;A->cMJ^rSS; z-@utXq1#OZ3}3g{(CF|ur%shQ@Cfm+zu1b9j^*1Q&1;=Ke7Jx&l2P5j2L+d9l2fwD z$S5n!^U_e$-1hS&CSKVgEV9h(azd!7jwq}358~^Ic(06AU2?9(BIS(#ZOX;Pp{p~m z`~JP%9ZbdvBR1l_b_XZ5AM}1uP5XxvRN#ce+Jm<`V_)xUK^tztXnh&{6>S9+pX{)@ zYj*IhGyy}RN|>QAQUE!3zucQ3;Z3d;PU5gU4lNxYD0SEj&tBnnKb+~?{_@j-!eAKB zsxSWe;kWy5@yF=n-FQ80Z{dbJ?h+dompviQqOn8y84E2FqCqSdS9e{-r$z!CZvuC>E3Jv$_GwV%Fq0^$$v`O%+ zA|om7K}o6y3-dI%epXwXSVWuyd3JUt+A8)9X*9KFZ{`8qIFx6{rbi)y8vl{3fcm^% ze|^vFO%}io)JztM$YqpzQSY_p2WfBAA2=b5qfg;T6t<6RoY8rI%cQ zb})<=Ply?jj$l3n`3AU487t!wf8MG_bTpB#pSY#vmUL5$^y`j&khLhba8efazNzhH zGC^=~Fvxc^tlH7KprD|$wUftr*U3`skRc0UWT&t5e9vs!ii*RgMU50|0esZ0aYR-B zp=M-$NNSaB?~|-ut$)qG>b3Jwn2YDjny+aSk^t5oCHm|Zs_)cT-3h4bLllVd1G&wW z%)7%>bV!`v2#cvY$XSkrqmVW%T_8ZM#W=07$~`Ir>uqH zQg3AoO2QQ5Yt7bTHGo3driqCm;&U>BWVhX(Zs1|F;1Xz=<++_WaB>c5iHf!|P+D4A zb}NpI4hIn8n1YUtTIoOrUMCZ+_Tb{rw7-2gOR$y(kw_u3jPbj{p1(0hi(9loAAse| zP{xns(85AX3=Bsly!rg@U0%)D>nCycO`$+OiMcJwKvTgA9Qy$I~5CPvFr1@+uDvrKTMpSAPT?P}7H<^BW(lOlOmoCA6312ueI&jf5yRYY7 zv#U{0k>giy?KV2`dKEU&@7-3e~^ z>%-qY_u-pIebG_$g|{q+jxpYN@h(U%jZ@3(=`@W#Lc7y{oNPQfGp2?9tODLD^oMW%xMlG;K#>M&8qf+=Mf~U2 z65wPCwkxuKOG$cs;T9n>)(Jg{^>l@68RU0xOUD} zrhg+CpBFytnxJH04rd0B*+rJoQ!WCO2`DRM_vo{OP=NXKztIS1a>vm&NJ#OW75)ZR z4dDEB8UVKfNj;!NMr!>71$3k|PY1IGQGOh%DKa-8JFwW~futNdQ~Cv89~@tTrC<3s zOiX6hBO)zuoCGjbl3a#|4N$oe5|L3#`7G}OLH`PuW2%deFnsqT-UKBjEw+!4X>R8% zP8j$qQ5HN8`8BBcmyb8QFIc>x;J0V0*<~Yw%#B&=fD!(Ug2TTi!+>vV2E_mX<~L<& zch$pi)*nei8KyH-j5NtxmaaU&c`Zy%KBj|4vF=MP z9XkJ|`#|qDrGh#-glXi{F(pYQ>1&N84cH7@uTgBX`uy{`@RYh2(D;t|Y_A5HCV}I@T6?K3`>nE6v#ip4)lNSL zr)CtnH5@!=QL?DqOz-WbL&xU`m2J&2H#?;LmZjYcg8*6@iVi)iRbmyBdT7tym8d4+ zGsf=$&~14nCwoR)_zo*2?*lwi$HVCS&6O(eE6|CvCytLYx2eJhIkCgIsIokjS|7AD zIJ>!Big-<-*0tQxN_7tu3rOWGJvU-Dl(xeQD73GAU!eRfSad%E9Rh##2M}UlEPKqC z@F%!_q~&|N!hU-VL^$D3h(}9|mh&cgEmqd~*vx>b5S4npmBliMj?fX!g5)73P%MHb z9Kx2?6!GbJX+1a&erGIFEsXW{1cSfjrw^s#Y`BFoK=%}lf~A;@*hPW~t3gjNmp>TyG$lGA9l_@^ZI z>Qdzd8MvK-YLia(d%29ZF$ z0?Kd&1x>2m9C|4!Nw#!_jQ$K3Jd1wZzYn}A1n@gnRs=2YQznrVI3YI_3J=K#{-iop zE{FRm$P^Oz3)gP;MOe+w;=C(hFP47+;N^?!O)GeIZ8R$@Cr1)QFsXF+4^ahaT@5j{ zl-bs61er!e-0@{XWG(Hj!v!>|gXul`E~h$4`9G74(pM8N5DqZV*VMU+(5#ZM{?DVP zhnrDQoDy(rn7GQc@=O`UhckNT2ae$}`DJE!Luq1FZfWOAHGhcSb?A#+YD8VyY0EqLR?tU4K5MEplC`=mXT|NwazI5WS`sp*^ zq!M))iD|C0gAMV0UKw(hm9XwYkktY1{O1w77fduFjfv;XkJ{O1O77z+NfPz|Hs?^ydQiL@&E4rKV$ws zyB+%fO}$hr@EOpTRoDftjgN7R7^vp>Bmk44_`hb>#r*$orgQ$jH>S+Ew>%Sfs?!?7*uF+H>A#?-SdPY%qk=(Xkl^E)8d7h=4t>%ps@HeY_D_usCw(%z3~^z;n;5F z$PlWR`AU-_2K6~wV3epzq7TjhaoUY@TxVz*DFM3pHYG6UbNUZ#@I>bIC@ zc&|4P0UPG&Wqr@DVNm}1FRy=h!Ko2`go>8)R&T(k@h0im0exn{wuFUR_mS)BY7uY7 z^9VKVPpXTH@1Qgor1zAx>|I7W9^7eVH9kH%2~=WMz5eadpFPSSg4SHwE71TlDKT*+ zO}eC>Rxct|wf9}G&et0Wsi~-_fO@f=(3A4HvWhKS)xICg0wz}&iwYUktG2t7woRHwn!9ujrmFTS1)K>K{ z!W!$_WiA%|8DqN)yFf(h1kR*St^ zYF(x=MhHa(IMC3%Ai@n3{Eb??&s(kqN&$9IMZoP-o0%xG*bpp0cq3(qKY$JUDZHkc=r=e zU8|>`&!NtIr1C+7sNB7L;^Ywqa*A3mXUlSOOMYGLVe1X- z7jsc5b?T2@a0dwS*Wd2Ic+Obb*QJE|Ln1`H1ONfm;yde08UbNS#dePLPo}~ZV-wR@ z_DU1}sN;``9coW&Pd3;h+FuCEP(lyICl1+}xSGCCn701t0mz-;#w3?5v?ov8Ai3tg zt_~`fVu(;pfh2R2jUvx*b#!OZ#oVWYmz(aV_lmS>M%fjHU zwAUvTxl35S4u#oCH#KNkkp?uy-Vg8M+0j=P#TcI}8)K%gjSqTxv#+ITl*w}0tTub{ zQ8hzqemZ-(#wf;G)cq0$;SbF{-vyNB78f^0N6c(d6Qkz%u%`x>dh_Bw@z6L;_`OO* zm*jO!UAuC$6KUH!RI6jV+B+1kv!{fch!splNYE6fNqg5JvM2uvVmEY^>(FBWH(=k*Z!%a0^T%oAodhZ1S zKVM&yGad`;7#-N>WRRLy3FOI=Q9sQb1Ceg)57}uEtOq_qhb*8L4YwcRVw+W6fk<<- zm5N25u7kF}eIjJxl(eBmx)Zp+=GFAvR-bYB6dyj^2d5s6u*~A%cp>|7&mJT#XH|Qr z?tO0+t1CR)wI&Y0e`QY=$7NI|`D7s2O&>myXSf>PRMI{fM`GNH2D zGHy39axTwv>=Prct+oE*va+wiq(QM*8V$s5JJnt!%|q66^q!7^0WaLPbfAQ-tR&YI zx^iPl!B&+Ye140Has*E2I8_O3aI*3NF6RmW?nY;277npH1mwnh^XfB7P><+yxB zBO)dSb+kr{Pr@q-sbOFo^+Cnk!fmieubCP)$a=Oj9)Natj;9xb3gz` z;YJ=Yn>0~QR=v_{1A?o(J;m0K&Q&Y{Y{k19j` zf3&@ISd?qmKFr>3MMPi=3IZY`2ucb_r=p@rN%u&1cMP|Mh?Ix}(hbreU4w#j=g=eF z%`h-S{4PMB$NhZA`+e{G{f_rO{6jt9jw{x+)_I<5EgGRE*!svDjtfQpBE)s()F9vJ z>iX*2z1-!BoHR1$E!M(Xpjd~1ym8UMfX806_0|3BSHbBKL*_oOOIzk={OZiM*p+>3 zuq6uUiS?L6U6#o`0L)qLIO~ktp5Mi$I%}6aI66ljfw(aIZ5?0`<*MuY>deQYHCYiJ zv$6X%gazJLxJn9DZhl0lI`55Zt0L}cXf@H|&gMGK2oi6Kk%dE>{>Pqe(dZrfUf8f} zN{T6R{lGL>>`qT@n<`dOMynp!UO;*E@&h3D02?F*{I;BfT`YBvZTG=G4v@6f2l*-6 zHVuI^<$MMm^98fef%7C7$-V7?GicE2=K7?L9!AlO=qT|-TRztjJ$ZzvZ>}> zXee|_nquZIbLjp?PcyA5^V;Sn+~r{8VT%{MGD-<}ammbmQ2;hw87TmVb7of7Nd3&l z(j6dbypdwOyX#Z5xVb_6IwdJd>?BNnex|fcP-1gxNgzC1(Cg0jw}#kF%gM>9DMcNf z51~=HF)vLd(Xm6#KjNR8hc+DGW}z^U-?8V}U~|YXzI)oIDSaM*XdnZm%>6L(dJ(+h zYZOZm4oJRMQF}=D#ddz=7&!caO!Z#A4kwkjH7ahl$`9UYcBk5UN4s=zNCO;Yz>-iQ zJ~G`%$>5_gWodX?`$-9~o=Oq@bqY(%)}fXGQW1t{M82KI+zqiA=GkThX?6X0J|Qk`AA_I&R9FI0aE0dV{RO~q@M z5kOvpN}8-iT8HGh4Tc?RDrzz}KD5USIRarBxXe`Ip{VfK2<@j&Bcdbq)bv(>U~gu@ z%GOl9Vx#yLg-zsb)^#Nr;4?zg;~+!m`ki*TnTh)Yu8@_>CUMIh9TjAucyG;QsceJ{9GR<__vKyzgSHDclf z`S@giJU1H)7GDINWgD${Et8@y%>Nl$&4qSu2yj6?VQ0so(}hr=Ov*-sGjxuv%G0H& zr)A~%{UvL&vZj`YPxEUrmlQ4mA$#?W!h_B6^Y0k3OX?XaW3elknAWk6;sYAmr*S3Y zZhsK8J_H)YRe0r>yw9b8cZyxfX>XUW!2z8y$FQrani^3Yns@b6B5Y}4;*RmeC+BKg zz1`)=)+{JmPplkN*SFrWlm{LouI?{X zIM+0P9&g7FssG~-4b%&6>(j~8*aLc zr&<>lf9Qbr?t)c*<-ySm5k$5CE$=D{+Ox5=oLru_F|>{e3yDzXm6IDLs7?>9=Me|u zwL1z|B%qa3C1av1U%wdwAG>V#GR zEo#6(EHIkcKz%n?XJFZ&*BEwK`aUlZ23G@Vepx_5(9r=E0-x&5 zFSzhI?vA)-TP|cJ1qG{zA=Gbw{PmpP zv=BjYF-Eo<9JLKC;E(dK!-k5Enu`t40RXzLEA3d_tF5Br@Z-nR-=9U^U^+c@?g_yB zbzA%fcxxP{@?O*^F^Bp9nk*~2+RZQF1XKeYj(Fhr^&2;gYm`{4c%Zty2uG^d#M+ z{!x@tFH-jhG4WN;-L~K{x7D5$F7YCB%c*uTyl!-i9<1_fLrdu0lU0=3G%SZ)`F1zk z>+dfP8$KyOPdxcm5C}lzs{XRX<|J&+iYsbrVt`7(Wo2v@XoTx0P9uLwIl&m+kssRX zW;lhv2vJ_j)xabbgibE{$8BTQ4R*lR4Mb>PA9QL>xP_Q3r*s$mOp_xN$}Hj0q?Jxi zw5Y~|+4A98Oy?(m-Q0aGC=vO1nx@<~fFE;>@&^;-*e)VZbiv0)*{xs;gb-K$OSpU2o3h)qI~3#irB3i@RgT!jJ1SETvh5Ct6H|+>@!;d{np#MEr2tjtUh?* zqRT{XekW2`pwfEbHAt4}-z#5Fz_0vhvm;?oq0F|9bUP(h?mBLLG&lsQ50B7u+4td) zFFw*%+@_0*0nl~d7+F@z3^NWuO}B_bV6{L2{`pa{rP>ap5FHh4w%CTV9q?|}Lp1J- zAJ<#Y9v>ag-An5Q#ks0~OwBt2WLz564n&gyjKIp)_&!kPyq4Z&CexBJ+tb;e;!V=i zuA+LYnT_@Jr4%JiC6C<=1f=HhBwfYgwszH;PK|CELI>)0(Ps-OLK4vuMJLKXo3#7i-e2VNDr z&wKgG9R1!)$kT<4UdRbe->`jEl;09(ggkLqoK9I&T?GGCj{VxTK3>}pOr7)Ydvw(r zE+bg=0R^K}?QE>p-$MeQa{eMP8-i|5q>15qcpNRJ9W2>e;$yveQ{}d>7z=LA9L1~y zyV~)$%AxgvI%_{fp{lSMAKRW{`faOGk45)3kOr-K4>(Q@%vb&y@qTs5OhVl2yLAsUC=Hq!JQ5qt3~GD@UkyILw&{{r zP*qI2M0fF-u{m;;toa1>aiW;u=XKfJvGg)

hdTfDCov`-F=E07%_+D+5HJ3aYoV zVy2)u!=jv=Abz{HqdND%Ov*vVBU{JD?3tpDlE>ElDNlN+%fJDUfL~B*^g)8ovsv0# z;X5{M7Z&LZo?Qb)3CY!%aJW?F`iuvTOrAD+2L(^i0YPG1syj#uy8 znW4*cAU!=UB4%-6X>MZb^Sfw~xc4wcK{rWsN{2XWcuSndm+Ksii&OJ++>E~6Q3M1C zfG`m+Cy}A~Jih-8=Bt~{7)bsUAn6#(*jsGuTF4FfxM>>&@hYu=@REnT@MeG0DZH;J zZfgVd*DTp#MSXHyukyfgxIN49jmj*|pF7;f#~B$-%hCu~;ChKTbReCT@`=XK78geJ=y**guOY``QI68y;;}4@h7%JMwad9UDHmuw(OB+!)t{$! z&m21j@vnw1zxc#uc{yZu?paz8C=p3Dl%M2Fl&!)_cZvUzMwydPAXsCcA2X3m_OJxe zrAO|LcGq|f{OD|;8X0Ed;1k=Nr-F5Gz30&08*uKQ7?vtcri#^@n;uWisDs^NlOVSj z1{q~9Dola0P)!vDsi>p2?CiDHzQ)w_Po^0TMf4EaH1T)@%NM1nl`nBuLqvI z8?Bk4lb5@DZRQ#i4Kz<5_ldpmuGnF8#Y4iyw5iR?u6=va_*Un9BR- z=-i^v;#KZhEPF=msEkR-Fwlpb1Roy`Fadd>k6@U(*uB4zjei+z)}Qq9e()GtuY z#JgrzE;12Mktk?3W63I;Pa;hH&K@U#FfKnzX$w{Twmnt&h>xHei9GP&EX~Q$gWo)U zd}m+L(;)1+Q@G5H`G~;=`Dh|qa@>ALHL7`fU+QHq5szl=S4l!o?W0jKANZlDpPrfd zq@^(^(Ere8cBOI7ksi7KAp#nVJ5^6eHn%vlur;4*8wc&iTzW#KsjX>on0P;uUPTH; z=)0}^#q6NZ#Y{oopS_jVJ5JXd4vD4WCVAT~?3-0*H)KyalPHJ^p(QV+KYc4ezrU%@ zL*WrFq{&*oz6_@P&|{Ib&%;=?_Jvd z3l)&a4lrpXBvgAA%*&8zWC_tBaS+`~EURVU*<(xe7tgh~>xISPZ%M=NP}h zmgXi$XBVq=h;k^jXW|=}wk;aHZ%R&LB}tLzrJZj&^;T35srMZ$lM%lTwp}?W;Pot5 zab0!rdi=PFxCMp^s4SeI>yvq=AaP>YI&U#qn~1t~(0RyI!F_}JdV7UMh9tbuS4ysO zX2=240niS*<>iAIA?ObR-8v{%OWext@ckI{LY{Q{dj8EDe3M_tCp}2|c2hTVWMglG ziXu6hCFV^V0x<3EE$4$Wgxuucc+dp$hOWG`QyzuicTlh}(pIv|$i#xjexz}2h-8M2QoafDX|=EOyC80I(_`!MeRyPNE%YMI$1dwJ+3mddDN`GA*psEqymt z2rH2wl(zR_+BRK{Qe4n><8{|U>Ai$Ul}>{_Ne$IJfx#>19I}ZVI7;6@baCa6T^O-Z zMMZs!x#Atw$H#5FGShM^g@2u##rmHVhKvfX>cSTYI=>IO!Umz6*Jnn2Dn%pKD87SM z>sF8EcDQR*mDLv)7o$KoXd0T&o6;oo>zx${y@OeJ%#+O=$(s>o3y| z@trdqB3w*Nh;a3r0RQT9+ng~~;lBMIxivEB%P4mabGl_GB0v^6FXFZe^J{Ic%GzInsj>cI2Mt_c8)phfg^RU zfC`ATfOo}r_+76KHEbwaiv>CppnMYc`2HS-rHkc*e~E}N_uXy-qF`@Uk8NvZVW^z( z#F}X?vo_d{A}xopo)*&^nKuG1^S^fhIx)EsUQ-VztC3thDcXBcj&5kZK@5L|ZplP8 z<`Mnzhjv;Q>IR|LdvakAHYtzi=AHZOlyQP|%kwM3c~Il7e>kU3rID4DQ_Gb7nBEF7 zA_Q^+gF{}J`31`XduaUC=G;kBnDPkz=5+_3iHcS1S+>p9I1yu|odm-)`RE{*#c_v@ z5$bz(C@wq-%I6c~H65cX=?2#Q8wb*aN|pm}L&D>*Yc_Dlr;pP+%Q1FcR02jBT0Zv@Rks%*AqkyfdbiYBG&Hr&T0gZX()|_we(fuK8dOqJ za%t#K&5(s8|AJSYsVSxzgpFybrbOBIuP3nmilj>;`J172;5h&x^ zcJ)+Mi$I^S#=N_~|I#)SuWMVZxb=o$U)K7|r~0oz6D;H?wC<=TAFIMj-Ip3?L!lhiy>hl zh!6y0mRlLt&e^_POY5PDiw&~1^Xu21BwwF0>W*Wn40uH64INu*v2C5o&LJ&ry^g!A znHk}^UJ7a&>{)0D+nE4!j*pXE%*@bE(@gIcd!IBMzR%5gM#5|L3I&B>?DgZ91kaaj zC{hdxekWG#z9`@4;r<|{^I=0QWYCl;*Zon(SpVtc$Mcu;tDr|~7@hA_Fze7ul(aOL zdwctfKiP3}?*jHS9nm5uC+C54S65NHp;KPD8vJ%CdJh#QR;PW`g6Q}X&hc^Z(sX~& zJ$E?A*>gI`L>Fq;7c{a+~I#;jE zRXeRGr7|<$TdS#i@&u+0XJ9Z#aS^^I{&Dws5Wy3iG>NBo#P9yWC2{!LVt>Tv^|Hhs z`A7>ilN}nB>{oZ4S6R8_YWu@CZ%|&iA+6m==cra$xvS-_(02EwOzl>`cYaSgTKiEb z*w(Gt? z`R6bE#Yv4e*I^m#Id&~!SCFlMt?*u@=yklev!>R~na1DxtfvQ^_IPBdxx1-LTp(BS zQT&Nw3VeE}@kG0@y{_&k;UfQnf8C+`PEJ4XCh<9+{;sj&U1s0^eH7@?}#chAzwV8kob&t57oqr4n3O2w(w&(zI!yK`)eAEq|K z5ohLF#WiyF%2L++PbASvJ%(3UoQ=hHu~-+#13`>u#aotxlsCiFap+_qy`}O%LC=iLZicgESXxBcX5^$_&@yC9QgeY-EFhRe=qk1s zozeEC0bEBa-=mAJmuPw?G=$rJ-4B?AFNqULEZ6(io10k^P5<;+k{W_pZaxbX=<*W&?4Xt%h|7i~96&qdvnKYAd-a=%u6KQ7V}2Y3suZ zfRYcfAwWqy$W&0Q7K?~s%W^9%+xX}2lWrH+p7G@mZiuZ=e5yG;zp^l~WMpZW66Cu@ zD(Z2ahm~1ANAVFKbBSMo5KB1Q`Q&arF}VA`RI9rE;48%Miut6t??L&ddh(B3&e98I zHQ{xANmi=ITUKRqKf+5Ku`vZUPfurNS|V}Vg!(tE9+@pV9wSdjBhfj}R1W!tQO*?D z#^LJJhf8eG2t*9k5Lanp;zoFtFV7bTS-F7K%1O$f8~tdXKQ8~SI>v9UG>`bQ>bmbO zIPozL`utKYjPglr^y<{J4&|Q{OE4!$$J&5h>}eTn`qo#MUzd&Th@uvEnCqGhao^=h zS6NtG=H{__ACXQJ{8NNT5RZ5Id<8nx<<(YLb6RC!7d>gz10EaBm`Z%~%12ixIO0zJ z21qKpX}4ecPlCdrqr>*((vlYXml!lV>0j&odH>UYX*4AlZ~kj2;+@IMfBkA_Dh`LE zVCevHV@G2v&+XfQjj1;8)=;JhWj431EMS&^^)mq-&_TDq0<5-61@!+kz|xBNa^kEI4>xxq1nnQ9~aaGCH$shntUG^m>UY({w5t~ec%xl5;8vO+P-DyDne;*(P4|H51|AiH1F?Mt9o@RDY1iD zPlj}!qSvP!@;_O3Ki;?q@$K7zD$N6%?z@1Hm2u?l`AUGMR%SVa-LU--@gZz|6ze&` z$rScc5BSPM!$W6<0EV@J*`8AjiU(9q!zdo~(Fu)-QcwF_P?Uf5@{no&Rufnxhdb0g zJYVSH9%YD>x>U^S-p=M&6=AT+?97uJAE2)Uah@~npyFkAB3-s#4B`yCF{oGxJ3)h4g=)}}TFXG`ot=Ab^_J|j z=eE+hn($&*@tq~R@3%1AptcLz%eKm{d)@G0N$D};gGmDS^sU^5*A`}8_*G(pPCv@( zyzA>q8y`%sZf@i7t1=G~6J8OXKTBfol&8?qh#OCXZ4etdu0ncwt0?Oie|XagYdrw2 zPh?aS-(!9O%eNoH)t$1h{s<)-xXY~A&KU-#2rK>lADZ`!Vh z=jF?nq2v@(n;{(BX=ztLJy*K|r@CKXrg@OgoAFWamWVsEOZ&2NgD+W0{+G&+(YO`g zIFJLC<%X9^h`D1k6<|&QVtQce(VzpZ{$gsj2-fZ8)X}5D z%c*H`8l6TxVRDvmnnRq7@!+Pv2^!Pb^;S7^42GQ_ausp9)8}Q3!C>S7t;8u1MGY)E zxW1Asw}n$#dT-e|!o9v#fuwd8d1-n-K`~5feukTgT23b?fD!#elUOI9g{7P96g7t) z1v;J%AUiZNB8K+IY}G{Fg9V~~c$HWQN0m%PgF^3K)4!&Dcm%I>311v15ZoyE1HuIHc+x)t0-Y%nJ|KCowz3(&a1OxRA!Ypnmwj}bpEa1 za)>sa`%5EHC%dF~duqBW~rCU&}DUK_DERAKg}Rdjz=VPRI1^rpqgx>ENvv%Tu~ z7O1Dn(wQ#l zL?#YJYPlcnWKIl|ghj?%+$Sj8@(^yW^~>Kgol@PnLHHc%{6Gc6Pi=BF1k)AG7m%Gl ze}1S;eQ9v>-r|pEsIDZE^UG|7)vF>~8{MY8-?jo;LD>V4U4#s@JzTl>(=)Kq0R>|_ zI6Zk;1P*f#6~Z3fzb?(|}nNz$}Df*^%oh=JRD-!zKeJx||) zuME;T80kjTtdz*dFoihfpuoi_ItidPAJ)(o^FlA_{ec8-TE;C)GH^x!;Qh{H zQhFg34TsHh6BHjfK?A;k08k_B{?;HdeHH$T#}D@?A?EsrVB@t=O7u%GeQs;POSil2 z3p=$PeuT!6z?1&svR@wv^M*2Wb8T&H@gv=+I|7*VYw(>H6)KA?3ep_Vw6z31m#r-s zU~qsEl_D(>FfgyGMwMxD&8=s{ADT$LDt3UI zzilR24TFXvLqeRnxo>g7Ow9GKcBh3y723g;B9YBfz{7SPE=;hw3WR`n2K z-w`Z{<*FI5--J}=X63Y9&iS)`!C|PXszwfDz_|Z_J^VFf4vx;=uG7TYCHqb*_wK1W z2O@mjSL0(#(89O)#4J0;B$z%(KrH!PRQ?kC&!+Q3O%prx8e75;H* zz&Dt=MuB$%D)Dtugru8mg7QC zK_^FX;g(P6vC}rEbsxy$9#+hP5SVz`TDhS(^A*yq+;TK#1L?Y>b?R%N><%~Zbmuu2 zca`A6Ny8sd^B6eyHLm?=P=M>MjS#H=wp`w;7);>iN87L3Nc~ls=f#ITHyLv`G(M2% z{y~sEy!~(>(0%d{+JIheuo?e~pZ>4HuzyDxox=YS=KZ|?|MT`g@sIyjnD_tZ{wELq zuW$b|{s2D~@BF`TeJp(pTK4CCSK6A! zl-OrFt{fpu@oWho=9P4voSu#d_g<+lLwjlZbqWt?jw+V0Q~Co4a<}+cfI@X-9G=@q zA4R9O(ZdLt8-dFtt}RDTc%I`^ZT|~3`Dd{F&#(N2Kb=J5N52ZLe@qig!b$Wl={jv? zTDtr#kG7Ix!?^C=;xkh@iB2-7KldeBoH!r;7LNb+g!1<2tcMAT$;rkuR2Rb#;ZYHh z9-d`of2E}V7+c8v`;k-2PY&<@yn^LkGl zp54nXwj@n^Q&(O(Mz_FQ`Kf)nF!_+cj-L+}P=*`-N`w{8ab(VHYHE?VVlR{MnRsmz z2)lojcGbQ-wO@C=tM2<$T?<;aqt)gEJ#@*_yKWe?=?Mo_yU#UA5n_1X%XEDI&Y_#1 zkx?F*_6zClvESqI zfW^gkzMrLRQ(yiN8?nFCE$~{`}^F1RC-0cP{HlI^z5rB#o9ep9d|u+$;&~;{)VU-#FWz*GD;=z@p1Q`fA}zt zV#u4}E&zoh*Si?c_}@=tGoB{$YZ85QGOYy9IeQWuyi}brNQodKHmNh)JpoW!vhlg> zxM0&zFGpTpDi*@3H<~s)zhPC(^CLYVB_^>lQAv-jgY7+?$OTO$Eq)_OG<6TN#mvP< zfHQsrIOEE;3%@21DSWwpmFef>B*UE@gbs%30Dv)@(GLcGB%;Rz&HsS?VEEI1Mg~pQ z8+R)xAad2l%BuI1;%M_N%Kr}Js1rmlX7I@VF*x(&^pH#U?sbP*OFXEHg~sQ7`qjb~ zuB0g4D!l%bxWfOw!3h&}^g6U*>gx|C>KgI+FqPE6vY&!2*L0 zwaH3oGvgQNM&J}?gP1IE++3t(d0BE>K$Q@WolimCHagI8b`CVzSnufqxPT#JEpP^Z zo6^N-2d{hG#$`HVOd3x$ce~PH3w-k8lk;Ozbo9cXtIHzH0WC)bvsEX20$@c<8`+V5 zS?c`x%LKF zX=u_u;8~~#sue(So}gUV3Hu)kypf-y6q zZ0eJmLO$TlaMT2yR|EbFJn}aK`2Tef9*zJ181MWa-2eaW_K)ZO1m9myn-Z(TS#58l z&Z&N)50&`}`zL!^y?!4+o% zTmJ|i%`WN1YC$^c7z7k!X=pvYI$Q8#+!MF8;%sdW{Dr$e@o&k+F}Uq$bIF)Gbcc`I zoSTE0eoa>-t4=)^fhM@D-uNNuI(bGhW2-<-nXp=S4 zW1{e}zHU8@gNK@$?q-MYXJg%4Q4Iy4^cDnSSDOQmVKz4PwsLr6RgAT7H?WpIKhgN- z{bHD%JyWXvw6?;}p9=2br{C~_%AZ)?jdpX$-3wS5wx;^5PZ}`yZCJ8~jvKxZ! z`0Jnd&o3^Pu(b8}KV+wjiiof{)5g(C%Ri{$dgtiJeojs&gqdnN2hQ6E(eGqRoyxXk z2K&0-J&{Q==*a8_1Z^bCSwf%*RCzp)m}#JznB`G*oor-hPJuhb_XE4n;0*_qcB zEu=j6{sGj%&h`G%IeJS=O9^!8bYm9<9RHWk7SVzk4Y1wEWn#hS=spIN9^edtY_0D? z{-fSp+s-OS+szF-Wx~x3V|wTS6Y)tfMJfH=r>KZ_Bz%D*w88wU^(0CJU`uM85Mb@@ z@m`I4zXu}sEi$5oQ8r1sJ$)?lZKx>(^dYPFjavJyD*%1_`Ue0ZJTY4oY6k6f(0X6- z)JG?0q)*wP+@$mE_x66=L4`@s75+!O)(r~P-X)H%4~!AQ z28Pyr?lp6RC35p)9oPsKDw!>e#DS5rLQW#wcy zh#@0tO(BVuu{6z~!^ycOReEnd7>sLaWqgH|Db3i3sARL>GfqWSRpE+=5~6q!bqeNH zq^n_{SK_=l&Q9qEMjt36C6)(K%Ps58XFNdAu3uHSo=eF@4o|S!L1?AXzc zY_^q9mTZl{iVf;4@#hrCTM zgWEj1pzpIzTVasB3;NG!eaNOUvIAr z_Lu@yUR8yiwDfg(SG@GzcitA{&=+`Pd?%S&)y(3gs)owmaxuO5=o%)yaW07v;^F06 zR{X~Yg5o=W4*_m*`Hmjy7P;R&{FzLNnU-@3(lM9Ru2#;Hs&eREqx7F0wYOAR+69%` zARyeTjCH8Y4#eyc&SY4wzXbRnS2&xfn1|>l+L!4=wG0Ueb0Zxm_x4&|L`DwTC)JwB z#buG6qjCXl%sO-*iGnkCVk!dEi%hq5+-hWqoGmK;61|O$aEa9TH=3$R<$+8$r-s>% zjxQ~hS}&#@(fQwNt03t^WEERxhR;qo%oet`Cbq0_ieQdLZ~FLvkqjRPevFv{pdEy$ za$_$d{dW8x*D$j^+06TvAjdx{s$bYyUk59XwYH*LAGu>*Z)6OQlv|^KT!?mIS5LMSViK+JhQCyO;8HX$Ma$e@ZO5H2 z1#GZx?c7Oamh$rw^d0$0Z(4+Sj-cgyFzX75y3A2`FW;Y7~6!lRiWoGipBUy4llh=}_{6XME z^3zo*{{R4eftz=xfQk?3dV^!%UJO$Z@CF4x=oxIx^*Nb(`}&BjL;qM}&Dfn@L^Lnn zY7mh4l9&Slw*Ce#d-Z0xK&5dZ>=EE}ANv#i;XRgd$yTAHC=&*XDV_>>YRYvZg-NG; zueQSzz(HUk1pxrHV!f7Rncl#_%IntrMxtrBMD>onr^ql^B-SFX+51LGdCog|-{@qH zCiEvB;B@19$W5z9pYv`annAVNPX!Q_Hy9MB-a)c{Yc1STk_QQIu|Q=#Z?64!4~nu~ zo&-K_ezgu(MoNKz>>S&yAC|0(jqKcNVKZN9CSJZwB!I1|b+dDDfLgi@fuyvT_&>K9 z4h?FcMX>xw+RS$fvf||S9ze}Lrj7Rj8R#I;QTMXTH|;=kuU^13G-p5=vG6yU{jSR+ zOQ->7u+q8M-6wpeb8T@8BKrAxw?Qw7&$+g4!i>IT5j^gdtkp3`i$m(`S3sy8Lolzc*N z_mpJtxyhjJV@Fp_V^RaSJzb|(WHmM)G)RsMuuu@v+dL}qz@Ro|Xg2iE$ zo8sG#{G+U_v-l+BtK47aopsexQ?oHLO2#1`J1)zDWkmXG8Bu$ZUD?;ij&;6&2HKN| z48NhVd8pQo1kJh-2Ml)~2HNfb**w{(2tEac2GXX(DMz*%X;MA`0l={DTOWo1wpoH+ zp=}-off^6EO1BepFJ`QB-0q&Q)(`jqtEEZZIea|x=h*P$$)5SGQp6==Z@tQf41pMK0bA`Tzcf~ z6*J4>qNBP#{(BwDEu)^LueV5@@=A)J+X8HCY!dFw+aLlO?>|GfgnAGc14`{W<*h0r z+t4jrs8Hb7o&M?9&GAHES+2YO^n3bE{jLDQ9e0WRHE~P>(6IU+I35mlR3oD!8Q2%x z?BWj{6*v?Xu5hhkF_9vULn!pVoA|*8kXr%jNhH;r$=zSd+=dNILc}Y4e0+nO!{2zh z?Q0T)JUz?JQ#riMVgXkTL_+nk{#2B0azO9`+QfPUR?Dy1Im)#($^fmD?Ctk5^p&P+ z@zrFoQv{M!4}R?l!@W=A;0e-kVvrDDlBAl7y1J>UWsk_sYd{9&kwJ|B8425~v|da1 zr>1+M)}bLSTFVwB{3RGdEALhi<}ITIZcf}+0g^-zyvD^pznj27^!OIz8xkueS+a%s z%v>&)-NPwfD3j|E(~(ZndSieh%>d+~fg}*RC2Qb6kq{jWQ)B8TDooJA#@(NQV(6B6 z1VXj%Ux4i%yh=sLAb%$ba)p5sv4O|!Umg@vsH1C2eS&U5hx07~!zPJ=GhF9S<{ z>eT#?Kj1zoZL&{@H=8Kx3uiz5fgGzzq$b+Jnv0X}1MN4ud%<2O`iiM+)WFPqRyI3J zAZ?#V5P%Ev)=K{%7$sUA`cF*mIR5!x34fSk+*W|I+gsQQoH-qalA?T|W(F!Bw2b!y zV`HPmU z!>wtc-ji0iKQlF5y*ae0)P&|-dIWu+$4ZQSR$+x#j^KU%I?B`z0cd{;n_4xt-hal# z24R6Q_ZWV{_#bXV#L4n@W!2Z?}-mQdXZ_S;#`&zJgTay-ZEPQn-&l6ZPq` zWxX#M(Ix-7O9<6x?rQ3Dll{s!j&ETScAx{C=c?CQx>9;H zw{s#G{x&&~7jn8xr{-3crAn$@>=s#D5$YMPPEykBVElzv%^}0p|wRaVoUz%v1xP8-*n z1Y#{Y@&JIQSy*OfW=5@Mddz3O?`&R2<#?n#HfqS(_ZW4(!Z~?T^8%n_gpl*b+)i`G z!Ze#YOjee3ZPR2VQ<=r~tA4{k`-P_6R5`@QyN&6&upm7pA|kTzgQ%M%jMg#} zthrzbd`8Yehil2q43!sV831hG*DjFzz5WfTjZtp&Hf=K{3g`5{e;tD5(tD2$8)X6; z3jv*Wh6D}GWl;xow18T!2ncp9Jk)?c0a}1E%0C1=pODZ{N2GYfKM^qyv=V_TTzkf= z2{YQDgzcS{^_O`J2}h>@$V%wu`SSX~C6EN?J+toH+A8b*?&sztphF2w?|`Jbi{UzD zLuzbKpv{?Yz#fy*^-8eva|*yrt|4$9o?Bbk@!fx!gV*i18bJKDL9@-n!*hjS@T2K3 zD>mow{fOW=xrTlcZ%6<7uL+m`;0aFfAA$7Si&}?!s+xl6vxeO! zoLV7MOFfWigX}`y&gp-L;C1YycYUD!xn;fNb=VkpX`vWl$ikwSQ2H|!$+9@*<;~$} z?|?6rUFF=jJE3K~Z2$?qzOCtSL5JOmXBba>Gz#Si5KZjuf9;~Q963Y=!d7o{^G7=L zXgE)SL@g@yEDOg5Anq2jo7-`PfN;}w#agt-^4Ybg;({4-4_96=yuuL^qmK>U37~dZ ze41?J7W?Qya=3C22(6+W@hTuK$61k`61ZT#+a#pRRC^$r2V(n@(%ht1qo8Ap_fVBB z?jsQ(W4ug{SLfWB zGWPfd;ys@^f!4LBbCHf-c27PF}(yjgKYuHunAx4eL|!sjUYPdX?L^a3{W zFlk48aR46G0mC;(n>Y(g!@lNk+R^o5A_BHOVKLmZ6N?v}#m81$+sjjSgRG(0l^{?0 z%+FcmY6tiwwv4v`2uO9dg)Jw8lLWmo0>`xu)=ZF77$yO&m1YkM;m*Eui!HL`;L-Q_ zVD)Mbp&Rho95hJvm6Z)R&(+Tr+jJ8SO(Ma$ztqLxp)EV!Su06m?F@x zNriNoqCE}M5h)Mw3gL)QY^#4RNbggJi_$~$b?WMYm-+J^4@o$eOSpz#60g`I(bUgh zNV0Po4`WQ~;P9Rp2MMWsywFIMf?$Y6Rk6we{srS8B;;;exo5>j7_9IpCnKjU7FE9> zAaCg?ypWakd7!;Ma+T?L_H>WL%96yO!KEPpyaXqDUtz}>5PQ3|6Ed2+^Pe48#B}z& zL{%>U`TDI;AP}KnnwWZKX7(NtgB`_I1jh&#>^H!!R^o%U<81cdaMXMQY^JEwDjig_ z?&v^_+_4x2HF=|wi`>g?1#-*}Gj&kjI7C+7Kuhg%2!6)Rvy6)J5>zm0I-b0mJvzW5 zsOY$cUnLA`ErTzVcIi`6n%Gm3t3!nhoZ{;S1`cd}+r(5{MA?UjA`3%1c#oJMVNcKu z+^$8vMe`RXgZz5*VGzOGwHbhp6-r(Y2{=cqqNZoh%A-an3i7b;C7h+lpp51+G%f$! zhqH6Eg0w5(rBO+&zx6581n|*zT;}IeCfw3#{agT;iGD>54>UFo_gD{3G+bF;jx;)8 zOjEwac}t9f!qm!Wrp+!n*LXq*I zGux|2O{#sKff$*sp`(rMZ>=abmSyu$3-Z{Jy+4sBI(ny5N9;?X)zVy^ZZ!}9-Mntq z7{^<3Meri10p1X>lX)dqh1KWBfmDNX4{u8g+X6yHwGycD_F2k7h;+qNaqe)Sf-DcZ z8N~m5(r#vTU+s9)#xC0#)OTQ`k?2UuC#QTepTeK3%yc@<~4ui4Kg?8 z8MSPg2gHuNn5{6mMKmuz(3 zGx&-=nNeu^_yxYYzOzGh<#@^nE(*K`l*-_yovXbwCwxClL1z!#8Fc)X^4dRTsgnKq z;8Xu}rof7b)Ahx|baV4E*!zv~Of}7N>5fRHYyNobJ>C+%l42RyT&6noc)<^B*u!!c zldUnv@D_CG$HPOt_X_78;I7}{;}fDeKHl93{|19ysa8;rQQJq1`Fi4mwXaGz-;GGb z-C^dpIUElK{?|UvDRXYx+{7Yr$r#!3H6-7Dy$tp)B;m#7hYjCNWI8@7DIPt0k|%{N zZpq@)`aGm>@&UT~O`$RW^git^76D$q8O2s#xPV;u)O@v(sDJ;{)MS|(BDshAjp#G6 zxmL19t7|jWUKooGWNq_$@{|$xNHIJ&C)Ng6Qu_E5=++!`oQ!bO<7jwt`wI1AifCP~ zNWc?DecKdbDy)pm6xFkb!#dX<`P_S&>k`OpK}G|Scx{Vb=h2d8AwyBQsq7HrHSkM_&0ymhP=)-3|Jy`fLzOqN}-#*s<@O)B| z;X#Y+$7;%f;UT-5GquUO+mF)^Js!#%WEhPUD>bk8!Q%!$F;&$@(Nj?1_uq3b+>(-( zR|;B!1}fj#i4y#%BBw!gyZ5^H4QKN%Dl-#2FY!70?;;R!WvH(sJto_8;L=Hk2E&DB z$(fl+xZeIDne(|i)$qlYSLddinOM?Bw%zkQMvE&s>$+=3}^FObK>&D20 zpCbv@DDd`lyZ_OnSckK{Mzm^obGHQ*69Zj2zlCwsFvDbwp=@!opNjKfaa_%%LedT2 zt;h1wv)>F)o!UR+Y;n+#xXB7rJ5P@%IkI_OFK0 zI&|9?CoEL5OYMip=sZdN!yi5*+_DUBYjR0Ag z`B$dmSdT7DVf8?&I*T6Oje)c=-m}oqKp;ymjlA4^#JkS3j?meTT*rISjj9gJf zk_bppa+aJy1w?Xea=4N+l4DbP5lI3fNwVY&k~1hdBRMzF&3wyIW<*rlA{tOtk6f+BL^N-ed1`e|3bkrF3Fcq>Ca??Px%0U%MzT|Yc)wgi zj6S*4y|uNQUxOpj*39b5Sod_Zw2vs;*@i_iAwt`dDmL$;MGq{meMp zwqQ0*Jh;w3{sLjg`>rVz>!u$=5~@@^qb?cV?(|Q_W5%0Z?_V;~FFoG?h&qlGDgVlu zi`8QRM`lhY!6InrqqJmVqx)7wSlkNlpALdpL2ON+N{A@Bs0NCTOejIk71t?Nd9;g3 zNci?qh^1b&ip0WlH{WFhI*>TQpuV^ew|k#+9`Ca8B1L@ZcdF-yu-_;+$De~p=<0m` zx3B;3Kd;{UfB4Y<>C68g{2#<0|NmV0-|PMt|NNhCVym*s22^~VFRsmag*7$N@bVDZ zPVqRox%54OBux(ey2qMnQk`8~Kwm$rk^I!Gyl(d2lkaLsA$zvR;vUP$S`;NHH_TQQ z%Qm3%e|u4(kBynElcSHwQH6lwMEeELw@>lhUig2W5SDUJeR-di-AsKleN;5d23q@^ z&)#OyXLogtm%O6eq=XE{a(r20jq0D8n$+@>mQ+Y+pr+>wt^bd!#kh>n**FPIyB))^ zQCL)z6^?JKBm=kY(+mu>=C>iCNv^OxRVZTbYH_uVgDwR;tt{LZ&)B?fA8oXT{<-yA zETD)_Kw$XCj~{8D@+w>vpAf~$KYVb)%EUA=Gt=7f>vQT2ER&4_zlA)F+|S(!B9 zr_2N69pBBz?&nGx8>b~1I)AO-MGdX6L+3C}q43|666+R_7@8fi=HH%hqkYOH8*5|( z*$n4-t#H!Z29^6JT?uh`Eh3u{X-wA0Y4Zb1qE`5!PSgJsNqwhqp*E%lE)y&JQBuKAdfQgWjs zdhkexm$}kHHF>{tBHGqVqYwd6zXH0Ak7XyoIDYh#X4h_LTR}7@>S|b9S>-xrW@fr> zh|aa-I2qNcGu5iPR`mb#4{B3buW=ju%8YAAy?c96y7JR#b$dOM3E0*C;xhPL>0DO- zKX?8F`ri)QlXYFUg;PB@V0|=~%R-DNO-mc28V@7;$vvV_2@tn&6N9ZTpt&=tRcwt5 z|Gm4)y~0i*=2@bUXO^1)lyHO{))Rm@$G)LC%MdY6E84gH-n~PnS6I(A=%^o45t4I! z)QFxKek5rGuKTsh-Sv;Z{_-ODgZ_D#m6esN#Z=c!uQt%A9B%n))ENO(GEHSq^Nnk{ zS2&8Mpzm94Y*a7-v{X|!UAIPX_${Z*Yyzpy%FApz&;zwY6L!{nQrc) zqN1>mW2DcVUuF^&;Sk%{rpk`7IhmxpA*JKN)Enl5at&*b126*0-|L@E?2n_Rw5 z`D|@+vETD&PS&mb^N$2(vWx}YdY6-uoJ1Us7|i#u8p%$`;U52C-rt|lNPtgjYhf`5 zYp}T#;=M=~nJGpAWEL>Xr}9b)&aADqB-K&<;tmeX1!AX{{tdOqR4fb7@^FT+E({r3 z?@uM>v;U_16^vP<%JpD-s}|2_!N5JzVQsvyoJ->E>+x0=eqV{^p}|2gVOn%x9jhwR zi9m#P=xiE#OhKMilx1#gt3A+NYILi=5>^MA1`vZxF;E0eNFXrJ~M zIKZNkd^&FdW;B#{^TPrd4mDNPh%aACebVK7s--LF4zf&SC%~rFRy%b+=D79BFzotx zhVSl!$pe|Y#g@>o8YyGWt>6zgj}rKIrt#d# zhAR|?h`_k?qj&cR!&lelYaFH;&hg+-RG{S`Pl#)QDc3|-c|3|rEC;b} zw^4ZGXk+-q;cqHZT8eON4ISYeHxc3$f`QltSfQC^XnXaMEpF7!m6a{r0&K{qBJQjeg{yau z#{)}g+7ypib{{ZByKP8K>fc5hgU;M57bu2EX zYxC}!=I!l1*N{<6`g&6@o1YGHR|%BR{NA>A1l0%^fI~?}P8Q@KApzasEJRyA|NHZq zuHVaISGXr@gJoKcI2=>XrSeI$2$^Wh!GHQW;R~1J>&A~qV#*oQdmiqa!?$Q?nzbXO zYlJXEYrXyjaTfxf>z$=BLOz3&y>#^)iR6_#f%DlZv#EFf2~s#|nJ}kWIq@7sHXxIP(Zlc8BPz${RwD$eoB!IKYSg$mwFsB2u5PVcW$60ak=0nX za$%3JG#UfPH&_xGHr{Ij&!FYITJ7k1vHMR5*a+~{&hIByd0vKcJuXU1q%8Vg_N<#n z$xAo>akRkSs~>tclvDosnCZ~KKfk&+RNr{|vv1T4HOiY&B64f&=TC!s)}z^6-)3XF z0CcxMQ&|rTwD;+Hh&8`mx#h?w8bG%yl>D&3GBGvr0a$pw63?FR*F0-#p6CFNn9=`$Y76|J_H*JohyA&|DcDjgC!zu2C=B3XFM7|Aq4?cG!@)*H(0x;g3>z zDkcoN(CWgVpk(ae^LI-ppN-)Lg`-^J(n1sbaksfu{#o4e((;K?+k&(-iP3veb=}tK zn3sRYGtD5xM3-Pj1pcI0)De}?!%%TY+q6g)LwG^=9*11on#K6jbeu`A}bmt6VbSnu>?U%NAXB@Dk_}b|1Lix3cqoG04nK~2G1I%9y zpiZsd7&`PKa6f9B!4A8iJQWjkSWa~x$=w@Oo%QJ?XKbGKRO*ww++`yo-C$!T0bM|KcmDG!#@@#Z#G z$0G1y@O#4i*kFt=^#5pnU^4uD zH#rcw0UBw{c*?VQS4e8hZyFtQi>a9&+v)#Y%Y9fjhn>hFuQGZ*UXUz4>(-Of8QC#4@8k5qW|^%ujKPE7Eh1d zL9l%(Ce}vsHpb=zJ`(eygjDlh5yMhnrc^mzx^9!`)CajKzg_*AmZiflx3f%#BD9jB zrBVlhwqmbDyW;t&INRo1Z&J6hNh)<{Mta>I)V@~O>s7;7&_E#bm6Q}M=U1bRa{I+G83Z$MNB zg`Gi1t-m;jl7PCPu)F*%B|K44eM8K`xH;q@#L*xKg%JRUdSMn;hoPFFcDD3bD<-{l@6;V_hvZhs{y z#X=O68&zLWFi1XQKVBnU2LQ!Z+DwSP?GA_G+1V{wQ!Y+KorCBx1=hC@PydKM{5a1p z^Svy))BKQKoA`wFq_5upsg#l?D!6}gVoHZYt7P)Wb3@Tmz>jQE5%0M)HFcL9uE%b8 zDrK@dRU2GgJ&_>Q2qqxz`fyo!g-+?4a}dFR1*4Jy;s&r_X4)oSLNsah9+iE`dR5-Q zXvo_naQS9tnnlSHgC^IYM6a*m23Dl9mcHyi{8x5d1YA|q|AjBHFz-&osbbbZUo znJ745K+-)VdWcU_UgU?nr5;?F1%u5=ODIqQnOJ+ccI**Q1}yUQY>Z%w32jr#^siM>G8bB}jqv zDSu)rckwL%CUP=qAMtT=TFwvh_DeJYFMMM(SMd}mIe)jH_?^T0;kAz9jn}v=FE8`0 zFOQOPlnqLT{#JDidX32(Tlx6$b|m=NtzZCbT^&0)SMe8xu#4E7J^%w1-r6>M#; zQX3F~oRwdQ^#9bQP+t9gd+(4^^bws)jz^JYJeF@bO8@ zjXLWd@83gP);(g6LmfcLFtJ6)HE`}9y3)Tp%e4l`)92|Gys$8_!RM%q3~5b_UqQ%9 zUhNj39W*81Vy!GO3sJMEPF@qg7u zX^qy>h161HBbj*2W=h2g%;*P{<+F(fTeg?TR|u2xMHl1LEK2N{E{FYieK!v?1Wlru zd2}D6h=P4s=1=_b;DA;>SgN}sy9pAWSaXT5{@i5Im)x9RXpBUt2~LP0X|X$-LTkG0 z6NN1&;$Y04rzb1oC-qSb9^yf;5`(5Pv<$wAg>$Lsf29ZYyqHeSqGz{QoG7kR>*8CU z;uD$b&{k`U1RVZ86j^3^`ZmIAb^S=!VQ+6#L!&p2$6}%`d_OI#KS}ArBeXhpErFEK z+1Xh=LO#BHwYUhv*n5r?s5uQ+Su!XklLQ`+u$@~sYjCD!WRzQ-UWYBO^g|M^n}Uq{ zazp9c+w6&AmsOu?^VWxO_g^Q%uze2H+0c#sNXnvyMv~|T_1_7 ze5);K68V}5(rr^%z|73P8maHqYILrnO3~qLmA@w=%2NBWgLiG4T<>6ynNYQdw8Njl*VkjWf)(vW6|I^ z;SDPjonPwqx#3SGv|n@|nboW3e!hO^2$b(qitJxPbyJlN4tse%hS@=_9A_@AW!k~% zgHWjd!~qm9?xl(eqRm?Lq+@?(d-I{g~ zt9C=Cimwn`JipILhy7udB@m7EG0U6G*llt@7J&5vYH;wTfxhtzxS87M!i*rnF4uDjX-9GjQz~< zM2T(xcpx`9e}2pu-@cJUL}s7uWEI-KAimS2WX226u+F;xgKFkT5>%f2iFQ$g&+N1V zm7>7wYX?gtp&4@e^*-hc%M@4T^*k?Yhw? zS}e#FD`L%qzVKraJ@YQg%BwX$pR|HuwnGeUh8Hal0HLkc+0L+UNJ33*VRXxjoNw#I z;iQE*s;*T877>;#(*5)2w*R$R0byYsZJk|SfTcbE`FksWXCysg=snPaG{x!FINp

e;XP!lF@xhYJON<#!s?tgp=_Xj~hrm+20L3TA~oE}l_PP=GQ@@n{Z^7!HAy^4xa^UxhGJ}%W{5nkRc@J!pGP_k6-sfA$0qxiCeBfP)7$rjg< z>GFEdyhxG2`LB2kK--4l|9bhus&JP)xJtjMovV^W5$p zK=ne&FO4fA0&qc!>LZ}`Oc2)E<`+oH1}`h<^4U)xH~x$N*1 zXujB2x7#CHMM2nZ_<{aR&(9NzTD*IYMY=i%-gKW4=h~_>5)GNLl~t68z-`Q#yo61S zz?#tEvmY)uhl;;db2H>F&uwnHBY#yquKD>JeY*SGxnp!STe_SmI1O#BWFi)yW?`k8 zKUrYt&+K36IR#E5qnCw>$!i5j!7J+PqX&-`wMX@7K`$uQucFmE@7{#yIzLzw$K9Sy zSE=Xk6QSkl=@bI-M%=G7DH`J|taXASr zVoJ(uNSe8xH?InZ7^V_;_w#2fG*;Uw090C~fX|a9s_Ep@B=;Irjj?`Zo z&NlE^m^9VU!0^~WhvVX_9s$0-msK*Q!*Z$<{$ZZtf=zC-$Mjc;g?CL7e>O-(6@cYv z-aCivS3IBt5drmkC<=kNt*pSnz}!7Wipj#Ms>{QP;3U)TPh?_@{M>lBxJt8S0Ucn8 zoA4jd9#vA3UbdPWnOeRvx+sz!1~FH_8(FJdX4AZy!&2^EFQp`@^D)oF~1S0tRxa896!C&U>t)@R|JN zT%B)sN!Sfw6*o}_cvbGFI=-IevwGJ_MY}cgR8{KzYJV2`^3@0%`b77=h~ zu&71C2N)TG|GDOFxy46>gs=U->M{JZDi| zeQ)+(Fh?2GyXM&?#^pu^#|)jfxVgD`%+Q*0i4o)t6v#=#QuWxO#3Bp;8nD z8YC+!G*Nx_0%mRKbbA20^>I*$W2~myqL!B-lvuh=I)`kDLbsXZ6j*4xt}?&-1QGEo zYzBF;8l2P(F&GJ$jDL`eDay+NLJ_rATd$UY?(d^#?&=m6C!cbRg)1-?XZ^+i5t9t^ zJa8h6ID|5{m98h{_P0#;2s)f?wqguA=t&hil5e6* z^~5pnR2DLqy*{S3Dc$W{HJ1MCFK<4}bv@4gHyg`E-Lo~OgludDy;H}V(@x5W=T%fVb9;>T;n<zq*iSUAzNQKm;)}!h*jlo zp>kha8}rkr>oRktfEp3OaL<1F`!=M90iGNt;a-8nCpEKSrOW8`#nrc%p9{af zAHG9J-l2({83z5P6v)JF7RPhM`?bD%FBvX%d!%Q`?1+1_;-u)tgpz8bm;B$)u5N6E zP>R>cDK~X@PiO3eNH_?VdlS+TsZ>`h zT)phKKT}KtvyfdPEAqOxM@q z>T*z_WMnibT@cT2skPHcQev)+@^kXxi)aKO)h1Ad8XJw%n} zIX_t5t!Yn7$1s)yGwwc{b#ZZ8rvfeF=b>i*`wxc1j*m!gJtBK(XKGoz88eY6R0dmG zo62e*Y%P_Yn0h7oDpAmF4Y_2^#KdF>O(;xMduz=4%s8PpJHsf?%Qwc=CUgvK){1ZF z#_?KGkFK}%3L6>Q&s4`GBusBFXtW@~YE_Q>69pPcJ`Im7DYkH!=H`!;h=5#b6Idd$ zPO0v26BgyK8Ku~JA6se&2?-g$x_D;u8Up}5FoA#oPGTa3obWh{cvP23BS)-SXYD^{ z5mK!u6(KA;2rudw%#3zjX6Ew#?D8j4!IJ^wMi7PUX+%v-bUG{~cc-QqgZ(n{>sPHV zL|<>O6x?p6)-_+P*3$#(FNxK#8K_GxEwo)x+{)G1J3eO4l3hDH&7y*jBhi=A03L~U za05*@WfTy1(pLqeGQ%R19a#B6z|T46bx@JnPkdFA#i>GlA}UFKvD2D8K{#)E-o z=;X~Df2G}N-D~%pf;1M5{+|f?p%3~(m6&{E-DEF3iRLCYHi)gu%kyho+~W6Q;OZDO zBuHzye{dI!EE#&elZAj0d6}pm)}M!|s;Y9y4oo7fKR=j6h(0kXu8uVIb6@rGqf*j2^ptC;1=FK&Wp9Rs=X@aT?dtWFL zin8@#VF6zHy2-hU)XqLKinrdSJ5!smzRSxDHylWskLBn~z;!1pZPSucEEmq~S+9*C zB}p6X@|?a#g{Qo^^epFcb+);W{VJ*p0bs6>g3A|2`GKHwlLDD3rv0B>kk9H}<-Ni& z7)AcDI27nMa?@Siw`wP=u0oPagcWTsM?KcD`Dva$LccdAKa9~{^RS|=3a7E|1aJSbG+qfWMdM^B&DV{ zG&3Ug>CvNAHL-s$?HD+^_Z*QR=H-A%F$PLxA!NfTK;nY&9tarOWAd_q5#;`8=~t;7 zC<*4l(P-sN`84+Zj7^S&$?i8-=cw}nztP@xc{)Qzyg$5iX?{K&pg9J_kO-tJ&^$c+ z;mb8;8yV9$GBTX5`9wtH4r{EuM#d}O&%5JK4GMh|Temh=_KCuw-Hu9?1EpFY^cu%O z7V-G9oGDjMr0-*J1$PGU+z_mM?AxP>UR^*BIZ=lqMjTP`o*aFRZ1Ct_~YkLuG z@=rbZy(l8Pm!8ZQk3sa}qod3<$&u+LhxQ>_&Ph#sIDSv-pESz1UAu$Ts?B6wUobnI z1uoDE(pq!9Wc=A)7jvULIXnWiHdCSXz@_&-S?%O*cLj~il?AYZp78UJ?V7vICU1oV z1QAN;S|6aSBIpQuk-ti1pJ=F-6A4Mxz%Syjcm~hsXV)O*n?J~7QY`3O90 z3=g^pR~I~fv*}eTrpH*V;h7&pL;KE4tA&InNt;o25qqw^NtmgGQ@NIo%!1lV8%%8; z3f4_hc3u?)q-Xo!d6;QuObVbsvMns{pPFx-AFmY_T=q9abT4&lT%3-5wCzv!AD`5L z&=U-h1hg8nRe})sUA0elJ<6TBU(9l99ozzZTRfc+jH^8W&9q3Ea+~dV87Pi?aAGq$ZAc}y-(B-u9K6sQj(YgEl@tfZ8iQa zG}#A8C%J(j09-U+eh11w-=0D|?GIF20pT`qWyCEnz5}O*DW=Q9R!dm=${nM?O!f9nj^tO@)bJVtMBg}xYEuPo%`6*a`yJ(HHTLp(TWq|B( zj0toIFR@$T(c|(8EBea5gu=yLtK0b&s&^{?=5MzXX$T z`(VcHV~kbHulKRhU*ff|LNG=qu-Pn{0IUVX17kx&#QdGe+rm6NNSIOT>moyxubsY- zGFV&GdC-*GcHkS3E*K%?rWEJi2B4XvV^Bi_BC!@FWi-~8oV#+GD(9bU&f2k&&-fOc z`&;@q7m?2qG7J8d7Nf7(HBZDn)>o)bm^lZE%bLI9`&Qas_^xDK!+K{^@4lz1rdAqR zZ<~2Ict3*(rxoPIO(|(4B)GZNj_!eyrq~+Dx4$?4YN#TKS_t1Kc#qe_y31q?uKY{^ z(GZZ!1!xEn@bo&~=j1p@ny72Jr-NU7bolH0TTUKMK1HgyM}Z<72DD%Y$>x%LC6$+i z6Mz2i5d77?Fhf-Hs`PL3?3g_QRC|iw_lk@L_3zzad^~t9!H_%4T0vHshd~J9i;EkT zHdCS|zbbQEF0sSv-NThfn`eGdUs~=)F?*hhd(?moOOdA=Y-yqdt|Vr5Rh9q-qLd7a z49D*!sHs?T47!f>ZW;)?5H#nF+1Rgu4kgrL`Tx@zYDmC?4ogS?~1!3H$hTZF|qpd z=iL&C^Dwj)U$W6PEb17bjhCyL=e@9;$Xp&LM@s;*O@EdmEZ7X!3w7eXRl{aXFHgb& zL!Y^bpA_12EV<`BR1n+vF0FI@8I-1Qv#lj!sse-pO}3zc>a9=MTdR zv^KP$2)kZ%YPyjOlyEOVDnB?{js4{TF;5Bb%a} zKlNSV(?x~FD^fR2P9rFU072Lm*77)Y+)J}S-V^|_{R! zs*4CZ&p+(4iRLLN1hZ1l%a;XeF?K^GYtdU?eMz;$M+Ztw8A!{v&DN_>zka5C<(qpT zK7+LXl%^7NWrnVf=huFpr@aAfq?KdjC%=Y3EVcX^+qom9?%jydAt>0ius)FZ(^3{o z`Nv;1^Nm3vAtVyI|86+=rwOB2ENG4J2Me?-p!xI#*@va0akk`ds#JJnwX+%87-u_KF5 z9xAMdJlfwv%^o{DIRFBY<9fm@ATeHjInmZLiwFAoBqerG;$;)JSb0jopS#BT1~^Fz zhO{<_2E$K>>yvLf-iZOI-D)UxZ`v|VU;i;-#255rSw#JbE1<9uHIwu20Q+-v2w_OsO-0<^B_QEOx+m_ocOp%%rWygmpbiYCi@HQ6iu^M>Utm0h75SIjHi9iFMqJUzhAX~@zH|iZf4%K~aKZ~o(9E0d z<2=5Kwzw)i=nJmg{O?1;A~FXw!kuj->qk7Qb933S&_K113p4tO zO!bhG^fwD-3#m5=xVV@9nO;WxGP%)pFc<+OwYY!2hoxNd`Zr4V`%fJ5dG~++kA*{V z_4?oV=g%L1-oi=blTxMz7RPnJr*)@s3v?}(wXDNz5MP z^QRM%rB@Pc_eyw`uKO;_Qxv2X=pV}U`(uY)#uQEd%GLG_)8Sn@xwbC}dY&A@2_6$< z(uG$E#JIY-u%8JMi9&`LK;D#S2{5)icck2|PTuYDW7*>{FgsW2&III=!=RPpE)%Xs z2BpgzNu7(1Bn0csoGD9 zujiKb(+wn#$&$9DU94fH6=jS`Q{*SC0$j{OeC&wlhO@oUNaw`B6RLwd!2#ZzB%c!! zfO?enK%U-DjT<-@({v53V1{uPgS2rCyCQONAu=oCoq&Vsxe6VJ5>US>ypQEbtyqmL zVT_WMo0tM+42;WTE&Q?y3c}t(^UZj=V_YJ;n9&ucNL(bEq)IW zF=gk1z*dS!c?$A#EId598o6`oHR^paljmU<03-$E#TmT>y+pajs2;y3%tCD|@9jZ> zgHfIBMkYYrCq^F>BC6uI76pgd3>##L&PCb9e)bHfr)p{w8ZSK7pizpw8-2na?ByHK zu?psHNXRF!{UF=}#Bd21-md`2Gh-97VTnY60(Dwt_t?7{7r{4GstIP*a@5UiO;sB5 zYg1BQR2Z&EwD*|^V*|;ZOx%sSZCa2+T?ojw5EXSGcuNd=jr2dN?$?`>bP-zh$j;X6 zUc@1{zUxr=)mAWjr4jQ>$=erDU6?&$2w{7bU$lMSdcRL3Ofzcfbuv=rl7duvC}!4~5XQ?@ z$}4=4NY?u7F?2S)L}<>AfS}d*$b?jR%b`>UgXX;_Z|03S0CZg>%VcVPu<%37V5$3I zv%dG~WI=O#KS=(wuxEBfg2I)>wd)Rti`9Tq(y;xpw90Rxm2C*{Q}r5`XzZv*hJd+g zIM0;cwx$P86T0N0=x$6AhVUOw&TWiK<*Oqr;!1ti3g@F5TWHJ`BQo9FI3jsujM8)tZALY?=>M;mcR zSH}*%#~kdDau}^5pS2v#yt8lU`V!ofWhSSA@jBz=s?d|)1S8X~&({o>9)vb79K4+$ z9KjDC$b}+<43Z?UeQd3bVUb`4NU9%Z+Ij&yCpT_WcUIVe zIFsLIrqmYoBu710eqeNP{a2w&SYfhR2*(PO)B9O6oAHvB<(YFCOlVzT{j=S8O2r9Y zy9Z13iq})8)tUH4qmd^$s>Lr=%-FU3LleayM`S_~^zG%I647fVCDek3f|v6tP)v~) zb&{1`-?)EU8vER)8HhmW88l6-EG!c7FUS#0>DvFn9q zB}jPJ*Sd76-zOt&G)3>NMEkZY1{-U^EK*y9%50-tmi=P1L1o##eOJDrL4#}pBnm{< zbXHcsRvl6)5=4x{v1C63)3T7&rzES0&l?Yt23_e~4!Td7sf;swbe|dLXZU&G_oM{) z`ALVyifB&bl97^*=4;4sr%@aZeVZ#*Tkee)Yx{{{k2%2SeaVv7!>L_wG{Ivqjb#Yt zM)>QlXa{0)O3Lt+SxKN}*A6Krk&sdte=B0ZQIM8i0pwa_g>6%rUuoti)&-4C^gRbK z^*}j;91ky;Q<75091>nGU*F#E+f1)$$m2TxZjv`|*---s97WZ2naVoTJj9#h zMHw_R zExZoMPkxq`sw&k58WSvnbW3LnRKTpZE5-^#ZR7Id%tQvu*GUKnd=^^-O^-4o>qE4* z$r*~vp5^K0d1gxP;7pYDpDYv=)S{?}X_iF01%#s^g0WR#Y%8{ffbE|yQmVv!ss8&^ z)Pp(nm94F<3ap}{X&9@+wLCsiabT)nsnTU7>;(uBo?bC>@sKNRD%(ATm0NQEVl*EGE(MZ-vSKYi_Pt1UG6C29k3eblD; zs$F|?k#xB`fQt_dS@OtAsd7t=3%BBe@)4z3NWavR?kN6v?(=sCwUuDqD$_jv z3nBACa`O_*uR9JqDOq{KaO5*n0U7!j4xN&-me}BKw3cC-RVzQ5zk)F~mJAQQ-5oc= zoHM&GB`FnrK4Sg?dSa9XDc3u)3y1D9YaZ7Y)jA1ceU>Ay*u>XV1#Mg>)Lt>2m%LV# ziRC(o5k^s>E09G}rSvA`#mp}pon+-@6(7dj2AXHsr*+xhND6(B#rrz_~ckX|0Y$R#( zvKB9%g%FQQ{GfYZ97QQKkgh^;)bJEs`$x5qaOmk;tre_pBu@xAm+;K?XSZe`GSuuM zUsdZc;c`?9647X?_ z)xV>D?I|Xgbs8=={YrnaRhT$gBM_y5NwS;pfc~CY)(3?5>V<>vj;#>-FUQkKHq#bw zz~u&|1IqM;vn5^&74;i@J9b*A8$g=!X-9sjboH_{sBZ$?;IJeS^UeMPKAV{XW=_s? z2|+3f_+S!@n0#xg6 zyNx8Kq3cMMv!KkC01s29GHjXXS|z^Wv#= zI6b@+GrH*6Q(M85TbirO+hhv9&QNuTjTL~;cLbJ37wS=6#Qk_-XBOoZQ;105gmMH# z%s4JCM-$JhV6sf) zRHuRXDt9$s4c-D8Yz_6b0pUX{MRj{I0~j&?r+_-kX2sW~t6<7%K$8XuhrWk<1>vg< z+`|oaN4b_|cZ1!m^A~k;vAPPmpvQ9RqtlwPm0WR!;A&j-ww%ffA`~<7lm0ZHHfbn7 zAD6dMLpM+-QXn7c9e!5`+^<;&g@IfnR zBJ%dQsobc%6V#BfLGgSY!xvqknmhmdD^KjI;w}aWfawspA=?!bNoWjRG4dltWerJ5 zRrDaVVtD@oMS0SIw?xg3gB%2{6RE9~b6@LCaDge6<=@#7V%xw?k^yM79#I^5iNO`8 zZad-xs9YnGV2XlIUs*UBkk1~_NGm1C$tbY>)8m)o%mmer_L485cT%buwnxFH)QXdo zyIVsY4J!z8Ys+FdNUZ+hQB^yfr%$6+dP9Zc3|Z<*wuVVNQrRh!+zd}6#W1YS6r|-& zA3D2WL#QZ$x>f|ydD&lFBp>NA zw2F5IRm7~*D!27kJNbu2d5O=)-dz)WDf*Bjz}{j^(KZS?KZX;0UHe7c{OZU$=jrlG zK-CTqd|m2XWau6{G9V<@|FWq`S|Ne$zVOo2lp9ron492+p$`Z&+R!2#l57?}OJ`d* zq!kqscO!Bch5&v3nv(A^l~}dS$nBB91!NQy5BGH41|@D*O)V|yIZi&cPCF^gQlF>< zysgIP8Ya-|Fq4G+qs8LNRc+<(j_os9TWRw^dL&)(dR_#h=23MFzW z)oU+hSXbVs6cu&3P#(E_+teKctPqSeNv#GH;`PC$(RyNI!?f|33DJG{?^P2SO0`$N z?!jw4YClzR{+YQSz(DPa`X^V{xq1D6Fsr;QeDJrX{)!B=i`Lhv;xmD=oH^pAV z=pqG08DKz_hkfn}I{~S+IFoAp2bH(^?PqIqAC7idFFytKDs(g_)WjSxY$h=jRq$nM zLB`1UZR~X@);6E5o+Xm_e50#;D%S^bm%}N1V0f5yVVP)`XSH!SKk~d%BZN6N>EmR+ zN}@RbcEG0~4}wRFpx`yEN2-&00h^nqyqw@%jx?&vxtRs00Pxq1Y+H#_K>3$nq99?@ zo<{`;(LfBXm>~y0Uh_wuKYZ}03}_Y@XR(`fL3}%7QBP9lUHL{6g@Gtgm&Pai$%Y+| z*W(w4n6F}QzIyf%Tl4YS>lN7*Sb2JS0%JPuw%XEG;3Z}?A$gRxVb3xac=jQiz079v zdBPgpUdX+u@b4>4YATZI|Yt>lPz?i)EgJ%RUDBf+8WR_4=bf#a}n^r6+SZN$uZMl^8;U-}{ zhY#?8D(6FyDb$SDoVAI`c)8jXmtCLAKrxU#nJzhao_UpntEP?W;{AB}uX zOHwo~;)u*Yk=4vmPq85#8&iOI%?EH)beT=fhpL;~k6lOmeI;6p8JLxVqYG7BcAjCz zU`A1&LhcDySQpfJo??p#zghbk?&rktlzI8^lm*~@nVD@#ly;-7MdEwc zS;~GKa^aUF@_XAQQ$&yN{^fmqp+9pBSUv1|4N2(qYZ4M}{40YQo3f*4{w5bw5}N|# zEZx)Iypk~54Zfgyh?TvU&XmLd0)xk%fu6NG3i%QdQG}?XStkHmdVQhy?bt)<>eKFE z`RRwhbzv{a%8Ng;Sv#0>c;4Aae+K;YzEj8Qrpc~UZ*7!3KIf6+gQ$A>wc`rDWixu@ zE0}t=I{q=hS^CaJHxf%gJKp%+?BeOFDb zNx{>$wpM`hzAkd{sxrIG444NV6R5lZv8zX~USuRDuDqm7c_&|l`WZY%P*5oQx-~QM z{Hg4#%PSBixWX|CVbuH+!8fIpeDk&_E>2%BDa{7k*YccrLs#!*8N) z%{WS&O!JuCa3F6wj1=CNO|jbVy)Dki$CV%sQfYeZB_p{i8MB`q)(rcyUaW5s3YXgt zm0DvjxsS*ax2)}hplO!jeOEgykh?n;I?x3ADoH$^-*cJV0JvUIzy@$TutASh*cX)! z=fm3pJ56u;Y2k&_0azeWOvoh^aq$niYw`Vy^vYyO=e zr}t^T^#a#haSPtRN2h8Ns*CRg`{qj2YJ6dJtH2IRGB|7ZbQlr_vDUi6G2hO(y_Hej z^Q|0SWI%zz#71ojL>QVRVhGeriucd)l8NZ}OgA4ag;BknnH<5sm-r3+%oEF=cDX|#*DZ_CmIfS-N!SrxHK$GOWH9D-3C|fR$8RReSdZfzi#% zzke1~mAx*0>UTXu%YpHJE0gNUySZ&uk(<|PwWvQoaC7U|KlIvpSlF%ai0{agfthtu zO6uO8p6uznI^I@KD{kq=kj};U%YHY#^8VuaUk{j8~}FqXbEWAD87>-fypO-N>ZWDn$qOp2f; zM8!|JWVba^Dxv1Rqq~f3Y$2f`$(7Fa3EzF9*G4}wJpMsjaeUO0_$AI~EAxB~ExPtF zxOpDof=c(wYG&vg`?!9x-_k;7va+ZuQew#TWyKL2K4g43T2EwndG4Ghkr)xbVq)Lj zP@mP4r-=%cR?^QMOwBE`HM+5UZ*>Z~Y#Py(zs|xzlG!*T?Dn@?tDK;Ubc}}Ew5a6` z)*g9pKl`D>&B+&=Fkz`cX;<)V5%|22)KZh+(72@-CllQhPHlaC+twP2O%s`_$M+*x zfs%7+iPI+eD9K1`_Z=^rO+aSaYVk0k$BK$c?vEgJz0NKu2NU2-ZjU%3Xn+pc54vXl z<|Q`sJ&n@79|fcu3XF|ResC}b1cjyruiZ_q1Q_Qj{HO`izp*Y$S;pnRXA(*qaj!va ze}DfdpY4{ydx6I_hd2FS{kmj`iql|ETH3VGJg>y;9|;#BP6p^)Hang~_2PDgnKB=e z5D}H=`j5N>G6NY+PiAlwA4BK1bWqd$+jXrtpemZ3%g8`WhkeseMXv}AA3N-vMXitC ziMfNt$X%%(rMlm_g2{7Jmewt`qma7+um9AO|7*9S?64;fnz?%?SgX{zaFsxrR(;z+ z`855eR>qI=dRAIf5-rmw6SoTKlFTb*Kp8rABTSrCpz45{}DuO`?2&hPRON&T1 z49$Racjr+Qk(TaehVIT$Qo6gO8M-^}=KFi^yVjk*SvqIV-cNkudBPPHzj-%NXJ5GP ze1f#J8iBU8$LaJy`1893w>r~c{tn*8Kd+;QsXJS{KE z-{*I}q5|_a5*T=|Re^mKCC8Opau$I!kOvjdPW6QEtA&Z~bC{8Ktw+HsURqe#3jx#z zH*TdebLGeMO0d@%-xEBST&s;dj`P&TlEv(E3u#kRb-(`= z?S(D1kyE--2mT7b1(V3uq6sG#@6nqvp`l?21;^@u5pwyh+WNr0>eaG_(%eT;?Dg59 z@vqaXP3JA^mbFh6zB-OJq{GfGposyA>QMT&(`TIso!N1rr64k@K!Zvz7c$(MrI5Q| za`2>c?dEiYhC-@Jv?G~&J)h0uljzRG%faug#61z7W)XPZX-S%9lW6)MkGpuU`J{Ec z;($v&k(AxaV0)s0qxK0tadMS@Vf^g*ze}r3O?LY$f8NwPUx-kYtQ5%u&_trLq!8CP zsTDcv@hptI{D0_yLuyd>uO|abpMc}{ zFib@mFTIQ|oCVg(kDqPEeCR4;zS{V31u8Ba57zq8-jAa3!W5qq{XXX7$B!|GhlevFdON0i=;t0V+eb!5eve+<@J~~`3Eq_#wx4%9y1|Gt zL)Be2v2SRs4|@>U%^u)X5y;TqBO-f8o}$aW?V^Mzo|!=&7O8Ad|DCC!)7!$oZ6#WL zK7uY2%R8nbzuCRmWIC<=pk05&a(rQ(U?;@E@>B#{L+BJk=5(ExH;x0JETMI#;%dHH zP$s#SkA=IXS1`-3R5*XyWt#wG+=h^bt~kylLVLu#8l>{PnH6=R^sRI5=OIi!x9;$1 z25)O)W17T5!}{7pK^UsUxZbFZnTtc42Wo9bV`jx1oS*}4d)6*U-o6!9O%b9huiY*t zBC@1YYu=$3{X|5hW$xQKWhU~{v2Ee1DKl-&C61AaNk>boYhjm;o)|{6gDR$fN4Q;r zc@csW1RYbRrRQ@Pw>D?Rob3Ev1x3iuoQ2p+Ju94M9-PeP52NNz*|m~iQD2BFr$N}v z&&aAWxN^|+V|g0k(~@epMd{L&eQ^IyClM;;^U{yR?YqL3Ynn_t%)%vv{v}~yYQ0_# zB~MAlcqtezK~y^CPXhn^Qni+d$bEk?FA6K@PY$jcGE15|Cz6(d!4X<4*AJPT!yy2v zM}+HJ2W`iIHN*ISfvKtrrplmG`fSl0gmq85h|>$_?QayNf<`I2&m{@FA|7i}l!7IM ze;=La73z%o%Q6?s84@Xa>bliTIJ_aCWBL0C<3%=Bd>i}QXT(KAja}cZbiRFTgjzHb z1)(K6e1;>$*@5vO1f0O9JoB&8H(e_U)_ zUqAN4$KB@+Cfnd9y`{RH^}+whE0+e(>>-WJ~SFol#lXEg)j@8*@~Gy^r8ikGa&_)`+~*IT1AZ^yt<5o6z7g zBPT;N9v%78f!3dOn(3Ju^5p(~RQ~MTI)Qxegx&DGPV?;3UrG99FBV_!`nCJ`_~>*? z{F$@KlJ7pi@LK4(TPy>0_pbO2%lNT!G5V&br&Fx|p3i*8IIKKS6+nlT-?z3NR(KH}KHjFF2hTftMJc7G@$AR^v7AmNY@Dn+ zPoIo~c9c3qpRr6#R{LkRr)S%!opi`r>YT!cUxX!x<%W1>ACd$*k1cLo`c4wBU5jR< zdi68%(O~oV8#X%k^#b9pSQQNoQ2TWoGtI&r6_o8($JCL$E+5gx*KJ!nOKukSxAY`eiJ4KbCIasEPf@1+h24U76WS& zJbpP~SvDPw83kEq_%P#ufXrKN9@)E4I90?3SQOkCZ+NZi?BKdjLJ^8tL%~UQD+-Qb zzrPnl?u4GOG)lm_l~uh?{2bY?*c0<=n ziHt^q920RQ=3q&7}0 zeQ3w6jhT-z(-HKyk;VqBktQ22K1bdeU3PgIPW?gA^~+;%y zd^V>j7Wkp}FdHN_9i1N!cb4MkSTGgJv-j%)EJugkSYw}3R`GGG0*kG5P{QH}Yb9|B zHa#y4`C_OAg;`XfU+ji{g@tu!_U=LPfSqgX2NfsTFdHqE_VBc@uc&v(Q`3)nB_{rn zJ~V@lfs4}=N?x3toVfLH+S=dq3SXogBl~OI`T0=9EqKI>;^HIMvgC|X*u zEZ2&8`;ni++dL}`NB;nB5RO7ynPf#rL}l&>q%s>MX%+1>lTu_ zqS5HA>>OZ_Wi9+vjwjIX-BD^i_T`FH(EhZc4ad=uqvH=(q53B!%3#7RvX}D<>w-od zS`dzY6Me@=yZan}>ua4~)E|FR5afD&;&43aFVvJo({? zu2NvP8``e=mC?nm__LAe4hW?wqii@oHV2wk0nfQJ#JclpSbGLPqwoI0)AShUXk0uz zW109OyC&IVF>TVMnXaAL2z^C#zT8ttwcW77Y{q50kWIyUs(iI_R+S+Zjw-;DxzO#} z`SOpsW~NJ{NJC0VSXGRdoGl=Gij3EedB15l-M3-?>7Lsb+Zu)r53VUX*7fu>Ozg?6 z4D*Sx-L4j5hnjxZ4pOUT!PhZ_IvLre>3PvROU|)ZtyJ%PL_Up{Ub*k=$5h>ekoESZ zl*3d1KroL<`uJl#{=d1KYjfpX_#C3+o~H6mL2GpE-xJOgA8?0w`%kk_c2AW$e2Ck zlZn)gbQMB%KJ2KvGT!~YV($c%EP&warGBM%MOZv}oK2dNdYat_MX}Qq9_J9LL81k-cC+p~-}$0z z|3%Ymlxp(f*pw;#iq0&-K$tmMGjMZP!|0zI-I3LKT@<);auK=iJg}BW-#tQ~*d!aS zEcpbMW1x!9<>-|1>fd-L7pz0^f*l8Rk%)5kENgiEZ=Dx|X0e{qL&HNB#6wt5u+uHg z7yIN{(ItfsOEBplK=})~dB&r`Y{o||IMXk+I$TG6^ z!u!r+6?ruiU8540tuon74Y1Dp2#eSPUI{$@lv=K*5km)AG4D0XeX84mOB>Ckmm#?d zJ9VRZ1OMlnkBTj^!onJf6LI03^X1qbST~aouqmx$g z>Pll{qXFwwKX#swO?RxcUQ;A$iT*EuGBdv8PWD=F}qH^ z1Lvh-uo{?`(D$|8_vu=I!$WxrY$r~K`>A}!vwyA76n zrIXt*$4A%iJGh7@hSHhNr)2$SQZmwJk}>4uN6a9lS&fY!eV3sGne&>P6k2WC<~3o9k&ld)!-0`^20u(4NAx@MFMv|-D6HyM5fHeH#XX@pr%kU`QEn}1D zX)*yknf9Lc=k*CxVZ950+<#6>@+d&mB>v8=`PHicG{;F_v3-|&ja_5Z!uhmpfA@1d z_qZ$b+=aW|AIB4=eljwGIm1Y%%}I91SZ;fDAdC?UL=dW8IG&8Ogbc6a6_F=;c_J9@ zssVavC6EUmoYmmAz?$*u>BS#?3HzE*BuM6jsW*QA%D8{|^!~kodOC?E$X{$(Jz~`g z4s-wg?v2y)^~1^IB)MeFWv5N~&EF+9mr*HjK8lu8l=SoU7w{*Mci|eCvJgfk&Pwc2 zXxE?9+0a24W&h})Ay;lR{ImMA;g-S8t_eeXL0Lk~G`;(|NB!s^sj=C0eMVS; zB;_Ce-kj3)m(SR~P470CoJb4UKIdNkm}Ayd+k9o1P1E6O95&thgK( zs!JMEw7NRCh6d4t7bM~$EzSLCvwbu zvRf@CGSPm*GSj4^a=|DoHNL=41-^TH{{)T()^B6RqyHh zo~c6x?6kahF5NkGcQB@}m9$xxGJu=FQULvf`F>yy}dGot>T104qEDQ1bwl`{u^|0i&!e>eAd| zNl8g+^#O{nBZ@Oh>W`rh&OHvVR(~!nEw!~$MY%TLO5tPDL7iJpAB7DJoH^!Clv`=T zv^M6@GMO(DmEuPkU%2DgEIuuzwohVq8v}4yZsqrhFCD=s##uEGtPOQV5&RwVeD&I& zKDAO)Q|D<4j9v|9{uAuc_!v9ace@_Am^-Ao-PYz$JGl}QH=r(=pu7oW$;#oB zZwWn!SISaGyyKH274YC7DVL@7UJumz79kPo(r07g^B!r_;jY-rtL+QK`sH0=HwRO} z{WjY)KcE*ZCCG1I+0C{Ys1_y`5fPbKO8I5O=v#U)N~lQm>{2OAQL$dfV|9OXcaUIv z24lFiV&ke98k@)n)O$26YMQZw%7I~A6lkg$rqjmI z5wBRzhP`h-UUM)RLlk48*1Nl9R{b%TR~=;M!83L1{e4K%eacy1OlPZ^-hEX{%FI5F z%6M~LOUrKOE1@pfl}$Sbiwh=96(#7+pT?;S>t4pN+i~2y@~7$tOo`Q* zOj=wVoKq-!5;h9@w`R*1kt2@u7VD!NeWS{XFlli~{52JpruN=ArxSI=u+JSZsu=6bDkm)L`R81bQ2n-0D(qc~ zIt-E>D{(4To1OxY0yZgcCrnnMnwpL798^plAJIh?vsHoUj&irJsFyWeK4LA*vvLn> zRJ#!$gY%1~f21(xNnubRP&yBHJkIn0Hdh1p2S^zqWmh@c`! zRwBT(Xt=CcU!SUWV`IwQTl#gtkwaUnq28f_@?)(9imq%Fy%?&2^GbTaBYQ(me}y@);1YSNRM zmU8PPwv()c-mzSLuysuO7(#WtNJ9~F6D0%G582&OPsBH52Lwv`*KQ>sh&D(=?=|gDD)9%8Si(qcV+(usq9(S|AinNAsTDd)iyHxvTjnoJ8lyM-y(jG_MZ2 zbHAVd^)`o5s~Q)Z{~F5k8MS3k_>$TP4hqF9*Y#{KQ?*Ucpr^}512cyiX|T&0l?Zc@ zPpw6nx!E~hycwKZtha{1OtMxw{Hr=JEVp?P7wK7$7>*i3)?inINp{xWhVIwAclT1he9y~ef!!W?fok~Qhrab%P3xV|8sJtGv-;| znlb~`wo1r)QqOJc7$j+cdOz^+unybOg%UYLN?NsUE=^x?X>~F-L3QsY?mj|YUrfP~ z&$r3euILaV^XOf8#9QR;{c(FM5d_t@Z%&s3JyF1*@{kAO>6ktAf8V!s!h4^D}+mpuQyCvCt8EI*O-yYnOw;bo@mPryk{hh7N1nbYHvweWPSDj`eI zs&V2KU>R@Fy*ql8-|}%FsO&g^t8sT-vP&8mAuMPwDMp}**S4F3#1<8dB$J#MV@EQw zW7L2%udyQO)Sl)k`R4OO?7r_0VEnCb!Ap$BX|d>P z<1;|Wy>9<;7>tjPmgD;DxUR--y72m+v)H0Hyd?p#4Ffj_sfE;m=s&Q8RuL|CdVa~~?8fqhJ|^NzSRi*+ihnpXZk z@@aCt8xHB4zxSkrXmCv45-wxjD-fdU(bj#iNf&NV9oo(^{~}CraKMNAhY-2%j{#!g zhuWV#ui{GX@Kjr@X(~~^Ml>zbg6~wO*zt{U!4kXz< z*E9$~-E=tg$!&D>w0aoS^z`(MG@&JFzy;dhC?zRr@|8~@`O-e_tp+87zY2B5y|d~A z<-U(O`~?5kdvhy=Q~vjx6Ae!;FQaLmC^~}PiEXEe^A>Y2p|7E-S6>9C9~^8l5?UM% zi~|nab8kc0OTZceC=oc5%S-#RiKw-?Q+f`D#HP8SV)k>SYBV@##ZY`wsO!JKv6;P| zCl_~wJ?pPLYdIX9Vg#H;Mr9dM7wlgj*)JS@yY!XojpNi-625F6r<;4;zxQ~LAY>b% z{z#%dvj^^ZG{8m2>)SaulxxmW1@#6q=6~vZ8t@Z&3$nhvTWW8otQ?U~4^Y(Y#fj9Q zx;hUZ`y0pi|2}4Tur{b0oGbjosaJS7R7_00b2edNN&vV{j7wtCJzy^W9N;yHOGwhW zWH&qv!^6XK@wjTlBNlE-H(|W2$jEs2mi8#bQ>=+x9|>@UQ)MgdnF~W=Pzc;>hs3o7 zxH&d|hq5u|$3<3C^lI0+?rq?CDyXO^Dl4Qiz0rLxrD6$~jk1VL4tzX3Fq_+Is)Fgw z8`qP64eX&&eAZ|l0o0dAzZVGZJ@D<%^x`)+r?W@fKR*OtKADDtft;5L7!Ra0rIajG zG~7g0oW@HHWbB3&D{#Ocq(niTT^-1QpeSTbQh}_iQbP&Jy<0brtG9qdgVHSCS!oU0 zz{GsKUFA;7S#YLaK9ctmb^|6A%j&>&E zsQUANyMOe-8ZkR()9Jk8aDE?d&nq{Nk70;vG#FS1aD(p7&RQpE@aBaXTD6B87uHl) z5Oa9kVo;JDGwg2NJa^^0I_Vd#b+7=tao}pO#r;<6wi)19(YD>}YyB$^kj(qZI(-E) z`x5#nP(>~f-~wqW9ObAZB-D2TKhpvacp}}_t5xGUYrpaLPt${==kbnbjgGW00jG#`=kBe zgYlh9j0tSRGZhVd(Ye{FEP5uW9+HWOYC=xvFyB-vrp9G&VpI^U`HG)`pB>nMg{eKA zD|q<0JmZrCFZ=@nl2Ln+I29e4sA;W9!c|N!L|(=x%vcUfZ=YC+|L3R@mjz$PW#?vd zTF*Bp4HQ`JW2##Fh>4a`nVDtrj*b7k>h9=~kdWbG@6(riS3ok60P#Q!d~f|9g-0dn z8k!mz>K?W4O|l#3?6s7c&BQd#{k+0@{3sIzI_uOBgMW)d>}Tr(HRFD4s!rG2YA9`Z zVj|HU!$x0ALka#+5-q|_)=y)+IttzHz z;hz-zTDZ4!#nTQl&MoY`if&uL{3vN~*%fi&(Lf!zxJQx6{6qjQA}h;ysjQ$970tA8 z`NXv#WvSLLbnYmqvq!tmYi=zcp!Hy(=KJwU>@z+}I_YLbCO04foUuOT8rn0;ey&&jdOHFB!J zg38zhaXHiU=l!K5WdgC9{VDKz4EYc>-6vQM*$>YogaUfy#HAG!6`9#sq7-fNFGz0>X=H&zfXF}&TE3}Ock?frtNzhS zrA{gq+;jzI6f6G%^A`PAVj5|>tcQo7d%8fR-7*>zrM(~7gh#HY zt#D91#&bAnRi6-4%kkP|N?jmH&cXXJmb*@sYOBzrb6Y2?a#=DWvFvVg=km+H1WJsw z#}06k&6#OlEaD-t;DW zN1sI7ww!qnm*KaRO3a|O4QBHR7pF;|sk*xKOvY-#y$(EVB_=A9wv^3@2AOLCcx>=Z zEWJf5oL742X_e{WaoGkaundT)_!sEUm5iB`6k&W75T<)g=4ilvwDphgBCIWZ3^G9g zx)H_RF3d@LYgn(Wpl;c(SRki}PfqS0F zcIn%bj86iRhZnr!Psf18@p7b|VD`+-i*<=R;8@r_b_(xY*bg{c3~gA?I=f)nJEqN$`h* zXk8O1WwakL+$$Iyc|KNHe^3n`^qK>hE(E=n?O+ZGX<=#E#i=!Mw0L*-^O(=4cv!|X|i!~ZJ#k5gMv@t7MRmf^E}PGKH;CEDTC0gLt{v2 zs!(jMkriU9v*uo{Anp^@!D6_%Q6B-fF64eH!+R}e9`e&4fk#M8n3a5@OUIV~We=k2@z3O#Cz}Dq5y~g?3rn?N5BaaH~q3dC5 zNc}yYjLs2@TBm&h64A5N*lZS(YoMnAp^*Q`&9C_;*bh@8QXF;E`plr|l1GDjVa2nX zLjd^)v|3h-v4zP83IveC1dLiSsWlXL5yh52)I+DkPqd@4^u`yu7h+z29b7mcUw=4eQl$wI{&>Ne6HU?SLW}}iOe5^yCy>HJI zn*fm1l7N#TxhIIvzoDSiuQOGl!dkp#_~k9wj{=|*{n`FsMSq*97`tMxzELm6Vvh3H zYxiAt{nfoZKSm>wv?j`*REIG_>dMOQ!OK_I(Hk>$c5T{~lgIoR2b-d*X6Rfe%h~0f z-F-lHyVayBty6^>5#qjBf_et%W+j=au8Ty91$9Q}6Ofu>NsB zXckCHn%<_>>onMcZ5YV9{ijc!U@NJrYQA?j)2HHHM*I%!q6f58n(f9-A(!*@F>!5V z2vB?g841vW(Zp$MXsqpyX{teeXSG46{2^4))>F#L3Rk*am2fyb#`!h?m9xchh)5kb zx)6T)4f&yX=v9-TmlGjKfTaSF4KP*X^ z(DbxABusFQ5yOY5!fu%Bpa|YN0IR@ax--zQJ2bV!+Nd%g z^4zr;FH;B?J%VPDX69fU(#^ool#eio)pRl^JLS$|s_OLibpge(J6S?RLQXzHf&#}# z($o~)Zm){&>>X^av>yD3JJ53@fEh&gXgjUXr0=Hx@Sjo7JMxdYgXcLWSXsub$-14n zs{|ofb&XUw-L=xp&b#>UWR=U4Typ?M4#kk*IKL4Ii1)?C<=A+_r>m2RP_;F#f~u0T zGTydDz+TA`rqPw-@my2e%v^;}UM(^xn#L?C>2Dt0%Wpr|3XDf~c6*># z#yHw+owKo?IqVy{aQ9#X`V)X%->(PV&Jk9p)zowLtW0H_>R04wrZ)>hZt!0=z% z7f|+gMV=dntdxub5Eqw@oa*GC9;D7TV^<96z<*iQco*~W02}M+#T}#$7Z5=$4wY4% zd_t1Rp}zd9yYK*b-&MW&-fgFqV9Wc%yCHQz_$mF=ZbMrLfX5~F_k=S5*yzV&Z){?g zr4;*yg$9A#n$0OLsK}4=3qqN}+vcawU(?$CGF;Os7^>*?+p+`)EGfy;%G4{2_q!I!-@VClb4rgV5CQ{auSLYYzC@FLX!B; zmWEfcbb6qFl$7dB*)x{#OIbBF(*rtmJVLy6aN16JuTI_X$@8Z2T+xX+N8&V)i9*SH zcM_Z?)tx%zyL2MJnEZ08&OH4@pN-Q%DGO8a&2y$J8>FP zAVB@xmq~Qfrd@s8s&dyxEs^4rDmN18`h+HicY63qiHG3~QEJlA(2V@cfopJ0UZV7M zpZACM7dz75cmAY4&Mjbl@%|MQZ0%(|)O^pVili%OW2+|n|*x_Eh(LHPE)HzK8s!>VmY6D?yWtHSp?I+`DUfdW0lv>5bLaqUd zim;0dyioh-Ydc=Aq|<`*{OkfGW#MaTCJHJjb9w(nFtjnH7UZ? zWr1Bpvp;%BCIxm5j&?jaofe->yfE^J?jaE{6-T6|ib6qHhu32V@p&Qb(OS_7|DZ>{ zd!Q3cm6OzP`z$}UUU}uf&Zt!#86PjG-|_OD!p(c_@*Zh1uT181=dc_wF{GeCrVNYJ zrBF~WXXnsv&(tX*={h<)LBUj#U?cq}^{%_Qso75>hBM3XLY)tWD(_A?jlBfq(%!4y z1!o`A*7jk8ilNhEol^YX_>oFTY~{h>+Nz0p>`P)N8`IPCoco)T6~F(08+a1tEle&c zCAF@uiA^@-3nxg9?v)akH1f^J<-82?_cs(Ji}WUcr4FAL&DN4hAaQUk3s)KhXgp9u z+<)LpLql6(Jy;i5F4hJV4x>X2(UplMYO01WSJ4&bgBJjaFVwgQfD%$_hpAdPhCVz2 zk9t)K${$O^q~zXxx8?Bg@bku%{35B}a%bE5O!OB921cO%X?GCO4ebZjk6R57xWTOj z0hz3Fz$@yLE3f<s_49Z|@|))ccRkl}!g8fG64l{=2TBwrs^EzVFLd#x z=2fNrG*noMWG~D1B?tLn$EV*iETL$x9#9DUz*~mtPKs!h}w>T-QTV;G+u$(5#O`sswvm>eP=pa5*yZnT$Hm2lL}@c=mr@xSu7Il|MsFaq z`TT?Bw;nU_d)~U52KX!8-_ACwclk3XVzWt!*HS~ zPE-nbT=3ypC@Kzi(3^y|FEqs0y1-3bhPqr)H$}y_vk}-QX{lauaY@&sJsDl0xmION z0Z8O zke_q2QEiT`YBsF>>+`{ zP|&i?_qC!@qXMb7m#D$ZMi{!-$KFjyHgo6}}r1$cI_=MXmKmFO z)AMPn2>}Gw;IuB_X*!Ga0cdD!{6voH!d)m2b-a_kU3*NEjn-m}#y2rCnnD5~z`Z_P zt2|;e@1o|(;h{;f#*6MT$Fg#hnj}CvI$$|+#90~a%)!6@xpc2E>jx^d+?}6U84+dk z^LjNQ5ZIu(IfLN0Bz-{ze$NPBNe)<6%AQ{Y&D;7KXUWcI4%J)0dZA7D&ENjqxd32p z>*a-lcv2uP9y&Nm2V!N(FB({Dw&&;Ah=4*gHlAp<>hd)YXOh2%cmKV~3KK9f575{H ztx6ISXIs)Q`MWBF03cBZ#&#(yE59P5;#?iAHP>*X^vU>`O%KkR%3Oqs*UuX)>;HbT z@c$p!yb{3X?{7o0w|&7M{6oEdN1v9JrKTx0IMlpKHWm2$AG4t$1vmqR7VwgghQ9!e zVl`Hei8V3F8jQ{WU{Topu%FYeMC#!@#BG8Kg%v&K@U<0?#f7H8Nu+H z8m6p_2u8TN-lj%UHw+CA(|5}V+A9pppO!;}0G@)tI{-C|1MAwF3Bda#e)!3WcXV}d zarh05Pd6Us9swW#RI}p}o7< z;7R05LE;4m`q2IKY#N|Es{!WcBVfY$*&TEG!X(?7fkSrNmF-T{Z%*+7Dg1IMS*O+C zagvR-v?HM$1QS40A(&22>+aKejmoP|mO3JkKOXRQZSx+Z4qO3KVgjC+l0w{htq)tH zN3gsUJUDoKu(U=&KFg6}cz%8!#66C(>`K?}w7td~-%*TF5Q!?|)e;+lGFGJ4Il-@p zFt?-mAfSylG%yN)C_#x^VKR~RO{$ZVA>Nj@ky0&sxyz?zQFc$bbRzQ%f^YhfH#2bzXXhWOrzyE2c4o8{aY|Unn_6j)sYS zK)Sz6AHL}whh+b|dnX-u`1o7=E+3UexRcwiAy)GU#aJe_^HYlP zp~|Q{tRY;^AS%!U`eJQ(?oAEf73%@gd?DO_N1naW0^3AfCEL^WN}p|BlHBgU0V@G; ze*eA?kfn@uMnqpz#kUR~0rj}M!>kcFDG8`+MNsHXf^Y&V|FZ%_c4}d!lpn;vPtHcg z)w{4;dMuO*$V^V~qMg8`QYkqA?}8XkQPlL;VS%Ar1_*rdaIe5TAy5EsUwrE!Jr!0K zSqdO&f(#H#B)4O?LIYK9>iwItoD`gBI_IMS9iRb9|8XqnG2bZIz(TCba6&tJXqczL zs_B+~2%G={yO5XT@t<(udIKWC3>unjQ{aLGtk~&ko9!#?HXxWdH~r)$x_lhIvWhml zb|+K#zyJsm0LfShpi)fr+lt8_aAX-X#6?E;y3Z{TzrrD!pPB~xizPxA@yM+Nf=C`htZVoCK@yih)C(h%HE)<$QC}Pgl5F5Dg_Ws_y zIH1VQhC@z4(X!`_0KO3eBS7>A5V0B?t~574cBATD*#)U4*6+asw+D@_`5RYL`+Kgy z;cV!1H!&t6EGd*`4G;c~0uU!`n!@nzj2AK+Zs*HE<-)YIn40kjJLRT{4 zlFPd%0D0?>l!UH18TjewCY%5M^7gZo03X1#7^ze=Bn3;m>&wIBV@?;wfWj7t+WO(~ zkdlJJ%;`$#RNVzS9L2>tbgh?NP*wF_fF-t<3FKeHZEM&1Nios3<&}NVWO8r}(Tv#H z?=OCzUSh+#R))dy@&zEY4mB4O6AgD{=aS+)40iF(mbG4&*>e`$oFZJFhoTS8HnxUn z9Q$8%1kyYR>CI$55V1Q|R@Tsv3`on64M=0izPKGIB~D9E&(7Y5v+n#eNRckl6ZjhN z8bm)ldjF2yoAbvDX7@sSJG*ej0bJZg8WuXQ365vA-}W{V5;CAgc1I57@Q(TkyPYw0 z|4VQO8ROWLH$1U9lp_JXwz~gx*&i@NfRvR!3yqGJla-K?kuoMDbW&Z$CHRy2JVLOtZ|z0!^V$gsF)%5)IaBAr?;!tripOn6&W2mE0J&K?w84XkZ9H#a5f!%f}rd(>h( z6^DmMV-_2|zTb)p6LdXU96!|Dy1M-aL{je(FwGsrI-tv@QgKGlp*{C5hGWL=E$ir$ ze|s)`?r3khvao&)W>&CM(DGPqg*kRj_0K}s21J$qhbs*(!$tt={Z`9aMCb6ghQYn@ zZQ9e#nXEq=)4Qdrz{DC0VI4S{#+LEW@HT2kAAme{dV0!t78(9+vzeu-a)U=4Gu02s zL&Mv=CWp)67Fu^?o;<;4KLM}#C5OjvATAS$w_R$TwvV<(B8-`@iHKIQHYPnGKtpw* zq?J<|k{D$QLtk2(C=~*oDL}FK-PY0~5#zw*5XWA!4X#nNgf*YQ^=f-S5wU4 zHS{aL78~HUosW1HL$?Q;9OuWZ$G?r#2<`z{{-x(XUkpy%csOsh^s9Z#CG&yti@V;H zo3{Ed@GNyvGw{~Hm-O}fy8v-kS}9PB0|bmQb_(#ohCn!?{z08?y$kJV%u$i{;h|;| zpd3i#Eu&#l^qOs$1&+Mio;CeftT-U5gXu*JD`d^DB=8KEZaP>(7V(3VujrJE)rM~O zE9a^8>ZkYD`w^F2WNpe-qX0=~@|@|nk|h-M);xc3ufg{GHX`WN961?@Wj>%95eFZP zSD`hE()jnh%(`RnSS4|VIIS`}oPD1X??bxf=<17qeLuk@Awkb8{}-jJxL|t?do87B zOU-7)1@`u3;$>4V*A~3a$$fLLx+mBvQk@3zs~$tvMZmMCIVHS)69}c(*U?4Mrw6gU zsS7J(XoHKaH;no50e2z@I`tB(XKRjEp~dQ4!>w%}efO9eM>H>^XGq5eZ$lU{H2=He zamdZ2-VgcSYhV`Um%w}*>p3C-ya&|)PLG+~G{sGwUsVSyON5@M=!QnZ% zCG6oP1X>}wQl){)oUqQ`zkf9eYNID-~}`MBL=N zK|S-}`-si+=ilnI>l)5R`9XE2tQ;Au)TT_n3oHx@^stv*V}CykvDX$A*#IY>!>^^K z`}z5%n-UV)KvCJPQ=vOqL&4rMHFN@cX6uRKWC8Q2ZYXlaj;*u=`0jLsw68|hl}tu~ zB!xOoZ~g@c)oY{WdOz&S*||=O&UK$-s$L}ZE&u@*5#bsrA}HP=v}#=iIR$K&($qJo zcu?D8WY2IVi%nAsjg9u^rUN3%940`nCvCw^R{bCmhaCzq`+r{iB7!$IQIXu5nz+!f zke5J+SoRsvA7+~pb`o^ut=-OeSEe_OO`0s1+u75D;79yK;&K2Aj0QDPv7Q2uJY8=f zR#y#?5_j7_x&%Vg$^sody@OOko}CbL(Eu9PQfq}HyfC*{`; zS?d!rfL$A<0f$4$u;%Ig`(>F9&^;iCxRE_-*Czgor~?(?MsneJS$uE?2uO+al&zwX z>K|VRw?7h#7zWcq#KmXI_90-%l&n=( z{FO-oN;8Fy=U6kZ5vCV9&ZK-Ed(sLDU*{9a+fwTyWN%AW)^c8pLa`qVxBF;|2g>7} z2E%By0cJ>@*mo?$dt9g~Acy}WHyw?|dLV6m8&e<*hD@%$sj6k z<}LT`MC`SnOe$8A!09<6w-O+8-d1K$at38&oMa2KIC627Cy(S1TuBx>(Yq%?cA1ckSMPbnq{ z$<8O4XCC^v-*mpO=nrTQ*3ge%{AgP+)^sMRbLy;@d#c)LXZFVP@(#d{6b|}}%_b#n zCInU zE|RI7ZA{yxfo=-81fM0M_KjrCwgC5#qcnSk5A7HUfgIa33vwzfXh0vHPzmy*Gh;LgV<6(4J;{-|8-+|{2C~&9GLs$r9 zU5B9IA!->$wc_1GsBdD)p6-?MaD`Ak0Qfn8@?+@z%F;eHaq5$~LS9nxA4j>11<7Fd8LMfH^rJs?Fu=yk0Wu3T2|EzS!@R;4y!2yv@B_E9B4sqGEO8dk8&25+|$q z*Q-cXm;QKTB2-`_Jk<3;wcRd#Kjt%Nz`0J+w^)*x>GFR!=HvgmB%VY*XAU~v6|rm? z+s*?3)2>=;YkMh5(kAQ3YiLX$4a2{r$OESWlINKG&x$j}& zT1`UY2EY-LTu^a5wzx_{N?MA0|3WKk;>;#P`1?zSNV*&3 zWv03PGE`OG<~_rrJvMA9x-c(~-MRG>lGmLvn%BZ6dA)KPT?t;u3ywuHfvjFoH0aL@ z9L%MszIo|7-2O+Uj0cQf0I%2Y3elw2{^07CqLt#;ulIby3_3qkQI=f}U9{ zuWde;t^-#nEFz)9;uc9evIiSXyFK1L#JL~Os{84)_!xSB6lnQwn9NzENHS>wUI74F zSl_ZDmG@ggQwUUGlK0^>kIs-bfHnoMFPF<}W9>wt38QE|g3v4{2Q=L;f_x^^rF=my zdt76}MEk-qUchSc8Mkg5Q9DgG=oK5Ap;`8K^kFX`<=h3l#if<5n7>0UpdHh=f8Q70 zU^_avzVtKIP_@lnz*7kFeX`mz1jc4)kM_kL?=R8_GI?DMMC-_>{t0ZpRKI4_mQ%~XUAtY zqI5Je$rtqi9Jiy;7*|y+wDLAv|EPO>0L*>}3IuMv_ZI4ouoLL%4)6qctASl(h+?zmy+7&3It^Zxw!?Rh?v@fw z?b2G-9W300^cy7lzxEcF1jqUGEb)2MnO{(>I^osrKS1V*6XZ-A78~2`0g!71YIB19 z`eJj@t~!yog4=$#6{j#7?n6?JIy#h->RlZ#PtW3knCRb1%8mty$4Y%B5`cm`m}Up2 zBKXVlfc`1E>3JYV?zMB~D}R4~o{6|L9RQC8fRKY?oE$5L(*b~a@@!<|w{p2zyM=;f z39dwYHb!ugvZ|^QO#Z$5u4pYML-UW1`A=l9f*$KweO}l-0a^I~(ZzZA3Of^MM_i*) zptQV;^*kEj4%yu~;AG)U*jDN+}Dv2g4t@j5usBgn>a}36{s! zRl3)k7uUT=vT51!qBOxKBH*d1#L1^3AQCSg{H3URgPHhO*n#i~brKZJ)PO=_+TT_0 zaGIlPL3zRj{lB&kA5dtZ?Nc>Lo}>ZHmKd+5LcW_tU5Yn3(10Sda3d(DaQKBbQ(@0UnMj>>@SY&-;$hx+;TAH_`Y|LBVE^R-J9{0*w+0wEZ<@ z*P2U}08+RGWS3VXr4_hXtnIA@WQ9o`Ogee9PC~vzuYwJXw_agao-7pS7+aPbujdd~ zlGPm@qHHG5i0H5)=@$4YN2(4SZ%hZSL z7khKBh32w?j$1gaXO5VR^nDOVJw2Mw*T8*bTnc?O#Y6dz#+O$5ASq^IQ;Mk`@m~TO z5tKrXwr7PLWW%M=(;;eR3+YsUaH&t44@UODpL)qaOky(ug~q@gAO}H!=OY|UHc{FE zgVlP@mKGJ6{^ptP2q9+7N%Gp?9GR4>71xismIBPWoPlM{a6pFva3|N7&SwC<`21`Rkbd%~>sC%wFzJVo z5?q&#D?j5aceEgGdt*wQGS}g+B%>;vE|%0P;O7EZH=1knbd~w%y5XFg=9HA+sgtDx z7ZBQ>pPuVgxycN>`VX7prPTnZ!2yxeq?RrpUrCC$wE>x3N-mF2Sr??Z@;kH6n}zLz zfF0bE(0rJqcPxxy($-D`vayoPjy9S>JqjkK(FrhOk~F1abEv!5<$AZPXVhA&!gym) z&!nirt79Q? zgm+>e;8eVMoEQ&3nLb5`RNq$mjwBMs*H6VNdRbZ?Y9?jL&mmqr=-pjXzw26SxHC5~ zL<4c!I%Um-+<24yA#%KT|7iPX7+-_%(S{~KaXYSADKJgV%zrsE*bnV{bvkTsPBxTh zF*lTz9Uj+9ZnDbituo&zMN@Qz;UNy*Oc5qQ1v?7%$k|H;k6ax*`oH5lqWRcsFsDevT;s1)!!|-)pm`FJg@W>Y^B7;x@1T-{ zvzU_%)+=gFj-sQZllFGIN*j4f(iU?*KF~W(eaVK?tl@BL!pa?xs#dFkJedhg6ml2;Qn)xQ*4C7O;Yv+M9YrCjqG=WzW3&uf+NP%oJsVlb-J6mx->4@65sV|P zKs?frU4!MNEuRb=!WRng!P1^zv>q#9-6ijkBenaU>TE{5B`bcWD3#`zG z+UsWBPCk~ia)do`3cVg>`9A6bppR~MOSY!=aDF@1{CW29*U-p2X-K--wj-5Y=r~de znDrhi^yjJbmigbe9ZCj|w}4#xFf7hZcgeL}Skzka!?nX&s@w9?l4M&eTbQD5jK24I zo!bJ@BYb}Mfw&N6rg&~A&R=6p%*2U1t0%(h?|51FP1@CfWh&I_7);i`d(-q? zkFbCe)tH(5*$cR75KjyD8Xx68@9KQ@{beYkdN5mu0ivcB=i*WU8f6y-FTs|GIPR|r zL2Ii8eM}C@df9uQ|2f;M^^dl9+bXbm|8Q?%(0W)r90vhfO zjm`h+^{`%CUJ~FNS9R2$UmZVSkSw}8kt|tQ*As84a@8DEom}B`S zG7?{nF`*iuF)`y`0UR-2iggG%=9W3btW~g?^}^$F0zfp5ke{j_EwmWPKkzee1=Vgu zt%v>UC*1`J?EH^rGO9LV3AFqbW7gGxvS>L{XA2KO2>yv>y*mtTsPt}XaVgcZ&r5PZ z5!^(M?4|=D3E;8}Bu(kp7N(4}46E6NjDigQo0Trvu|e*m{0=CD+yivw0M&n%IF6`? zsA8QS0BY8Vct}(6%{}@tHOi9CA`(QBl$Z#j*9pKcn81XC%R^cL`^wF^-gB0NmvjDP z>q1xpu?{NwKrAJhRL0)PlD1>+z1zh_tAc9W$$=5Uk{yrA$<5X@nmYa(75c~VbS2W) z_a3L=D?rNk2=E1PgcdfNwi^Scr)@K5CT1q4x1Q0z^{2gKp#S63Cx0p0vzwo*RZZKQ zk8)S0PY!l>8A~~AE@wWuI>%(E39Uh5(FHTat}DWx#Z&QBBRj*%OYprccK2>jsu>&n zpzqH9M5@?KQ<+PL-)2APESIqG(r~{j*_s)R>OK@ijG_T0I9|@n-^Hck@38Rk(+a{x zJPsqzpH0yk(9zQ}Ch*OjFx3`l*KV&+ZK{ojq_mi$Xl;N{tc^(Hl}*}7S#{3>9$q#T zyH-vNG0#V*!+N~5v`$CO#eu4yJ~G$6L_;I+{zPp%9&;ZF?8Lib)&cK=?p`=emEXjy zT044wRD8eP`bRBIeB`fG8ifW+d2qm`D0%=71LaM}v1rLY9Wk33O0 z`SDR6W3A(+(*f1jpOfbv4;1P)f;K32_Fv8pMUL4!tqSmPS1J0a;X@zM*U(QnFKO*g zDGg$L&%<=>xo%BeUCINu>!7{I&f7jSL9()LQhRCSiHYXb)x!iBKD{^-6KKc>DoRQV z)dn!-CY2qN>)3Vb@O8{3RwBeZAgf2<~2}( zQwk?BWaty>8lW<^XVYjgObu>apKNff5CH-H%PSkxDOEWTXUv05fCh?0K*v##w6;i| zI-NV^yP`Y6nR{hXoH6zGz#A>V1zpowmIQ?+M+OZ3n}zWe$hF#aq_=Of()xJ&3Y~xS z)0O{Q-6Stgo`ssXv0Wk3^V69d6M<^Y&vynNq&!;~DR2!VYY3Fd@zUwQwjfM6sV6;H z)NoR*NdByw>f6tJ6mV#t!@jlPex^A*jZP(;rUZ8?Dz&~|@{SIV$~Q2mvz%EhCvzqj zS+KA)`e3G@0%=H@wYMq^b9jqdx6!;@T3I1m-%zbA+ch2jbk&F*n3I84U)IXCFu^Aq z3fO|W1OzYyn|JLMih1=sze66Oz(){;Z-$etGgP4O&A++ziFlYHt5=hh)S#B9Uga9M zoj4;3Q{udA}4_?qm$+kzl8H5!63-Ee?4GsqM}MY<-DcV9|LfE#)ca8NNV& z)V1KNFoqo(l#yg;nYFVvC(+^SE57&4?AeB zne*Z1kTUyk7_&zg@9G+%!!yI@o|h!x=-j$_wC}S0=MXiWu$zsB(-hlkjWK06aoOIQ zn6Te%ZD`3fmL;bi>L}}e>*U0Jfc4Nx2Ap8Fh~*qwhMkETiKk6U<$qr1)$+%GC{HJQ zbPNp3d!@(u@%l`ZUib>mRO&HmW}VPkDBX>^?Za;jV^&rJWeOwoFF{>V-f$rVaV8)j zu#(B$S4{XPG`}@_Kl)GDcB>(j=v$X+^sM= zP$ksAr=M2mfJh7cy6;! z;hsN_y!C4IpAQX82#s>S0s9X%qSn9vqk#zZ;!o4lzt^kfBJ#gK+CP7PYOnr3cBwM{ z*St^gfKKv%KK{SYHNxrnlMI6T|34SI-_oDp^gmZ@i`}2Q|L^swSO1SC`_J|N#ryvI zTI|sOPw)GmxBd6~|1bN#|9RhkzJF{B^uOpn{~zA>;xl{!bSvB0L-qq-MIM>6n3q#k zHC`3@+`xnLsA`^j%79aV&v4#vkA#*IS~AbGM9ng54bsnvTF)-HzvkUN)iZti_jmY2 z<@0eHRZU+Fd1JT?5-9+S&^O>SVwx#6NU5oV*L(}yQ z9_FJ2uO9*L8=)C!@i1<-(q`6iwP}DusP_7>5P{z?);T|mgjz!mb-tpiga%D2(#2Ic zhPuO#s|m>~9QxYx1#mWn97?=&B!#A>XPsxhvnW5y7M~3z;dILZWRa-!Gu+89#mFb; z*arje)k^afuC$f~potnL2*5eT*a$_>LQ$_CW||?0zq=;&8uR`?Q*WCZzlrjNLSd1D z4x2%RT@QVtg0Y22YaRB?^Ru%2PXoHUySGs$Z{i6lK>;J`Na3b84g=##J%5>sxMt<| z!|LYHxUZrp)EX5VzZdxM$jI+8bp~66#^T|kU{>&cpewn6*JQK)UH7Zu_9t=&cayCl zrE%9qs4Dzm$4o-CoSyWMa{4PuCZ(`$V*3 zTbfh9ip~LJ44mDsFRO2F*nyv~uC88?aC5JzoqvKuOJ>h81YLc3c z;&C45Sfyn3*CAHVDg@KzzOJoUcBGCGxcEgYjx3FA@EoQ@*USBc>KC%#;!idEe#m@j zY)EonZNflt_lu{)?M9N4;qnaa{R0^lO{oLLVyDad3OQNX@=h9nx6|;NiMJ-fRJS+( zG{$UY*5R`U$-lqgE#(J30dnu;rTK(sSGCTTv*CRw$Z0>iSs|@4QwuX` zdfzX1W(nbwf4yK`T<)@O2nE>*yr(itGyOn5pYDBlw=4_)G5KhB>E8I&g17!hz zX+Yq#w(`4OUjr1cWCG@jm5A1Mj~G^beIyg>U3@~~_{{K^#IJ&VUme!_#Zrk3%qe&D z>spOduc>7h$|Fu{DG9@35&%Z9syA38&+=nXUiBL+nS@ z+q+kw9UL5FE9umEF!7y5rKH>`Z&14+w8mZ!=xaNXe!3)@oW$wGn%;EKl(ra|d=o#= z;8;3$*ZrHdUWwa6-;{y?cY)D4pEDoB3u%MBUbW2dV+c&pS-=YiK{#kiN-0&d;PyG8 z8m!piCkBI26B7+>EVw-0?L<#S3&tBqrQP7ETJA!iP>n6I z-aEX?$(milId`%)GcYr%sYnP5p_G&euOW|zzw8tBd)MAZ?^)P94QQGLl`k^7|V)oBTMJ z0KeQ3iIBh08}kOVu7+=6C(3Ud>m5P#Y47PCwox!vpo7cSvNkbnSkEj(B7Js=KnYi@Zy!=R9`3b71Y}Ch_VQcHFF&7I*wbah5x7 zB1Vl@L_Q;d<3P6v$h*c>r;1N#<%Aa`aEvNIrzOveP40s@Crljd&<3JXuO;mOL10_f z0;ilPFFjoBx+jALct?k0VoKGn3Go+tQXd1OBZsKm7dK$vN=?awWWm^!qByGwkVq>3 zoCJ}YkZpcn{3V}^f20;5MK1*tOjRdLfyXG;v`7U#lqGGqL9`yAKot0AndvhrG>S^@3lf1=^ z`PiMUEO$WG=q)tvwz$|91CoDCaoj0XrI?e>BN5}gPDnF*wT~V#t--NPS9c3>*QbJY z2<4HTRg{DSP1W2NXTQ43wXVEHMVyyWC?MuNw9?j#fY@e=GngdjVg7%g7471>R$Y;I zQ)ULq8X;$F0s?;5Ij*E+mF_*}#ndaMB9dfM!Rp!FO3R=BITp7FQVgOO4jNGt&zb6f z=LtMzRTk692Qy@QYn^vGv};Jy^3(au-y(o{1-Wl1xrk(DAAnFNii~gDr^wLk_CZ40 zR(ZLgsrbdqdLZfL@_scR%<~?*@3^YN&xiE@hvU}}1xH<@cj6I%`JA={oVZxa7OUMr z1OzmR$A*Wi@E(;|5B0+=xYyU#84)6FH8tX*2K&tfjTh*?fIyb0ga5n##J;3*htoEH ztWk{!WIOIBRQc0o&v`i$`V*iDB;Ffbq6Q? zo)Tr&+q#?jQRJ}SyaDlP2D|c~1=!^gZCfICn!q_g$o>NCb#v3KHsMeayZJliFOlpZ z2+}#Mb>e%Prj_Z`8|{uu)%aF+du+jExnpII<(OPh-_nS+w>Hxo(!oxC|BAwFtDOSc z;zN~v3(N@!U^JK0$@-?{N$(w&EL_UT;kdounifexrWPr}?m@3INx!?+#A7X9VY)t= zx*^X=n6lHxsBzTcBXfJkvqllkHH5MFGyRDALIJNcf^LE~rRpwopbH@bvLX@P5JAIf zF}MpyOsQ~lr}6DVs97~kk!QSoZl)py#4{nJ0(FgAuVY6_OG?m#lccH1g^P%m(hD-A zA5T?zB6gyFIOWHU6w$}X(A9={Ukuxp0SOZKgY7E-3*u4=xbNrfe45dTt#&^7ya4#r zpuOF4_!l3@W9teVS68UoKrJZ&)fhVuhUc7ZHL?&~LCTEB8iHK&>unrh5Hqc+aw0H& zIWeS4j;5;eu2D5sW6v-)5w*DT2;OiZ`cny)8{I@LZi~FEtg4&cY#{W|8`C(LQ`2@y zvLH)OHqZ;^qty)fMwAF} zF_aEpRO&T>gUTRdPwKX8oWUuW)r{E{CGQLRyY^n1gsOWSub4j1e!QTOw2yo5-~5}lF`$}b zXT)Mm;Fw%OuO2;AuXc}xKQ`g8F&VVA^wvJ(!6;ZIU!NZ8Ytv8>7uQoW8E?{cXn&h| zwfFE_uW;$h&<_8aZ9v$uZ|p*Sv9bpQjQ zmpn1Y_;}~1a7Vi`DvrDl-?q+!KNctd8XZ&yw4w1IA#phvFwH)cW$W)5t}Tm?Z$jls zq;s*_XjQT=@7gS`tl+NArizitPx}33Xb9t*Ho130K#nc|Szv6!ZXP)W&W=0mh~47>D=}1H(^2Ma`~GaZtr1+Vj(x8wDLx2sySsUzlU5+ zp9WC~f077J`SL}HsTRzO1&u8Y(w?9(BT9L)Sk-rQae@~0PguEP{@XFGOSbpC4+V_a!y z^6O)+yc{-!)>r&~N2Ly;yiw3l9T!I|50QxUzLJMywCnb;sGZ&mE1;NCY@!>!a=Qjc z?0bB?WhG4mr2(;_6o~cDAigY^Xq-S_mMI}G(fKy6MhF8XCh)j zQ&+3}!IHETsSxS1EF+Yx#>0|Bf&EC{S*?536}57-w4wjl;rH;vmV8F07&cMc_X?`a z2R2kE5=Yr9r#jfz{FN8zN_z>3<|Pp8`!v6-oZDLv}!87rwg)3ioIJKqxr%JXqtd5q#W+1h%L8V%bXjLW{XPz$5* zmgmX!-C7?3Jzwwq@|4b{vdF?$Pe0JBC}tirJx`^=BH^glP1GD6VKV0%2jz^V5K07l zq$){(LXKknTWMzPlpLKT;f1yBnw{P3n2uuQX3*hb@l(H?+!7G+VkiI=Vf7-dSu21j z;`}^6$lS`rkQzENKi_{NVV*9ng(jTXF;q@-*zGlRmr5Fdr+wt5o4{ z9}*D#Cockl?O~gs&a22DyT#Gup;8VGr*3*7r$lIb6D13`Jf#AVJ(0?!ES1l9G8$Yj zmbxnK?w=zH&E=UB!pw=T;1o2H?#K6uE=fq%$w;BqViCpe8vaQ_mE0n&$f!v3$+0T( zN1_BKCuNUa2_Lj&hF3&aL}#qF+&w&Uh++O2Vw?0u)Fx4Q<>Yj&^e{-)={c;gZgn=d zr+I0^FCx3YwuZl8zkx|yXt_RCAl@vg$~Lj7wxDhi8}8;a+~Rtl7|L6PKb;=LCG!@qts z-WIiy0A%&|BOseXx$kBHo8mP53k+~bY!R;cVQK@3MSM$(lp(8UH&NaVocZaypcCgr z>rBQVlf*XncLbIH0CC7*^z`q|1>JgerqXLyCz*QxNe50B5-Bg`S#L4RHp}qYA*5IM zNw`Mnc}dxkMgwXLK_%qMZPmY@&^$HW=~)d~fJ#Xs6tJnc_`UhwK>$JO_<{=Pg(JZK zDjwY;41a-u8d5GwsrB)Q$S}Ytgr$Q_!Ec_Ft}7siF9tNV0!vF?*d_E)%H&$G=Y7zt z_ut)h2fn~yFnBdFj6tQiGg8$axAIEz-NhNYmRe<%Eyui=rki~j>`q2?o9?4TI9zM; zw@fgxSn28$4F*AW#lO>SjhOHSQ3t_&5)JAor@W|St}`AC$tKui`GI}bBLf>(y*l%= zGtsV|y_C=!SU1ZvNEodzEG&%Ab~$`ZPnh__wv(KsRBWCLXZqo%o}Q?n?UkT6^qFZ{ z@AG76=>lA59KCecUIyoWPG<0$Vq^z8PWZ>om-H{&8w-=5ou8>-i1KtDMLi(*)Ra?r zNZ=aRW?6XG{7kOIN#tEwG-Yw?D;&NXXf}P`@tV_~;-{Lp9QV}o%O1K^ z+9QQ|b9*N3FmIXqkkN4vzJuGbi31;#&c=6*rfg(CO-uu-l2|5^tsrcf{ipBc2dk z{YXHIj$>KqulJjnmgp=nMO|E$C9P-Y6Y6?M$!4}PllnvgH{3+^Yap|* zG?VvuYK-5Pmx0>J^5Owrk6JQ`3GhhxKf@u0xMP{YE^;xTuA>ees)2Iw$eElrw=&nq#Cd+UZn7II+)aLxT9d~ zR5NhYg_@l8mwO~!Y4RJ(+Xl{61dp%I`=*Zbu7Bp>B9&CT4qQjZ`r!Li7xDYpSahOfjLH`*wkcx52rH8~~nR z%QA0xMA-~MQ;Y66W4WNJ7*?rsMtjFYu4_3`7U1bka>d)`{Y z7h>(+HgCD(QYwtYiXk(GW}2%2(*Ixy@4jHjhxKpx;%-JkaQMl}c1theGM&HH_2do} z{`oT|COW>Q(EQpRl&5-mz{65q7Co2X^!K!n2=gGAkiTn*%Mw)z(Fs@H&YRgWn;A!; zO(&oTsVzDmpvOb=GBdI?OYs1fR_9i4*_TmuN)&knip3k;TSkG(gX>3A#}vXxt6ji! z1_2Kx3R!t$ z0~}0d^siriB&%zN33Mz>C&$H~vNW2uLFd3hWrI@f<>|MaHBeyZFtqK;olGCqG;gKP zL4K43uBnQwz>*@yJ4ixRN|B+z5sBYIv1{{49x(nV?UyE*a4ChPWMnc~a@qivWY)ew zMU|;}T*Dt#vCaE&y-shi2R{IpApCqCIEWpTdnbOuhXz6n07+~nkjI5U+?G>fN9S>( zasn8C=JfPyS6wb|3B;8H(I-l_Wxa;ld5|oRt?c zZQ7ya1XrKD^6tB`<)UE1B=J=3r3QF@ns#=bu0 zhy{h#kg?BM=*pqi17}V&@Duya&(L@HTv8UZtF-5q%pCZIKAgU&ZkF8Lnm%R#;f`Q{ zu1mQ^@kP@f*k0@xS=l*?ih#M`5wHh%5-XmC%n5uyy9OF8!`OH@wytC(=|7Fv>d4UB zOUY?(*j(G+WQd%AF$CNvAD?*k|O=q6rJ>K@sPaTVGe-&lKTu9Kc zw5?k5e|DaSW`qfX&R5{~?9(KW+KJBwCKC_BSDgl+`{B}3QnvFkKM$e-tl+tmu_ZpU zKAsH749^*|UI5kF6c*#!BoG{%Bbniz0;w=ZN15W zUHL}Q|5QdIC0z?c;Bba2=E6lLIgm{Q>8Ucva7SSoZxfJKaLC4Ul`e{3;+^t&=o?qHeI?Q@35bjlVWM2O;VBkFhVv_AY zd$6qUMTO!%Lh^QD7`1UH+$?X;tS{WtH&wyq{v}y!#CnOtcG~UwTIE93D1*0Wlj>R^ z!GQtDAJ34McJLk+{gh3Qv63v-p(E(^4Z=Di8?}$)axDWnhV~q^fAS^v$vW@k(4~|H zr?tAxL>=9sCg-L|`5w8*;1sGiGD8Dn=TzY?<~7}G?OTD8ifOc8njxG=(|#o}Jf7Mo z{S*LqAwd0*kenKtjv@{5oJsZ2=O6wp9zwtR9GjOdPv4F~?Ae=MP8w6a5*X zY<_$b`detDweo7vfP{vi$-KN(yi$JbS?ux?bDZT6p*9R-m|3yuacd}*NPkS@I!v0r zs4`%uj_U?(3IMmQs5Xq>`;P^9;7nwqJpr3Pa_CCc_WdhF?6n-;E zYe==w4?Osjc(pEm#ggafClv+OuZPgEF-Y@EGlS+%4Zj&e*;cJI2A4xDH~(cOzq717 znYx}H9OQ6w2|Y3&Gss$uDyLFt)E6$dm^=$233_?|)^D4=ZjqFoj>~depOi0mQb4)& zpZb3~^{WTh76u3!BV(`4u~H7v+qaq|y}Z1foQ_ADegPPI+xqBV8VE4UFmEoO8*)I+ z@pHNk0bN_W4x7VmpcY1O*^Q4OP-YeK=^xvD>(#^uzNEl3Ew#HqF>=-U^DA9i{Pm5s z!<~_dvXaMl{?wdde;x|+cK;u#8ry#)ME-OAzv6ZL`(*YL$iOFM9_jPG}e{R*tt zV2Oey2G$#}#KDpPOA@TNU`c@`4VDa8vS7)9^$sj~u-=2EPwmBnow zJXb@gieB`Dc616n`~m5L(`AuKM3Kxp7ZJ2SC}cr`U0zY#&50YiYFB77<}0x^W{<@ zP1IL0X@2W<9RcKxQ4`z}Qn-vJPPkU=>nl&w$Y5Vq*Z!i?md<--Z!!xWAW+VeRy$y)&Bf3lY0V5&S&LEYT;yi< z$tU$>NI)ah4mRl1&XYnYzHW;fem{D7P%qpRchqhm=R~>_!JpXE))&ZCU7IBgqx2!w z%j;Sy;evW>Dl)SBg1VF~2ED63nJb7gerOjGLrR-fIWArpxvx0oG3is_-K-sjbYNY! zC_bBM`gOfLx=*dlvoCo4w%b5?@W=NPw`i}WAz7lh-)81#mqzQ9GuEA2`&F-^%MO?> z`7li<##RNj5=<)93!hfnN*V3iQByWK(o*A8U+zLn>2arXctWu6(we?XJEWCBEZTB1h#?bo*$X z$P$okFk`(pOZKRUfswvf|kSu8_syS;NPinr-fK7EXpbyq5f=1Tlfv*z1c`6H?! z2E9I^SK1x!Tk0CyZS%6h75whE96w^85guD1Fdlz|8P*(yy={uD!jDP$w%tj754}A( zxYC;m+AF;;1B=_I*?V@Qx->p&KSKDqD-24KzSi?X<2x9oy5BM=*(zw^Y`439b6A-D zgyYwtazSvfNpPMl=XtGAWQfvM%;f8%gyL1k&{!uIoC6# zPMm0^8ZZzP)X9|cW#e<{QIp-h#Vo@v=u6az-QzeMr{59ys#zrDtJ1_JcM(tX_(D{B zX?*sxlaol-J6{XJr3XHCXi@PVu_u)ZAAedr#oZr%JMfDrlMX#-z85{R$!-}=>nr|* zUTg4ID(4r)+A{lCoRVNh$&3m`4p;01ZAKcxohM9+^fGtR`TZ>&{(6tVQW@U+Yyu_o zP`#N!yhd-;W&8E|sYvRJYTljqoAa@^(K95j+uoScq4m^0JJX{7PEhaR+Z(}mt#*Ku zkQpxdI2cVWT(6?rD&G8TeQj@Pxfam?(S5t1YFKPMyX$!{Rl8z%b{NRwb#aD(P7r!G zYg3t2+mv$Wn^OxH2w{}xVo&46*%`ES+CoLSgXqp1x$s+ohl)V^XLD;83zq-Wa?M0I{c{gl7#nb4b}yR;n2#7JnAwL zS=EI#dYTlEM4V_({%FBQ2u1TuEPXtn7} zM_&(#-`3`umrM9BjOX1_rF0#zuT>+^jHeV?$xV|GP)P+maQfAYNf%40{JVn7$Q; z%Ouw%Ea*j_>#g5gLX#~*Ksv5g=o-g4^vvlbCidU!j8VK|alFKUM%94&>@iJP{&*O* zq*Ov|CO<@3@3xT`=bN*>9s3T*E1mV|DJo$^Nx8PcH=C*Uw3_AVvSsyeFB_6#W@Mz{ z%S(M5Y_Fq|8+|eVwt*J2Cykxzp!8OY2_{=`a>Ikb65_zpdqW!V!W2acNSz0yVWB5w zNvV<41<3oBof4M4FD<)`dq3D$e0Zr<#7SSxm#Np7$H@fu z*ofuWnCckTNSl7T=mK3#&2G}B+g-Pk$ghp@QxZ_C2c^eQro-y|MYx(WYkJ_YHCiTn zWfcE=i%{D9r5Ik0w=XS?kRO&=;jZM($>i;Ga-0GS?<}gjH{5Pe^3xI9yOw_1LN`?CC%&7%+u-P&+Bo^`#-zI z#Z3SG78{G3`3ba2zMKl{dhBfId~Ae!Y+!S&uXk(+`IfSN_vYo5t`(rZjwYXFf0NLm z?b{1q%2MRRTBqHDU7wof*>W~t!-G>5NuUMY?Umb3%tzs!{yla*FKYM-jwkM8bL7^xMJTh`<8Kd5x|b9N0qIs1fr z<{x_YRX&xJo~tdas{1R+VaQHuF9#H)C1Q$Qx(0R%N8_$Vq-!>$ zeIs}Eh4x^exwW$Fd@YIhM#oEkc%6tKLM^2ozTn|PsE%L0CQKV<(`fC?4k6;mIW+vB zC7dGJ?!m09_8Tc9{_+@tN{RiRly~`fW8GOjkxI70DqcRkf6gdo$)h3l7;*K;HuVH* z_3$;uwn4_0zxP*DKwfKiUUN=f`+DBbI0ZO~(tLduj73#Khx-Lm4PKucE}8nep{l%M zlxzNf-)PW(JIpf1|Fgqv%oZ;+!PISD%P~{9eQc3fh3;O=bMZ#Q`-ZJxGuO_$it57t z7P{RdbV-dCug4I*b`SSA``C_Wy%^Y#S16DHZHEsA~$WYwf9P&aG?TsQVcY)wNC4wba&iTzEA3d9-$TG-rCWuXy}~ zd$f&tv{ZU@oLn}AUbZ4Hn+q@7cVzR*vK!?KUNrJ;*4+sIH!JWgNAwgM-7(lnu(_~9p~-Gl(v$0xs9H48$aQOaTpiz zX|YByO$6jm_*zYbY@OW|D0(hX%u!p!P#V)`z1dH>*`>SLC%oC?wTZk<-uarmmwcH$ zRE#&Wg*O2%l%x0@LSR;X53Gujsfz5WN?@o8&#j8HsEXdGN+PQYi?52+s*0MbN))Jy zsI7|Eci37I{+}HEWs;;NzN8g2-9>8MWdhwLu9@GCGmA_!Aten20VdsGIHl!{(9Dvj z@NY5Uc|+kjCE*48hTk#`^A-(rkXFNjyVq|TuSFBDIW?~Z=a;Ib9^c$AOn)Jvc+~Y9 za<-_m7zCR7cb}N{X2#|e#`bx}pAOMRbkW8n(J;Pf6EsyLYE@$bRT!753HGQF)2K1U zC`@S7qm%cTjYx8LRXfvxgEu-&Wkh7 z>6O;gD(L9T=&u%W-FcYqoQm#(l3K!aj!$7dk`N@)$m+z zKA-x?yzX2knu7FN5otgBA2;n2?9E}??Pr3edXF0a{2G#80r!tX@n!`+nx330Ahy*fJnROA< z4(3u%{Cm&C9n}#>40ZCHWv`JV=IWcKS)-B1lxT%M`%7Sr*yxiSen(PHqd2?tt;oA0i zGMm_7b3-%6y!#7nz`5X_y~TwQON`yK1zQmY_MtBuhl1dzIa9$Eu*J0+R$jbYmh_tA zxqd4mXE|~=mTo?G-{AjQsHQb zPpl#GlFgP^+Lb-NO+T$Ka;-yRzprg-!dD+E=)=4DVf@Xd%q7AFtz`nibQ8%mE^Faz z8(qcQa*AlFX5Xk*SmL%_WePeoNubg5qcPIWeYo&KO;FTBYQrOzA1jtd)43~Xp7#uG zUk4V)e#ouR;OCNeT66wqJ8SwCvx<}?oZp`>3SoNeXV@Ti`4;tFTbI0jQE^ZC-3p{D ztf1k)qs;t7Z`qQ|vsHhwOds8suOWfKC?IF2J;S5@Ay)3WbQBWS98=vchQ)F2AIB=7 z+I0wn`7&+NZ}*kctM%AKuXVB3G;0(8g;Oe!w{B6Tnju7%vxdaK3b!+bRV+iwH7OPG8(@onzyvA%CR7ubQ*h}XrhaFLtfR7(T|)?{v!5dl^lyL zb?*))tmd(0qaL2%3c#YfIG^|pIL2HUHDAfbc(Y{VOBQaGBxxU%^0iQni63)~_!GI16c{A?`C)DJa`-A- zjgVk1&}%1$>{ZD=%;11};YH^vMpZ)s&kEY(&M%q!jbthesxtZMaO{&Q{&)d4?Vfkv zlz72}LW|CiAx2dk5AA6ZvP<_Dh*d8>a^a56`p;2bBvjpM!MKdBg*Ed&bjF#}mS4vA z-MDt>Uu2BZ7wm;qX)+B@DcbTn&+$_hI1{-__1P5|sXMCKV?$%vb&NO%WjrmKYEfM3 z@%qePqmn6CShW*dqMU{0v+L-ETRvieZ+Jmja5^_xtm#R-qEB8hr+GxJ;X8*xob+3a-_aMLQ&(-;nP$WH|+_R0W-IUm;v-3=J$@qFA(jw^ASCnXT8ohIxG`D^hHhUVY zcdBBe-|*Fy;`;TwZ~Mo6p*Jiz@)>iDzdd@9yRvcZqKr^;Gh_QMr{2g`}?%}%C(Wfs5UFq&lXf)J&0wnb}YX-8;9)k%I{&c zTyV^(Wy{p{eLYdJ=BQ({3&kb&VsXP;+J4ut6kc1YIOy3Z?N;=9utg?#8 z%soY8FrsT#*ADrD8ZwEH1D81w010TFSpeoPYebqv>;1Nc96529#1Z>HrMZ`oq)`-?{IOCD3^Fk`I4}w&B}Gq0 E3QH5sn*aa+ diff --git a/output/pdf/Active_Skillference_combined.receipt.json b/output/pdf/Active_Skillference_combined.receipt.json index ce220c1..c8462dd 100644 --- a/output/pdf/Active_Skillference_combined.receipt.json +++ b/output/pdf/Active_Skillference_combined.receipt.json @@ -1,7 +1,7 @@ { "schema": "active-skillference-pdf/v1", "output_path": "output/pdf/Active_Skillference_combined.pdf", - "size_bytes": 5425948, - "sha256": "92e00688fba67c672fac6ec65d347464f5fe8beff9d613db606298aaf73145c7", - "source_revision": "f2b77fa0fbcd8745b6564b33a8e6bad70d85471b" + "size_bytes": 5419540, + "sha256": "821802178b229ae4089d72b2c910aafac0485934556594866b5157615009d94b", + "source_revision": "222c1f06951c59fe97df4167d359222f2e5941dc" } diff --git a/tasks.yaml b/tasks.yaml index c9b3eca..09727c8 100644 --- a/tasks.yaml +++ b/tasks.yaml @@ -384,7 +384,7 @@ tasks: containers and the dashboard process were stopped and the compose volume removed after recording evidence, per the "disposable outputs" convention. - id: asf-skilltree-460-live-reverify - title: Live re-verify local SkillTree stack against skills-service 4.6.0 + title: Live re-verify the pinned SkillTree stack (4.6.0; upstream is 5.0.5) status: todo scale: minor surface: skilltree @@ -397,6 +397,7 @@ tasks: prerequisite_edges: 1199 quizzes: 630 evidence: + - gh api repos/NationalSecurityAgency/skills-service/releases/latest --jq .tag_name - uv run python scripts/reset_local_skilltree.py --confirm-local-reset --timeout 1200 - uv run python scripts/validate_docker_digests.py notes: >- @@ -421,6 +422,14 @@ tasks: That requires the destructive reset (docker compose down -v), which would also discard the 1199/1199 parity that took roughly an hour of per-edge writes to reach on this Apple/ARM host, so it is deliberately left for an explicit maintainer run rather than folded in here. + 2026-11-12 currency check, recorded without claiming the phase exit: upstream's latest release + is now skills-service 5.0.5 (published 2026-11-12), so the pinned 4.6.0 is a major version + behind and validate_docker_digests.py can keep passing on a pin that is itself stale. Settle + the order before spending a live run: prove 4.6.0 as pinned and bump in a separate change, or + retarget the pin first (docker-compose.yml tag+digest, docs/gates.md, vendor/README.md) and + prove that. Either path needs a Docker-capable host and a maintainer's explicit destructive + run (docker compose down -v) to reach the fresh-volume bootstrap this task actually asks + about, so the maintainer is the only party who can close it. scholarship_plan: lanes: - delivery_and_reporting @@ -480,29 +489,34 @@ tasks: - The capture provenance record names the system state the images evidence. - id: asf-public-release-metadata title: Publication DOI and repository target decision - status: blocked + status: done scale: external surface: metadata owner: manuscript/config.yaml evidence: + - git ls-remote https://github.com/ActiveInferenceInstitute/Active_Skillference.git (27 refs — + HEAD, 3 branch heads, 2 annotated tags with peeled commits, 19 pull-request refs) + - git ls-remote --tags https://github.com/ActiveInferenceInstitute/Active_Skillference.git + (v1.0.0 = 5f32086, v1.0.1 = c4e34e7) + - curl -s https://zenodo.org/api/records/21865644 (DOI 10.5281/zenodo.21865644, version 1.0.0, + published 2026-08-10, concept DOI 10.5281/zenodo.21865643) - uv run python scripts/validate_live_counts.py - uv run python scripts/validate_manuscript_variables.py - - uv run python scripts/z_generate_manuscript_variables.py - - uv run python scripts/validate_rendered_freshness.py - - git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git - HEAD notes: The public repository target is owner-confirmed as ActiveInferenceInstitute/Active_Skillference and is set in manuscript/config.yaml. The owner also confirmed MIT for authored code/content and generated artifacts, Daniel Ari Friedman as the copyright holder, and the need for a - counsel-review checklist before formal release. The private docxology/Active_Skillference - mirror is current and local offline validators are green. Probed on 2026-08-09, the public - target now exists and is public, but git ls-remote returns zero refs and the GitHub API - reports size 0, so the target is reachable and still unpublished. Existence is not - publication, and no commit, no tag, no archive deposit, and no DOI exist for it yet. This task - stays blocked rather than todo because the remaining step is an owner-executed push, release, - and Zenodo deposit performed outside this repository, not repository work that can be - implemented here. DOI, release-date, and repository-URL fields stay blank rather than - carrying fabricated metadata. + counsel-review checklist before formal release. Resolved on 2026-11-12 — the owner-executed + release path this task waited on has happened. The public target now carries main plus 19 + pull-request refs and the annotated tags v1.0.0 (5f32086) and v1.0.1 (c4e34e7), and Zenodo + record 21865644 returns DOI 10.5281/zenodo.21865644 (concept DOI 10.5281/zenodo.21865643) with + version 1.0.0 published 2026-08-10, whose deposited file + Active_Skillference_v1.0.0_DOI-10.5281-zenodo.21865644.pdf (5731097 bytes, md5 + 7e43004dbef2ebc7ade2b71197b49ec5) is byte-identical to the tracked PDF of that name at the + repository root. manuscript/config.yaml already carries that DOI and repository URL, so the + decision is recorded from observed events rather than asserted. The 2026-08-09 probe + (reachable, zero refs, GitHub API size 0) is superseded. The version-1.0.1 deposit that does + not exist yet, and the metadata that still claims it, are tracked in + asf-public-archive-provenance-release. scholarship_plan: lanes: - delivery_and_reporting @@ -521,8 +535,8 @@ tasks: - scope_limit - literature boundaries: - - The empty public repository remains an external blocker, not a failed scholarly claim. - - DOI and repository metadata remain blank until real GitHub and Zenodo events are observed. + - The empty public repository was an external blocker for months, not a failed scholarly claim; it cleared when the owner pushed the repository and deposited v1.0.0. + - DOI and repository metadata were recorded only after the GitHub push and Zenodo deposit were observed. phase_exit: - The public target carries published content and a real release, and the publication environment can complete the actual release path. - Any release metadata is copied from observed repository and archive records. @@ -1238,7 +1252,10 @@ tasks: notes: >- The current tracked and untracked release-facing paths were re-inventoried without staging or committing. docs/release_readiness_inventory.md was regenerated from live git state and - validated so the release boundary stays explicit while this pass remains unpublished. + validated so the release boundary stays explicit while this pass was unpublished; the + repository and its v1.0.0 deposit were published afterwards (2026-08-10, DOI + 10.5281/zenodo.21865644), which is why that snapshot's "not a published release" probe text is + superseded. - id: asf-doc-command-drift-audit title: Documentation command parity and stale-count audit status: done @@ -1616,19 +1633,32 @@ tasks: deps: - asf-public-release-metadata evidence: - - git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git - HEAD + - git ls-remote --tags https://github.com/ActiveInferenceInstitute/Active_Skillference.git + (v1.0.0 = 5f32086, v1.0.1 = c4e34e7) + - curl -s https://zenodo.org/api/records/21865644/versions (a single version — 1.0.0) - uv run python scripts/validate_release_readiness_inventory.py --write - uv run python scripts/validate_release_readiness_inventory.py - uv run python scripts/validate_manuscript_variables.py - uv run python scripts/validate_rendered_freshness.py notes: >- - The public target exists and is public but currently holds zero refs, so reachability is no - longer the gate; published content is. After the target carries a real published release, - complete the publication metadata chain: release tag, Zenodo DOI, CFF/CodeMeta/manuscript URL - fields, artifact hashes, third-party notices, counsel-review checklist, and release inventory - refresh. This task must not fabricate DOI or repository metadata before the upstream blocker - is cleared. + Upstream dependency cleared on 2026-11-12: the public target carries main, 19 pull-request + refs, and the annotated tags v1.0.0 (5f32086) and v1.0.1 (c4e34e7), and Zenodo record 21865644 + holds version 1.0.0 (published 2026-08-10; concept DOI 10.5281/zenodo.21865643). The release + chain can therefore be read from observed records; what remains is making it agree with + itself. (1) Version drift: the only deposit is 1.0.0, while .zenodo.json (version 1.0.1, + publication_date 2026-08-09), CITATION.cff (version 1.0.1, date-released 2026-08-09), + codemeta.json (version 1.0.1, datePublished 2026-08-09) and manuscript/config.yaml + (paper.version and book.edition 1.0.1) all claim a release Zenodo does not hold, the v1.0.1 tag + has no deposit behind it, and manuscript/config.yaml keeps publication.version_doi and + version_record blank. Either deposit v1.0.1 and record the version DOI it returns, or downgrade + the claim to the observed 1.0.0. (2) docs/release_readiness_inventory.md is stale at HEAD: + validate_release_readiness_inventory.py fails on the tracked-diff summary, on the listed + manuscript/README.md modification, and on the recorded figure hashes, and + src/release_readiness.py renders "Public DOI/repository remains blocked until git ls-remote + succeeds" unconditionally, so a --write refresh would publish a sentence the current probe no + longer supports. Fix the generator before refreshing the snapshot. (3) The third-party notice + sweep and the counsel-review checklist are still outstanding. Every value must come from the + observed records; nothing here may be fabricated. scholarship_plan: lanes: - delivery_and_reporting @@ -1647,7 +1677,7 @@ tasks: - scope_limit - literature boundaries: - - DOI, public URLs, and release timestamps are blank until the repository and archive events occur. + - Deposited version identifiers and dates must match the Zenodo record that actually exists (1.0.0, published 2026-08-10), not the locally claimed 1.0.1. - Metadata standards support release reproducibility, not scientific validity, legal advice, or learner outcomes. phase_exit: - A real public tag, archive deposit, DOI, CFF/CodeMeta metadata, hashes, notices, and inventory agree. @@ -1747,6 +1777,23 @@ tasks: not have found either. The regression tests added here are the mechanically checkable part (browser-session restart, post-timeout confirmation, retry bounds, CSP directives, deferred parity, gzip), and the live gates remain the proof for the rest. + 2026-11-12 re-measurement of the current tree (`python -m pytest --cov=src`, no floor + override): 989 passed, aggregate src coverage 90.92% (10348 statements, 738 missed, 3228 + branches, 355 partial), i.e. 0.92 points above the 90% gate. The four modules this task names + are unchanged (full_app_stack.py 63.77%, form_login_validation.py 49.76%, + source_health_network.py 72.39%, dashboard/payload.py 70.37%), but they are not the whole + sub-90 set: 23 src modules sit under 90%, and the extremes are src/publication/zenodo.py + 0.00% (98 statements that no test touches, although scripts/publish_zenodo_deposit.py calls + it), form_login_validation.py 49.76%, src/screenshots/capture.py 47.79%, + full_app_stack.py 63.77%, and src/curriculum/materials/* between 75.48% and 84.38%. The same + measurement found a second illusion: dropping the last blocked task from tasks.yaml moved the + blocked-status branches of src/roadmap.py (94.23% -> 92.99%) and src/figures/plots/roadmap.py + (92.62% -> 89.34%) out of coverage, because those paths were exercised only by the real + roadmap data and have no fixture, so the next blocked task silently restores the appearance of + coverage. A roadmap-state fixture is therefore part of closing this. Blocker for the + live-gated share is a Docker/browser-capable host, and the maintainer who can provide it is + the owner; the per-module coverage exception record this task keeps pointing at still does not + exist in the repository. scholarship_plan: lanes: - delivery_and_reporting @@ -2102,3 +2149,41 @@ tasks: - "Subprocess returncodes are already checked at the flagged call sites." phase_exit: - "PLW1510 resolved via explicit checks or a justified targeted noqa; ruff check returns 0." + - id: asf-gnn-upstream-5-1-adoption + title: Adopt the v5.1.0 GNN upstream pin (currently v5.0.3) + status: todo + scale: minor + surface: validation + owner: gnn/upstream.json + evidence: + - git ls-remote --tags https://github.com/ActiveInferenceInstitute/Generalized_Notation_Notation.git + (v5.1.0 = 75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf) + - GNN_ROOT= GNN_PYTHON= ./gates.sh gnn + - uv run pytest tests/test_gnn_conformance.py -q + notes: >- + gnn/upstream.json pins Generalized_Notation_Notation at v5.0.3 + (8f293d9ba810e6eda5c09842d9c7ed9ffa5f92c1), while the estate cutover moved the importer + repositories to v5.1.0 (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf); this repository was not part + of that cutover, so its pin is accurate but behind. The tag is duplicated in five places that + must move together: gnn/upstream.json, gnn/README.md, docs/gates.md, .github/GNN.md, and the + pinned-record constant plus the scratch-checkout tag in tests/test_gnn_conformance.py, with + README.md linking the v5.0.3 tag at three places. scripts/validate_gnn_spec.py extracts the + pinned commit with git archive and parses the specification with that checkout's parser, so a + metadata-only edit would be drift rather than proof: the bump counts only once + ./gates.sh gnn passes against a v5.1.0 checkout. Blocker: the arm fails, never skips, without a + prepared v5.1.0 checkout and interpreter (GNN_ROOT/GNN_PYTHON), and the hosted + `gnn conformance (pinned)` job cannot stand in for it while Actions is billing-locked + org-wide. A local Generalized_Notation_Notation checkout at v5.1.0 already exists on the + working host (75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf), so the maintainer who runs it can + close this. + scholarship_plan: + lanes: + - delivery_and_reporting + existing_source_keys: [] + planned_source_targets: [] + claim_support_types: + - scope_limit + boundaries: + - "A pin bump is specification-conformance maintenance, not scientific or learner-outcome evidence." + phase_exit: + - "gnn/upstream.json records tag v5.1.0 and commit 75c834bd8519dc7b5adc4d24f16bf393cfd1ecdf, the five duplicated pin references agree, and `./gates.sh gnn` reports clean against that checkout with no skips." From 3dc47a816072feff55e5a7585acee9e8e663495b Mon Sep 17 00:00:00 2001 From: Daniel Ari Friedman <6911384+docxology@users.noreply.github.com> Date: Mon, 16 Nov 2026 13:08:49 -0800 Subject: [PATCH 2/3] Repair Skillference software contracts and isolate source-writing smoke tests - Adopt individually provenance-bound original software and regression paths after independent core/runtime review. - Repair low-precision normalization, rendered-name identity, native quiz settings, supported learner routes and real bootstrap fixtures. - Keep generated/scientific output, K4 convention, GNN pin, browser/runtime and release-owner holds unchanged. --- README.md | 2 + docs/gates.md | 4 + docs/learner_dashboard.md | 15 +- docs/output_inventory.md | 2 + docs/pilot/README.md | 2 + docs/publication_readiness.md | 25 +- docs/quickstart.md | 2 + docs/rendering_pipeline.md | 10 +- run.sh | 32 +-- scripts/README.md | 6 +- scripts/bootstrap_local_skilltree.py | 16 +- scripts/capture_dashboard_screenshots.py | 2 + scripts/render_manuscript.py | 99 +++++--- scripts/run_kernels.py | 5 +- scripts/run_learning_dashboard.sh | 17 +- scripts/run_skilltree.sh | 17 +- scripts/serve_learner_dashboard.py | 2 + scripts/validate_dashboard_console.py | 94 +++---- scripts/validate_rendered_freshness.py | 13 +- scripts/z_generate_manuscript_variables.py | 12 +- src/artifacts/json_io.py | 29 ++- src/curriculum/_ext_dynamical_geometry.py | 9 +- src/curriculum/_ext_learning_hierarchical.py | 2 +- src/curriculum/_ext_precision_exploration.py | 20 +- src/curriculum/catalog_data/core_strata.py | 15 +- src/curriculum/graph.py | 90 ++++--- src/curriculum/materials/builders.py | 15 +- src/curriculum/materials/github_sources.py | 34 +-- src/curriculum/materials/helpers.py | 39 +-- src/curriculum/materials/validation.py | 28 ++- src/curriculum/model.py | 13 +- src/curriculum/quality.py | 38 ++- src/curriculum/recommendations.py | 10 +- src/dashboard/artifacts.py | 12 +- src/dashboard/assets/app.js | 25 +- src/dashboard/browser_validation.py | 231 +++++++++++++++++ src/dashboard/config.py | 3 + src/dashboard/live_validation.py | 43 +++- src/dashboard/payload.py | 37 ++- src/dashboard/server.py | 28 ++- src/figures/plots/registry_io.py | 6 +- src/gnn_conformance.py | 12 + src/integration/full_app_stack.py | 121 ++++++--- src/kernels/active_inference.py | 34 ++- src/kernels/distributions.py | 62 +++-- src/kernels/inference.py | 24 +- src/kernels/predictive_coding.py | 21 +- src/live_counts.py | 5 +- src/manuscript/pdf_receipt.py | 185 ++++++++++++++ src/pilot/contract.py | 21 +- src/pipeline/generated_outputs.py | 2 +- src/provenance/binding.py | 50 ++-- src/publication/zenodo.py | 68 +++-- src/release_readiness.py | 42 ++-- src/scholarship/source_health_report.py | 13 - src/screenshots/capture.py | 239 ++++++++++++++---- src/skilltree/client.py | 23 +- src/skilltree/content.py | 239 ++++++++++++++++++ src/skilltree/deployment_matrix.py | 4 +- src/skilltree/local_auth.py | 84 ++++--- src/skilltree/transport.py | 5 +- tests/README.md | 13 + tests/conftest.py | 184 ++++++++++++++ tests/test_atomic_json_publication.py | 57 +++++ tests/test_course_materials.py | 183 +++++++------- tests/test_curriculum_graph.py | 50 ++++ tests/test_dashboard_console.py | 7 +- tests/test_dashboard_live_validation.py | 54 +++- tests/test_dashboard_mount.py | 82 ++++++ tests/test_dashboard_server.py | 62 +++++ tests/test_figures.py | 22 +- tests/test_full_app_stack.py | 127 +++++++--- tests/test_generated_output_cli.py | 20 ++ tests/test_gnn_conformance.py | 20 +- tests/test_kernels_distributions.py | 11 +- tests/test_manuscript_hydration.py | 30 +++ tests/test_material_claim_fields.py | 153 +++++++++++ tests/test_pdf_provenance.py | 121 +++++++++ tests/test_pilot_schema_boundaries.py | 48 ++++ tests/test_pipeline_generated_outputs.py | 46 +++- tests/test_predictive_coding_diagnostics.py | 30 +++ tests/test_probability_support.py | 87 +++++++ tests/test_recommendations.py | 49 ++-- tests/test_release_readiness_inventory.py | 35 ++- tests/test_render_manuscript.py | 101 +++++--- tests/test_roadmap.py | 37 +++ tests/test_run_entrypoint.py | 61 +++++ tests/test_scholarship.py | 22 ++ tests/test_screenshot_context.py | 146 +++++++++++ tests/test_screenshots.py | 3 +- tests/test_scripts_smoke.py | 45 ++-- tests/test_skilltree_client.py | 4 +- tests/test_skilltree_content.py | 251 +++++++++++++++++++ tests/test_skilltree_local_auth.py | 55 ++-- tests/test_skilltree_seeder.py | 12 +- tests/test_skilltree_transport_boundaries.py | 89 +++++++ tests/test_zenodo.py | 132 ++++++++++ 97 files changed, 3876 insertions(+), 931 deletions(-) create mode 100644 src/dashboard/browser_validation.py create mode 100644 src/manuscript/pdf_receipt.py create mode 100644 src/skilltree/content.py create mode 100644 tests/test_atomic_json_publication.py create mode 100644 tests/test_dashboard_mount.py create mode 100644 tests/test_generated_output_cli.py create mode 100644 tests/test_manuscript_hydration.py create mode 100644 tests/test_material_claim_fields.py create mode 100644 tests/test_pdf_provenance.py create mode 100644 tests/test_pilot_schema_boundaries.py create mode 100644 tests/test_predictive_coding_diagnostics.py create mode 100644 tests/test_probability_support.py create mode 100644 tests/test_run_entrypoint.py create mode 100644 tests/test_screenshot_context.py create mode 100644 tests/test_skilltree_content.py create mode 100644 tests/test_skilltree_transport_boundaries.py create mode 100644 tests/test_zenodo.py diff --git a/README.md b/README.md index de05abd..feb0a3a 100644 --- a/README.md +++ b/README.md @@ -40,6 +40,8 @@ it is doing and what to do if it stops; a run that refuses with a reason is work intended. The live path needs a running Docker daemon because SkillTree owns quizzes, completion, points, and ranking. The offline path needs neither Docker nor a network. +Help, readiness, and artifact-only exits do not initialize passwords. Live launchers share the password owner; bootstrap may skip seeding only with `--skip-seed-if-live-content-ready` after semantic content verification, not matching counts. Offline artifacts and controlled regressions do not prove a live official SkillTree display. + ## Read it without installing anything Generated outputs are committed, so the artifacts are browsable straight from the diff --git a/docs/gates.md b/docs/gates.md index 6565086..1ef0fe4 100644 --- a/docs/gates.md +++ b/docs/gates.md @@ -90,6 +90,10 @@ between them fails instead of silently changing the resolution. The live half of the gate set (Docker, a real SkillTree, browser proofs, captured screenshots) stays local and explicit. No arm here simulates it. +Strict live proof requires semantic content parity and actual official SDK project/subject/quiz views with visible same-origin frames, ready/accessibility evidence, and matching project/instance/export digest. Capture consumes a successful v2 `--proof-path` and matching PNG associations. Pending/fallback/unchecked views and controlled loopback fixtures are not live proof. + +The 2026-11-13 full non-browser census failed: its native source binding changed, seven tests failed, and coverage was below the unchanged 90% floor. The separate seven-node continuation and appended coverage database are narrower observations, not a fresh immutable whole-source certificate. The complete per-module census still records under-floor modules with no approved exceptions. Script smoke tests now run real generators in a disposable copy of public source inputs instead of rewriting the audited checkout's tracked output; final qualification must still verify the native unchanged-source receipt. Source-bound machine reports remain authoritative. Keep science K4, the pinned GNN origin, real stack/browser, and roadmap-owner holds separate; generated `tasks.yaml`/TODO/readiness projections are not hand-updated. + ## Measured numbers Per-arm wall times and the test/coverage counts are generated, not written by hand: diff --git a/docs/learner_dashboard.md b/docs/learner_dashboard.md index b0c9765..61717c1 100644 --- a/docs/learner_dashboard.md +++ b/docs/learner_dashboard.md @@ -95,14 +95,9 @@ uv run python scripts/validate_full_app_stack.py --keep-running Use this when reporting that the app is live. It starts or reuses the Docker Compose SkillTree stack, regenerates project artifacts, bootstraps and seeds local SkillTree auth, reconciles missing live prerequisite edges, starts the dashboard, requires `/healthz`, -requires live admin count parity, requires live dependency parity, checks -figure-registry/dashboard chart-contract parity, proves the raw SkillTree `FORM` login -through `chrome-devtools-axi`, and runs the Chrome console gate. Docker daemon -availability is a hard precondition: if Docker or Colima is not running, this command -fails rather than silently downgrading to artifact-only mode. `--app-ready` is the faster -local usability proof and is explicitly not a full dependency-parity proof. -`--keep-running` leaves the dashboard process open for manual inspection at -`http://127.0.0.1:8765`. +requires semantic admin content parity and complete live dependency parity, checks figure-registry/dashboard chart contracts, and proves real FORM login plus the official SDK's project, subject, and quiz views. Visible matching-origin frames, ready callback/accessibility snapshots, and matching project/instance/export digest are required; pending, fallback, unchecked, or count-only states are not green. `chrome-devtools-axi` and a real Docker/SkillTree stack are prerequisites, not substituted boundaries. + +Successful v2 browser/full-stack proof supplies `--proof-path` to screenshot capture. Use explicit task-owned screenshot output with matching PNG associations. `--keep-running` retains only a successfully admitted dashboard server; failure must settle it. `--app-ready` remains a narrower usability scope, not full dependency parity. Artifact-only mode: @@ -163,7 +158,7 @@ lesson. are absolute, credential-free HTTP(S) URLs; unsafe schemes remain plain text. - SkillTree deployment matrix: exposes only browser-safe deployment mode status for local FORM bootstrap, ignored local state, token-backed admin seeding, and externally operated - SkillTree. Admin secrets, client secrets, and bearer tokens remain server-side. + SkillTree. Root credentials and trusted-client secrets remain server-side. The temporary learner bearer token reaches official browser-client memory, never persistent browser storage; it is not accurate to call all bearer tokens server-only. - Official display host: imports `@skilltree/skills-client-js` and configures `SkillsConfiguration` with `serviceUrl`, `projectId=activeInference`, and an absolute dashboard URL for the local `/api/skilltree/token` authenticator when bootstrap is @@ -176,6 +171,8 @@ lesson. proof that the live SkillTree process is currently running or accepting browser passwords. +Public artifact/status projections remove quiz answer options and correctness keys recursively; the raw validated export and semantic digest remain server-side authorities. `/api/status.project.project` supplies the nested public identity used by live validation. No public projection or fixture response is a scientific, security-review, or live-service qualification by itself. + ## Accessibility The shell targets WCAG 2.1 Level A keyboard and structure basics: a visually-hidden diff --git a/docs/output_inventory.md b/docs/output_inventory.md index fc0eabb..112c053 100644 --- a/docs/output_inventory.md +++ b/docs/output_inventory.md @@ -8,6 +8,8 @@ Paths in the table are relative to the project root (`Active_Skillference/`). +The complete registered figure population includes captured artifacts; it is not the deterministic rendered subset. Registry-owned counts and generated provenance must describe the right population without inventing a fixed total. + ## File inventory | File | Role | Generated by | Step | diff --git a/docs/pilot/README.md b/docs/pilot/README.md index ec3a830..ad7e676 100644 --- a/docs/pilot/README.md +++ b/docs/pilot/README.md @@ -17,6 +17,8 @@ The contract requires consent and data minimization, prohibits direct identifier analysis export, names the unit of analysis and missing-data policy, and keeps SkillTree as the system of record for progress and quiz state. +Four nonempty linked populations establish schema readiness only. They do not establish recruitment, participant data, demographics, collection completion, or educational efficacy. + ## Protocol sequence 1. Pre-register recruitment, inclusion/exclusion, outcomes, observation windows, stopping diff --git a/docs/publication_readiness.md b/docs/publication_readiness.md index 0013d9a..1ee30d5 100644 --- a/docs/publication_readiness.md +++ b/docs/publication_readiness.md @@ -29,11 +29,7 @@ roadmap tasks are complete. - License, third-party notice, citation metadata, DOI, and public-repository boundaries are signposted in `LICENSE`, `THIRD_PARTY_NOTICES.md`, `docs/ip_license_status.md`, `CITATION.cff`, `.zenodo.json`, `codemeta.json`, and `manuscript/config.yaml`. -- Renderer-facing repository and release-date metadata remain blank or omitted until a real - published release exists, so rendered PDFs, CFF, and CodeMeta do not advertise an - unavailable release endpoint. Probed on 2026-08-09, - `https://github.com/ActiveInferenceInstitute/Active_Skillference` exists and is public but - holds zero refs, so the target is reachable and still unpublished. +- The 2026-08-09 observation found a reachable public target with zero refs; that is historical evidence, not today's remote state. Later v1.0.0/DOI records below remain authoritative for their own identifiers. Current metadata must agree with actual records and a newly admitted observation, never a guessed blanket unpublished state. ## Release-Allowed Claims @@ -43,12 +39,7 @@ roadmap tasks are complete. - SkillTree is cited and discussed as an NSA-origin open-source training platform, with explicit non-endorsement and deployment-boundary language. - Project code/content metadata currently declares MIT, while third-party materials are either inventoried as vendored notices or cited/linked without relicensing. -- The owner-confirmed public target is - `ActiveInferenceInstitute/Active_Skillference`, and that target now exists as an empty - public repository; the claim becomes release-allowed only after the repository carries - published content and a real release, and the release inventory is refreshed. -- Platform-facing source-repository links and release dates are not release-allowed while - that target holds no published content. +- Public repository/release claims must match the actual selected release record and newly admitted remote observation. A reachable target, an empty target, advertised refs, a failed probe, and an unrefreshed report are different states. Do not substitute historical reachability for current publication or infer current emptiness from a failed command. ## Claims To Avoid Until Future Evidence @@ -69,11 +60,8 @@ roadmap tasks are complete. ## Current Blockers -- DOI/Zenodo release is blocked until an actual release tag and deposit exist. -- Public repository release is blocked until the confirmed target - `ActiveInferenceInstitute/Active_Skillference` holds published content and a real release - tag, and repository metadata is refreshed. The target's existence cleared the - reachability question only; the publication question is still open. +- Subsequent DOI/Zenodo reservation, streamed upload, and publication each require explicit authorization and actual receipts. Existing v1.0.0 identifiers remain preserved; this work authorizes no new deposit, tag, upload, or release. +- A current repository/readiness projection requires its owning generator and an admitted fresh probe. AS-O14's first-page/title identity and product-selection hold, plus legal/date/DOI decisions, remain owner obligations. - Learner-outcome claims are blocked until a real study produces evidence. - Live deployment claims are blocked until Docker SkillTree, dashboard handoff, privacy, security, trademark, and access-control reviews are complete. @@ -101,11 +89,10 @@ uv run python scripts/validate_web_math.py uv run python scripts/validate_release_readiness_inventory.py ``` -Also re-check the public-target blocker. An empty result means the repository exists but -nothing is published there, which keeps the blocker in force: +An explicitly admitted future remote check may inspect advertised refs; empty success is not the same as failed or unrefreshed evidence. No new remote observation is claimed here: ```bash -git ls-remote --heads https://github.com/ActiveInferenceInstitute/Active_Skillference.git HEAD +git ls-remote --refs https://github.com/ActiveInferenceInstitute/Active_Skillference.git ``` For a DOI-backed release or deployment publication rather than a static public-review diff --git a/docs/quickstart.md b/docs/quickstart.md index f16c477..38d8180 100644 --- a/docs/quickstart.md +++ b/docs/quickstart.md @@ -13,6 +13,8 @@ screen-reader affordances. Use `./run.sh --offline` when Docker is unavailable a `./run.sh --check` to test readiness without starting anything. The step-by-step commands below are what `run.sh` runs, for when you want to drive them individually. +Help/readiness/artifact-only paths never initialize passwords. Live launchers use the shared password owner. `--skip-seed-if-live-content-ready` admits actual semantic content readiness; the removed count-only shortcut is not sufficient proof. A successful artifact build is not a running official SkillTree stack. + Get Active Skillference running in 5 minutes: validate the curriculum DAG, run the kernels that produce every claim, export a provenance-gated SkillTree project, and run diff --git a/docs/rendering_pipeline.md b/docs/rendering_pipeline.md index 9521ada..62fc76d 100644 --- a/docs/rendering_pipeline.md +++ b/docs/rendering_pipeline.md @@ -13,13 +13,9 @@ the manuscript's own configuration, preamble, and bibliography. It needs `pandoc `pandoc-crossref`, `xelatex` and `bibtex` on `PATH` and nothing from any other repository; the arm is `./gates.sh render` ([`gates.md`](gates.md)). -After finalizing the PDF, the renderer writes the tracked -`output/pdf/Active_Skillference_combined.receipt.json`: schema -`active-skillference-pdf/v1`, repository-relative output path, exact byte count, -SHA-256, and the checkout's Git `HEAD` source revision. Commit source changes before -rendering so that revision identifies the source used; commit the receipt alongside -the PDF. The render arm finishes with `validate_rendered_freshness.py`, which rejects -stale hydrated sections, graph counts, or publication-bundle roadmap text. +After finalizing the PDF, the renderer writes an `active-skillference-pdf/v2` receipt binding actual configuration, preamble, bibliography, ordered hydrated sections, literal TeX assets, and finalized output bytes. HEAD alone is not source identity. Legacy receipts require a new admitted render, never relabeling or restamping. Unknown prose placeholders fail while literal code examples survive. + +The configured atlas is physical page two, before the table of contents; its placement has a real-render regression. Input/content custody, retained-publication identity, and parent visual inspection remain separate requirements. The freshness validator still rejects stale sections, graph counts, and roadmap text. ## The flow diff --git a/run.sh b/run.sh index 80eb6f7..e39856f 100755 --- a/run.sh +++ b/run.sh @@ -22,22 +22,6 @@ DASHBOARD_URL="http://${DASHBOARD_HOST}:${DASHBOARD_PORT}" SKILLTREE_WAIT_ATTEMPTS="${SKILLTREE_WAIT_ATTEMPTS:-240}" ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR="${ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR:-.active_skillference_local}" ACTIVE_SKILLFERENCE_LEARNER_ID="${ACTIVE_SKILLFERENCE_LEARNER_ID:-local-learner}" -# Compose interpolates this; it must match the value the stack was created -# with, so reuse the stored one and generate it only on first run. This mirrors -# src/integration/full_app_stack.py::ensure_postgres_password. -_pg_file="$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR/postgres_password" -if [ -z "${ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD:-}" ]; then - if [ -s "$_pg_file" ]; then - ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(tr -d '\n' < "$_pg_file")" - else - mkdir -p "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" - chmod 700 "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" 2>/dev/null || true - ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(LC_ALL=C tr -dc 'A-Za-z0-9' < /dev/urandom | head -c 40)" - printf '%s\n' "$ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD" > "$_pg_file" - chmod 600 "$_pg_file" 2>/dev/null || true - fi -fi -export ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD MODE="full" for arg in "$@"; do @@ -121,6 +105,14 @@ if [ "$MODE" = "offline" ]; then --host "$DASHBOARD_HOST" --port "$DASHBOARD_PORT" fi +# Live launchers share one private password owner; help/check/offline never enter it. +if [ -z "${ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD:-}" ]; then + ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(uv run python -c \ + 'import sys; from src.integration.full_app_stack import ensure_postgres_password; print(ensure_postgres_password(sys.argv[1]))' \ + "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR")" +fi +export ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD + # ------------------------------------------------------------- live SkillTree # Run a long command with a live elapsed-time heartbeat, so a slow step is # visibly slow rather than indistinguishable from a hang. @@ -172,9 +164,9 @@ curl -fsS "$SKILLTREE_URL/public/status" >/dev/null 2>&1 || fail \ "Or run without it: ./run.sh --offline" step "Seeding the curriculum into SkillTree" -note "This is the slow step, and it is idempotent: if SkillTree already reports" -note "the exported subject and skill counts, seeding is skipped entirely and" -note "this finishes in seconds." +note "Seeding is skipped only when the actual project, subjects, skills, quizzes" +note "and answer keys match the export. Missing resources can be created;" +note "changed or extra content is refused rather than silently overwritten." note "A first seed writes every subject, skill, and quiz over REST. Budget" note "10-40 minutes on Apple/ARM Docker emulation; it is working, not stuck." note "Prerequisite edges are NOT seeded here. They are far slower and the" @@ -183,7 +175,7 @@ if ! run_with_heartbeat "seeding" uv run python scripts/bootstrap_local_skilltre --base-url "$SKILLTREE_URL" \ --state-dir "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" \ --learner-id "$ACTIVE_SKILLFERENCE_LEARNER_ID" \ - --skip-seed-if-live-counts-ready \ + --skip-seed-if-live-content-ready \ --skip-dependencies; then fail "Seeding did not complete." \ "The output above ends with the failing call." \ diff --git a/scripts/README.md b/scripts/README.md index ded6b87..77e401d 100644 --- a/scripts/README.md +++ b/scripts/README.md @@ -72,8 +72,8 @@ ls -la output/figures/ # generated PNGs plus figure_registry.json | `validate_web_math.py` | Fail if dashboard or rendered web HTML math surfaces lack MathJax support | `src.dashboard/assets/`, `output/web/` | none (exit 0/1) | | `publish_zenodo_deposit.py` | Upload the release artifact to the reserved Zenodo draft and publish it, minting a permanent DOI | `ZENODO_PROD_TOKEN`, `data/zenodo_reservation.json`, a release file | published deposit JSON on stdout | | `reserve_zenodo_doi.py` | Reserve a Zenodo DOI on an unpublished draft deposition; never publishes | `ZENODO_PROD_TOKEN` from a dotenv file, `.zenodo.json` | JSON reservation summary on stdout | -| `capture_dashboard_screenshots.py` | Capture the registered learner-dashboard evidence screenshots from a running local stack and write their provenance record | Chrome via `chrome-devtools-axi`, running dashboard, `src.screenshots` manifest | PNGs plus `screenshot_provenance.json` under `manuscript/assets/screenshots/` | -| `validate_dashboard_console.py` | Browser-console proof for the dashboard, with optional required-browser and required-dashboard modes | Chrome via `chrome-devtools-axi`, dashboard URL | none (exit 0/1) | +| `capture_dashboard_screenshots.py` | Capture from successful v2 browser/full-stack `--proof-path`, with matching project/instance/export digest and PNG associations | Real Chrome/official SDK stack and explicit task-owned output | PNGs plus bound provenance | +| `validate_dashboard_console.py` | Strict official project/subject/quiz view proof: `--require-browser --require-dashboard`, visible matching-origin frames and ready/accessibility evidence; no fallback success | Existing `chrome-devtools-axi` and actual dashboard/SkillTree | Explicit v2 proof artifact | | `validate_skilltree_form_login.py` | Browser proof that the raw local SkillTree `FORM` login reaches the authenticated administrator surface without leaking local secrets | Chrome via `chrome-devtools-axi`, `.active_skillference_local/skilltree_auth.json`, SkillTree login URL | JSON proof summary on stdout | | `validate_live_skilltree_dashboard.py` | CLI wrapper around `src.dashboard.live_validation`; fails if the local dashboard is not wired to a reachable SkillTree service, current export counts, complete artifacts, ready local auth, token exchange, and optional live parity checks | `src.dashboard.live_validation`, `output/data/skilltree_project.json`, dashboard `/api/status`, SkillTree `/public/status` | none (exit 0/1) | | `validate_full_app_stack.py` | CLI wrapper around `src.integration.full_app_stack` for the strict Docker + SkillTree + dashboard + browser proof, including dependency reconciliation unless `--app-ready` is selected | `src.integration.full_app_stack`, Docker/Compose, local SkillTree, dashboard, generated artifacts | JSON proof summary on stdout plus a redacted proof artifact under the state directory | @@ -85,7 +85,7 @@ ls -la output/figures/ # generated PNGs plus figure_registry.json | `z_generate_gate_metrics.py` | Render `docs/gate_metrics.md` from the recorded arm-timing receipt, or `--check` that it is current | `docs/_generated/arm_timings.json` | `docs/gate_metrics.md` with `--write`; none with `--check` | | `export_skilltree.py` | Provenance-gate the curriculum and export a SkillTree-native project JSON | `src.curriculum`, `src.kernels`, `src.provenance`, `src.skilltree` | `output/data/skilltree_project.json` | | `seed_skilltree.py` | Build + render the project and seed it into a live SkillTree server over REST | `src.curriculum`, `src.kernels`, `src.provenance`, `src.skilltree` | network (a running SkillTree instance) | -| `bootstrap_local_skilltree.py` | Bootstrap local SkillTree `FORM` auth, seed the project, and store ignored trusted-client state | `src.curriculum`, `src.kernels`, `src.provenance`, `src.skilltree.local_auth` | network + `.active_skillference_local/` | +| `bootstrap_local_skilltree.py` | Bootstrap shared live password owner, seed semantic content, store ignored trusted-client state; `--skip-seed-if-live-content-ready` replaces count-only readiness | Actual local SkillTree and current export | network + ignored private state | | `serve_learner_dashboard.py` | Serve the local learner dashboard shell | `src.dashboard` | local HTTP server at `http://127.0.0.1:8765` | | `validate_gnn_spec.py` | Validate the discrete-agent GNN specification against the pinned GNN revision and the kernel | `gnn/`, `src.gnn_conformance` | none (exit 0 pass, 1 conformance failure, 2 GNN unavailable) | | `run_skilltree.sh` | One command: bring up Docker SkillTree + Postgres, wait for the API, then seed | `docker-compose.yml`, `vendor/` | a running instance at `http://localhost:8080` | diff --git a/scripts/bootstrap_local_skilltree.py b/scripts/bootstrap_local_skilltree.py index fb7ddc8..4f63bed 100755 --- a/scripts/bootstrap_local_skilltree.py +++ b/scripts/bootstrap_local_skilltree.py @@ -43,13 +43,15 @@ PROJECT_NAME = SKILLTREE_PROJECT_NAME -def build_project(): # type: ignore[no-untyped-def] +def build_project( + project_id: str = PROJECT_ID, project_name: str = PROJECT_NAME +): # type: ignore[no-untyped-def] """Build the provenance-rendered SkillTree project (no network).""" graph = build_curriculum() registry = build_claim_registry() validate_curriculum(graph.skills, registry) rendered = render_graph(graph, registry) - return export_project(rendered, PROJECT_ID, PROJECT_NAME) + return export_project(rendered, project_id, project_name) def main() -> int: @@ -85,11 +87,11 @@ def main() -> int: ), ) parser.add_argument( - "--skip-seed-if-live-counts-ready", + "--skip-seed-if-live-content-ready", action="store_true", help=( - "After login, skip idempotent core seeding when live SkillTree already " - "reports the exported subject and skill counts." + "Skip core seeding only after live identities, content and quiz answer " + "correctness match the current export." ), ) parser.add_argument( @@ -173,7 +175,7 @@ def main() -> int: ) return 0 - project = None if args.no_seed else build_project() + project = None if args.no_seed else build_project(args.project_id, args.project_name) try: result = ensure_local_skilltree_auth( base_url=args.base_url, @@ -183,7 +185,7 @@ def main() -> int: project=project, seed_project=not args.no_seed, seed_dependencies=not args.skip_dependencies, - skip_seed_if_live_counts_ready=args.skip_seed_if_live_counts_ready, + skip_seed_if_live_content_ready=args.skip_seed_if_live_content_ready, verify_token=True, ) except (LocalSkillTreeAuthError, SkillTreeClientError) as exc: diff --git a/scripts/capture_dashboard_screenshots.py b/scripts/capture_dashboard_screenshots.py index 0fa0a3e..fb83378 100755 --- a/scripts/capture_dashboard_screenshots.py +++ b/scripts/capture_dashboard_screenshots.py @@ -37,6 +37,7 @@ def parse_args(argv: list[str] | None = None) -> argparse.Namespace: help="tracked directory receiving the captured PNGs and provenance record", ) parser.add_argument("--browser-command", default=DEFAULT_BROWSER_COMMAND) + parser.add_argument("--proof-path", type=Path, help="Explicit current v2 full-stack proof; older/unmatched proofs remain unassociated.") parser.add_argument("--viewport-width", type=int, default=VIEWPORT_WIDTH) parser.add_argument("--viewport-height", type=int, default=VIEWPORT_HEIGHT) return parser.parse_args(argv) @@ -51,6 +52,7 @@ def main(argv: list[str] | None = None) -> int: browser_command=args.browser_command, viewport_width=args.viewport_width, viewport_height=args.viewport_height, + proof_path=args.proof_path, ) try: provenance = capture_screenshots(settings) diff --git a/scripts/render_manuscript.py b/scripts/render_manuscript.py index 6d619dd..9879df0 100755 --- a/scripts/render_manuscript.py +++ b/scripts/render_manuscript.py @@ -31,8 +31,7 @@ from __future__ import annotations import argparse -import hashlib -import json +import os import shutil import subprocess import sys @@ -43,15 +42,19 @@ MANUSCRIPT = ROOT / "manuscript" OUTPUT_MANUSCRIPT = ROOT / "output" / "manuscript" OUTPUT_PDF = ROOT / "output" / "pdf" -PDF_NAME = "Active_Skillference_combined.pdf" -# Files the hydrator writes beside the sections that are not manuscript sections. -SECTION_EXCLUDE = {"config.yaml", "preamble.md", "README.md", "AGENTS.md", "SYNTAX.md"} # Log lines that mean the PDF would ship with broken references or citations. A LaTeX # "Font Warning: ... undefined" line must not be read as a broken reference, so a match # needs both the subject ("Citation"/"Reference") and the word "undefined". BROKEN_LOG_SUBJECTS = ("Citation", "Reference") +sys.path.insert(0, str(ROOT)) + +from src.manuscript.pdf_receipt import ( + PDF_NAME, add_pdf_assets, capture_pdf_inputs, hydrated_section_files, + latex_asset_paths, validate_pdf_inputs, write_pdf_receipt, +) + def _require(tool: str) -> str: found = shutil.which(tool) if found is None: @@ -167,12 +170,32 @@ def _titlepage(config: dict[str, object]) -> str: return "\n".join(lines) +def _page_two_visual(config: dict[str, object]) -> str: + front = config.get("front_matter") + if not isinstance(front, dict) or "page_two_visual" not in front: + return "" + visual = front["page_two_visual"] + if not isinstance(visual, dict): + raise SystemExit("front_matter.page_two_visual must be a mapping") + image_name, title, caption = (visual.get(key) for key in ("image", "title", "caption")) + if not all(isinstance(value, str) and value.strip() for value in (image_name, title, caption)): + raise SystemExit("front_matter.page_two_visual requires image, title and caption") + image = (OUTPUT_MANUSCRIPT / str(image_name)).resolve() + if not image.is_file(): + raise SystemExit(f"front_matter.page_two_visual.image does not resolve to a file: {image}") + return "\n".join([ + "```{=latex}", r"\thispagestyle{empty}", r"\begin{center}", + r"{\Large\bfseries " + _latex_escape(str(title)) + r"\par}", + r"\vspace{0.6em}", + r"\includegraphics[width=\textwidth,height=0.75\textheight,keepaspectratio]{" + + image.as_posix() + "}", + r"\par\vspace{0.6em}", _latex_escape(str(caption)), + r"\end{center}", r"\clearpage", "```", "", + ]) + + def _section_files() -> list[Path]: - return [ - path - for path in sorted(OUTPUT_MANUSCRIPT.glob("*.md")) - if path.name not in SECTION_EXCLUDE - ] + return hydrated_section_files(OUTPUT_MANUSCRIPT) def _assemble(config: dict[str, object], destination: Path) -> list[Path]: @@ -182,7 +205,10 @@ def _assemble(config: dict[str, object], destination: Path) -> list[Path]: f"no hydrated sections in {OUTPUT_MANUSCRIPT}; run " "`uv run python scripts/z_generate_manuscript_variables.py`" ) - parts = [_titlepage(config)] + parts = [ + _titlepage(config), _page_two_visual(config), + "```{=latex}\n\\setcounter{tocdepth}{2}\n\\tableofcontents\n\\clearpage\n```\n", + ] for index, path in enumerate(sections): if index: parts.append("```{=latex}\n\\newpage\n```\n") @@ -241,8 +267,8 @@ def _check_log(log: Path) -> None: ) -def _write_pdf_receipt(target: Path, *, root: Path = ROOT) -> Path: - """Record the finalized PDF bytes and the checkout's source revision.""" +def _write_pdf_receipt(target: Path, *, root: Path, inputs: dict[str, object]) -> Path: + """Record final PDF bytes and the actual pre-render input manifest.""" revision = subprocess.run( ["git", "rev-parse", "HEAD"], cwd=root, @@ -250,26 +276,14 @@ def _write_pdf_receipt(target: Path, *, root: Path = ROOT) -> Path: capture_output=True, text=True, ).stdout.strip() - content = target.read_bytes() - receipt = target.with_suffix(".receipt.json") - receipt.write_text( - json.dumps( - { - "schema": "active-skillference-pdf/v1", - "output_path": target.relative_to(root).as_posix(), - "size_bytes": len(content), - "sha256": hashlib.sha256(content).hexdigest(), - "source_revision": revision, - }, - indent=2, - ) - + "\n", - encoding="utf-8", + return write_pdf_receipt( + target, root=root, source_revision=revision, inputs=inputs, ) - return receipt -def _render_pdf(scratch: Path, config: dict[str, object]) -> Path: +def _render_pdf( + scratch: Path, config: dict[str, object], *, inputs: dict[str, object], +) -> Path: pandoc = _require("pandoc") crossref = _require("pandoc-crossref") xelatex = _require("xelatex") @@ -289,8 +303,6 @@ def _render_pdf(scratch: Path, config: dict[str, object]) -> Path: str(source), "--to=latex", "--standalone", - "--toc", - "--toc-depth=2", "--number-sections", "--filter", crossref, @@ -302,12 +314,17 @@ def _render_pdf(scratch: Path, config: dict[str, object]) -> Path: "--metadata", "documentclass=article", *_metadata(config), + "--variable", + "title=", "-o", str(tex), ], cwd=scratch, what="pandoc", ) + inputs = add_pdf_assets( + ROOT, inputs, latex_asset_paths(tex, build_directory=scratch, root=ROOT), + ) latex = [xelatex, "-interaction=nonstopmode", "-halt-on-error", tex.name] _run(latex, cwd=scratch, what="the first xelatex pass") @@ -315,14 +332,25 @@ def _render_pdf(scratch: Path, config: dict[str, object]) -> Path: _run(latex, cwd=scratch, what="the second xelatex pass") _run(latex, cwd=scratch, what="the final xelatex pass") _check_log(scratch / "combined.log") + validate_pdf_inputs(ROOT, inputs) produced = scratch / "combined.pdf" if not produced.is_file() or produced.stat().st_size == 0: raise SystemExit(f"xelatex reported success but wrote no PDF ({produced})") OUTPUT_PDF.mkdir(parents=True, exist_ok=True) target = OUTPUT_PDF / PDF_NAME - shutil.copy2(produced, target) - _write_pdf_receipt(target) + temporary: Path | None = None + try: + with tempfile.NamedTemporaryFile(dir=OUTPUT_PDF, prefix=f".{PDF_NAME}.", delete=False) as stream: + temporary = Path(stream.name) + with produced.open("rb") as source_stream: + shutil.copyfileobj(source_stream, stream) + shutil.copystat(produced, temporary) + os.replace(temporary, target) + finally: + if temporary is not None: + temporary.unlink(missing_ok=True) + _write_pdf_receipt(target, root=ROOT, inputs=inputs) print(f"rendered {len(sections)} sections -> {target.relative_to(ROOT)}") return target @@ -341,10 +369,11 @@ def main(argv: list[str] | None = None) -> int: _require("xelatex") _require("bibtex") _hydrate(args.skip_kernels) + inputs = capture_pdf_inputs(ROOT, _section_files()) config = _config() with tempfile.TemporaryDirectory(dir=ROOT, prefix=".render-") as scratch: - _render_pdf(Path(scratch), config) + _render_pdf(Path(scratch), config, inputs=inputs) return 0 diff --git a/scripts/run_kernels.py b/scripts/run_kernels.py index d84e940..1e83eb2 100755 --- a/scripts/run_kernels.py +++ b/scripts/run_kernels.py @@ -20,7 +20,10 @@ def main() -> list[Path]: - bundle = build_generated_outputs(_ROOT) + try: + bundle = build_generated_outputs(_ROOT) + except ValueError as exc: + raise SystemExit(str(exc)) from None print(f"registered {bundle.provenance_claims} provenance claims") print(f"validated scholarship coverage: {bundle.cited_sources} cited sources") print( diff --git a/scripts/run_learning_dashboard.sh b/scripts/run_learning_dashboard.sh index acecd8a..9d2074d 100755 --- a/scripts/run_learning_dashboard.sh +++ b/scripts/run_learning_dashboard.sh @@ -13,7 +13,6 @@ DASHBOARD_AUTHENTICATOR="${DASHBOARD_AUTHENTICATOR:-${SKILLTREE_AUTHENTICATOR:-} ACTIVE_SKILLFERENCE_LOCAL_AUTH="${ACTIVE_SKILLFERENCE_LOCAL_AUTH:-1}" ACTIVE_SKILLFERENCE_LEARNER_ID="${ACTIVE_SKILLFERENCE_LEARNER_ID:-local-learner}" ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR="${ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR:-.active_skillference_local}" -POSTGRES_PASSWORD_FILE="$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR/postgres_password" SEED_ARGS=(--base-url "$SKILLTREE_URL" --require-live) if ! command -v docker >/dev/null 2>&1; then @@ -34,19 +33,9 @@ if ! command -v curl >/dev/null 2>&1; then fi if [ -z "${ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD:-}" ]; then - mkdir -p "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" - chmod 700 "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" - if [ ! -s "$POSTGRES_PASSWORD_FILE" ]; then - ( - umask 077 - uv run python - <<'PY' > "$POSTGRES_PASSWORD_FILE" -import secrets - -print(secrets.token_urlsafe(32)) -PY - ) - fi - ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(cat "$POSTGRES_PASSWORD_FILE")" + ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(uv run python -c \ + 'import sys; from src.integration.full_app_stack import ensure_postgres_password; print(ensure_postgres_password(sys.argv[1]))' \ + "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR")" fi export ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD diff --git a/scripts/run_skilltree.sh b/scripts/run_skilltree.sh index 7d37407..9c3f9b3 100755 --- a/scripts/run_skilltree.sh +++ b/scripts/run_skilltree.sh @@ -16,7 +16,6 @@ SKILLTREE_WAIT_ATTEMPTS="${SKILLTREE_WAIT_ATTEMPTS:-240}" ACTIVE_SKILLFERENCE_LOCAL_AUTH="${ACTIVE_SKILLFERENCE_LOCAL_AUTH:-}" ACTIVE_SKILLFERENCE_LEARNER_ID="${ACTIVE_SKILLFERENCE_LEARNER_ID:-local-learner}" ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR="${ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR:-.active_skillference_local}" -POSTGRES_PASSWORD_FILE="$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR/postgres_password" if [ -z "$ACTIVE_SKILLFERENCE_LOCAL_AUTH" ]; then if [ -n "${SKILLTREE_TOKEN:-}" ]; then @@ -50,19 +49,9 @@ if [ "$ACTIVE_SKILLFERENCE_LOCAL_AUTH" != "1" ] && [ -z "${SKILLTREE_TOKEN:-}" ] fi if [ -z "${ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD:-}" ]; then - mkdir -p "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" - chmod 700 "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR" - if [ ! -s "$POSTGRES_PASSWORD_FILE" ]; then - ( - umask 077 - uv run python - <<'PY' > "$POSTGRES_PASSWORD_FILE" -import secrets - -print(secrets.token_urlsafe(32)) -PY - ) - fi - ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(cat "$POSTGRES_PASSWORD_FILE")" + ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD="$(uv run python -c \ + 'import sys; from src.integration.full_app_stack import ensure_postgres_password; print(ensure_postgres_password(sys.argv[1]))' \ + "$ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR")" fi export ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD diff --git a/scripts/serve_learner_dashboard.py b/scripts/serve_learner_dashboard.py index c2d6e86..55a540c 100755 --- a/scripts/serve_learner_dashboard.py +++ b/scripts/serve_learner_dashboard.py @@ -34,6 +34,7 @@ def main() -> int: parser.add_argument("--skilltree-url", default="http://localhost:8080") parser.add_argument("--host", default="127.0.0.1") parser.add_argument("--port", type=int, default=8765) + parser.add_argument("--instance-id", default="", help="Public identity for an owned proof run.") parser.add_argument( "--authenticator", default=os.environ.get("SKILLTREE_AUTHENTICATOR", ""), @@ -87,6 +88,7 @@ def main() -> int: local_auth_enabled=args.local_auth_enabled, local_state_dir=Path(args.local_state_dir), learner_id=args.learner_id, + instance_id=args.instance_id, ) if args.check: print(json.dumps(dashboard_payload(config), indent=2, sort_keys=True)) diff --git a/scripts/validate_dashboard_console.py b/scripts/validate_dashboard_console.py index b1dfd0b..e4c49c3 100755 --- a/scripts/validate_dashboard_console.py +++ b/scripts/validate_dashboard_console.py @@ -1,42 +1,32 @@ #!/usr/bin/env python3 -"""Opt-in browser smoke gate: fail on unexpected dashboard console errors. - -The server-side dashboard tests cannot see what the learner's browser logs. This gate -drives a real browser (chrome-devtools-axi) against a running dashboard and fails if any -console error appears EXCEPT the one known, cosmetic upstream probe: the -``@skilltree/skills-client-js`` SDK checks service status against SkillTree's built-in -default ``Inception`` project, which returns 401 under our project-scoped authenticator. -That probe is non-blocking (the real ``activeInference`` learner view still renders) and -is not fixable from our configuration, so it is allow-listed and everything else fails. - -Requires a running dashboard and the ``chrome-devtools-axi`` CLI. By default, if either -is absent the gate reports ``skipped`` (not ``clean``) and exits 0, matching the project -convention that optional local browser tooling must not break offline checks. Strict -callers can pass ``--require-browser`` and ``--require-dashboard`` to fail closed. +"""Opt-in real-browser proof of official project, subject and quiz views. + +The gate waits for the official SDK's actual route callback, a visible frame +on the configured SkillTree origin, and matching browser-visible content. +Fallback and unresolved pending states fail even when the console is empty. +Only the established cosmetic upstream Inception probe is allow-listed. +Absent tooling/dashboard remains an explicit optional skip unless required. """ from __future__ import annotations import argparse -import re +import json +import sys +from pathlib import Path import shutil -import subprocess -import time import urllib.error import urllib.request from urllib.parse import urlparse -# Console errors that are known, cosmetic, and outside our control are allow-listed here. -_ALLOWED_ERROR_PATTERNS = ( - re.compile(r"projects/Inception/.*token.*401", re.IGNORECASE), - re.compile(r"Unable to authenticate.*Inception", re.IGNORECASE), -) -# This script deliberately stays import-free from the project's src/ package (no -# sys.path bootstrap), so the http(s)-only guard is inlined rather than shared via -# src.net_guard; keep both checks in sync (see also src/net_guard.py). _ALLOWED_URL_SCHEMES = frozenset({"http", "https"}) +_ROOT = Path(__file__).resolve().parents[1] +sys.path.insert(0, str(_ROOT)) + +from src.dashboard.browser_validation import BrowserDisplayError, validate_dashboard_browser + def _dashboard_reachable(url: str, timeout: float) -> bool: if urlparse(url).scheme not in _ALLOWED_URL_SCHEMES: @@ -49,42 +39,16 @@ def _dashboard_reachable(url: str, timeout: float) -> bool: return False -def _console_errors(url: str, settle_seconds: float) -> list[str]: - subprocess.run( - ["chrome-devtools-axi", "open", url], - check=True, - capture_output=True, - text=True, - timeout=60, - ) - time.sleep(settle_seconds) - result = subprocess.run( - ["chrome-devtools-axi", "console", "--type", "error"], - check=True, - capture_output=True, - text=True, - timeout=60, - ) - errors: list[str] = [] - for line in result.stdout.splitlines(): - if "[error]" in line: - errors.append(line.strip()) - return errors -def unexpected_errors(errors: list[str]) -> list[str]: - """Return console errors that are not on the cosmetic-upstream allow-list.""" - return [ - error - for error in errors - if not any(pattern.search(error) for pattern in _ALLOWED_ERROR_PATTERNS) - ] def main() -> int: parser = argparse.ArgumentParser(description=__doc__) parser.add_argument("--dashboard-url", default="http://127.0.0.1:8765/") parser.add_argument("--settle-seconds", type=float, default=5.0) + parser.add_argument("--timeout", type=float, default=45.0) + parser.add_argument("--json", action="store_true", help="Emit the structured browser proof.") parser.add_argument( "--require-browser", action="store_true", @@ -114,18 +78,20 @@ def main() -> int: print(message) return 0 - errors = _console_errors(args.dashboard_url, args.settle_seconds) - unexpected = unexpected_errors(errors) - if unexpected: - print("dashboard console gate FAILED — unexpected console errors:") - for error in unexpected: - print(f" - {error}") + try: + summary = validate_dashboard_browser( + args.dashboard_url, timeout=args.timeout, settle_seconds=args.settle_seconds, + ) + except BrowserDisplayError as exc: + if args.json: + print(json.dumps({"ok": False, "checked": False, "error": str(exc)})) + else: + print(f"dashboard console gate FAILED: {exc}") return 1 - allowed = len(errors) - len(unexpected) - print( - f"dashboard console gate OK (no unexpected errors; {allowed} known " - "upstream Inception probe message(s) allow-listed)" - ) + if args.json: + print(json.dumps(summary, sort_keys=True)) + else: + print("dashboard console gate OK (official project, subject and quiz views ready)") return 0 diff --git a/scripts/validate_rendered_freshness.py b/scripts/validate_rendered_freshness.py index 32e89c4..d9f507a 100755 --- a/scripts/validate_rendered_freshness.py +++ b/scripts/validate_rendered_freshness.py @@ -5,8 +5,8 @@ Count-bearing prose is substituted from ``output/data/manuscript_variables.json`` at render time, so if the curriculum is regenerated but the manuscript is not re-rendered, the rendered ``.md`` (and the PDF) silently keep stale numbers while every other gate -stays green. This gate re-renders each source file from the *current* variables and -asserts the committed rendered copy is byte-identical, catching exactly that drift. +stays green. This gate re-renders each source file from the current variables, +then validates finalized PDF bytes and the complete bound render-input manifest. """ from __future__ import annotations @@ -23,6 +23,7 @@ sys.path.insert(0, str(ROOT / "scripts")) import z_generate_manuscript_variables as zg +from src.manuscript.pdf_receipt import pdf_receipt_issues ROADMAP_OPEN_RE = re.compile( r"(?P\d+)\s+open items across\s+(?P[^.]+)", @@ -174,6 +175,12 @@ def main() -> int: for item in stale: print(f" - {item}") return 1 + pdf_stale = pdf_receipt_issues(ROOT) + if pdf_stale: + print("PDF freshness FAILED (a genuine re-render is required):") + for item in pdf_stale: + print(f" - {item}") + return 1 bundle_stale = check_publication_bundle_roadmap_freshness() if bundle_stale: print("publication bundle freshness FAILED (re-render before publishing):") @@ -183,7 +190,7 @@ def main() -> int: rendered = sorted(zg.OUTPUT_MANUSCRIPT.glob("*.md")) print( f"rendered manuscript freshness OK ({len(rendered)} files match current " - "artifacts; publication roadmap text matches)" + "artifacts; PDF bytes/inputs and publication roadmap text match)" ) return 0 diff --git a/scripts/z_generate_manuscript_variables.py b/scripts/z_generate_manuscript_variables.py index 55411d2..9031c4b 100755 --- a/scripts/z_generate_manuscript_variables.py +++ b/scripts/z_generate_manuscript_variables.py @@ -40,7 +40,9 @@ def replace_claim(match: re.Match[str]) -> str: def replace_variable(match: re.Match[str]) -> str: key = match.group(1) - return variables.get(key, match.group(0)) + if key not in variables: + raise KeyError(f"unresolved manuscript variable: {key}") + return variables[key] return VARIABLE_TOKEN_RE.sub( replace_variable, CLAIM_TOKEN_RE.sub(replace_claim, text) @@ -52,15 +54,13 @@ def _render_markdown(text: str, variables: dict[str, str]) -> str: cursor = 0 for match in CODE_SPAN_RE.finditer(text): parts.append(_render_prose(text[cursor : match.start()], variables)) - parts.append(match.group(0)) + # Only code examples lose placeholder syntax; prose must resolve above. + code = CLAIM_TOKEN_RE.sub(lambda item: f"claim:{item.group(1)}", match.group(0)) + parts.append(VARIABLE_TOKEN_RE.sub(lambda item: item.group(1), code)) cursor = match.end() parts.append(_render_prose(text[cursor:], variables)) rendered = "".join(parts) - # Code-span examples explain token syntax. Keep them readable without - # leaving literal double-brace placeholders in rendered public surfaces. - rendered = CLAIM_TOKEN_RE.sub(lambda m: f"claim:{m.group(1)}", rendered) - rendered = VARIABLE_TOKEN_RE.sub(lambda m: m.group(1), rendered) leftovers = sorted(set(ANY_TOKEN_RE.findall(rendered))) if leftovers: raise ValueError(f"unresolved manuscript tokens: {', '.join(leftovers)}") diff --git a/src/artifacts/json_io.py b/src/artifacts/json_io.py index 0f31039..2fb8be8 100644 --- a/src/artifacts/json_io.py +++ b/src/artifacts/json_io.py @@ -1,6 +1,9 @@ from __future__ import annotations import json +import os +import stat +import tempfile from pathlib import Path from typing import Any, cast @@ -30,17 +33,23 @@ def write_json_artifact( ensure_ascii: bool = True, ) -> Path: path = Path(path) + content = json.dumps(payload, indent=2, sort_keys=True, ensure_ascii=ensure_ascii) + "\n" path.parent.mkdir(parents=True, exist_ok=True) - path.write_text( - json.dumps( - payload, - indent=2, - sort_keys=True, - ensure_ascii=ensure_ascii, - ) - + "\n", - encoding="utf-8", - ) + mode = stat.S_IMODE(path.stat().st_mode) if path.exists() else None + temporary: Path | None = None + try: + with tempfile.NamedTemporaryFile( + mode="w", encoding="utf-8", dir=path.parent, + prefix=f".{path.name}.", suffix=".tmp", delete=False, + ) as stream: + temporary = Path(stream.name) + stream.write(content) + if mode is not None: + temporary.chmod(mode) + os.replace(temporary, path) + finally: + if temporary is not None: + temporary.unlink(missing_ok=True) return path diff --git a/src/curriculum/_ext_dynamical_geometry.py b/src/curriculum/_ext_dynamical_geometry.py index f797b95..b996c54 100644 --- a/src/curriculum/_ext_dynamical_geometry.py +++ b/src/curriculum/_ext_dynamical_geometry.py @@ -178,12 +178,13 @@ "Measure local distinguishability of distributions with the Fisher metric.", "The Fisher information matrix " "$g_{ij}(\\theta)=\\mathbb{E}_{p_\\theta}[\\partial_i\\log p_\\theta\\,\\partial_j\\log p_\\theta]$ " - "is the unique (up to scale) Riemannian metric on a statistical manifold invariant under " - "reparameterisation. It tells you how far apart two nearby distributions are in information " - "terms.", + "defines a Riemannian metric for a regular model with nonsingular Fisher information. " + "Its line element is invariant under smooth invertible reparameterisation, but this " + "coordinate invariance alone does not establish uniqueness. The metric measures " + "the local distinguishability of nearby distributions.", ("geo_statistical_manifold",), _q( - "The Fisher information matrix plays the role of what on a statistical manifold?", + "For a regular model with nonsingular Fisher information, the Fisher matrix plays which role?", ["A Riemannian metric", "A prior", "A loss function"], 0, ), diff --git a/src/curriculum/_ext_learning_hierarchical.py b/src/curriculum/_ext_learning_hierarchical.py index 7cbd27d..9205d58 100644 --- a/src/curriculum/_ext_learning_hierarchical.py +++ b/src/curriculum/_ext_learning_hierarchical.py @@ -387,7 +387,7 @@ "Prediction-Error Message Passing Across Levels", _AI, "Propagate precision-weighted prediction errors upward and predictions downward to perform inference over the whole hierarchy.", - "Hierarchical inference is implemented by message passing: each level sends top-down predictions to the level below and returns bottom-up prediction errors to the level above. Under a Gaussian (Laplace) approximation these messages are precision-weighted prediction errors $\\xi^{(\\ell)} = \\Pi^{(\\ell)}\\big(\\mu^{(\\ell)} - g(\\mu^{(\\ell+1)})\\big)$, and belief updating ascends the free-energy gradient until errors are minimised across all levels simultaneously. Only the residual that a level cannot explain is forwarded upward, so well-predicted signals are 'explained away' and do not propagate. This reciprocal exchange is the computational core shared by predictive coding and discrete hierarchical active inference.", + "Hierarchical inference is implemented by message passing: each level sends top-down predictions to the level below and returns bottom-up prediction errors to the level above. Under a Gaussian (Laplace) approximation these messages are precision-weighted prediction errors $\\xi^{(\\ell)} = \\Pi^{(\\ell)}\\big(\\mu^{(\\ell)} - g(\\mu^{(\\ell+1)})\\big)$, and belief updating follows negative free-energy gradients to reduce the joint variational objective. Only the residual that a level cannot explain is forwarded upward, so well-predicted signals are 'explained away' and do not propagate. This reciprocal exchange is the computational core shared by predictive coding and discrete hierarchical active inference.", ("hier_empirical_priors", "vm_message_passing", "ai_predictive_coding"), _q( "What is forwarded UP the hierarchy during inference?", diff --git a/src/curriculum/_ext_precision_exploration.py b/src/curriculum/_ext_precision_exploration.py index 57022a9..7bb93bc 100644 --- a/src/curriculum/_ext_precision_exploration.py +++ b/src/curriculum/_ext_precision_exploration.py @@ -112,7 +112,7 @@ "Guidance Functions and Risk-Sensitive Policies", _AI, "Distinguish risk (divergence from preferences) from ambiguity in policy scoring.", - "The guidance view decomposes the pragmatic part of expected free energy into a risk " + "The guidance view decomposes total expected free energy into a risk " "term, the divergence between predicted and preferred outcomes, and an ambiguity term, " "the expected uncertainty of the likelihood. Shaping the preference distribution C is " "therefore how a designer steers behaviour: tightening C makes the policy risk-averse " @@ -146,9 +146,13 @@ "Rate-distortion theory asks for the least information an agent must keep to hold " "expected distortion below a bound; expected free energy makes the mirror trade-off, " "balancing the epistemic value of an observation against the pragmatic value of a " - "preferred outcome. Both say a bounded agent should pay for information only to the " - "extent it buys reduced cost, which is why precision behaves like the slope that " - "sets the exchange rate between bits gathered and value pursued.", + "preferred outcome. A rate-distortion multiplier can set a model's relative trade-off. " + "Policy precision is different: in $q(\\pi)=\\sigma(-\\gamma G(\\pi))$, $\\gamma$ " + "scales total expected free energy and controls policy-selection sharpness for fixed " + "scores, not the relative coefficients of epistemic and pragmatic components. " + "Under the kernel's signed-value convention, pragmatic cost is the negative of " + "pragmatic value. This policy-precision statement does not identify every model or " + "sensory precision with the same role.", ("ai_expected_free_energy", "im_rate_distortion", "ai_epistemic_value"), _q( "Expected free energy and rate-distortion theory share which structure?", @@ -160,11 +164,11 @@ 0, ), _q( - "In this reading, precision plays the role of:", + "For fixed EFE scores, increasing policy precision $\\gamma$ does what?", [ - "The slope setting the exchange rate between information and value", - "A fixed constant unrelated to the trade-off", - "The number of policies considered", + "Concentrates probability on policies with lower total expected free energy", + "Changes only the epistemic coefficient relative to the pragmatic coefficient", + "Leaves relative policy probabilities unchanged", ], 0, ), diff --git a/src/curriculum/catalog_data/core_strata.py b/src/curriculum/catalog_data/core_strata.py index c5c6d64..bcff3e3 100644 --- a/src/curriculum/catalog_data/core_strata.py +++ b/src/curriculum/catalog_data/core_strata.py @@ -84,8 +84,9 @@ "Eigenvalues & Positive-Definiteness", _M, "Interpret eigenvalues and positive-definite matrices.", - "For $A\\mathbf{v}=\\lambda\\mathbf{v}$, $\\lambda$ is an eigenvalue; covariance and " - "precision matrices are symmetric positive-definite.", + "For $A\\mathbf{v}=\\lambda\\mathbf{v}$, $\\lambda$ is an eigenvalue. Covariance matrices " + "are symmetric positive-semidefinite; a nonsingular covariance matrix is " + "positive-definite, and its inverse precision matrix is also symmetric positive-definite.", ("m_matrix",), _q( "A covariance matrix is always:", @@ -147,7 +148,7 @@ "partials and certifies local curvature/convexity.", ("m_chain_rule", "m_eigen"), _q( - "A function is locally convex where its Hessian is:", + "For a twice continuously differentiable function, which Hessian condition is sufficient for local strict convexity?", ["Positive-definite", "Negative-definite", "Singular"], 0, ), @@ -157,8 +158,12 @@ "Convexity", _M, "Recognise convex functions and sets.", - "A convex function lies below its chords: $f(\\lambda x+(1-\\lambda)y)\\le \\lambda f(x)+(1-\\lambda)f(y)$. " - "Convexity guarantees a unique minimum — and powers Jensen's inequality later.", + "A convex function lies below its chords for $0\\le\\lambda\\le1$: " + "$f(\\lambda x+(1-\\lambda)y)\\le \\lambda f(x)+(1-\\lambda)f(y)$. " + "Every local minimum is global, but convexity alone guarantees neither attainment " + "nor uniqueness. Strict convexity gives at most one minimizer; existence requires " + "additional conditions, such as continuity on a nonempty compact feasible set. " + "Convexity also powers Jensen's inequality.", ("m_jacobian_hessian",), _q( "A convex function's graph lies:", diff --git a/src/curriculum/graph.py b/src/curriculum/graph.py index 65bf895..cdb25f1 100644 --- a/src/curriculum/graph.py +++ b/src/curriculum/graph.py @@ -13,6 +13,7 @@ from __future__ import annotations from dataclasses import dataclass +from heapq import heapify, heappop, heappush from .model import ALL_STRATA, CurriculumError, Skill, Stratum, Subject @@ -147,9 +148,10 @@ def _check_subject_identity(self) -> None: "SkillTree resolves skills and subjects in one id namespace" ) - def _check_prerequisites_exist(self) -> None: - ids = self.skill_ids() - for skill in self.skills: + def _check_prerequisites_exist(self, skills: list[Skill] | None = None) -> None: + skills = self.skills if skills is None else skills + ids = {skill.skill_id for skill in skills} + for skill in skills: for prereq in skill.prerequisites: if prereq not in ids: raise CurriculumError( @@ -211,25 +213,26 @@ def _check_acyclic(self) -> None: # -- ordering / depth --------------------------------------------------- def topological_order(self) -> list[Skill]: - """Kahn's algorithm — every skill appears after all its prerequisites.""" - self._check_prerequisites_exist() - by_id = {s.skill_id: s for s in self.skills} + """Kahn's algorithm, choosing the smallest currently ready skill id.""" + skills = self.skills + self._check_prerequisites_exist(skills) + by_id = {skill.skill_id: skill for skill in skills} indegree = {sid: 0 for sid in by_id} dependents: dict[str, list[str]] = {sid: [] for sid in by_id} - for prereq, dependent in self.prerequisite_edges(): - indegree[dependent] += 1 - dependents[prereq].append(dependent) - # Deterministic: process ready nodes in sorted id order. - ready = sorted(sid for sid, d in indegree.items() if d == 0) + for skill in skills: + for prereq in skill.prerequisites: + indegree[skill.skill_id] += 1 + dependents[prereq].append(skill.skill_id) + ready = [sid for sid, degree in indegree.items() if degree == 0] + heapify(ready) order: list[Skill] = [] while ready: - sid = ready.pop(0) + sid = heappop(ready) order.append(by_id[sid]) for dep in dependents[sid]: indegree[dep] -= 1 if indegree[dep] == 0: - ready.append(dep) - ready.sort() + heappush(ready, dep) if len(order) != len(by_id): raise CurriculumError("topological_order failed — graph is cyclic") return order @@ -256,15 +259,16 @@ def core_max_depth(self) -> int: implementation and cognitive-security families — giving the depth of the actual concept chain a learner must climb, separate from review sequencing. """ + ordered = self.topological_order() core = { skill.skill_id - for skill in self.skills + for skill in ordered if skill.skill_id.split("_")[0] not in _SCAFFOLD_SKILL_PREFIXES } if not core: return 0 level: dict[str, int] = {} - for skill in self.topological_order(): + for skill in ordered: if skill.skill_id not in core: continue core_prereqs = [p for p in skill.prerequisites if p in core] @@ -284,12 +288,15 @@ def reachable_from_foundations(self) -> set[str]: correctly reported as disconnected, rather than being trivially reachable from that illegal root. """ - by_id = {s.skill_id: s for s in self.skills} + skills = self.skills + by_id = {skill.skill_id: skill for skill in skills} forward: dict[str, list[str]] = {sid: [] for sid in by_id} - for prereq, dependent in self.prerequisite_edges(): - forward[prereq].append(dependent) + for skill in skills: + for prereq in skill.prerequisites: + forward[prereq].append(skill.skill_id) roots = [ - s.skill_id for s in self.skills if s.is_root and s.stratum.is_foundational + skill.skill_id for skill in skills + if skill.is_root and skill.stratum.is_foundational ] seen: set[str] = set() stack = list(roots) @@ -330,19 +337,36 @@ def _check_all_skills_reachable_from_foundations(self) -> None: # -- reporting ---------------------------------------------------------- def stats(self) -> dict[str, object]: - levels = self.levels() + ordered = self.topological_order() + levels: dict[str, int] = {} + core_levels: dict[str, int] = {} + core_ids = { + skill.skill_id for skill in ordered + if skill.skill_id.split("_")[0] not in _SCAFFOLD_SKILL_PREFIXES + } + skills_per_stratum = {stratum.value: 0 for stratum in ALL_STRATA} + for skill in ordered: + levels[skill.skill_id] = ( + 1 + max(levels[prereq] for prereq in skill.prerequisites) + if skill.prerequisites else 0 + ) + skills_per_stratum[skill.stratum.value] += 1 + if skill.skill_id in core_ids: + core_prereqs = [prereq for prereq in skill.prerequisites if prereq in core_ids] + core_levels[skill.skill_id] = ( + 1 + max(core_levels[prereq] for prereq in core_prereqs) + if core_prereqs else 0 + ) + strata = {subject.stratum for subject in self.subjects} return { "subjects": len(self.subjects), - "skills": len(self.skills), - "prerequisite_edges": len(self.prerequisite_edges()), - "strata_present": sorted(s.value for s in self.strata_present()), - "strata_count": len(self.strata_present()), - "all_strata_present": self.strata_present() == set(ALL_STRATA), - "max_depth": max(levels.values()) if levels else 0, - "core_max_depth": self.core_max_depth(), - "roots": sum(1 for s in self.skills if s.is_root), - "skills_per_stratum": { - stratum.value: sum(1 for s in self.skills if s.stratum is stratum) - for stratum in ALL_STRATA - }, + "skills": len(ordered), + "prerequisite_edges": sum(len(skill.prerequisites) for skill in ordered), + "strata_present": sorted(stratum.value for stratum in strata), + "strata_count": len(strata), + "all_strata_present": strata == set(ALL_STRATA), + "max_depth": max(levels.values(), default=0), + "core_max_depth": max(core_levels.values(), default=0), + "roots": sum(skill.is_root for skill in ordered), + "skills_per_stratum": skills_per_stratum, } diff --git a/src/curriculum/materials/builders.py b/src/curriculum/materials/builders.py index 3b16b8f..abd7036 100644 --- a/src/curriculum/materials/builders.py +++ b/src/curriculum/materials/builders.py @@ -40,6 +40,7 @@ def build_course_material_index( registry: ClaimRegistry | None = None, ) -> dict[str, object]: """Build one modular material record per skill in the curriculum graph.""" + registry = registry if registry is not None else ClaimRegistry() source_index = { str(source["source_id"]): source for source in source_payload.get("sources", []) @@ -53,14 +54,14 @@ def build_course_material_index( for skill in graph.topological_order(): subject = subject_by_skill[skill.skill_id] - pointers = _pointers_for_skill(skill, source_payload, source_index) + pointers = _pointers_for_skill(skill, source_payload, source_index, registry) prerequisites = [ - _skill_ref(skill_by_id[prereq], subject_by_skill[prereq], levels[prereq]) + _skill_ref(skill_by_id[prereq], subject_by_skill[prereq], levels[prereq], registry) for prereq in skill.prerequisites ] dependents = [ _skill_ref( - skill_by_id[dependent], subject_by_skill[dependent], levels[dependent] + skill_by_id[dependent], subject_by_skill[dependent], levels[dependent], registry, ) for dependent in downstream.get(skill.skill_id, ()) ] @@ -72,9 +73,9 @@ def build_course_material_index( ) skills[skill.skill_id] = { "skillId": skill.skill_id, - "name": skill.name, + "name": _render_text(skill.name, registry), "subjectId": subject.subject_id, - "subjectName": subject.name, + "subjectName": _render_text(subject.name, registry), "stratum": skill.stratum.value, "depth": levels[skill.skill_id], "objective": _render_text(skill.objective, registry), @@ -87,7 +88,7 @@ def build_course_material_index( ), "prerequisites": prerequisites, "relatedSkills": _related_skills( - skill, skill_by_id, subject_by_skill, levels + skill, skill_by_id, subject_by_skill, levels, registry, ), "downstreamSkills": dependents, "assessmentSummary": _assessment_summary(skill), @@ -110,7 +111,7 @@ def build_course_material_index( "externalMaterialPolicy": source_payload.get("policy", {}), }, "summary": {}, - "sources": _source_summaries(source_index), + "sources": _source_summaries(source_index, registry), "skills": skills, } normalized = normalize_course_material_index(payload) diff --git a/src/curriculum/materials/github_sources.py b/src/curriculum/materials/github_sources.py index 54fa8be..dbbd62e 100644 --- a/src/curriculum/materials/github_sources.py +++ b/src/curriculum/materials/github_sources.py @@ -11,37 +11,11 @@ from src.net_guard import require_http_url -from .models import CourseMaterialIssue - - -def _required_string(mapping: dict[str, Any], key: str) -> str: - value = mapping.get(key) - return value if isinstance(value, str) else "" +from .helpers import _iter_configured_pointers, _required_string - -def _as_pointer_list(value: Any) -> list[dict[str, Any]]: - if not isinstance(value, list): - return [] - return [item for item in value if isinstance(item, dict)] +from .models import CourseMaterialIssue -def _iter_configured_pointers( - payload: dict[str, Any], -) -> list[tuple[str, dict[str, Any]]]: - pointers: list[tuple[str, dict[str, Any]]] = [] - stratum_mappings = payload.get("stratum_mappings") - if isinstance(stratum_mappings, dict): - for stratum_key, raw in stratum_mappings.items(): - for pointer in _as_pointer_list(raw): - pointers.append((f"stratum_mappings.{stratum_key}", pointer)) - topic_mappings = payload.get("topic_mappings") - if isinstance(topic_mappings, list): - for topic_idx, topic in enumerate(topic_mappings): - if not isinstance(topic, dict): - continue - for pointer in _as_pointer_list(topic.get("pointers")): - pointers.append((f"topic_mappings[{topic_idx}].pointers", pointer)) - return pointers def _validate_online_trees( @@ -117,7 +91,7 @@ def _github_tree_payload( with urllib.request.urlopen(request, timeout=timeout) as response: payload = json.loads(response.read().decode("utf-8")) if not isinstance(payload, dict): - raise TypeError("GitHub response was not a JSON object") + raise ValueError("GitHub response was not a JSON object") return payload except urllib.error.HTTPError as exc: last_exc = exc @@ -156,7 +130,7 @@ def _fetch_github_tree_paths( payload = _github_tree_payload(request, timeout=timeout) tree = payload.get("tree") if isinstance(payload, dict) else None if not isinstance(tree, list): - raise TypeError("GitHub response did not include a tree") + raise ValueError("GitHub response did not include a tree") return { str(item.get("path")) for item in tree diff --git a/src/curriculum/materials/helpers.py b/src/curriculum/materials/helpers.py index 8d52230..cff6fad 100644 --- a/src/curriculum/materials/helpers.py +++ b/src/curriculum/materials/helpers.py @@ -209,6 +209,7 @@ def _pointers_for_skill( skill: Skill, payload: dict[str, Any], source_index: dict[str, dict[str, Any]], + registry: ClaimRegistry, ) -> list[dict[str, object]]: raw_pointers: list[tuple[dict[str, Any], str, tuple[str, ...]]] = [] stratum_mappings = payload.get("stratum_mappings", {}) @@ -254,13 +255,13 @@ def _pointers_for_skill( repo_url = _required_string(source, "repo_url").rstrip("/") merged[key] = { "sourceId": source_id, - "sourceTitle": _required_string(source, "title"), - "title": _required_string(pointer, "title"), - "role": role, + "sourceTitle": _render_text(_required_string(source, "title"), registry), + "title": _render_text(_required_string(pointer, "title"), registry), + "role": _render_text(role, registry), "path": path, "commit": commit, "url": f"{repo_url}/blob/{commit}/{path}", - "licenseStatus": _required_string(source, "license_status"), + "licenseStatus": _render_text(_required_string(source, "license_status"), registry), } scope_sets[key].add(match_scope) term_sets[key].update(matched_terms) @@ -277,7 +278,7 @@ def _sections_for_skill( prerequisites: list[dict[str, object]], dependents: list[dict[str, object]], pointers: list[dict[str, object]], - registry: ClaimRegistry | None, + registry: ClaimRegistry, ) -> list[dict[str, str]]: prereq_body = ( "This root skill can be entered without a prior local prerequisite." @@ -308,11 +309,11 @@ def _sections_for_skill( }, { "heading": "Prerequisite Context", - "body": prereq_body, + "body": _render_text(prereq_body, registry), }, { "heading": "Downstream Role", - "body": downstream_body, + "body": _render_text(downstream_body, registry), }, { "heading": "Assessment Contract", @@ -366,6 +367,7 @@ def _related_skills( skill_by_id: dict[str, Skill], subject_by_skill: dict[str, Subject], levels: dict[str, int], + registry: ClaimRegistry, ) -> list[dict[str, object]]: related_ids = list(skill.related) for prereq in skill.prerequisites: @@ -381,17 +383,20 @@ def _related_skills( related_skill, subject_by_skill[related_skill.skill_id], levels[related_skill.skill_id], + registry, ) ) return rows -def _skill_ref(skill: Skill, subject: Subject, depth: int) -> dict[str, object]: +def _skill_ref( + skill: Skill, subject: Subject, depth: int, registry: ClaimRegistry, +) -> dict[str, object]: return { "skillId": skill.skill_id, - "name": skill.name, + "name": _render_text(skill.name, registry), "subjectId": subject.subject_id, - "subjectName": subject.name, + "subjectName": _render_text(subject.name, registry), "stratum": skill.stratum.value, "depth": depth, } @@ -410,18 +415,20 @@ def _downstream_map(graph: CurriculumGraph) -> dict[str, tuple[str, ...]]: } -def _source_summaries(source_index: dict[str, dict[str, Any]]) -> list[dict[str, str]]: +def _source_summaries( + source_index: dict[str, dict[str, Any]], registry: ClaimRegistry, +) -> list[dict[str, str]]: rows: list[dict[str, str]] = [] for source_id in sorted(source_index): source = source_index[source_id] rows.append( { "sourceId": source_id, - "title": _required_string(source, "title"), + "title": _render_text(_required_string(source, "title"), registry), "repoUrl": _required_string(source, "repo_url"), "commit": _required_string(source, "commit"), - "licenseStatus": _required_string(source, "license_status"), - "statusNote": _required_string(source, "status_note"), + "licenseStatus": _render_text(_required_string(source, "license_status"), registry), + "statusNote": _render_text(_required_string(source, "status_note"), registry), } ) return rows @@ -502,5 +509,5 @@ def _write_json(path: Path, payload: object) -> None: write_json_artifact(path, payload) -def _render_text(text: str, registry: ClaimRegistry | None) -> str: - return render(text, registry) if registry is not None else text +def _render_text(text: str, registry: ClaimRegistry) -> str: + return render(text, registry) diff --git a/src/curriculum/materials/validation.py b/src/curriculum/materials/validation.py index 3398efb..dacfd8b 100644 --- a/src/curriculum/materials/validation.py +++ b/src/curriculum/materials/validation.py @@ -8,6 +8,8 @@ from src.curriculum.graph import CurriculumGraph from src.curriculum.model import ALL_STRATA from src.curriculum.source_contracts import GITHUB_COMMIT_RE +from src.provenance.binding import ProvenanceError, render +from src.provenance.claims import ClaimRegistry from .github_sources import _github_repo_parts, _validate_online_trees from .helpers import ( @@ -217,6 +219,24 @@ def audit_course_material_index( ) -> CourseMaterialAudit: """Check that every skill has non-empty material and external pointers.""" issues: list[CourseMaterialIssue] = [] + registry = ClaimRegistry() + pending: list[tuple[str, object]] = [("materials", index)] + while pending: + location, value = pending.pop() + if isinstance(value, str): + try: + render(value, registry) + except ProvenanceError: + issues.append( + CourseMaterialIssue( + "unresolved-claim-token", location, + "learner-facing material contains an unresolved or malformed claim token", + ) + ) + elif isinstance(value, dict): + pending.extend((f"{location}.{key}", child) for key, child in value.items()) + elif isinstance(value, (list, tuple)): + pending.extend((f"{location}[{idx}]", child) for idx, child in enumerate(value)) normalized = normalize_course_material_index(index) if not normalized.skills_is_mapping: issues.append( @@ -282,14 +302,6 @@ def audit_course_material_index( ) ) continue - if "{{claim:" in str(section.get("body", "")): - issues.append( - CourseMaterialIssue( - "unresolved-claim-token", - f"skills.{skill_id}.sections[{section_idx}]", - "section body contains an unresolved claim token", - ) - ) if not isinstance(pointers, list) or not pointers: issues.append( CourseMaterialIssue( diff --git a/src/curriculum/model.py b/src/curriculum/model.py index f3615c0..7ccd368 100644 --- a/src/curriculum/model.py +++ b/src/curriculum/model.py @@ -15,6 +15,7 @@ from __future__ import annotations import re +import unicodedata from dataclasses import dataclass, field from enum import Enum from typing import Literal @@ -22,6 +23,16 @@ # ``{{claim:NAME}}`` token used in skill descriptions for provenance-bound values. CLAIM_TOKEN_RE = re.compile(r"\{\{claim:([A-Za-z0-9_.\-]+)\}\}") + +def _strip_format_chars(text: str) -> str: + """Normalize invisible Unicode format characters before claim admission/collection. + + Format characters can split an otherwise identical token or result number. + Rendering, validation, linting and original-token provenance share this policy. + """ + return "".join(character for character in text if unicodedata.category(character) != "Cf") + + SelfReportingType = Literal["HonorSystem", "Approval", "Quiz"] _VALID_SELF_REPORTING = ("HonorSystem", "Approval", "Quiz") QuestionType = Literal["SingleChoice", "MultipleChoice", "TextInput"] @@ -148,7 +159,7 @@ def claim_tokens(self) -> set[str]: texts.extend(question.answers) tokens: set[str] = set() for text in texts: - tokens.update(CLAIM_TOKEN_RE.findall(text)) + tokens.update(CLAIM_TOKEN_RE.findall(_strip_format_chars(text))) return tokens @property diff --git a/src/curriculum/quality.py b/src/curriculum/quality.py index 5dfee98..8250dbc 100644 --- a/src/curriculum/quality.py +++ b/src/curriculum/quality.py @@ -72,7 +72,8 @@ def _word_count(text: str) -> int: def audit_curriculum_quality(graph: CurriculumGraph) -> list[CurriculumQualityIssue]: """Return learner-quality issues that structural validation does not cover.""" issues: list[CurriculumQualityIssue] = [] - by_id = {skill.skill_id: skill for skill in graph.skills} + skills = graph.skills + by_id = {skill.skill_id: skill for skill in skills} for subject in graph.subjects: if _word_count(subject.description) < MIN_SUBJECT_DESCRIPTION_WORDS: @@ -84,7 +85,7 @@ def audit_curriculum_quality(graph: CurriculumGraph) -> list[CurriculumQualityIs ) ) - for skill in graph.skills: + for skill in skills: if len(skill.assessments) < MIN_ASSESSMENTS_PER_SKILL: issues.append( CurriculumQualityIssue( @@ -334,24 +335,13 @@ def _matched_placeholder_distractor(answer: str) -> str | None: def transitive_prerequisites(graph: CurriculumGraph) -> dict[str, set[str]]: - """Return each skill id -> the set of all skills reachable through prerequisites.""" - direct = {s.skill_id: set(s.prerequisites) for s in graph.skills} + """Return transitive prerequisite sets without recursion on a validated DAG.""" closure: dict[str, set[str]] = {} - - def resolve(sid: str, seen: frozenset[str]) -> set[str]: - if sid in closure: - return closure[sid] - acc: set[str] = set() - for prereq in direct.get(sid, set()): - if prereq in seen: # guard (graph is a validated DAG, but be defensive) - continue - acc.add(prereq) - acc |= resolve(prereq, seen | {prereq}) - closure[sid] = acc - return acc - - for skill in graph.skills: - resolve(skill.skill_id, frozenset({skill.skill_id})) + for skill in graph.topological_order(): + ancestors = set(skill.prerequisites) + for prereq in skill.prerequisites: + ancestors.update(closure[prereq]) + closure[skill.skill_id] = ancestors return closure @@ -360,7 +350,8 @@ def semantic_prerequisite_coverage( ) -> tuple[SemanticPrerequisiteCoverage, ...]: """Return concept-level reachability counts used by the semantic coverage figure.""" closure = transitive_prerequisites(graph) - by_id = {s.skill_id: s for s in graph.skills} + skills = graph.skills + by_id = {skill.skill_id: skill for skill in skills} rows: list[SemanticPrerequisiteCoverage] = [] for concept, patterns, definers in _CONCEPT_PREREQUISITES: compiled = [re.compile(pattern, re.IGNORECASE) for pattern in patterns] @@ -370,7 +361,7 @@ def semantic_prerequisite_coverage( if any(definer in prereqs for definer in definers) ) assessed = 0 - for skill in graph.skills: + for skill in skills: if skill.skill_id in definers: continue text = " ".join( @@ -417,8 +408,9 @@ def audit_semantic_prerequisites( for concept, pats, definers in _CONCEPT_PREREQUISITES ] closure = transitive_prerequisites(graph) - by_id = {s.skill_id: s for s in graph.skills} - for skill in graph.skills: + skills = graph.skills + by_id = {skill.skill_id: skill for skill in skills} + for skill in skills: text = " ".join( tuple(q.prompt for q in skill.assessments) + tuple(a for q in skill.assessments for a in q.answers) diff --git a/src/curriculum/recommendations.py b/src/curriculum/recommendations.py index d3f5d9b..5e903c4 100644 --- a/src/curriculum/recommendations.py +++ b/src/curriculum/recommendations.py @@ -14,6 +14,8 @@ from collections.abc import Iterable, Mapping from typing import Any +from src.skilltree.summary import SkillTreeSummaryError, validate_export_payload + _TOKEN_RE = re.compile(r"[A-Za-z][A-Za-z0-9_-]{3,}") _STOP_WORDS = frozenset( { @@ -164,6 +166,10 @@ def _index_project( dict[str, set[str]], dict[str, tuple[Mapping[str, Any], ...]], ]: + try: + validate_export_payload(project) + except SkillTreeSummaryError: + return {}, {}, {}, {} raw_subjects = project.get("subjects") raw_dependencies = project.get("dependencies") raw_quizzes = project.get("quizzes") @@ -256,9 +262,7 @@ def _depths( if indegree[dependent] == 0: queue.append(dependent) if visited != len(skill_ids): - # The export validator normally prevents this. Keeping the recommendation - # helper fail-closed avoids inventing a path if it is called on raw data. - return {skill_id: 0 for skill_id in skill_ids} + raise SkillTreeSummaryError("SkillTree project dependencies contain a cycle") return depths diff --git a/src/dashboard/artifacts.py b/src/dashboard/artifacts.py index 3cd5631..4a590cf 100644 --- a/src/dashboard/artifacts.py +++ b/src/dashboard/artifacts.py @@ -17,6 +17,7 @@ summarize_export_payload, validate_export_payload, ) +from src.skilltree.content import export_content_digest class DashboardArtifactError(ValueError): @@ -26,7 +27,7 @@ class DashboardArtifactError(ValueError): # Derived expected artifact count. Roster integrity is guarded by tests for unique ids, # unique paths, and registry/dashboard parity rather than a stale hand-maintained literal. EXPECTED_ARTIFACT_COUNT = len(ARTIFACTS) -_ANSWER_KEY_FIELDS = frozenset({"isCorrect"}) +_ANSWER_KEY_FIELDS = frozenset({"isCorrect", "answers"}) def artifact_inventory(project_root: Path) -> list[dict[str, object]]: @@ -62,7 +63,7 @@ def read_json_file(path: Path) -> tuple[dict[str, Any] | None, str | None]: def redact_skilltree_answer_keys(payload: Any) -> Any: - """Return a browser-safe SkillTree export projection without quiz answer keys.""" + """Remove answer options and correctness keys; official SkillTree owns quizzes.""" if isinstance(payload, dict): return { key: redact_skilltree_answer_keys(value) @@ -92,10 +93,12 @@ def load_redacted_skilltree_export( return redacted, None -def load_skilltree_project(project_root: Path) -> dict[str, Any]: +def load_skilltree_project( + project_root: Path, *, export_payload: dict[str, Any] | None = None, +) -> dict[str, Any]: """Load the exported SkillTree project, summarized for dashboard use.""" path = project_root / "output" / "data" / "skilltree_project.json" - payload, error = read_json_file(path) + payload, error = (export_payload, None) if export_payload is not None else read_json_file(path) if payload is None: return { "available": False, @@ -142,6 +145,7 @@ def load_skilltree_project(project_root: Path) -> dict[str, Any]: "id": payload.get("projectId"), "name": payload.get("name"), }, + "exportContentDigest": export_content_digest(payload), "subjects": subjects, "dependencies": dependencies, "quizzes": quizzes, diff --git a/src/dashboard/assets/app.js b/src/dashboard/assets/app.js index 4c2aad9..d619dff 100644 --- a/src/dashboard/assets/app.js +++ b/src/dashboard/assets/app.js @@ -1161,6 +1161,8 @@ function formatBytes(bytes) { function renderOfficialFallback(message, config) { const container = el("skills-client-container"); container.setAttribute("aria-busy", "false"); + container.dataset.mountState = "fallback"; + container.dataset.route = ""; const wrapper = document.createElement("div"); wrapper.className = "embed-state warn"; const title = document.createElement("strong"); @@ -1191,6 +1193,11 @@ async function mountOfficialDisplay(payload) { const config = payload.config; const container = el("skills-client-container"); container.setAttribute("aria-busy", "true"); + container.dataset.mountState = "pending"; + container.dataset.projectId = config.project.id; + container.dataset.route = ""; + container.dataset.instanceId = payload.instanceId || ""; + container.dataset.exportContentDigest = payload.project?.exportContentDigest || ""; if (!payload.skilltree.available) { renderOfficialFallback( "SkillTree is offline from this dashboard session. Start the local service or use the artifact panels.", @@ -1226,6 +1233,22 @@ async function mountOfficialDisplay(payload) { client.SkillsConfiguration.configure(skilltreeConfig); await client.SkillsConfiguration.afterConfigure(); const display = new client.SkillsDisplayJS({ + handleRouteChanged: (route) => { + if (container.dataset.mountState === "fallback") { + return; + } + const nativeLearnerRoute = route === "/" || route === "/rank" + || ["/subjects/", "/quizzes/", "/badges/", "/tags/"].some( + (prefix) => String(route).startsWith(prefix), + ) || route === "/badges" || route === "/tags"; + if (!nativeLearnerRoute) { + renderOfficialFallback("The official client did not reach a learner route.", config); + return; + } + container.dataset.route = route; + container.dataset.mountState = "ready"; + container.setAttribute("aria-busy", "false"); + }, options: { serviceUrl: config.skilltree.url, projectId: config.project.id, @@ -1257,7 +1280,6 @@ async function mountOfficialDisplay(payload) { }, }); container.innerHTML = ""; - container.setAttribute("aria-busy", "false"); display.attachTo(container); } catch (error) { renderOfficialFallback( @@ -1431,6 +1453,7 @@ main().catch((error) => { setText("service-state", "error"); document.body.classList.remove("is-loading"); el("skills-client-container").setAttribute("aria-busy", "false"); + el("skills-client-container").dataset.mountState = "fallback"; el("skills-client-container").innerHTML = `

Dashboard data could not load. diff --git a/src/dashboard/browser_validation.py b/src/dashboard/browser_validation.py new file mode 100644 index 0000000..e9bf79d --- /dev/null +++ b/src/dashboard/browser_validation.py @@ -0,0 +1,231 @@ +"""Real-browser proof of official SkillTree routes, not merely an empty console.""" + +from __future__ import annotations + +import html +import math +import re +import time +from collections.abc import Mapping +from typing import Any +from urllib.parse import parse_qsl, urlencode, urlsplit, urlunsplit + +from src.dashboard.live_validation import LiveSkillTreeDashboardError, _fetch_json +from src.skilltree.form_login_core import SkillTreeFormLoginError, parse_axi_result +from src.skilltree.form_login_validation import run_browser_command +from src.skilltree.local_auth_state import LocalSkillTreeAuthError, _skilltree_origin + +_ALLOWED_ERROR_PATTERNS = ( + re.compile(r"projects/Inception/.*token.*401", re.IGNORECASE), + re.compile(r"Unable to authenticate.*Inception", re.IGNORECASE), +) + +DISPLAY_PROBE_JS = r"""(activate = false) => { + const container = document.getElementById('skills-client-container'); + if (!container) return {state: 'pending', url: location.href}; + const panel = container.closest('.panel'); + const tab = panel && [...document.querySelectorAll('.tab')].find(item => item.dataset.panel === panel.id); + if (activate && tab) tab.click(); + const frame = container.querySelector('iframe'); + const bounds = frame && frame.getBoundingClientRect(); + return { + state: container.dataset.mountState || 'pending', + url: location.href, + projectId: container.dataset.projectId || '', + route: container.dataset.route || '', + dashboardInstanceId: container.dataset.instanceId || '', + exportContentDigest: container.dataset.exportContentDigest || '', + frameUrl: frame ? frame.src : '', + frameVisible: !!bounds && bounds.width > 0 && bounds.height > 0, + busy: container.getAttribute('aria-busy') === 'true' + }; +}""" + + +class BrowserDisplayError(RuntimeError): + """The real official frame could not prove the requested learner view.""" + + +def unexpected_errors(errors: list[str]) -> list[str]: + return [error for error in errors if not any(pattern.search(error) for pattern in _ALLOWED_ERROR_PATTERNS)] + + +def browser_output(arguments: list[str], *, browser_command: str, timeout: float) -> str: + try: + result = run_browser_command([browser_command, *arguments], timeout=timeout) + except SkillTreeFormLoginError as exc: + raise BrowserDisplayError(str(exc)) from exc + if re.search(r"(?im)^\s*(?:code:\s*(?:TIMEOUT|ERROR)|isError:\s*true)\s*$", result.stdout): + raise BrowserDisplayError("browser command returned an explicit tool failure") + return result.stdout + + +def display_probe( + *, browser_command: str, timeout: float, activate: bool = False, +) -> dict[str, Any]: + script = f"() => ({DISPLAY_PROBE_JS})({'true' if activate else 'false'})" + stdout = browser_output(["eval", script], browser_command=browser_command, timeout=timeout) + try: + value = parse_axi_result(stdout) + except SkillTreeFormLoginError as exc: + raise BrowserDisplayError(str(exc)) from exc + if not isinstance(value, dict) or value.get("isError") is True: + raise BrowserDisplayError("browser display probe returned no state object") + return value + + +def wait_for_official_display( + *, browser_command: str, dashboard_url: str, skilltree_url: str, + project_id: str, route: str, timeout: float = 45.0, + dashboard_instance_id: str | None = None, export_digest: str | None = None, +) -> dict[str, Any]: + if not math.isfinite(timeout) or timeout <= 0: + raise BrowserDisplayError("browser readiness timeout must be finite and positive") + deadline = time.monotonic() + timeout + while True: + remaining = deadline - time.monotonic() + if remaining <= 0: + raise BrowserDisplayError(f"official SkillTree display did not reach ready route {route!r}") + probe = display_probe( + browser_command=browser_command, timeout=min(remaining, 30.0), activate=True, + ) + if _origin(str(probe.get("url", ""))) != _origin(dashboard_url): + raise BrowserDisplayError("browser navigated away from the requested dashboard origin") + if probe.get("state") == "fallback": + raise BrowserDisplayError("official SkillTree display is in fallback, not mounted") + if probe.get("state") == "ready" and probe.get("route") == route: + if ( + probe.get("projectId") != project_id + or _origin(str(probe.get("frameUrl", ""))) != _origin(skilltree_url) + or (dashboard_instance_id is not None and probe.get("dashboardInstanceId") != dashboard_instance_id) + or (export_digest is not None and probe.get("exportContentDigest") != export_digest) + ): + raise BrowserDisplayError("official frame identity or export association is invalid") + if probe.get("busy") is False and probe.get("frameVisible") is True: + return probe + time.sleep(min(0.25, max(0.0, deadline - time.monotonic()))) + + +def _view_contracts(status: Mapping[str, Any]) -> tuple[str, str, list[dict[str, str]]]: + config = status.get("config", {}) + project = status.get("project", {}) + if not isinstance(config, Mapping) or not isinstance(project, Mapping) or not project.get("available"): + raise BrowserDisplayError("dashboard has no available exported project") + identity = config.get("project", {}) + if not isinstance(identity, Mapping) or not isinstance(identity.get("id"), str): + raise BrowserDisplayError("dashboard project identity is unavailable") + project_id = identity["id"] + skilltree = config.get("skilltree", {}) + if not isinstance(skilltree, Mapping) or not isinstance(skilltree.get("url"), str): + raise BrowserDisplayError("SkillTree target is unavailable") + quizzes = {quiz["quizId"]: quiz for quiz in project.get("quizzes", []) if isinstance(quiz, Mapping) and isinstance(quiz.get("quizId"), str)} + for subject in project.get("subjects", []): + if not isinstance(subject, Mapping): + continue + for skill in subject.get("skills", []): + if not isinstance(skill, Mapping) or skill.get("quizId") not in quizzes: + continue + quiz = quizzes[skill["quizId"]] + views = [ + {"kind": "project", "route": "/", "visibleText": str(identity.get("name", ""))}, + {"kind": "subject", "route": f"/subjects/{subject['subjectId']}", "visibleText": str(subject.get("name", ""))}, + {"kind": "quiz", "route": f"/subjects/{subject['subjectId']}/skills/{skill['skillId']}/quizzes/{quiz['quizId']}", "visibleText": str(quiz.get("name", ""))}, + ] + if any(not view["visibleText"].strip() for view in views): + raise BrowserDisplayError("official view contract lacks a display name") + return project_id, skilltree["url"], views + raise BrowserDisplayError("export has no quiz-gated skill for the required official quiz proof") + + +def validate_dashboard_browser( + dashboard_url: str, *, browser_command: str = "chrome-devtools-axi", + timeout: float = 45.0, settle_seconds: float = 5.0, +) -> dict[str, Any]: + if not math.isfinite(timeout) or timeout <= 0: + raise BrowserDisplayError("timeout must be finite and positive") + if not math.isfinite(settle_seconds) or settle_seconds < 0: + raise BrowserDisplayError("settle_seconds must be finite and non-negative") + parsed = urlsplit(dashboard_url) + base = urlunsplit((parsed.scheme, parsed.netloc, parsed.path.rstrip("/"), "", "")) + try: + status = _fetch_json(f"{base}/api/status?defer_live=1", min(timeout, 30.0)) + except LiveSkillTreeDashboardError as exc: + raise BrowserDisplayError(str(exc)) from exc + project_id, skilltree_url, views = _view_contracts(status) + instance_id = status.get("instanceId") + digest = status.get("project", {}).get("exportContentDigest") + if not isinstance(instance_id, str) or not instance_id or not isinstance(digest, str) or len(digest) != 64: + raise BrowserDisplayError("dashboard instance/export identity is missing") + allowed_count = 0 + results = [] + for view in views: + query = dict(parse_qsl(parsed.query)) + query["skillsClientDisplayPath"] = view["route"] + url = urlunsplit((parsed.scheme, parsed.netloc, parsed.path or "/", urlencode(query), "")) + browser_output(["open", url], browser_command=browser_command, timeout=60.0) + probe = wait_for_official_display( + browser_command=browser_command, dashboard_url=dashboard_url, + skilltree_url=skilltree_url, project_id=project_id, route=view["route"], + timeout=timeout, dashboard_instance_id=instance_id, export_digest=digest, + ) + if settle_seconds: + time.sleep(settle_seconds) + snapshot = browser_output(["snapshot"], browser_command=browser_command, timeout=30.0) + visible = " ".join(html.unescape(snapshot).split()).casefold() + expected = " ".join(html.unescape(view["visibleText"]).split()).casefold() + if expected not in visible: + raise BrowserDisplayError(f"official {view['kind']} view lacks its expected visible content") + console = browser_output(["console", "--type", "error"], browser_command=browser_command, timeout=30.0) + errors = [line.strip() for line in console.splitlines() if "[error]" in line] + if unexpected_errors(errors): + raise BrowserDisplayError("dashboard has unexpected browser console errors") + allowed_count += len(errors) + results.append({"kind": view["kind"], "route": probe["route"], "state": "ready", "visibleContentChecked": True}) + return { + "ok": True, "checked": True, "projectId": project_id, + "dashboardInstanceId": instance_id, "exportContentDigest": digest, + "views": results, "allowedConsoleErrors": allowed_count, + "evidenceKind": "real official SDK route, visible frame and browser accessibility content", + } + + +def _origin(url: str) -> tuple[str, str, int]: + try: + return _skilltree_origin(url) + except LocalSkillTreeAuthError as exc: + raise BrowserDisplayError("browser/configured origin is invalid") from exc + + +def browser_receipt_complete(payload: object) -> bool: + """Require all actual official view observations; an empty console is insufficient.""" + if not isinstance(payload, Mapping) or payload.get("ok") is not True or payload.get("checked") is not True: + return False + if ( + not isinstance(payload.get("projectId"), str) or not payload["projectId"] + or not isinstance(payload.get("dashboardInstanceId"), str) or not payload["dashboardInstanceId"] + or not isinstance(payload.get("exportContentDigest"), str) + or len(payload["exportContentDigest"]) != 64 + ): + return False + views = payload.get("views") + if not isinstance(views, list) or len(views) != 3: + return False + kinds: set[str] = set() + for view in views: + if not isinstance(view, Mapping): + return False + kind, route = view.get("kind"), view.get("route") + if ( + not isinstance(kind, str) or kind not in {"project", "subject", "quiz"} + or kind in kinds or not isinstance(route, str) + or view.get("state") != "ready" or view.get("visibleContentChecked") is not True + ): + return False + if kind == "project" and route != "/": + return False + if kind == "subject" and not route.startswith("/subjects/"): + return False + if kind == "quiz" and "/quizzes/" not in route: + return False + kinds.add(kind) + return kinds == {"project", "subject", "quiz"} diff --git a/src/dashboard/config.py b/src/dashboard/config.py index cdbdb9e..65243e5 100644 --- a/src/dashboard/config.py +++ b/src/dashboard/config.py @@ -59,6 +59,7 @@ class DashboardConfig: local_auth_enabled: bool = False local_state_dir: Path | None = None learner_id: str = DEFAULT_LOCAL_LEARNER_ID + instance_id: str = "" def __post_init__(self) -> None: if not self.skilltree_url.strip(): @@ -188,6 +189,7 @@ def make_dashboard_config( local_auth_enabled: bool = False, local_state_dir: Path | None = None, learner_id: str = DEFAULT_LOCAL_LEARNER_ID, + instance_id: str = "", ) -> DashboardConfig: """Build the default dashboard config, resolving the project root once.""" root = project_root or resolve_project_root("src") @@ -203,6 +205,7 @@ def make_dashboard_config( local_auth_enabled=local_auth_enabled, local_state_dir=local_state_dir, learner_id=learner_id, + instance_id=instance_id, ) diff --git a/src/dashboard/live_validation.py b/src/dashboard/live_validation.py index c1ef7e3..a866c09 100644 --- a/src/dashboard/live_validation.py +++ b/src/dashboard/live_validation.py @@ -18,6 +18,7 @@ from src.net_guard import NonHttpUrlError, require_http_url from src.project_identity import SKILLTREE_PROJECT_ID, SKILLTREE_PROJECT_NAME from src.skilltree.summary import SkillTreeSummaryError, summarize_export_payload +from src.skilltree.content import compare_live_content, export_content_digest class LiveSkillTreeDashboardError(RuntimeError): @@ -54,7 +55,7 @@ def _fetch_json(url: str, timeout: float) -> dict[str, Any]: return payload -def _count_exported_project(project_path: Path) -> dict[str, int]: +def _load_exported_project(project_path: Path) -> dict[str, Any]: try: payload = read_json_object(project_path) except FileNotFoundError as exc: @@ -68,11 +69,17 @@ def _count_exported_project(project_path: Path) -> dict[str, int]: except JsonArtifactError as exc: raise LiveSkillTreeDashboardError(str(exc)) from exc try: - return summarize_export_payload(payload).project_counts() + summarize_export_payload(payload) + export_content_digest(payload) + return payload except SkillTreeSummaryError as exc: raise LiveSkillTreeDashboardError(str(exc)) from exc +def _count_exported_project(project_path: Path) -> dict[str, int]: + return summarize_export_payload(_load_exported_project(project_path)).project_counts() + + def _require(condition: bool, message: str) -> None: if not condition: raise LiveSkillTreeDashboardError(message) @@ -284,7 +291,8 @@ def validate_live_skilltree_dashboard( material_index = _fetch_json(f"{dashboard_url}/api/course-materials", timeout) project_path = project_root / "output" / "data" / "skilltree_project.json" - exported_counts = _count_exported_project(project_path) + exported_project = _load_exported_project(project_path) + exported_counts = summarize_export_payload(exported_project).project_counts() project = dashboard_status.get("project", {}) config = dashboard_status.get("config", {}) dashboard_skilltree = dashboard_status.get("skilltree", {}) @@ -309,10 +317,10 @@ def validate_live_skilltree_dashboard( counts == exported_counts, f"dashboard counts {counts!r} != export {exported_counts!r}", ) - public_project = config.get("project", {}) - _require(isinstance(public_project, dict), "public project config is not an object") - project_id = project.get("projectId") or public_project.get("id") - project_name = project.get("name") or public_project.get("name") + public_project = project.get("project", {}) + _require(isinstance(public_project, dict), "dashboard project identity is not an object") + project_id = public_project.get("id") + project_name = public_project.get("name") _require( project_id == SKILLTREE_PROJECT_ID, "dashboard project id mismatch", @@ -429,6 +437,7 @@ def validate_live_skilltree_dashboard( local_state_dir=local_state_dir, timeout=timeout, required=require_live_admin, + exported_project=exported_project, ) live_dependencies = _validate_live_dependencies( skilltree_url=skilltree_url, @@ -439,6 +448,16 @@ def validate_live_skilltree_dashboard( timeout=timeout, required=require_live_dependencies, ) + live_content = live_admin.get("content", {"checked": False}) + _require( + not require_live_admin + or ( + live_content.get("checked") is True + and live_content.get("complete") is True + and project.get("exportContentDigest") == export_content_digest(exported_project) + ), + "live SkillTree content or dashboard export digest does not match the current export", + ) figure_contracts = ( _validate_figure_contracts(project_root=project_root, artifacts=artifacts) if require_figure_contracts @@ -449,6 +468,8 @@ def validate_live_skilltree_dashboard( "ok": True, "skilltreeStatus": skilltree_status, "dashboardUrl": dashboard_url, + "dashboardInstanceId": dashboard_status.get("instanceId"), + "exportContentDigest": export_content_digest(exported_project), "skilltreeUrl": skilltree_url, "counts": exported_counts, "artifacts": {"present": len(artifacts), "missing": 0}, @@ -459,6 +480,7 @@ def validate_live_skilltree_dashboard( "figureContracts": figure_contracts, "liveAdminProject": live_admin, "liveDependencies": live_dependencies, + "liveContent": live_content, "tokenKeys": token_keys, } @@ -548,6 +570,7 @@ def _validate_live_admin_count( local_state_dir: Path | None, timeout: float, required: bool, + exported_project: dict[str, Any], ) -> dict[str, Any]: """Optionally verify the live SkillTree admin project count.""" from src.skilltree.client import SkillTreeClientError @@ -585,6 +608,11 @@ def _validate_live_admin_count( try: client.login(state.admin_username, state.admin_password) live_skills = client.live_skill_count() + content = compare_live_content( + get=lambda path: client.request_json("GET", path), + base_url=skilltree_url, + project=exported_project, + ) except (LocalSkillTreeAuthError, SkillTreeClientError) as exc: _require(not required, f"live SkillTree admin count check failed: {exc}") return { @@ -599,4 +627,5 @@ def _validate_live_admin_count( return { "checked": True, "skills": live_skills, + "content": content, } diff --git a/src/dashboard/payload.py b/src/dashboard/payload.py index fd8dd19..3ec816a 100644 --- a/src/dashboard/payload.py +++ b/src/dashboard/payload.py @@ -11,6 +11,7 @@ load_figure_table_companions, load_scholarship_health_network, load_skilltree_project, + read_json_file, ) from src.dashboard.config import DashboardConfig, DashboardPayload from src.dashboard.graph_analysis import analyze_project_graph @@ -22,6 +23,8 @@ SkillTreeSessionClient, load_local_state, ) +from src.skilltree.local_auth_state import _require_same_skilltree_origin +from src.skilltree.content import compare_live_content DEFERRED_LIVE_PARITY_REASON = ( "live dependency parity was deferred for first paint; " @@ -41,7 +44,10 @@ def dashboard_payload( that otherwise blocks every panel on first paint. The default is unchanged, so validators and proof artifacts still get a real parity check. """ - project = load_skilltree_project(config.project_root) + exported, _ = read_json_file( + config.project_root / "output" / "data" / "skilltree_project.json" + ) + project = load_skilltree_project(config.project_root, export_payload=exported) public_config = config.to_public_dict() skilltree_status = check_skilltree_status(config.skilltree_url) public_skilltree = public_config.get("skilltree", {}) @@ -66,7 +72,7 @@ def dashboard_payload( "project": project, "liveSkilltree": deferred_live_skilltree_summary(project) if defer_live - else live_skilltree_summary(config, project), + else live_skilltree_summary(config, project, export_payload=exported), "graphAnalysis": analyze_project_graph(project), "claims": load_claims_summary(config.project_root), "courseMaterials": load_course_materials_summary(config.project_root), @@ -96,8 +102,10 @@ def deferred_live_skilltree_summary(project: dict[str, Any]) -> dict[str, object def live_skilltree_summary( config: DashboardConfig, project: dict[str, Any], + *, + export_payload: dict[str, Any] | None = None, ) -> dict[str, object]: - """Return browser-safe live SkillTree admin/count parity when local auth is ready.""" + """Return browser-safe dependency and semantic-content parity when auth is ready.""" from src.skilltree.client import SkillTreeClientError from src.skilltree.dependency_reconciler import ( dependency_pairs_from_payload, @@ -133,6 +141,13 @@ def live_skilltree_summary( state = load_local_state(state_dir) if state is None: raise LocalSkillTreeAuthError("local SkillTree auth state is missing") + _require_same_skilltree_origin( + base_url=config.skilltree_url, state_url=state.skilltree_url, + ) + if state.project_id != config.project_id: + raise LocalSkillTreeAuthError( + "configured project does not match bootstrapped local auth state" + ) expected_pairs = dependency_pairs_from_payload(dependencies, strict=False) client = SkillTreeSessionClient( config.skilltree_url, @@ -140,6 +155,17 @@ def live_skilltree_summary( timeout=15.0, ) client.login(state.admin_username, state.admin_password) + if export_payload is None: + export_payload, _ = read_json_file( + config.project_root / "output" / "data" / "skilltree_project.json" + ) + if export_payload is None: + raise LocalSkillTreeAuthError("private exported project is unavailable") + content = compare_live_content( + get=lambda path: client.request_json("GET", path), + base_url=config.skilltree_url, + project=export_payload, + ) live_pairs = client.live_dependency_pairs() parity = summarize_dependency_parity( expected_pairs=expected_pairs, @@ -147,8 +173,9 @@ def live_skilltree_summary( ) return { **parity, - "skills": client.live_skill_count(), - "subjects": client.live_subject_count(), + "skills": content["skills"], + "subjects": content["subjects"], + "content": content, } except (LocalSkillTreeAuthError, SkillTreeClientError) as exc: return { diff --git a/src/dashboard/server.py b/src/dashboard/server.py index 29da681..c18ce61 100644 --- a/src/dashboard/server.py +++ b/src/dashboard/server.py @@ -4,6 +4,7 @@ import gzip import json +import uuid from http import HTTPStatus from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer from pathlib import Path @@ -32,8 +33,18 @@ def dashboard_recommendations(config: DashboardConfig) -> dict[str, Any]: - """Return the read-only path prototype for the current exported project.""" - return recommend_next_skills(load_skilltree_project(config.project_root)) + """Validate the private export; return only answer-key-free path candidates.""" + from src.dashboard.artifacts import read_json_file + + project, error = read_json_file( + config.project_root / "output" / "data" / "skilltree_project.json" + ) + if project is None: + return { + "available": False, "error": error, + "recommendations": [], "personalized": False, "signalSource": "none", + } + return recommend_next_skills(project) ASSETS_ROOT = Path(__file__).resolve().parent / "assets" @@ -163,6 +174,7 @@ def _host_header_hostname(host_header: str) -> str: def make_dashboard_handler(config: DashboardConfig) -> type[BaseHTTPRequestHandler]: """Create a request handler bound to one immutable dashboard config.""" + instance_id = config.instance_id or uuid.uuid4().hex class DashboardRequestHandler(BaseHTTPRequestHandler): server_version = "ActiveSkillferenceDashboard/1.0" @@ -227,6 +239,7 @@ def do_GET(self) -> None: self._send_json( { "ok": True, + "instanceId": instance_id, "projectId": config.project_id, "skilltreeUrl": config.skilltree_url, } @@ -238,10 +251,13 @@ def do_GET(self) -> None: if path == "/api/status": query = parse_qs(parsed.query) self._send_json( - dashboard_payload( - config, - defer_live=_query_flag(query, "defer_live"), - ) + { + **dashboard_payload( + config, + defer_live=_query_flag(query, "defer_live"), + ), + "instanceId": instance_id, + } ) return if path == "/api/recommendations": diff --git a/src/figures/plots/registry_io.py b/src/figures/plots/registry_io.py index aa93272..6f02b62 100644 --- a/src/figures/plots/registry_io.py +++ b/src/figures/plots/registry_io.py @@ -99,11 +99,11 @@ def render_all( include_scholarship = scholarship_audit is not None include_claim_support = claim_support_audit is not None include_roadmap = roadmap is not None - planned_count = planned_figure_count( + planned_count = len(planned_figure_registry_entries( include_scholarship=include_scholarship, include_claim_support=include_claim_support, include_roadmap=include_roadmap, - ) + )) figures = [ plot_curriculum_dag(graph, out_dir / "curriculum_dag.png"), plot_dependency_strata_flow(graph, out_dir / "dependency_strata_flow.png"), @@ -233,7 +233,7 @@ def planned_figure_count( ) -> int: """Return the canonical number of registered figures for a render configuration.""" return len( - planned_figure_registry_entries( + figure_registry_payload( include_scholarship=include_scholarship, include_claim_support=include_claim_support, include_roadmap=include_roadmap, diff --git a/src/gnn_conformance.py b/src/gnn_conformance.py index bfcbd0c..5882bbe 100644 --- a/src/gnn_conformance.py +++ b/src/gnn_conformance.py @@ -315,6 +315,18 @@ def require(condition: bool, message: str) -> None: ) passed.append("dimensions match the kernel's canonical model (2/2/2)") + try: + DiscreteActiveInferenceAgent( + A=np.asarray(model["A_matrix"], dtype=float), + B=np.asarray(model["B_matrix"], dtype=float), + C=np.asarray(model["C_vector"], dtype=float), + D=np.asarray(model["D_vector"], dtype=float), + ) + except (TypeError, ValueError) as exc: + raise ConformanceError( + f"declared probability/transition model is invalid: {exc}" + ) from exc + require( np.allclose(np.asarray(model["A_matrix"], dtype=float), agent.A), "A does not match the kernel's informative sensor", diff --git a/src/integration/full_app_stack.py b/src/integration/full_app_stack.py index 985bf42..51b5b42 100644 --- a/src/integration/full_app_stack.py +++ b/src/integration/full_app_stack.py @@ -208,7 +208,7 @@ def bootstrap_command( state_dir: str, learner_id: str, skip_dependencies: bool = False, - skip_seed_if_live_counts_ready: bool = False, + skip_seed_if_live_content_ready: bool = False, ) -> list[str]: command = [ python, @@ -222,8 +222,8 @@ def bootstrap_command( ] if skip_dependencies: command.append("--skip-dependencies") - if skip_seed_if_live_counts_ready: - command.append("--skip-seed-if-live-counts-ready") + if skip_seed_if_live_content_ready: + command.append("--skip-seed-if-live-content-ready") return command @@ -254,8 +254,9 @@ def dashboard_command( port: int, state_dir: str, learner_id: str, + instance_id: str | None = None, ) -> list[str]: - return [ + command = [ python, "scripts/serve_learner_dashboard.py", "--skilltree-url", @@ -270,6 +271,9 @@ def dashboard_command( "--learner-id", learner_id, ] + if instance_id is not None: + command.extend(["--instance-id", instance_id]) + return command def live_validation_command( @@ -313,6 +317,7 @@ def console_validation_command( dashboard_url, "--require-browser", "--require-dashboard", + "--json", ] @@ -441,6 +446,7 @@ def wait_for_dashboard( *, dashboard_url: str, timeout: float, + instance_id: str | None = None, ) -> dict[str, Any]: health_url = f"{dashboard_url.rstrip('/')}/healthz" deadline = time.monotonic() + timeout @@ -452,7 +458,10 @@ def wait_for_dashboard( f"dashboard process exited before /healthz became ready\n{output}" ) try: - return _fetch_json(health_url, timeout=_REQUEST_TIMEOUT_SECONDS) + health = _fetch_json(health_url, timeout=_REQUEST_TIMEOUT_SECONDS) + if instance_id is not None and health.get("instanceId") != instance_id: + raise FullAppStackError("health endpoint belongs to another dashboard instance") + return health except ( FullAppStackError, TimeoutError, @@ -482,19 +491,50 @@ def proof_summary( dependency_reconciliation: dict[str, Any] | None = None, app_ready: bool = False, ) -> dict[str, Any]: - checks = [live_validation, dashboard_health] + from src.dashboard.browser_validation import browser_receipt_complete + + browser: dict[str, Any] = {"checked": False, "reason": "skipped by --skip-browser"} + if browser_console_stdout is not None: + try: + candidate = json.loads(browser_console_stdout) + except json.JSONDecodeError as exc: + raise FullAppStackError("browser validator did not emit a structured proof") from exc + if not browser_receipt_complete(candidate): + raise FullAppStackError("browser validator did not prove all official learner views") + browser = candidate + content = live_validation.get("liveContent", {}) + dependencies = live_validation.get("liveDependencies", {}) + if not isinstance(content, dict) or not isinstance(dependencies, dict): + raise FullAppStackError("live content/dependency proof must contain objects") + instance_id = dashboard_health.get("instanceId") + digest = live_validation.get("exportContentDigest") + ok = ( + live_validation.get("ok") is True + and dashboard_health.get("ok") is True + and isinstance(instance_id, str) and bool(instance_id) + and live_validation.get("dashboardInstanceId") == instance_id + and isinstance(digest, str) and len(digest) == 64 + and content.get("checked") is True and content.get("complete") is True + and content.get("exportContentDigest") == digest + ) + if not app_ready: + ok = ok and dependencies.get("checked") is True and dependencies.get("complete") is True if raw_skilltree_login is not None: - checks.append(raw_skilltree_login) + ok = ok and raw_skilltree_login.get("ok") is True if dependency_reconciliation is not None: - checks.append(dependency_reconciliation) - ok = all(check.get("ok") is not False for check in checks) - if ( - dependency_reconciliation is not None - and dependency_reconciliation.get("complete") is False - ): - ok = False + ok = ok and dependency_reconciliation.get("complete") is True + if browser_console_stdout is not None: + ok = ( + ok and browser.get("dashboardInstanceId") == instance_id + and browser.get("exportContentDigest") == digest + and browser.get("projectId") == content.get("projectId") + ) return { "ok": ok, + "runId": instance_id, + "dashboardInstanceId": instance_id, + "exportContentDigest": digest, + "projectId": content.get("projectId"), "skilltreeUrl": skilltree_url, "dashboardUrl": dashboard_url, "skilltreeStatus": skilltree_status, @@ -504,24 +544,17 @@ def proof_summary( "courseMaterials": live_validation.get("courseMaterials", {}), "figureContracts": live_validation.get("figureContracts", {"checked": False}), "liveAdminProject": live_validation.get("liveAdminProject", {}), - "liveDependencies": live_validation.get("liveDependencies", {}), + "liveDependencies": dependencies, + "liveContent": content, "dependencyReconciliation": dependency_reconciliation or {"checked": False}, "mode": "app-ready" if app_ready else "full-parity", - "browserConsole": { - "checked": browser_console_stdout is not None, - "summary": (browser_console_stdout or "skipped by --skip-browser").strip(), - }, - "rawSkillTreeLogin": raw_skilltree_login - if raw_skilltree_login is not None + "browserConsole": browser, + "rawSkillTreeLogin": raw_skilltree_login if raw_skilltree_login is not None else {"checked": False, "reason": "skipped by --skip-browser"}, "keptRunning": kept_running, "dashboardPid": dashboard_pid if kept_running else None, - "dashboardPidPath": str(dashboard_pid_path) - if kept_running and dashboard_pid_path - else None, - "dashboardLogPath": str(dashboard_log_path) - if kept_running and dashboard_log_path - else None, + "dashboardPidPath": str(dashboard_pid_path) if kept_running and dashboard_pid_path else None, + "dashboardLogPath": str(dashboard_log_path) if kept_running and dashboard_log_path else None, } @@ -530,7 +563,7 @@ def write_proof_artifact(proof: dict[str, Any], path: Path) -> Path: safe = _redact_proof_value(proof) if not isinstance(safe, dict): # pragma: no cover - defensive type narrowing raise FullAppStackError("full-app proof must be a JSON object") - safe["proofVersion"] = 1 + safe["proofVersion"] = 2 return write_json_artifact(Path(path), safe, ensure_ascii=False) @@ -601,7 +634,7 @@ def validate_full_app_stack(args: argparse.Namespace) -> dict[str, Any]: state_dir=state_dir, learner_id=args.learner_id, skip_dependencies=True, - skip_seed_if_live_counts_ready=True, + skip_seed_if_live_content_ready=True, ), ], timeout=args.timeout, @@ -626,6 +659,7 @@ def validate_full_app_stack(args: argparse.Namespace) -> dict[str, Any]: ) pid_path, log_path = dashboard_runtime_paths(state_dir) + instance_id = secrets.token_hex(16) dashboard = start_dashboard( [ "uv", @@ -636,15 +670,18 @@ def validate_full_app_stack(args: argparse.Namespace) -> dict[str, Any]: port=args.port, state_dir=state_dir, learner_id=args.learner_id, + instance_id=instance_id, ), ], log_path=log_path if args.keep_running else None, ) + proof_written = False try: dashboard_health = wait_for_dashboard( dashboard, dashboard_url=dashboard_url, timeout=args.timeout, + instance_id=instance_id, ) live_result = run_checked( [ @@ -713,9 +750,27 @@ def validate_full_app_stack(args: argparse.Namespace) -> dict[str, Any]: ) proof["proofPath"] = str(proof_path) write_proof_artifact(proof, proof_path) + proof_written = True return proof finally: - if args.keep_running and dashboard.poll() is None: - record_dashboard_process(state_dir=state_dir, pid=dashboard.pid) - elif not args.keep_running: - stop_dashboard_process(dashboard) + finalize_dashboard_process( + dashboard, state_dir=state_dir, keep_running=args.keep_running, + successful=proof_written, + ) + + +def finalize_dashboard_process( + process: subprocess.Popen[str], *, state_dir: str, + keep_running: bool, successful: bool, +) -> None: + """Keep an owned process only after successful proof publication.""" + if successful and keep_running: + if process.poll() is not None: + raise FullAppStackError("owned dashboard exited before keep-running finalization") + try: + record_dashboard_process(state_dir=state_dir, pid=process.pid) + except BaseException: + stop_dashboard_process(process) + raise + else: + stop_dashboard_process(process) diff --git a/src/kernels/active_inference.py b/src/kernels/active_inference.py index d69c4e1..eca1754 100644 --- a/src/kernels/active_inference.py +++ b/src/kernels/active_inference.py @@ -10,7 +10,7 @@ from numpy.typing import ArrayLike, NDArray from .distributions import log_stable, normalize, softmax -from .inference import bayes_posterior +from .inference import _as_probability_vector, _validate_model, bayes_posterior from .information import kl_divergence FloatArray: TypeAlias = NDArray[np.float64] @@ -36,9 +36,12 @@ def expected_free_energy( Tuple `(G, epistemic, pragmatic)` in nats. """ - state_probs = np.asarray(qs, dtype=np.float64) + state_probs = _as_probability_vector(qs, name="qs") likelihood = np.asarray(A, dtype=np.float64) + _validate_model(state_probs, likelihood, 0) preferences = normalize(C) + if preferences.shape != (likelihood.shape[0],): + raise ValueError("C length must match the number of observations") qo = likelihood @ state_probs pragmatic = float(np.dot(qo, log_stable(preferences))) @@ -64,7 +67,7 @@ def policy_posterior(G: ArrayLike, gamma: float = 1.0) -> FloatArray: Posterior policy distribution `softmax(-gamma * G)`. """ - return softmax(-gamma * np.asarray(G, dtype=np.float64)) + return softmax(-np.asarray(G, dtype=np.float64), precision=gamma) @dataclass @@ -77,12 +80,25 @@ class DiscreteActiveInferenceAgent: D: FloatArray def __post_init__(self) -> None: - """Coerce model components into float arrays.""" + """Coerce and validate the existing probability-weight model contract.""" self.A = np.asarray(self.A, dtype=np.float64) self.B = np.asarray(self.B, dtype=np.float64) self.C = np.asarray(self.C, dtype=np.float64) - self.D = np.asarray(self.D, dtype=np.float64) + self.D = _as_probability_vector(self.D, name="D") + _validate_model(self.D, self.A, 0) + n_states = self.D.size + if ( + self.B.ndim != 3 + or self.B.shape[:2] != (n_states, n_states) + or self.B.shape[2] == 0 + or not np.all(np.isfinite(self.B)) + or np.any(self.B < 0.0) + or not np.allclose(self.B.sum(axis=0), 1.0, atol=1e-9, rtol=0.0) + ): + raise ValueError("B must contain finite non-negative column-stochastic state transitions") + if normalize(self.C).shape != (self.A.shape[0],): + raise ValueError("C length must match the number of observations") def infer_states(self, obs: int) -> FloatArray: """Infer the posterior state distribution for an observation.""" @@ -92,7 +108,13 @@ def infer_states(self, obs: int) -> FloatArray: def predict(self, qs: ArrayLike, action: int) -> FloatArray: """Predict the next-state distribution under an action.""" - state_probs = np.asarray(qs, dtype=np.float64) + state_probs = _as_probability_vector(qs, name="qs") + if state_probs.shape != self.D.shape: + raise ValueError("qs length must match the model state dimension") + if not isinstance(action, (int, np.integer)) or isinstance(action, (bool, np.bool_)): + raise ValueError("action must be an integer index") + if action < 0 or action >= self.B.shape[2]: + raise ValueError("action index out of range") return self.B[:, :, action] @ state_probs def evaluate_policies(self, qs: ArrayLike) -> FloatArray: diff --git a/src/kernels/distributions.py b/src/kernels/distributions.py index d517d28..c87bd48 100644 --- a/src/kernels/distributions.py +++ b/src/kernels/distributions.py @@ -25,14 +25,15 @@ def normalize(x: ArrayLike) -> FloatArray: """ values = np.asarray(x, dtype=np.float64) + if values.ndim != 1 or not values.size or not np.all(np.isfinite(values)): + raise ValueError("values must be a non-empty finite vector") if np.any(values < 0.0): raise ValueError("values must be non-negative") - - total = float(np.sum(values)) - if total <= 0.0: + scale = float(np.max(values)) + if scale <= 0.0: raise ValueError("values must have a positive sum") - - return values / total + scaled = values / scale + return scaled / np.sum(scaled) def softmax(x: ArrayLike, precision: float = 1.0) -> FloatArray: @@ -52,14 +53,22 @@ def softmax(x: ArrayLike, precision: float = 1.0) -> FloatArray: ValueError: If precision is not strictly positive. """ - if precision <= 0.0: - raise ValueError("precision must be > 0") - + if not np.isfinite(precision) or precision <= 0.0: + raise ValueError("precision must be finite and > 0") values = np.asarray(x, dtype=np.float64) - scaled = precision * values - shifted = scaled - np.max(scaled) - exp_shifted = np.exp(shifted) - return exp_shifted / np.sum(exp_shifted) + if values.ndim != 1 or not values.size or not np.all(np.isfinite(values)): + raise ValueError("scores must be a non-empty finite vector") + # Scale small precision first so centering cannot erase finite scaled mass. + # At precision >= 1, center first to avoid overflowing large absolute scores. + with np.errstate(over="ignore", under="ignore"): + if precision < 1.0: + shifted = values * precision + shifted -= np.max(shifted) + else: + shifted = values - np.max(values) + shifted *= precision + np.exp(shifted, out=shifted) + return shifted / np.sum(shifted) def log_stable(x: ArrayLike, eps: float = 1e-16) -> FloatArray: @@ -91,10 +100,8 @@ def dirichlet_expected(alpha: ArrayLike) -> FloatArray: ValueError: If any concentration parameter is non-positive. """ - concentration = np.asarray(alpha, dtype=np.float64) - if np.any(concentration <= 0.0): - raise ValueError("alpha must be strictly positive") - return concentration / np.sum(concentration) + concentration = _concentrations(alpha) + return normalize(concentration) def dirichlet_expected_log(alpha: ArrayLike) -> FloatArray: @@ -110,10 +117,27 @@ def dirichlet_expected_log(alpha: ArrayLike) -> FloatArray: ValueError: If any concentration parameter is non-positive. """ + concentration = _concentrations(alpha) + scale = float(np.max(concentration)) + scaled_total = float(np.sum(concentration / scale)) + if scale > np.finfo(np.float64).max / scaled_total: + # At this magnitude the digamma asymptotic correction is unrepresentable. + total_digamma = np.log(scale) + np.log(scaled_total) + else: + total_digamma = digamma(scale * scaled_total) + return digamma(concentration) - total_digamma + + +def _concentrations(alpha: ArrayLike) -> FloatArray: concentration = np.asarray(alpha, dtype=np.float64) - if np.any(concentration <= 0.0): - raise ValueError("alpha must be strictly positive") - return digamma(concentration) - digamma(np.sum(concentration)) + if ( + concentration.ndim != 1 + or not concentration.size + or not np.all(np.isfinite(concentration)) + or np.any(concentration <= 0.0) + ): + raise ValueError("alpha must be a non-empty vector of finite positive values") + return concentration __all__ = [ diff --git a/src/kernels/inference.py b/src/kernels/inference.py index 6ea7770..160d30d 100644 --- a/src/kernels/inference.py +++ b/src/kernels/inference.py @@ -8,7 +8,7 @@ from numpy.typing import ArrayLike, NDArray from scipy.special import logsumexp -from .distributions import log_stable, normalize +from .distributions import normalize FloatArray: TypeAlias = NDArray[np.float64] _PROB_TOL = 1e-9 @@ -20,8 +20,8 @@ def _as_probability_vector(x: ArrayLike, *, name: str) -> FloatArray: values = np.asarray(x, dtype=np.float64) if values.ndim != 1: raise ValueError(f"{name} must be a 1D probability vector") - if np.any(values < 0.0): - raise ValueError(f"{name} must be non-negative and sum to 1") + if not np.all(np.isfinite(values)) or np.any(values < 0.0): + raise ValueError(f"{name} must be finite, non-negative and sum to 1") if not np.isclose(np.sum(values), 1.0, atol=_PROB_TOL, rtol=0.0): raise ValueError(f"{name} must sum to 1") return values @@ -43,8 +43,8 @@ def _validate_model(prior: FloatArray, A: FloatArray, obs: int) -> None: raise ValueError("prior length must match the number of states in A") if obs < 0 or obs >= A.shape[0]: raise ValueError("obs index out of range for A") - if np.any(A < 0.0): - raise ValueError("A must be non-negative (it is a likelihood P(o|s))") + if not np.all(np.isfinite(A)) or np.any(A < 0.0): + raise ValueError("A must be finite and non-negative (it is a likelihood P(o|s))") if not np.allclose(np.sum(A, axis=0), 1.0, atol=_PROB_TOL, rtol=0.0): raise ValueError("each column of A (P(o|s) for fixed state s) must sum to 1") @@ -103,11 +103,10 @@ def variational_free_energy( raise ValueError("q length must match prior length") mask = q_probs > 0.0 - energy = np.sum( - q_probs[mask] * (log_stable(q_probs[mask]) - log_stable(prior_probs[mask])) - ) - accuracy_cost = np.sum(q_probs[mask] * log_stable(likelihood[obs, mask])) - return float(energy - accuracy_cost) + if np.any(prior_probs[mask] == 0.0) or np.any(likelihood[obs, mask] == 0.0): + return float("inf") + terms = np.log(q_probs[mask]) - np.log(prior_probs[mask]) - np.log(likelihood[obs, mask]) + return float(np.dot(q_probs[mask], terms)) def log_evidence(A: ArrayLike, obs: int, prior: ArrayLike) -> float: @@ -128,7 +127,10 @@ def log_evidence(A: ArrayLike, obs: int, prior: ArrayLike) -> float: prior_probs = _as_probability_vector(prior, name="prior") likelihood = np.asarray(A, dtype=np.float64) _validate_model(prior_probs, likelihood, obs) - return float(logsumexp(log_stable(likelihood[obs, :]) + log_stable(prior_probs))) + support = (likelihood[obs, :] > 0.0) & (prior_probs > 0.0) + if not np.any(support): + return float("-inf") + return float(logsumexp(np.log(likelihood[obs, support]) + np.log(prior_probs[support]))) def belief_propagation_chain( diff --git a/src/kernels/predictive_coding.py b/src/kernels/predictive_coding.py index a46d98c..3a45c6a 100644 --- a/src/kernels/predictive_coding.py +++ b/src/kernels/predictive_coding.py @@ -50,9 +50,9 @@ class PredictiveCodingResult: trajectory: FloatArray # estimate at each step (length steps + 1) free_energy: FloatArray # free energy at each step estimate: float # final estimate - sensory_prediction_error: float # y - estimate at convergence - prior_prediction_error: float # estimate - prior_mean at convergence - steps_to_converge: int # first step within tolerance of the fixed point + sensory_prediction_error: float # y - final estimate + prior_prediction_error: float # final estimate - prior_mean + steps_to_converge: int # first update <= tolerance, or steps if none qualifies def gaussian_free_energy( @@ -102,13 +102,15 @@ def predictive_coding_descent( obs_variance: Observation variance (inverse sensory precision), > 0. rate: Positive learning rate for the gradient step. steps: Number of descent steps. - tolerance: Convergence tolerance on the distance to the fixed point. + tolerance: Threshold on absolute update magnitude, not fixed-point distance. start: Initial estimate (defaults to the prior mean). Returns: - A :class:`PredictiveCodingResult` with the trajectory, free-energy curve, - the converged estimate, the two prediction errors, and the step at which the - estimate first came within ``tolerance`` of the fixed point. + A :class:`PredictiveCodingResult` with all ``steps + 1`` trajectory and + free-energy samples, the final estimate and prediction errors, and the + first step whose absolute update is at most ``tolerance``. If no update + qualifies, ``steps_to_converge`` is ``steps``; the trajectory is never + truncated when the threshold is reached. """ if prior_variance <= 0.0 or obs_variance <= 0.0: @@ -126,9 +128,8 @@ def predictive_coding_descent( estimate, observation, prior_mean, prior_variance, obs_variance ) steps_to_converge = steps - # steps_to_converge is measured by the update magnitude (the descent's own stopping - # condition), NOT by distance to the analytic target — so this diagnostic stays - # independent of the closed form and cannot mask a closed-form bug. + # This update-magnitude diagnostic is independent of the analytic target. + # It records the first qualifying step but does not stop the trajectory. converged = False for step in range(1, steps + 1): sensory_error = (observation - estimate) / obs_variance diff --git a/src/live_counts.py b/src/live_counts.py index d993c49..b36cfb0 100644 --- a/src/live_counts.py +++ b/src/live_counts.py @@ -165,7 +165,10 @@ def audit_live_count_literals(root: Path) -> list[LiveCountIssue]: for path in audited_count_paths(root): rel = path.relative_to(root) - text = _CODE_BLOCK_RE.sub("", path.read_text(encoding="utf-8")) + text = _CODE_BLOCK_RE.sub( + lambda match: "\n" * match.group(0).count("\n"), + path.read_text(encoding="utf-8"), + ) for line_number, line in enumerate(text.splitlines(), start=1): for key, pattern in patterns.items(): expected = counts[key].value diff --git a/src/manuscript/pdf_receipt.py b/src/manuscript/pdf_receipt.py new file mode 100644 index 0000000..3d9b633 --- /dev/null +++ b/src/manuscript/pdf_receipt.py @@ -0,0 +1,185 @@ +"""Content-bound PDF receipts for repository inputs, not release authorization. + +Primary inputs are captured before assembly; actual LaTeX image inputs are added +before compilation. Every bound file and the section set must still match before +publication. Historical receipts are never rewritten by validation. +""" + +from __future__ import annotations + +import hashlib +import json +import re +from collections.abc import Iterable, Mapping, Sequence +from pathlib import Path +from typing import Any + +from src.artifacts.json_io import write_json_artifact + +SECTION_EXCLUDE = frozenset({"config.yaml", "preamble.md", "README.md", "AGENTS.md", "SYNTAX.md"}) +PDF_NAME = "Active_Skillference_combined.pdf" +_PRIMARY_INPUTS = { + "manuscript/config.yaml": "config", + "manuscript/preamble.md": "preamble", + "manuscript/references.bib": "bibliography", +} +_IMAGE_RE = re.compile(r"\\includegraphics\*?(?:\[[^\]]*\])?\{([^{}]+)\}") + + +class PdfReceiptError(ValueError): + """A PDF or its actual repository inputs are missing, changed or unverifiable.""" + + +def hydrated_section_files(directory: Path) -> list[Path]: + return [path for path in sorted(directory.glob("*.md")) if path.name not in SECTION_EXCLUDE] + + +def _relative(path: Path, root: Path) -> str: + try: + return path.resolve().relative_to(root.resolve()).as_posix() + except ValueError as exc: + raise PdfReceiptError("PDF inputs/output must remain inside the project root") from exc + + +def _record(path: Path, root: Path, kind: str) -> dict[str, Any]: + relative = _relative(path, root) + try: + with path.open("rb") as stream: + digest = hashlib.file_digest(stream, "sha256").hexdigest() + size = path.stat().st_size + except OSError as exc: + raise PdfReceiptError(f"PDF {kind} file is unreadable: {relative}") from exc + return {"path": relative, "kind": kind, "size_bytes": size, "sha256": digest} + + +def capture_pdf_inputs(root: Path, sections: Sequence[Path]) -> dict[str, Any]: + if not sections: + raise PdfReceiptError("PDF input manifest needs hydrated sections") + rows = [_record(root / name, root, kind) for name, kind in _PRIMARY_INPUTS.items()] + rows.extend(_record(path, root, "section") for path in sections) + return { + "sections": [_relative(path, root) for path in sections], + "assets": [], + "files": sorted(rows, key=lambda row: row["path"]), + } + + +def latex_asset_paths(tex: Path, *, build_directory: Path, root: Path) -> list[Path]: + text = tex.read_text(encoding="utf-8") + text = re.sub(r"(? dict[str, Any]: + validate_pdf_inputs(root, manifest) + rows = {row["path"]: dict(row) for row in manifest["files"]} + for path in assets: + row = _record(path, root, "asset") + rows.setdefault(row["path"], row) + result = { + "sections": list(manifest["sections"]), + "assets": sorted(name for name, row in rows.items() if row["kind"] == "asset"), + "files": [rows[name] for name in sorted(rows)], + } + validate_pdf_inputs(root, result) + return result + + +def validate_pdf_inputs(root: Path, manifest: Mapping[str, Any]) -> None: + sections, rows = manifest.get("sections"), manifest.get("files") + assets = manifest.get("assets") + if ( + not isinstance(sections, list) or not sections + or any(not isinstance(path, str) for path in sections) + or len(set(sections)) != len(sections) + or not isinstance(rows, list) or not rows + or not isinstance(assets, list) + or any(not isinstance(path, str) for path in assets) + or len(set(assets)) != len(assets) + ): + raise PdfReceiptError("PDF input manifest is missing its complete section/file set") + expected_sections = [_relative(path, root) for path in hydrated_section_files(root / "output" / "manuscript")] + if sections != expected_sections: + raise PdfReceiptError("hydrated PDF section set/order changed") + by_path: dict[str, Mapping[str, Any]] = {} + for row in rows: + if not isinstance(row, Mapping) or not isinstance(row.get("path"), str): + raise PdfReceiptError("PDF input manifest contains a malformed file row") + name = row["path"] + path = Path(name) + if path.is_absolute() or ".." in path.parts or _relative(root / path, root) != name or name in by_path: + raise PdfReceiptError("PDF input manifest has an unsafe or duplicate path") + if ( + not isinstance(row.get("sha256"), str) or not re.fullmatch(r"[0-9a-f]{64}", row["sha256"]) + or type(row.get("size_bytes")) is not int or row["size_bytes"] < 0 + or row.get("kind") not in ("config", "preamble", "bibliography", "section", "asset") + ): + raise PdfReceiptError(f"PDF input digest row is malformed: {name}") + current = _record(root / path, root, row["kind"]) + if current["sha256"] != row["sha256"] or current["size_bytes"] != row["size_bytes"]: + raise PdfReceiptError(f"PDF input changed: {name}") + by_path[name] = row + if any(name not in by_path or by_path[name]["kind"] != kind for name, kind in _PRIMARY_INPUTS.items()): + raise PdfReceiptError("PDF input manifest is missing configuration/preamble/bibliography") + if {name for name, row in by_path.items() if row["kind"] == "section"} != set(sections): + raise PdfReceiptError("PDF input manifest does not cover the complete hydrated section set") + if {name for name, row in by_path.items() if row["kind"] == "asset"} != set(assets): + raise PdfReceiptError("PDF input manifest does not cover its complete image set") + + +def write_pdf_receipt( + target: Path, *, root: Path, source_revision: str, inputs: Mapping[str, Any], +) -> Path: + validate_pdf_inputs(root, inputs) + if not re.fullmatch(r"[0-9a-f]{40}", source_revision): + raise PdfReceiptError("PDF source revision must be the observed full Git commit") + output = _record(target, root, "pdf") + if output["size_bytes"] <= 0: + raise PdfReceiptError("finalized PDF is empty") + receipt = target.with_suffix(".receipt.json") + return write_json_artifact(receipt, { + "schema": "active-skillference-pdf/v2", + "output_path": output["path"], "size_bytes": output["size_bytes"], + "sha256": output["sha256"], "source_revision": source_revision, + "inputManifest": dict(inputs), + }) + + +def pdf_receipt_issues(root: Path, target: Path | None = None) -> list[str]: + target = target if target is not None else root / "output" / "pdf" / PDF_NAME + receipt = target.with_suffix(".receipt.json") + try: + payload = json.loads(receipt.read_text(encoding="utf-8")) + if not isinstance(payload, dict) or payload.get("schema") != "active-skillference-pdf/v2": + raise PdfReceiptError("missing/legacy PDF input provenance; a genuine re-render is required") + manifest = payload.get("inputManifest") + if not isinstance(manifest, Mapping): + raise PdfReceiptError("PDF receipt has no complete input manifest") + validate_pdf_inputs(root, manifest) + output = _record(target, root, "pdf") + if ( + payload.get("output_path") != output["path"] + or payload.get("sha256") != output["sha256"] + or payload.get("size_bytes") != output["size_bytes"] + ): + raise PdfReceiptError("finalized PDF bytes do not match the recorded render") + revision = payload.get("source_revision") + if not isinstance(revision, str) or not re.fullmatch(r"[0-9a-f]{40}", revision): + raise PdfReceiptError("PDF receipt has no observed source revision") + except (OSError, ValueError) as exc: + return [f"PDF freshness failed: {exc}; re-render before publishing"] + return [] diff --git a/src/pilot/contract.py b/src/pilot/contract.py index 0f9bb10..154dd57 100644 --- a/src/pilot/contract.py +++ b/src/pilot/contract.py @@ -32,8 +32,7 @@ def validate_pilot_contract(payload: Mapping[str, Any]) -> None: raise PilotContractError("contractVersion must be 1") _require_text(payload, "status", expected="protocol-only") _require_text(payload, "purpose") - _require_mapping(payload, "claimsBoundary") - boundary = payload["claimsBoundary"] + boundary = _require_mapping(payload, "claimsBoundary") _require_list(boundary, "allowed") _require_list(boundary, "prohibited") if not boundary["allowed"] or not boundary["prohibited"]: @@ -50,13 +49,14 @@ def validate_pilot_contract(payload: Mapping[str, Any]) -> None: if participants.get("directIdentifiersAllowed") is not False: raise PilotContractError("directIdentifiersAllowed must be false") fields = _require_list(participants, "requiredFields") - _unique_texts(fields, "participants.requiredFields") + _unique_texts(fields, "participants.requiredFields", {"participant_id_hash"}) events = _require_mapping(payload, "events") event_types = _require_list(events, "allowedEventTypes") _unique_texts(event_types, "events.allowedEventTypes") event_fields = _require_list(events, "requiredFields") - _unique_texts(event_fields, "events.requiredFields") + linking_fields = {"participant_id_hash", "event_type", "event_timestamp"} + _unique_texts(event_fields, "events.requiredFields", linking_fields) outcomes = _require_mapping(payload, "outcomes") primary = _require_mapping(outcomes, "primary") @@ -87,7 +87,7 @@ def validate_pilot_contract(payload: Mapping[str, Any]) -> None: export = _require_mapping(payload, "export") _require_text(export, "sourceBoundary") export_fields = _require_list(export, "fields") - _unique_texts(export_fields, "export.fields") + _unique_texts(export_fields, "export.fields", linking_fields) def _require_mapping(payload: Mapping[str, Any], key: str) -> dict[str, Any]: @@ -115,7 +115,11 @@ def _require_text( return value -def _unique_texts(values: list[Any], location: str) -> None: +def _unique_texts( + values: list[Any], location: str, required: set[str] | None = None +) -> None: + if not values: + raise PilotContractError(f"{location} must not be empty") texts = [] for index, value in enumerate(values, start=1): if not isinstance(value, str) or not value.strip(): @@ -123,3 +127,8 @@ def _unique_texts(values: list[Any], location: str) -> None: texts.append(value) if len(texts) != len(set(texts)): raise PilotContractError(f"{location} must not contain duplicates") + missing = (required or set()) - set(texts) + if missing: + raise PilotContractError( + f"{location} is missing linking fields: {', '.join(sorted(missing))}" + ) diff --git a/src/pipeline/generated_outputs.py b/src/pipeline/generated_outputs.py index 539fe6d..87b8ec7 100644 --- a/src/pipeline/generated_outputs.py +++ b/src/pipeline/generated_outputs.py @@ -98,7 +98,7 @@ def build_generated_outputs(project_root: Path | None = None) -> GeneratedOutput banned = audit_manuscript_banned_phrases(manuscript_dir) if banned: detail = "; ".join(f"{name}: {hits}" for name, hits in banned.items()) - raise SystemExit( + raise ValueError( f"manuscript endorsement-overreach phrases found: {detail}" ) scholarship_audit = validate_scholarship( diff --git a/src/provenance/binding.py b/src/provenance/binding.py index 1686a8d..a08b10e 100644 --- a/src/provenance/binding.py +++ b/src/provenance/binding.py @@ -20,12 +20,11 @@ from __future__ import annotations import re -import unicodedata from dataclasses import dataclass, replace from pathlib import Path from src.curriculum.graph import CurriculumGraph -from src.curriculum.model import CLAIM_TOKEN_RE, Skill, Subject +from src.curriculum.model import CLAIM_TOKEN_RE, Skill, Subject, _strip_format_chars from src.provenance.claims import ClaimRegistry # The shape a computed *result* takes: either a decimal with >=2 fractional digits @@ -66,17 +65,6 @@ class UnbackedNumber: context: str -def _strip_format_chars(text: str) -> str: - """Remove Unicode *format* characters (category Cf: zero-width space U+200B, - ZWJ/ZWNJ, BOM, word-joiner, …). - - These are invisible and neither ``\\s`` nor ``\\b`` match them, so a token or number - split by one (``{{claim:e}}``, ``1.6094``) would slip past both the - malformed-token check and the unbacked-number linter. Stripping them before any - integrity scan closes that bypass class; they have no legitimate place in curriculum - prose, so removal is safe and makes the gates see the real glyph sequence. - """ - return "".join(ch for ch in text if unicodedata.category(ch) != "Cf") def render(text: str, registry: ClaimRegistry) -> str: @@ -104,21 +92,26 @@ def _sub(match: re.Match[str]) -> str: def validate_skill(skill: Skill, registry: ClaimRegistry) -> None: - """Raise if any claim token in the skill is unbacked.""" - missing = sorted(t for t in skill.claim_tokens() if not registry.has(t)) - if missing: - raise ProvenanceError( - f"skill {skill.skill_id!r} references unbacked claims: {', '.join(missing)}" - ) + """Apply canonical fail-closed token admission to every learner-facing skill field.""" + texts = [skill.name, skill.objective, skill.description] + for question in skill.assessments: + texts.append(question.prompt) + texts.extend(question.answers) + try: + for text in texts: + render(text, registry) + except ProvenanceError as exc: + raise ProvenanceError(f"skill {skill.skill_id!r}: {exc}") from exc def validate_curriculum(skills: list[Skill], registry: ClaimRegistry) -> None: - """Fail closed if ANY skill references an unbacked claim. Used by the exporter.""" + """Fail closed on unbacked or malformed learner-field tokens before export.""" problems: list[str] = [] for skill in skills: - missing = sorted(t for t in skill.claim_tokens() if not registry.has(t)) - if missing: - problems.append(f"{skill.skill_id}: {', '.join(missing)}") + try: + validate_skill(skill, registry) + except ProvenanceError as exc: + problems.append(str(exc)) if problems: raise ProvenanceError( "unbacked claim tokens (export refused):\n " + "\n ".join(problems) @@ -127,7 +120,7 @@ def validate_curriculum(skills: list[Skill], registry: ClaimRegistry) -> None: def render_graph(graph: CurriculumGraph, registry: ClaimRegistry) -> CurriculumGraph: """Return a copy of the graph with every claim token resolved in EVERY learner-facing - field: subject description, skill name/objective/description, and each assessment's + field: subject name/description, skill name/objective/description, and each assessment's prompt and answer options. Fails closed: :func:`render` raises :class:`ProvenanceError` on any unbacked OR @@ -160,11 +153,14 @@ def _render(text: str) -> str: new_subjects.append( replace( subject, + name=_render(subject.name), description=_render(subject.description), skills=rendered_skills, ) ) - return CurriculumGraph(subjects=new_subjects) + rendered = CurriculumGraph(subjects=new_subjects) + rendered.validate() + return rendered def _strip_protected(text: str) -> str: @@ -266,10 +262,10 @@ def lint_curriculum( def lint_graph( graph: CurriculumGraph, *, allow: tuple[str, ...] = () ) -> list[UnbackedNumber]: - """Lint every learner-facing free-text field in the graph: subject descriptions, - skill names, and skill descriptions. Empty list == clean.""" + """Lint subject names/descriptions and learner-facing skill prose.""" violations = lint_curriculum(graph.skills, allow=allow) for subject in graph.subjects: + violations.extend(_scan_text(subject.name, subject.subject_id, allow)) violations.extend(_scan_text(subject.description, subject.subject_id, allow)) return violations diff --git a/src/publication/zenodo.py b/src/publication/zenodo.py index 239ff74..9a6e673 100644 --- a/src/publication/zenodo.py +++ b/src/publication/zenodo.py @@ -1,19 +1,15 @@ -"""Reserve a Zenodo concept DOI without publishing anything. +"""Explicit Zenodo reservation, upload and publication operations. -Zenodo allocates a DOI to an unpublished draft deposition. That allocation is -real — the identifier is issued to this account and nobody else can receive it — -but it does not resolve until the draft is published, and a draft can still be -discarded. Reserving is therefore the one DOI action that is not irreversible, -which is why this module can perform it and deliberately cannot publish. - -There is no publish call here on purpose. Minting a live DOI is a permanent, -outward-facing act and belongs to a human running the release tooling, not to a -build step. +A reserved DOI belongs to an unpublished draft and does not resolve until that +draft is published. Publication is a separate irreversible operation guarded by +the expected DOI. No build/import path publishes or contacts Zenodo implicitly; +release authority and credentials remain with the operator. """ from __future__ import annotations import json +import os import urllib.error import urllib.parse import urllib.request @@ -162,37 +158,33 @@ def find_draft(*, token: str, title: str) -> ReservedDoi | None: def upload_deposit_file(*, token: str, deposition_id: int, path: Path) -> str: - """Attach one file to a draft deposition and return its Zenodo filename. - - Zenodo will not publish a deposition with no files, so this runs first. It - uses the modern bucket API, which streams the body rather than building a - multipart form in memory. - """ + """Stream one real file to the draft bucket with an explicit content length.""" deposition = _request(f"/deposit/depositions/{deposition_id}", token=token) - bucket = str(deposition.get("links", {}).get("bucket", "")) - if not bucket: - raise ZenodoError( - f"deposition {deposition_id} exposes no upload bucket; it may already " - "be published" - ) - request = urllib.request.Request( - f"{bucket}/{urllib.parse.quote(path.name)}", - data=path.read_bytes(), - method="PUT", - headers={ - "Authorization": f"Bearer {token}", - "Content-Type": "application/octet-stream", - }, - ) + links = deposition.get("links", {}) if isinstance(deposition, dict) else {} + bucket = links.get("bucket") if isinstance(links, dict) else None + if not isinstance(bucket, str) or not bucket: + raise ZenodoError(f"deposition {deposition_id} exposes no upload bucket; it may already be published") try: - with urllib.request.urlopen(request, timeout=600) as response: - payload = json.loads(response.read().decode("utf-8")) + with path.open("rb") as stream: + request = urllib.request.Request( + f"{bucket}/{urllib.parse.quote(path.name)}", + data=stream, method="PUT", + headers={ + "Authorization": f"Bearer {token}", + "Content-Type": "application/octet-stream", + "Content-Length": str(os.fstat(stream.fileno()).st_size), + }, + ) + with urllib.request.urlopen(request, timeout=600) as response: + payload = json.loads(response.read().decode("utf-8")) except urllib.error.HTTPError as exc: - detail = exc.read().decode("utf-8")[:500] + detail = exc.read().decode("utf-8", errors="replace")[:500] raise ZenodoError(f"Zenodo file upload failed ({exc.code}): {detail}") from exc - except (urllib.error.URLError, TimeoutError, ValueError) as exc: + except (OSError, ValueError) as exc: raise ZenodoError(f"Zenodo file upload failed: {exc}") from exc - return str(payload.get("key", path.name)) + if not isinstance(payload, dict) or not isinstance(payload.get("key"), str) or not payload["key"].strip(): + raise ZenodoError("Zenodo upload returned no confirmed filename") + return payload["key"] def publish_deposit( @@ -261,7 +253,7 @@ def reserve_doi(*, token: str, metadata: dict[str, Any]) -> ReservedDoi: ) if result.is_published: raise ZenodoError( - f"deposition {deposition_id} reports as published; this module only " - "reserves and must never mint a resolving DOI" + f"deposition {deposition_id} reports as published; the reservation " + "operation refuses an already-published record" ) return result diff --git a/src/release_readiness.py b/src/release_readiness.py index a221ee7..77ffe91 100644 --- a/src/release_readiness.py +++ b/src/release_readiness.py @@ -115,13 +115,18 @@ def build_release_readiness_snapshot(root: Path) -> ReleaseReadinessSnapshot: ) +def probe_public_repository( + root: Path, repository_url: str = PUBLIC_REPOSITORY_URL, +) -> subprocess.CompletedProcess[str]: + """Observe all advertised refs without fetching or mutating either repository.""" + return _run_command( + ["git", "ls-remote", "--refs", repository_url], cwd=root, timeout=20, + ) + + def collect_blocker_probes(root: Path) -> BlockerProbeSummary: """Probe volatile external blockers for inventory rendering.""" - public = _run_command( - ["git", "ls-remote", "--heads", PUBLIC_REPOSITORY_URL, "HEAD"], - cwd=root, - timeout=20, - ) + public = probe_public_repository(root) docker = _run_command(["docker", "info"], cwd=root, timeout=20) return BlockerProbeSummary( public_repository=_summarize_public_repository_probe(public), @@ -132,12 +137,9 @@ def collect_blocker_probes(root: Path) -> BlockerProbeSummary: def _summarize_public_repository_probe( result: subprocess.CompletedProcess[str], ) -> str: - """Report reachability *and* whether the public target has any content. + """Separate transport failure, an empty target and advertised repository refs. - ``git ls-remote`` exits 0 for a repository that exists but holds no refs, so - a bare exit-code check reads an empty placeholder repository as ready. That - is the difference between "the target exists" and "the release is published", - and DOI/archive metadata must stay blank across that gap. + Neither transport success nor advertised refs proves a release/archive/DOI. """ if result.returncode != 0: return _summarize_probe(result) @@ -158,6 +160,20 @@ def render_release_readiness_inventory( public_repository="not refreshed by this validation run", docker_daemon="not refreshed by this validation run", ) + if probes.public_repository == "reachable (refs present)": + public_observation = "The repository is reachable and advertises refs." + elif probes.public_repository.startswith("reachable but empty"): + public_observation = "The repository is reachable but advertises no refs." + elif probes.public_repository.startswith("not refreshed"): + public_observation = "No fresh public-repository observation was made by this validation run." + else: + public_observation = "The public-repository probe failed or was unavailable." + public_release_line = ( + f"- {public_observation} Advertised-ref probe: " + f"`git ls-remote --refs {PUBLIC_REPOSITORY_URL}`. " + f"Current probe: {probes.public_repository}. " + "Release approval, public archive availability and DOI publication require independent evidence." + ) status_count = len(snapshot.status_lines) tracked_count = len(snapshot.modified_tracked_files) untracked_count = len(snapshot.untracked_files) @@ -269,11 +285,7 @@ def render_release_readiness_inventory( "", "## Remaining Release Blockers", "", - ( - "- Public DOI/repository remains blocked until " - f"`git ls-remote {PUBLIC_REPOSITORY_URL}` succeeds. Current probe: " - f"{probes.public_repository}." - ), + public_release_line, live_stack_line, ( "- Cross-vendor security audit evidence is closed as a point-in-time " diff --git a/src/scholarship/source_health_report.py b/src/scholarship/source_health_report.py index 260738b..9f74053 100644 --- a/src/scholarship/source_health_report.py +++ b/src/scholarship/source_health_report.py @@ -35,10 +35,6 @@ def scholarship_source_health_network( bot_blocks = _issue_rows(source_health.issue_list, _BOT_BLOCK_CODES) role_gaps = _issue_rows(source_health.issue_list, _ROLE_GAP_CODES) metadata_issues = _issue_rows(source_health.issue_list, _METADATA_CODES) - bot_blocks.extend(_online_status_rows(source_health, "http_error:403")) - dead_links.extend(_online_status_rows(source_health, "url_error:")) - dead_links.extend(_online_status_rows(source_health, "timeout:")) - dead_links.extend(_online_status_rows(source_health, "tls_error:")) drift = { "singleSectionWithoutClaimSupport": list( advisories.get("singleSectionWithoutClaimSupport", []) @@ -104,15 +100,6 @@ def _issue_rows( ] -def _online_status_rows( - audit: SourceHealthAudit, - prefix: str, -) -> list[dict[str, object]]: - return [ - {"code": "online_status", "key": key, "message": status} - for key, status in audit.online_results.items() - if status.startswith(prefix) - ] def _boundary_source_row(row: dict[str, Any]) -> dict[str, object]: diff --git a/src/screenshots/capture.py b/src/screenshots/capture.py index d4cbdbc..dbdabc0 100644 --- a/src/screenshots/capture.py +++ b/src/screenshots/capture.py @@ -15,6 +15,7 @@ import shutil import subprocess import urllib.error +import urllib.parse import urllib.request from dataclasses import dataclass from datetime import UTC, datetime @@ -22,7 +23,11 @@ from typing import Any from src.artifacts.json_io import write_json_artifact -from src.net_guard import NonHttpUrlError, require_http_url +from src.net_guard import NonHttpUrlError, is_loopback_host, require_http_url +from src.dashboard.browser_validation import ( + BrowserDisplayError, browser_receipt_complete, display_probe, wait_for_official_display, +) +from src.skilltree.local_auth_state import LocalSkillTreeAuthError, _skilltree_origin from src.screenshots.manifest import ( SCREENSHOT_PROVENANCE_FILENAME, VIEWPORT_HEIGHT, @@ -54,6 +59,7 @@ class CaptureSettings: viewport_height: int = VIEWPORT_HEIGHT #: Directory the browser automation server is permitted to write to. staging_root: Path = Path(DEFAULT_STAGING_ROOT) + proof_path: Path | None = None def _run( @@ -110,13 +116,49 @@ def ensure_browser_ready(*, browser_command: str) -> bool: return True -def _fetch_json(url: str, *, timeout: float) -> dict[str, Any]: +def _capture_url(url: str) -> str: try: - require_http_url(url, context="dashboard status read for screenshot capture") - except NonHttpUrlError as exc: - raise ScreenshotCaptureError(str(exc)) from exc + parsed = urllib.parse.urlsplit(url) + except ValueError as exc: + raise ScreenshotCaptureError("capture URL is invalid") from exc + if parsed.username is not None or parsed.password is not None or parsed.query or parsed.fragment: + raise ScreenshotCaptureError("capture requires a credential-free dashboard base URL without query or fragment") try: - with urllib.request.urlopen(url, timeout=timeout) as response: + require_http_url(url, context="screenshot capture") + port = parsed.port + except (NonHttpUrlError, ValueError) as exc: + raise ScreenshotCaptureError("capture requires an absolute loopback HTTP(S) URL") from exc + if not parsed.hostname or not is_loopback_host(parsed.hostname): + raise ScreenshotCaptureError("capture requires an absolute loopback HTTP(S) URL") + if port is not None and port <= 0: + raise ScreenshotCaptureError("capture URL port is invalid") + return url.rstrip("/") + + +def _capture_target(url: str) -> tuple[tuple[str, str, int], str]: + normalized = _capture_url(url) + return _skilltree_origin(normalized), urllib.parse.urlsplit(normalized).path.rstrip("/") or "/" + + +class _LoopbackCaptureRedirect(urllib.request.HTTPRedirectHandler): + def redirect_request( + self, req: urllib.request.Request, fp: Any, code: int, msg: str, + headers: Any, newurl: str, + ) -> urllib.request.Request | None: + _capture_url(newurl) + if _skilltree_origin(req.full_url) != _skilltree_origin(newurl): + raise ScreenshotCaptureError("capture redirect changed the requested origin") + return super().redirect_request(req, fp, code, msg, headers, newurl) + + +def _capture_opener() -> urllib.request.OpenerDirector: + return urllib.request.build_opener(_LoopbackCaptureRedirect) + + +def _fetch_json(url: str, *, timeout: float) -> dict[str, Any]: + _capture_url(url) + try: + with _capture_opener().open(url, timeout=timeout) as response: payload = json.loads(response.read().decode("utf-8")) except (urllib.error.URLError, TimeoutError, ValueError) as exc: raise ScreenshotCaptureError(f"could not read {url}: {exc}") from exc @@ -125,53 +167,118 @@ def _fetch_json(url: str, *, timeout: float) -> dict[str, Any]: return payload -def capture_context(settings: CaptureSettings) -> dict[str, Any]: - """Return the system state the capture run is evidence of.""" - status = _fetch_json( - f"{settings.dashboard_url.rstrip('/')}/api/status", - timeout=_STATUS_TIMEOUT_SECONDS, - ) - project_counts = status.get("project", {}).get("counts", {}) - live = status.get("liveSkilltree", {}) - proof_path = ( - settings.project_root - / ".active_skillference_local" - / "full_app_stack_proof.json" - ) - proof_mode = "not recorded" - if proof_path.is_file(): +def capture_context( + settings: CaptureSettings, *, browser_probe: dict[str, Any] | None = None, +) -> dict[str, Any]: + """Observe current status and associate only an explicit, matching successful proof.""" + dashboard_url = _capture_url(settings.dashboard_url) + status = _fetch_json(f"{dashboard_url}/api/status", timeout=_STATUS_TIMEOUT_SECONDS) + project, config, live = status.get("project"), status.get("config"), status.get("liveSkilltree") + service = status.get("skilltree") + if not all(isinstance(value, dict) for value in (project, config, live, service)): + raise ScreenshotCaptureError("dashboard project/config/live state must contain objects") + project_counts = project.get("counts", {}) + public_project, skilltree = config.get("project", {}), config.get("skilltree", {}) + if not all(isinstance(value, dict) for value in (project_counts, public_project, skilltree)): + raise ScreenshotCaptureError("dashboard project counts/identity/target must contain objects") + instance_id = status.get("instanceId") + digest = project.get("exportContentDigest") + project_id = public_project.get("id") + skilltree_url = str(skilltree.get("url", "")) + if browser_probe is not None: try: - proof = json.loads(proof_path.read_text(encoding="utf-8")) - proof_mode = str(proof.get("mode", "not recorded")) - except (OSError, ValueError): - proof_mode = "unreadable proof artifact" + browser_matches = _capture_target(str(browser_probe.get("url", ""))) == _capture_target(dashboard_url) + except (ScreenshotCaptureError, LocalSkillTreeAuthError, ValueError): + browser_matches = False + if ( + browser_probe.get("state") != "ready" + or browser_probe.get("dashboardInstanceId") != instance_id + or browser_probe.get("exportContentDigest") != digest + or browser_probe.get("projectId") != project_id + or not browser_matches + ): + raise ScreenshotCaptureError("browser capture state does not match current dashboard content") + association = _proof_association( + settings, instance_id=instance_id, digest=digest, project_id=project_id, + skilltree_url=skilltree_url, browser_probe=browser_probe, + ) + content = live.get("content", {}) return { - "dashboardUrl": settings.dashboard_url, - "skilltreeUrl": str( - status.get("config", {}).get("skilltree", {}).get("url", "") - ), - "skilltreeAvailable": bool(status.get("skilltree", {}).get("available")), - "viewport": { - "width": settings.viewport_width, - "height": settings.viewport_height, - }, - "exportedCounts": { - "skills": project_counts.get("skills", 0), - "subjects": project_counts.get("subjects", 0), - "dependencies": project_counts.get("dependencies", 0), - "quizzes": project_counts.get("quizzes", 0), - }, + "dashboardUrl": dashboard_url, + "dashboardInstanceId": instance_id, + "exportContentDigest": digest, + "projectId": project_id, + "skilltreeUrl": skilltree_url, + "skilltreeAvailable": bool(service.get("available")), + "viewport": {"width": settings.viewport_width, "height": settings.viewport_height}, + "exportedCounts": {key: project_counts.get(key, 0) for key in ("skills", "subjects", "dependencies", "quizzes")}, "liveDependencies": { - "checked": bool(live.get("checked")), - "matched": live.get("matchedDependencies"), - "expected": live.get("expectedDependencies"), - "complete": bool(live.get("complete")), + "checked": live.get("checked") is True, "matched": live.get("matchedDependencies"), + "expected": live.get("expectedDependencies"), "complete": live.get("complete") is True, }, - "fullAppStackProofMode": proof_mode, - "evidenceKind": ( - "captured browser recording of a live system; pixels are not " - "reproducible byte-for-byte and are not a deterministic render" - ), + "liveContent": { + key: content.get(key) for key in ("checked", "complete", "exportContentDigest") + } if isinstance(content, dict) else {"checked": False}, + "officialDisplay": { + key: browser_probe.get(key) for key in ("state", "route", "frameUrl", "frameVisible") + } if browser_probe is not None else {"state": "unobserved"}, + "proofAssociation": association, + "fullAppStackProofMode": association.get("mode", "not associated"), + "evidenceKind": "captured browser recording of a live system; pixels are not reproducible byte-for-byte and are not a deterministic render", + } + + +def _proof_association( + settings: CaptureSettings, *, instance_id: Any, digest: Any, project_id: Any, + skilltree_url: str, browser_probe: dict[str, Any] | None, +) -> dict[str, Any]: + if settings.proof_path is None: + return {"checked": False, "reason": "no explicit proof path"} + path = settings.proof_path + if not path.is_absolute(): + path = settings.project_root / path + try: + raw = path.read_bytes() + proof = json.loads(raw) + except (OSError, ValueError) as exc: + return {"checked": False, "reason": f"proof is unreadable: {type(exc).__name__}"} + if not isinstance(proof, dict) or proof.get("proofVersion") != 2: + return {"checked": False, "reason": "historical/unsupported proof schema is unassociated"} + content = proof.get("liveContent", {}) + browser = proof.get("browserConsole", {}) + try: + same_targets = ( + _capture_target(str(proof.get("dashboardUrl", ""))) == _capture_target(settings.dashboard_url) + and _skilltree_origin(str(proof.get("skilltreeUrl", ""))) == _skilltree_origin(skilltree_url) + ) + except (ScreenshotCaptureError, LocalSkillTreeAuthError, ValueError): + same_targets = False + complete = ( + proof.get("ok") is True and same_targets + and proof.get("mode") in ("app-ready", "full-parity") + and isinstance(instance_id, str) and bool(instance_id) + and isinstance(digest, str) and len(digest) == 64 + and proof.get("runId") == proof.get("dashboardInstanceId") == instance_id + and proof.get("exportContentDigest") == digest + and proof.get("projectId") == project_id + and isinstance(content, dict) and content.get("checked") is True + and content.get("complete") is True and content.get("exportContentDigest") == digest + and browser_probe is not None and browser_receipt_complete(browser) + and browser.get("dashboardInstanceId") == instance_id + and browser.get("exportContentDigest") == digest + and browser.get("projectId") == project_id + ) + if proof.get("mode") == "full-parity": + dependencies = proof.get("liveDependencies", {}) + complete = ( + complete and isinstance(dependencies, dict) + and dependencies.get("checked") is True and dependencies.get("complete") is True + ) + if not complete: + return {"checked": False, "reason": "proof does not match this successful current browser/content/instance state"} + return { + "checked": True, "mode": proof["mode"], "runId": proof["runId"], + "proofSha256": hashlib.sha256(raw).hexdigest(), "proofPath": str(path), } @@ -182,6 +289,11 @@ def _capture_one( ) -> dict[str, Any]: _run([settings.browser_command, "eval", view.setup_js]) _run([settings.browser_command, "wait", str(view.settle_ms)]) + try: + probe = display_probe(browser_command=settings.browser_command, timeout=30.0) + except BrowserDisplayError as exc: + raise ScreenshotCaptureError(str(exc)) from exc + context = capture_context(settings, browser_probe=probe) # The browser automation server runs under its own filesystem sandbox and # silently reports success while writing nothing outside /tmp, so captures # are staged there and moved into the tracked tree by this process. @@ -192,6 +304,12 @@ def _capture_one( f"browser reported success but {staged} was not written; the browser " "automation server may not be permitted to write to that directory" ) + try: + after = display_probe(browser_command=settings.browser_command, timeout=30.0) + except BrowserDisplayError as exc: + raise ScreenshotCaptureError(str(exc)) from exc + if any(after.get(key) != probe.get(key) for key in ("state", "route", "url", "dashboardInstanceId", "exportContentDigest")): + raise ScreenshotCaptureError("browser state changed while the screenshot was being recorded") target = settings.output_dir / view.filename target.write_bytes(staged.read_bytes()) staged.unlink() @@ -203,14 +321,16 @@ def _capture_one( "settleMs": view.settle_ms, "sha256": digest, "bytes": target.stat().st_size, + "context": context, } def capture_screenshots(settings: CaptureSettings) -> dict[str, Any]: """Capture every manifest view and write the registry plus provenance record.""" - settings.output_dir.mkdir(parents=True, exist_ok=True) + dashboard_url = _capture_url(settings.dashboard_url) + with _capture_opener().open(dashboard_url, timeout=_STATUS_TIMEOUT_SECONDS) as response: + _capture_url(response.geturl()) restarted = ensure_browser_ready(browser_command=settings.browser_command) - context = capture_context(settings) _run( [ settings.browser_command, @@ -220,9 +340,20 @@ def capture_screenshots(settings: CaptureSettings) -> dict[str, Any]: ] ) _run([settings.browser_command, "open", settings.dashboard_url]) - # The page defers the live-parity read, so give the swap-in time to land - # before the first capture claims a parity number. - _run([settings.browser_command, "wait", "12000"]) + initial = _fetch_json(f"{dashboard_url}/api/status", timeout=_STATUS_TIMEOUT_SECONDS) + try: + probe = wait_for_official_display( + browser_command=settings.browser_command, dashboard_url=dashboard_url, + skilltree_url=str(initial["config"]["skilltree"]["url"]), + project_id=str(initial["config"]["project"]["id"]), route="/", + dashboard_instance_id=str(initial["instanceId"]), + export_digest=str(initial["project"]["exportContentDigest"]), + timeout=_STATUS_TIMEOUT_SECONDS, + ) + except (BrowserDisplayError, KeyError, TypeError) as exc: + raise ScreenshotCaptureError("official display did not provide capture-ready identity") from exc + context = capture_context(settings, browser_probe=probe) + settings.output_dir.mkdir(parents=True, exist_ok=True) staging_dir = Path(settings.staging_root) / f"asf-capture-{os.getpid()}" staging_dir.mkdir(parents=True, exist_ok=True) diff --git a/src/skilltree/client.py b/src/skilltree/client.py index a8cccc3..876c908 100644 --- a/src/skilltree/client.py +++ b/src/skilltree/client.py @@ -8,6 +8,7 @@ from __future__ import annotations import json +import math import urllib.error import urllib.request from typing import Any @@ -35,6 +36,7 @@ SkillTreeSkill, SkillTreeSubject, ) +from src.skilltree.content import missing_quiz_questions from src.skilltree.transport import ( NoCrossOriginBearerRedirect, SkillTreeClientError, @@ -49,12 +51,15 @@ class SkillTreeClient: """Thin admin/report client for a single SkillTree project.""" def __init__( - self, base_url: str, project_id: str, token: str | None = None + self, base_url: str, project_id: str, token: str | None = None, + *, timeout: float = 30.0, ) -> None: if not base_url.strip(): raise ValueError("base_url must be non-empty") if not project_id.strip(): raise ValueError("project_id must be non-empty") + if not math.isfinite(timeout) or timeout <= 0: + raise ValueError("timeout must be finite and positive") try: self.base_url = require_http_url( base_url.rstrip("/"), context="SkillTree client" @@ -63,6 +68,7 @@ def __init__( raise ValueError(str(exc)) from exc self.project_id = project_id self.token = token + self.timeout = timeout self._opener = urllib.request.build_opener(NoCrossOriginBearerRedirect) def _get(self, path: str) -> Any: @@ -72,7 +78,7 @@ def _get(self, path: str) -> Any: headers["Authorization"] = f"Bearer {self.token}" request = urllib.request.Request(url, headers=headers, method="GET") try: - with self._opener.open(request) as response: + with self._opener.open(request, timeout=self.timeout) as response: status = response.status raw = response.read().decode("utf-8") except urllib.error.HTTPError as exc: @@ -82,6 +88,8 @@ def _get(self, path: str) -> Any: status=exc.code, body=raw, ) from exc + except TimeoutError as exc: + raise SkillTreeClientError(f"GET {path} timed out") from exc except urllib.error.URLError as exc: # pragma: no cover - network-dependent raise SkillTreeClientError( f"GET {path} transport error: {exc.reason}" @@ -101,7 +109,7 @@ def _post(self, path: str, body: dict[str, Any] | None = None) -> Any: headers["Authorization"] = f"Bearer {self.token}" request = urllib.request.Request(url, data=data, headers=headers, method="POST") try: - with self._opener.open(request) as response: + with self._opener.open(request, timeout=self.timeout) as response: status = response.status raw = response.read().decode("utf-8") except urllib.error.HTTPError as exc: @@ -111,6 +119,8 @@ def _post(self, path: str, body: dict[str, Any] | None = None) -> Any: status=exc.code, body=raw, ) from exc + except TimeoutError as exc: + raise SkillTreeClientError(f"POST {path} timed out") from exc except urllib.error.URLError as exc: # pragma: no cover - network-dependent raise SkillTreeClientError( f"POST {path} transport error: {exc.reason}" @@ -152,13 +162,8 @@ def create_quiz(self, quiz: Quiz) -> Any: return result def _ensure_quiz_questions(self, quiz: Quiz) -> None: - if not quiz.questions: - return existing = self._get(quiz_questions_path(quiz.quiz_id)) - existing_count = min(quiz_question_count(existing), len(quiz.questions)) - if existing_count >= len(quiz.questions): - return - for question in quiz.questions[existing_count:]: + for question in missing_quiz_questions(existing, quiz.questions): self.create_quiz_question(quiz.quiz_id, question) def create_quiz_question(self, quiz_id: str, question: QuizQuestion) -> Any: diff --git a/src/skilltree/content.py b/src/skilltree/content.py new file mode 100644 index 0000000..b2c5a62 --- /dev/null +++ b/src/skilltree/content.py @@ -0,0 +1,239 @@ +"""Semantic, multiplicity-aware comparison of exported and live admin content. + +This module never updates or deletes live resources. Missing resources can be +created by the existing seeder; changed or unverifiable resources fail closed. +Only identifiers, counts and a digest leave the admin boundary, never answer keys. +""" + +from __future__ import annotations + +import hashlib +import json +from collections import Counter +from collections.abc import Callable, Mapping, Sequence +from typing import Any + +from src.skilltree.admin_ops import ( + project_create_path, + quiz_definition_path, + quiz_questions_path, + skill_create_path, + subject_create_path, +) +from src.skilltree.local_auth_state import _skilltree_origin +from src.skilltree.model import QuizQuestion +from src.skilltree.summary import validate_export_payload +from src.skilltree.transport import SkillTreeClientError + + +class SkillTreeContentError(SkillTreeClientError): + """Live content cannot safely be identified with an exported resource.""" + + +def export_content_digest(project: Mapping[str, Any]) -> str: + """Bind a receipt to the complete private export, including answer correctness.""" + validate_export_payload(project) + raw = json.dumps(project, sort_keys=True, separators=(",", ":"), ensure_ascii=False) + return hashlib.sha256(raw.encode("utf-8")).hexdigest() + + +def _object(payload: Any, resource: str) -> Mapping[str, Any]: + if isinstance(payload, Mapping): + data = payload.get("data") + if isinstance(data, Mapping): + return data + return payload + raise SkillTreeContentError(f"{resource} returned no verifiable content object") + + +def _rows(payload: Any, key: str, resource: str) -> list[Mapping[str, Any]]: + if isinstance(payload, Mapping): + if key in payload: + payload = payload[key] + elif "data" in payload: + return _rows(payload["data"], key, resource) + if not isinstance(payload, list) or any(not isinstance(row, Mapping) for row in payload): + raise SkillTreeContentError(f"{resource} returned no verifiable {key} list") + return payload + + +def _question_key(question: Mapping[str, Any]) -> str: + text = question.get("question") + question_type = question.get("questionType") + answers = question.get("answers") + if ( + not isinstance(text, str) or not text.strip() + or not isinstance(question_type, str) + or question_type not in {"SingleChoice", "MultipleChoice", "TextInput"} + or not isinstance(answers, list) or not answers + ): + raise SkillTreeContentError("quiz question response lacks verifiable semantic content") + normalized_answers = [] + for answer in answers: + if ( + not isinstance(answer, Mapping) + or not isinstance(answer.get("answer"), str) + or not isinstance(answer.get("isCorrect"), bool) + ): + raise SkillTreeContentError("quiz answer response lacks verifiable correctness") + normalized_answers.append((answer["answer"], answer["isCorrect"])) + return json.dumps( + [text, question_type, sorted(normalized_answers)], + ensure_ascii=False, separators=(",", ":"), + ) + + +def missing_quiz_questions( + payload: Any, expected: Sequence[QuizQuestion], +) -> list[QuizQuestion]: + """Return missing question occurrences; refuse extras or changed answer keys.""" + current = Counter(_question_key(row) for row in _rows(payload, "questions", "quiz")) + wanted = Counter(_question_key(question.to_dict()) for question in expected) + if current - wanted: + raise SkillTreeContentError( + "live quiz contains extra or changed questions; existing content was not overwritten" + ) + missing = [] + for question in expected: + key = _question_key(question.to_dict()) + if current[key]: + current[key] -= 1 + else: + missing.append(question) + return missing + + +def _index(rows: list[Mapping[str, Any]], field: str, resource: str) -> dict[str, Mapping[str, Any]]: + result = {} + for row in rows: + identifier = row.get(field) + if not isinstance(identifier, str) or not identifier or identifier in result: + raise SkillTreeContentError(f"{resource} has missing or duplicate {field}") + result[identifier] = row + return result + + +def _quiz_completion_contract(payload: Any, question_count: int) -> tuple[int, int]: + """Resolve native 4.6.0 length/pass-count settings, never an export-only percent.""" + rows = _index(_rows(payload, "settings", "quiz"), "setting", "quiz settings") + + def integer(name: str) -> int: + if name not in rows: + return -1 + value = rows[name].get("value") + if not isinstance(value, str): + raise SkillTreeContentError(f"quiz setting {name} is not a string integer") + try: + return int(value) + except ValueError as exc: + raise SkillTreeContentError(f"quiz setting {name} is not an integer") from exc + + length = integer("quizLength") + passing = integer("quizPassingReq") + if length == -1: + length = question_count + if length <= 0 or length > question_count or passing < -1 or passing > question_count: + raise SkillTreeContentError("quiz length/pass-count settings are unverifiable") + return length, passing if passing > 0 else length + + +def compare_live_content( + *, get: Callable[[str], Any], base_url: str, project: Mapping[str, Any], +) -> dict[str, Any]: + """Read actual admin content and return a browser-safe comparison receipt. + + A count-only response is not evidence of content parity. Server-generated + identifiers and quiz ordering are ignored; duplicate question/answer + occurrences and correctness flags are not. Global unrelated quiz definitions + are outside the project boundary; every exported quiz is checked explicitly. + """ + digest = export_content_digest(project) + project_id = str(project["projectId"]) + root = project_create_path(project_id) + live_project = _object(get(root), "project") + if live_project.get("projectId") != project_id or not isinstance(live_project.get("name"), str): + raise SkillTreeContentError("live project identity/name is unverifiable") + missing: list[str] = [] + extra: list[str] = [] + changed: list[str] = [] + if live_project["name"] != project["name"]: + changed.append(f"project:{project_id}") + live_subjects = _index(_rows(get(f"{root}/subjects"), "subjects", "project"), "subjectId", "project") + expected_subjects = {subject["subjectId"]: subject for subject in project["subjects"]} + missing.extend(f"subject:{identifier}" for identifier in expected_subjects.keys() - live_subjects.keys()) + extra.extend(f"subject:{identifier}" for identifier in live_subjects.keys() - expected_subjects.keys()) + live_skill_count = 0 + for subject_id, subject in expected_subjects.items(): + if subject_id not in live_subjects: + continue + subject_path = subject_create_path(project_id, subject_id) + live_subject = _object(get(subject_path), "subject") + fields = ("subjectId", "name", "description") + if any(field not in live_subject for field in fields): + raise SkillTreeContentError(f"subject:{subject_id} content is unverifiable") + if any(live_subject[field] != subject.get(field, "") for field in fields): + changed.append(f"subject:{subject_id}") + live_skills = _index(_rows(get(f"{subject_path}/skills"), "skills", "subject"), "skillId", "subject") + wanted_skills = {skill["skillId"]: skill for skill in subject["skills"]} + live_skill_count += len(live_skills) + missing.extend(f"skill:{identifier}" for identifier in wanted_skills.keys() - live_skills.keys()) + extra.extend(f"skill:{identifier}" for identifier in live_skills.keys() - wanted_skills.keys()) + for skill_id, skill in wanted_skills.items(): + if skill_id not in live_skills: + continue + live_skill = _object(get(skill_create_path(project_id, subject_id, skill_id)), "skill") + fields = ("skillId", "name", "description", "pointIncrement", "numPerformToCompletion", "helpUrl", "selfReportingType") + if any(field not in live_skill for field in fields): + raise SkillTreeContentError(f"skill:{skill_id} content is unverifiable") + if ( + any(live_skill[field] != skill.get(field) for field in fields) + or live_skill.get("quizId", "") != skill.get("quizId", "") + ): + changed.append(f"skill:{skill_id}") + for quiz in project["quizzes"]: + quiz_id = quiz["quizId"] + try: + live_quiz = _object(get(quiz_definition_path(quiz_id)), "quiz") + except SkillTreeClientError as exc: + if exc.status != 404: + raise + missing.append(f"quiz:{quiz_id}") + continue + fields = ("quizId", "name", "type") + if any(field not in live_quiz for field in fields): + raise SkillTreeContentError(f"quiz:{quiz_id} definition is unverifiable") + if any(live_quiz[field] != quiz.get(field) for field in fields): + changed.append(f"quiz:{quiz_id}") + question_count = len(quiz["questions"]) + percent = quiz.get("passingPercent") + if type(percent) is not int or not 1 <= percent <= 100: + raise SkillTreeContentError(f"quiz:{quiz_id} export pass percentage is invalid") + # Normalize defaults for the intended completed quiz. Missing questions + # still make the receipt incomplete below and can be seeded without + # misclassifying native default settings as an existing-content change. + length, required = _quiz_completion_contract( + get(f"{quiz_definition_path(quiz_id)}/settings"), question_count, + ) + if (length, required) != ( + question_count, (question_count * percent + 99) // 100, + ): + changed.append(f"quiz:{quiz_id}") + current = Counter(_question_key(row) for row in _rows(get(quiz_questions_path(quiz_id)), "questions", "quiz")) + wanted = Counter(_question_key(row) for row in quiz["questions"]) + if current - wanted: + changed.append(f"quiz-questions:{quiz_id}") + if wanted - current: + missing.append(f"quiz-questions:{quiz_id}") + scheme, host, port = _skilltree_origin(base_url) + return { + "checked": True, + "complete": not (missing or extra or changed), + "targetOrigin": {"scheme": scheme, "host": host, "port": port}, + "projectId": project_id, + "exportContentDigest": digest, + "subjects": len(live_subjects), + "skills": live_skill_count, + "missingResources": sorted(missing), + "extraResources": sorted(extra), + "changedResources": sorted(set(changed)), + } diff --git a/src/skilltree/deployment_matrix.py b/src/skilltree/deployment_matrix.py index c3de30d..f4b2e33 100644 --- a/src/skilltree/deployment_matrix.py +++ b/src/skilltree/deployment_matrix.py @@ -44,7 +44,7 @@ def skilltree_deployment_matrix( else "ready", }, { - "browserExposure": "same-origin token endpoint URL, never the token value", + "browserExposure": "same-origin endpoint and a temporary learner bearer token in browser memory", "evidence": ( "trusted-client bridge configured" if local_auth_ready and authenticator_configured @@ -52,7 +52,7 @@ def skilltree_deployment_matrix( ), "id": "token_backed_admin_seeding", "label": "Token-backed admin seeding", - "secretBoundary": "Client secret and admin credentials are exchanged server-side only.", + "secretBoundary": "Client secret and admin credentials stay server-side; the learner token is issued to browser memory.", "status": ( "ready" if local_auth_ready and authenticator_configured else "blocked" ), diff --git a/src/skilltree/local_auth.py b/src/skilltree/local_auth.py index e2e2f54..ad7b640 100644 --- a/src/skilltree/local_auth.py +++ b/src/skilltree/local_auth.py @@ -1,15 +1,15 @@ """Local-only SkillTree FORM auth bootstrap and trusted-client token bridge. -This module intentionally keeps credentials server-side. It creates or reuses a -single local root/admin account, stores local-only secrets in an ignored state -directory, and exchanges the SkillTree project trusted-client secret for learner -tokens. Browser code only ever sees the dashboard token endpoint path. +Root credentials and trusted-client secrets stay server-side in ignored local +state. The dashboard token bridge issues a temporary learner bearer token to +browser memory for the official SkillTree client; that token is not persisted. """ from __future__ import annotations import base64 import json +import math import os import sys import urllib.error @@ -31,6 +31,7 @@ subject_create_path, subject_payload, ) +from src.skilltree.content import compare_live_content, missing_quiz_questions from src.skilltree.local_auth_state import ( DEFAULT_LOCAL_ADMIN_PASSWORD, DEFAULT_LOCAL_ADMIN_USERNAME, @@ -69,7 +70,7 @@ SkillTreeSubject, ) from src.skilltree.seeder import seed -from src.skilltree.transport import SkillTreeClientError, quiz_question_count +from src.skilltree.transport import SkillTreeClientError, parse_json_body # Backwards-compatible re-export surface pulled from local_auth_state so downstream # callers can import these names from src.skilltree.local_auth. Declared explicitly @@ -104,6 +105,8 @@ def __init__( raise ValueError("base_url must be non-empty") if not project_id.strip(): raise ValueError("project_id must be non-empty") + if not math.isfinite(timeout) or timeout <= 0: + raise ValueError("timeout must be finite and positive") self.base_url = base_url.rstrip("/") self.project_id = project_id self.timeout = timeout @@ -133,7 +136,7 @@ def request_json( form_body=form_body, headers=headers, ) - return json.loads(raw) if raw.strip() else {} + return parse_json_body(raw) def request_text( self, @@ -432,16 +435,11 @@ def create_quiz(self, quiz: Quiz) -> Any: return result def _ensure_quiz_questions(self, quiz: Quiz) -> None: - """Create any exported quiz questions missing from the live quiz.""" - if not quiz.questions: - return + """Create missing semantic question occurrences, never infer a prefix by count.""" existing = self.request_json( "GET", f"/admin/quiz-definitions/{quiz.quiz_id}/questions" ) - existing_count = min(quiz_question_count(existing), len(quiz.questions)) - if existing_count >= len(quiz.questions): - return - for question in quiz.questions[existing_count:]: + for question in missing_quiz_questions(existing, quiz.questions): self.create_quiz_question(quiz.quiz_id, question) def create_quiz_question(self, quiz_id: str, question: QuizQuestion) -> Any: @@ -550,17 +548,20 @@ def verify_seed_landed( "server-side rejection swallowed as idempotent). Inspect for duplicate " "names / invalid ids before trusting the live instance." ) + if not live_project_content(client, project)["complete"]: + raise LocalSkillTreeAuthError( + "seed verification failed: live content differs from the exported project" + ) -def live_project_counts_match( +def live_project_content( client: SkillTreeSessionClient, project: SkillTreeProject -) -> bool: - """Return whether live SkillTree already has this export's core resources.""" - expected_skills = sum(len(subject.skills) for subject in project.subjects) - expected_subjects = len(project.subjects) - return ( - client.live_skill_count() == expected_skills - and client.live_subject_count() == expected_subjects +) -> dict[str, Any]: + """Compare identities, learner content and answer keys, not resource counts.""" + return compare_live_content( + get=lambda path: client.request_json("GET", path), + base_url=client.base_url, + project=project.to_export_dict(), ) @@ -651,19 +652,25 @@ def _seed_project_if_requested( project: SkillTreeProject | None, seed_project: bool, seed_dependencies: bool, - skip_seed_if_live_counts_ready: bool, + skip_seed_if_live_content_ready: bool, verify_seed: bool, ) -> tuple[bool, dict[str, int] | None]: if not seed_project or project is None: return False, None - should_seed = True - if skip_seed_if_live_counts_ready: - try: - should_seed = not live_project_counts_match(client, project) - except (LocalSkillTreeAuthError, SkillTreeClientError): - should_seed = True - if not should_seed: - return False, None + try: + content = live_project_content(client, project) + except SkillTreeClientError as exc: + if exc.status != 404: + raise + content = None + if content is not None: + if content["changedResources"] or content["extraResources"]: + raise LocalSkillTreeAuthError( + "live SkillTree content differs from the export; existing content " + "was not overwritten. Review the reported drift before reseeding." + ) + if skip_seed_if_live_content_ready and content["complete"]: + return False, None seed_summary = seed(client, project, include_dependencies=seed_dependencies) if verify_seed: verify_seed_landed(client, project) @@ -723,11 +730,15 @@ def ensure_local_skilltree_auth( project: SkillTreeProject | None = None, seed_project: bool = True, seed_dependencies: bool = True, - skip_seed_if_live_counts_ready: bool = False, + skip_seed_if_live_content_ready: bool = False, verify_seed: bool = True, verify_token: bool = True, ) -> BootstrapResult: """Bootstrap local root auth, seed the project, and store client credentials.""" + if project is not None and project.project_id != project_id: + raise LocalSkillTreeAuthError( + "exported project id does not match the requested bootstrap target" + ) client = SkillTreeSessionClient( base_url, project_id, timeout=default_local_request_timeout() ) @@ -744,7 +755,7 @@ def ensure_local_skilltree_auth( project=project, seed_project=seed_project, seed_dependencies=seed_dependencies, - skip_seed_if_live_counts_ready=skip_seed_if_live_counts_ready, + skip_seed_if_live_content_ready=skip_seed_if_live_content_ready, verify_seed=verify_seed, ) state = _project_client_secret_state( @@ -851,6 +862,8 @@ def exchange_trusted_client_token( status=exc.code, body=raw, ) from exc + except TimeoutError as exc: + raise SkillTreeClientError("POST /oauth/token timed out") from exc except urllib.error.URLError as exc: # pragma: no cover - network dependent raise SkillTreeClientError( f"POST /oauth/token transport error: {exc.reason}" @@ -862,7 +875,12 @@ def exchange_trusted_client_token( raise LocalSkillTreeAuthError( "SkillTree token endpoint returned an empty response" ) - payload = json.loads(raw) + try: + payload = json.loads(raw) + except json.JSONDecodeError as exc: + raise LocalSkillTreeAuthError( + "SkillTree token endpoint returned malformed JSON" + ) from exc if not isinstance(payload, dict): raise LocalSkillTreeAuthError( "SkillTree token endpoint returned non-object JSON" diff --git a/src/skilltree/transport.py b/src/skilltree/transport.py index 6c629c4..84afd8b 100644 --- a/src/skilltree/transport.py +++ b/src/skilltree/transport.py @@ -49,7 +49,10 @@ def redirect_request( # type: ignore[override] def parse_json_body(raw: str) -> Any: - return json.loads(raw) if raw.strip() else {} + try: + return json.loads(raw) if raw.strip() else {} + except json.JSONDecodeError as exc: + raise SkillTreeClientError("SkillTree returned malformed JSON") from exc def quiz_question_count(payload: Any) -> int: diff --git a/tests/README.md b/tests/README.md index fe61174..6c0c64d 100644 --- a/tests/README.md +++ b/tests/README.md @@ -16,6 +16,11 @@ uv run pytest tests/ --cov=src --cov-fail-under=90 Tests import from the `src.*` namespace (`pythonpath = [".", "src"]`); `conftest.py` also puts `src/` on `sys.path` and forces `MPLBACKEND=Agg` so figure tests run headless. +Script smoke tests copy the public source inputs into a module-scoped temporary +project before running the real orchestrators. Generated artifacts and web-math +fixtures stay there; running the suite must not regenerate the audited checkout's +tracked output or alter its source binding. + ## Coverage facts - The authoritative command is `uv run pytest tests/ --cov=src --cov-fail-under=90`. @@ -24,6 +29,14 @@ puts `src/` on `sys.path` and forces `MPLBACKEND=Agg` so figure tests run headle - **Zero mocks.** Real numeric computation everywhere; `pytest-httpserver` provides a real local HTTP server for SkillTree REST boundaries. +Distribution regressions include extreme finite scores at small inverse temperature, +where centering before scaling can overflow. Claim-rendering regressions also +require the rendered graph to retain unique subject and skill names before export. + +The admin-content fixture serves the pinned SkillTree quiz definition and its +separate settings endpoint. Pass requirements are native correct-question counts; +missing/default settings must not be mistaken for a missing export-only percentage. + ## Test Files | File | Covers | diff --git a/tests/conftest.py b/tests/conftest.py index f3f8308..4bd7b0b 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -3,6 +3,8 @@ import os import sys +import pytest + # Force headless backend for matplotlib in tests os.environ.setdefault("MPLBACKEND", "Agg") @@ -15,3 +17,185 @@ for _path in (ROOT, SRC): if _path not in sys.path: sys.path.insert(0, _path) + + + +@pytest.fixture +def skilltree_admin_content(): + """Serve actual model content over loopback, including private answer keys.""" + def serve(httpserver, project, *, initially_missing=False, quiz_settings=None): + payload = project.to_export_dict() if hasattr(project, "to_export_dict") else project + root = f"/admin/projects/{payload['projectId']}" + import json + from werkzeug.wrappers import Response + + root_requests = 0 + + def project_response(request): + nonlocal root_requests + root_requests += 1 + missing = initially_missing and root_requests == 1 + body = {"error": "not found"} if missing else { + "projectId": payload["projectId"], "name": payload["name"], + "numSubjects": len(payload["subjects"]), + "numSkills": sum(len(subject["skills"]) for subject in payload["subjects"]), + } + return Response(json.dumps(body), status=404 if missing else 200, content_type="application/json") + + httpserver.expect_request(root, method="GET").respond_with_handler(project_response) + httpserver.expect_request(f"{root}/subjects", method="GET").respond_with_json( + payload["subjects"] + ) + for subject in payload["subjects"]: + subject_path = f"{root}/subjects/{subject['subjectId']}" + httpserver.expect_request(subject_path, method="GET").respond_with_json(subject) + httpserver.expect_request(f"{subject_path}/skills", method="GET").respond_with_json( + subject["skills"] + ) + for skill in subject["skills"]: + httpserver.expect_request( + f"{subject_path}/skills/{skill['skillId']}", method="GET", + ).respond_with_json(skill) + for quiz in payload["quizzes"]: + quiz_path = f"/admin/quiz-definitions/{quiz['quizId']}" + httpserver.expect_request(quiz_path, method="GET").respond_with_json({ + key: quiz[key] for key in ("quizId", "name", "type") + }) + httpserver.expect_request(f"{quiz_path}/settings", method="GET").respond_with_json( + (quiz_settings or {}).get(quiz["quizId"], []) + ) + questions = [] if initially_missing else list(quiz["questions"]) + + def read_questions(_request, rows=questions): + return Response( + json.dumps({"questions": rows}), content_type="application/json", + ) + + def create_question(request, rows=questions): + rows.append(request.get_json()) + return Response("{}", content_type="application/json") + + httpserver.expect_request(f"{quiz_path}/questions", method="GET").respond_with_handler( + read_questions + ) + if initially_missing: + httpserver.expect_request( + f"{quiz_path}/create-question", method="POST", + ).respond_with_handler(create_question) + return serve + + +@pytest.fixture +def content_project(): + from src.skilltree.model import Quiz, QuizQuestion, SkillTreeProject, SkillTreeSkill, SkillTreeSubject + + quiz = Quiz( + quiz_id="q", name="Quiz", + questions=[ + QuizQuestion(question="First?", answers=[{"answer": "one", "isCorrect": True}]), + QuizQuestion(question="Second?", answers=[{"answer": "two", "isCorrect": True}]), + ], + ) + return SkillTreeProject( + project_id="p", name="Project", + subjects=[SkillTreeSubject( + subject_id="s", name="Subject", description="Subject content", + skills=[SkillTreeSkill( + skill_id="k", name="Skill", description="Lesson content", + self_reporting_type="Quiz", quiz_id="q", + )], + )], + quizzes=[quiz], + ) + + +@pytest.fixture +def pdf_bytes(): + """Build a structurally valid one-page PDF, not a copied historical output.""" + def make(text="Fixture image"): + escaped = text.replace("\\", "\\\\").replace("(", "\\(").replace(")", "\\)") + stream = f"BT /F1 18 Tf 20 120 Td ({escaped}) Tj ET\n".encode("ascii") + objects = [ + b"<< /Type /Catalog /Pages 2 0 R >>", + b"<< /Type /Pages /Kids [3 0 R] /Count 1 >>", + b"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 300 200] /Resources << /Font << /F1 4 0 R >> >> /Contents 5 0 R >>", + b"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>", + b"<< /Length " + str(len(stream)).encode("ascii") + b" >>\nstream\n" + stream + b"endstream", + ] + content = bytearray(b"%PDF-1.4\n%\xe2\xe3\xcf\xd3\n") + offsets = [] + for index, body in enumerate(objects, 1): + offsets.append(len(content)) + content.extend(f"{index} 0 obj\n".encode("ascii") + body + b"\nendobj\n") + xref = len(content) + content.extend(f"xref\n0 {len(objects) + 1}\n0000000000 65535 f \n".encode("ascii")) + for offset in offsets: + content.extend(f"{offset:010d} 00000 n \n".encode("ascii")) + content.extend( + f"trailer\n<< /Root 1 0 R /Size {len(objects) + 1} >>\nstartxref\n{xref}\n%%EOF\n".encode("ascii") + ) + return bytes(content) + return make + + +@pytest.fixture +def pdf_project(tmp_path, pdf_bytes): + """Create disposable source inputs and observe their real local Git revision.""" + import os + import subprocess + + from src.manuscript.pdf_receipt import PDF_NAME, add_pdf_assets, capture_pdf_inputs + + root = tmp_path / "publication" + manuscript = root / "manuscript" + hydrated = root / "output" / "manuscript" + manuscript.mkdir(parents=True) + hydrated.mkdir(parents=True) + assets_dir = hydrated / "assets" + assets_dir.mkdir() + assets = [assets_dir / name for name in ("cover.pdf", "atlas.pdf", "body.pdf")] + for path, label in zip(assets, ("CoverGraphic", "AtlasGraphic", "BodyGraphic"), strict=True): + path.write_bytes(pdf_bytes(label)) + (manuscript / "config.yaml").write_text( + "paper:\n title: FixtureTitlePage\n" + "authors:\n - name: Fixture Author\n" + "book:\n cover:\n image: assets/cover.pdf\n" + "front_matter:\n page_two_visual:\n image: assets/atlas.pdf\n" + " title: FixtureAtlasPage\n caption: FixtureAtlasCaption\n", + encoding="utf-8", + ) + (manuscript / "preamble.md").write_text("```latex\n\\usepackage{graphicx}\n```\n", encoding="utf-8") + (manuscript / "references.bib").write_text( + "@book{fixture,\n title={Fixture Publication},\n author={Example, Test},\n" + " year={2020},\n publisher={Fixture Press}\n}\n", + encoding="utf-8", + ) + section = hydrated / "01_fixture.md" + section.write_text( + f"# FixtureBodySection\n\nA cited statement [@fixture].\n\n![Body image]({assets[2].as_posix()})\n", + encoding="utf-8", + ) + target = root / "output" / "pdf" / PDF_NAME + target.parent.mkdir() + target.write_bytes(pdf_bytes("Finalized fixture PDF")) + git_env = {key: value for key, value in os.environ.items() if not key.startswith("GIT_")} + git_env.update(GIT_CONFIG_NOSYSTEM="1", GIT_CONFIG_GLOBAL=os.devnull) + subprocess.run(["git", "init", "--quiet", str(root)], env=git_env, check=True) + subprocess.run(["git", "add", "."], cwd=root, env=git_env, check=True) + subprocess.run( + [ + "git", "-c", "user.name=Receipt Test", "-c", "user.email=receipt@example.org", + "-c", "commit.gpgsign=false", "-c", "core.hooksPath=/dev/null", + "commit", "--quiet", "-m", "Disposable publication source", + ], + cwd=root, env=git_env, check=True, + ) + revision = subprocess.run( + ["git", "rev-parse", "HEAD"], cwd=root, env=git_env, + check=True, capture_output=True, text=True, + ).stdout.strip() + inputs = add_pdf_assets(root, capture_pdf_inputs(root, [section]), assets) + return { + "root": root, "section": section, "assets": assets, "target": target, + "revision": revision, "inputs": inputs, "git_env": git_env, + } diff --git a/tests/test_atomic_json_publication.py b/tests/test_atomic_json_publication.py new file mode 100644 index 0000000..5799fff --- /dev/null +++ b/tests/test_atomic_json_publication.py @@ -0,0 +1,57 @@ +"""Atomic publication preserves the previous complete artifact after real I/O failure.""" + +from __future__ import annotations + +import json +import os +from pathlib import Path +import subprocess +import sys + +import pytest + +from src.artifacts.json_io import write_json_artifact + + +@pytest.mark.skipif(os.name != "posix", reason="RLIMIT_FSIZE is a POSIX I/O boundary") +def test_failed_write_preserves_existing_bytes_and_cleans_temporary(tmp_path: Path) -> None: + artifact = tmp_path / "artifact.json" + original = b'{"generation": "previous"}\n' + artifact.write_bytes(original) + program = """ +import errno +from pathlib import Path +import resource +import signal +import sys +from src.artifacts.json_io import write_json_artifact +signal.signal(signal.SIGXFSZ, signal.SIG_IGN) +resource.setrlimit(resource.RLIMIT_FSIZE, (1024, 1024)) +try: + write_json_artifact(Path(sys.argv[1]), {"data": "x" * 16384}) +except OSError as error: + if error.errno != errno.EFBIG: + raise +else: + raise AssertionError("expected the actual filesystem write to exceed RLIMIT_FSIZE") +""" + result = subprocess.run( + [sys.executable, "-B", "-c", program, str(artifact)], + cwd=Path(__file__).resolve().parents[1], + capture_output=True, text=True, timeout=15, check=False, + ) + assert result.returncode == 0, result.stdout + result.stderr + assert artifact.read_bytes() == original + assert sorted(path.name for path in tmp_path.iterdir()) == [artifact.name] + + +def test_atomic_replacement_preserves_content_format_and_existing_mode(tmp_path: Path) -> None: + artifact = tmp_path / "artifact.json" + artifact.write_text('{"old": true}\n', encoding="utf-8") + artifact.chmod(0o640) + write_json_artifact(artifact, {"z": "λ", "a": [1, 2]}, ensure_ascii=False) + assert artifact.read_text(encoding="utf-8") == json.dumps( + {"z": "λ", "a": [1, 2]}, indent=2, sort_keys=True, ensure_ascii=False + ) + "\n" + assert artifact.stat().st_mode & 0o777 == 0o640 + assert sorted(path.name for path in tmp_path.iterdir()) == [artifact.name] diff --git a/tests/test_course_materials.py b/tests/test_course_materials.py index d3e1728..52841b3 100644 --- a/tests/test_course_materials.py +++ b/tests/test_course_materials.py @@ -4,7 +4,6 @@ from copy import deepcopy from pathlib import Path -from typing import Self import pytest @@ -363,124 +362,114 @@ def test_course_material_normalizer_surfaces_malformed_record_shape() -> None: assert any(issue.code == "record-type" for issue in audit.issues) -def test_github_tree_fetch_retries_transient_5xx(monkeypatch) -> None: - """A transient 5xx is retried with backoff and the fetch then succeeds.""" - import io - import json - import urllib.error - - from src.curriculum.materials import github_sources as gs - - body = json.dumps({"tree": [{"path": "a.md", "type": "blob"}]}).encode() - state = {"calls": 0} - - class FakeResponse: - def __init__(self, data: bytes) -> None: - self._data = data - - def __enter__(self) -> Self: - return self - - def __exit__(self, *exc: object) -> bool: - return False - - def read(self) -> bytes: - return self._data - - def fake_urlopen(request, timeout): - state["calls"] += 1 - if state["calls"] == 1: - raise urllib.error.HTTPError( - request.full_url, - 503, - "Service Unavailable", - {}, - io.BytesIO(b"boom"), - ) - return FakeResponse(body) +def test_github_tree_fetch_retries_transient_5xx(httpserver) -> None: + import urllib.request - monkeypatch.setattr(gs.urllib.request, "urlopen", fake_urlopen) - monkeypatch.setattr(gs.time, "sleep", lambda _seconds: None) + from src.curriculum.materials.github_sources import _github_tree_payload - paths = gs._fetch_github_tree_paths("o", "r", "c" * 40, timeout=2.0) - assert paths == {"a.md"} - assert state["calls"] == 2 + httpserver.expect_ordered_request("/tree").respond_with_json({}, status=503) + httpserver.expect_ordered_request("/tree").respond_with_json({"tree": [{"path": "a.md", "type": "blob"}]}) + result = _github_tree_payload( + urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, + ) + assert result == {"tree": [{"path": "a.md", "type": "blob"}]} + assert len(httpserver.log) == 2 -def test_github_tree_fetch_raises_immediately_on_permanent_4xx(monkeypatch) -> None: - """A permanent 4xx is not retried (it can never succeed).""" - import io +def test_github_tree_fetch_raises_immediately_on_permanent_4xx(httpserver) -> None: import urllib.error import urllib.request - from src.curriculum.materials import github_sources as gs - - def fake_urlopen(request, timeout): - raise urllib.error.HTTPError( - request.full_url, 404, "Not Found", {}, io.BytesIO(b"missing") - ) - - monkeypatch.setattr(gs.time, "sleep", lambda _seconds: None) + from src.curriculum.materials.github_sources import _github_tree_payload - with pytest.raises(urllib.request.HTTPError): - gs._github_tree_payload( - urllib.request.Request("https://api.github.com/x"), timeout=2.0 + httpserver.expect_ordered_request("/tree").respond_with_json({}, status=404) + with pytest.raises(urllib.error.HTTPError) as error: + _github_tree_payload( + urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, ) + assert error.value.code == 404 + assert len(httpserver.log) == 1 -def test_github_tree_fetch_retries_transient_network_error(monkeypatch) -> None: - """A transient network/timeout error is retried and the fetch then succeeds.""" - import json +def test_github_tree_fetch_raises_connection_error_from_real_loopback() -> None: + import socket + import urllib.error + import urllib.request - from src.curriculum.materials import github_sources as gs + from src.curriculum.materials.github_sources import _github_tree_payload - body = json.dumps({"tree": [{"path": "a.md", "type": "blob"}]}).encode() - state = {"calls": 0} + # Reserve a non-listening endpoint without stopping the shared HTTP fixture. + with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as endpoint: + endpoint.bind(("127.0.0.1", 0)) + url = f"http://127.0.0.1:{endpoint.getsockname()[1]}/tree" + with pytest.raises(urllib.error.URLError): + _github_tree_payload( + urllib.request.Request(url), timeout=0.2, attempts=2, base_delay=0, + ) - class FakeResponse: - def __enter__(self) -> Self: - return self - def __exit__(self, *exc: object) -> bool: - return False +def test_github_tree_fetch_raises_final_error_after_exhausting_attempts(httpserver) -> None: + import urllib.error + import urllib.request - def read(self) -> bytes: - return body + from src.curriculum.materials.github_sources import _github_tree_payload - def fake_urlopen(request, timeout): - state["calls"] += 1 - if state["calls"] == 1: - raise TimeoutError("timed out") - return FakeResponse() + for _ in range(3): + httpserver.expect_ordered_request("/tree").respond_with_json({}, status=503) + with pytest.raises(urllib.error.HTTPError) as error: + _github_tree_payload( + urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, + ) + assert error.value.code == 503 + assert len(httpserver.log) == 3 - monkeypatch.setattr(gs.urllib.request, "urlopen", fake_urlopen) - monkeypatch.setattr(gs.time, "sleep", lambda _seconds: None) - paths = gs._fetch_github_tree_paths("o", "r", "c" * 40, timeout=2.0) - assert paths == {"a.md"} - assert state["calls"] == 2 +@pytest.mark.parametrize("malformed", [[], None, {"tree": None}]) +def test_malformed_successful_github_response_becomes_one_source_issue( + httpserver, monkeypatch, malformed, +) -> None: + httpserver.expect_request( + "/repos/testorg/testrepo/git/trees/0123456789abcdef0123456789abcdef01234567" + ).respond_with_json(malformed) + monkeypatch.setenv("ASF_GITHUB_API_BASE", httpserver.url_for("")) + issues = validate_material_sources(_online_payload(), online=True, timeout=2) + findings = [issue for issue in issues if issue.code == "github-tree"] + assert len(findings) == 1 + assert len(httpserver.log) == 1 -def test_github_tree_fetch_raises_final_error_after_exhausting_attempts( - monkeypatch, +@pytest.mark.parametrize("include_second", [False, True]) +def test_offline_and_online_pointer_owners_agree_on_whitespace_and_all_list_entries( + httpserver, monkeypatch, include_second: bool, ) -> None: - """If every retry fails with 5xx, the final HTTPError is re-raised.""" - import io - import urllib.error - import urllib.request + from src.curriculum.model import ALL_STRATA - from src.curriculum.materials import github_sources as gs - - def fake_urlopen(request, timeout): - raise urllib.error.HTTPError( - request.full_url, 503, "Service Unavailable", {}, io.BytesIO(b"boom") - ) + monkeypatch.setenv("ASF_GITHUB_API_BASE", httpserver.url_for("")) + payload = _online_payload() + source = payload["sources"][0] + source_id = source["source_id"].strip() + for field in ("source_id", "repo_url", "commit"): + source[field] = " " + source[field].strip() + " " + source["path_globs"] = ["*.md"] + payload["stratum_mappings"] = { + stratum.value: [ + {"source_id": f" {source_id} ", "path": " first.md ", "title": " First ", "role": " Context "}, + {"source_id": f" {source_id} ", "path": " second.md ", "title": " Second ", "role": " Context "}, + ] + for stratum in ALL_STRATA + } + payload["topic_mappings"] = [] + assert validate_material_sources(payload) == [] + assert httpserver.log == [] + paths = ["first.md", "second.md"] if include_second else ["first.md"] + httpserver.expect_request( + "/repos/testorg/testrepo/git/trees/0123456789abcdef0123456789abcdef01234567" + ).respond_with_json({"tree": [{"path": path, "type": "blob"} for path in paths], "truncated": False}) - monkeypatch.setattr(gs.urllib.request, "urlopen", fake_urlopen) - monkeypatch.setattr(gs.time, "sleep", lambda _seconds: None) + issues = validate_material_sources(payload, online=True, timeout=2) - with pytest.raises(urllib.request.HTTPError) as exc_info: - gs._github_tree_payload( - urllib.request.Request("https://api.github.com/x"), timeout=2.0 - ) - assert exc_info.value.code == 503 + if include_second: + assert issues == [] + else: + assert issues and all("second.md" in issue.message for issue in issues) + assert len(httpserver.log) == 1 diff --git a/tests/test_curriculum_graph.py b/tests/test_curriculum_graph.py index 2924a25..73fd68f 100644 --- a/tests/test_curriculum_graph.py +++ b/tests/test_curriculum_graph.py @@ -196,3 +196,53 @@ def test_connectivity_fails_closed_when_no_foundational_roots() -> None: ) assert g.reachable_from_foundations() == set() assert not g.is_connected_foundations_to_applications() + + +def test_smallest_newly_ready_id_precedes_a_larger_ready_root() -> None: + graph = _graph(_skill("z", _M, ()), _skill("b", _M, ("a",)), _skill("a", _M, ())) + assert [skill.skill_id for skill in graph.topological_order()] == ["a", "b", "z"] + + +def test_reverse_stored_deep_chain_has_exact_stack_safe_prerequisite_closure() -> None: + from src.curriculum.quality import transitive_prerequisites + + identifiers = [f"k{index:04d}" for index in range(1200)] + skills = [ + _skill(identifier, _M, () if index == 0 else (identifiers[index - 1],)) + for index, identifier in enumerate(identifiers) + ] + graph = _graph(*reversed(skills)) + graph.validate() + closure = transitive_prerequisites(graph) + assert closure[identifiers[0]] == set() + assert closure[identifiers[600]] == set(identifiers[:600]) + assert closure[identifiers[-1]] == set(identifiers[:-1]) + assert graph.levels()[identifiers[-1]] == len(identifiers) - 1 + + +def test_diamond_closure_deduplicates_shared_ancestors_and_refuses_cycles() -> None: + from src.curriculum.quality import transitive_prerequisites + + graph = _graph( + _skill("a", _M, ()), _skill("b", _M, ("a",)), + _skill("c", _M, ("a",)), _skill("d", _M, ("b", "c")), + ) + assert transitive_prerequisites(graph) == { + "a": set(), "b": {"a"}, "c": {"a"}, "d": {"a", "b", "c"}, + } + with pytest.raises(CurriculumError): + transitive_prerequisites(_graph(_skill("a", _M, ("b",)), _skill("b", _M, ("a",)))) + with pytest.raises(CurriculumError): + transitive_prerequisites(_graph(_skill("a", _M, ("missing",)))) + + +def test_call_scoped_snapshots_do_not_cache_mutable_subject_membership() -> None: + graph = _graph(_skill("a", _M, ())) + before = graph.skills + graph.subjects.append(Subject("extra", "Extra", _M, "d", (_skill("b", _M, ("a",)),))) + graph.validate() + assert [skill.skill_id for skill in before] == ["a"] + assert [skill.skill_id for skill in graph.skills] == ["a", "b"] + assert [skill.skill_id for skill in graph.topological_order()] == ["a", "b"] + assert graph.stats()["skills"] == 2 + assert graph.stats()["max_depth"] == 1 diff --git a/tests/test_dashboard_console.py b/tests/test_dashboard_console.py index 43cd07e..e171552 100644 --- a/tests/test_dashboard_console.py +++ b/tests/test_dashboard_console.py @@ -12,6 +12,7 @@ sys.path.insert(0, str(REPO / "scripts")) import validate_dashboard_console as gate +from src.dashboard.browser_validation import unexpected_errors _INCEPTION = ( "msgid=89 [error] Error getting service status Error: SkillTree: Unable to " @@ -21,12 +22,12 @@ def test_known_inception_probe_is_allow_listed() -> None: - assert gate.unexpected_errors([_INCEPTION]) == [] + assert unexpected_errors([_INCEPTION]) == [] def test_unexpected_error_is_surfaced() -> None: other = "msgid=12 [error] Uncaught TypeError: cannot read property 'x' of undefined" - unexpected = gate.unexpected_errors([_INCEPTION, other]) + unexpected = unexpected_errors([_INCEPTION, other]) # The cosmetic probe is filtered; a real error must remain so the gate fails on it. assert unexpected == [other] @@ -37,7 +38,7 @@ def test_inception_allow_list_does_not_swallow_other_401s() -> None: "[error] Unable to authenticate using " "[/app/projects/activeInference/users/local-learner/token] Response Code=[401]." ) - assert gate.unexpected_errors([real_401]) == [real_401] + assert unexpected_errors([real_401]) == [real_401] def test_dashboard_reachable_rejects_non_http_scheme() -> None: diff --git a/tests/test_dashboard_live_validation.py b/tests/test_dashboard_live_validation.py index 2316698..4d13992 100644 --- a/tests/test_dashboard_live_validation.py +++ b/tests/test_dashboard_live_validation.py @@ -1,17 +1,20 @@ from __future__ import annotations import json +from copy import deepcopy from pathlib import Path import pytest -from src.dashboard.artifacts import EXPECTED_ARTIFACT_COUNT +from src.dashboard.artifacts import EXPECTED_ARTIFACT_COUNT, load_skilltree_project from src.dashboard.live_validation import ( LiveSkillTreeDashboardError, _fetch_json, validate_live_skilltree_dashboard, ) from src.project_identity import SKILLTREE_PROJECT_ID, SKILLTREE_PROJECT_NAME +from src.skilltree.content import export_content_digest +from src.skilltree.local_auth import LocalSkillTreeState, save_local_state def _write_project_export(project_root: Path, payload: object | None = None) -> Path: @@ -43,6 +46,7 @@ def _serve_dashboard_payloads( *, counts: dict[str, int] | None = None, token_payload: dict[str, object] | None = None, + project_summary: dict[str, object] | None = None, ) -> str: base_url = httpserver.url_for("").rstrip("/") project_counts = counts or { @@ -54,9 +58,9 @@ def _serve_dashboard_payloads( httpserver.expect_request("/public/status").respond_with_json({"status": "OK"}) httpserver.expect_request("/api/status").respond_with_json( { - "project": { - "projectId": SKILLTREE_PROJECT_ID, - "name": SKILLTREE_PROJECT_NAME, + "instanceId": "live-validation-fixture", + "project": project_summary if project_summary is not None else { + "project": {"id": SKILLTREE_PROJECT_ID, "name": SKILLTREE_PROJECT_NAME}, "dependencies": [], "counts": project_counts, }, @@ -173,3 +177,45 @@ def test_live_validator_rejects_token_secret_leak( def test_fetch_json_rejects_non_http_scheme() -> None: with pytest.raises(LiveSkillTreeDashboardError, match="only http/https URLs"): _fetch_json("file:///etc/passwd", timeout=1.0) + + +@pytest.mark.parametrize("changed", [False, True]) +def test_strict_live_validator_binds_actual_admin_content_not_just_matching_counts( + tmp_path: Path, httpserver, content_project, skilltree_admin_content, changed: bool, +) -> None: + exported = content_project.to_export_dict() + exported.update(projectId=SKILLTREE_PROJECT_ID, name=SKILLTREE_PROJECT_NAME) + exported["subjects"][0]["skills"][0]["skillId"] = "live_validation_01" + exported["quizzes"][0]["questions"][0]["answers"].append({"answer": "other", "isCorrect": False}) + _write_project_export(tmp_path, exported) + public = load_skilltree_project(tmp_path, export_payload=exported) + base_url = _serve_dashboard_payloads(httpserver, project_summary=public) + live = deepcopy(exported) + if changed: + answers = live["quizzes"][0]["questions"][0]["answers"] + answers[0]["isCorrect"], answers[1]["isCorrect"] = False, True + skilltree_admin_content(httpserver, live) + httpserver.expect_request("/", method="GET").respond_with_data("Session", content_type="text/html") + httpserver.expect_request("/performLogin", method="POST").respond_with_json({}) + httpserver.expect_request( + f"/admin/projects/{SKILLTREE_PROJECT_ID}/dependency/graph", method="GET", + ).respond_with_json({"nodes": [], "edges": []}) + state_dir = tmp_path / "private-state" + save_local_state(state_dir, LocalSkillTreeState( + admin_username="fixture-admin", admin_password="fixture-password", + learner_id="fixture-learner", project_id=SKILLTREE_PROJECT_ID, + skilltree_url=base_url, client_secret="fixture-client-secret", + )) + arguments = { + "dashboard_url": base_url, "skilltree_url": base_url, "project_root": tmp_path, + "local_state_dir": state_dir, "require_token": False, "require_live_admin": True, + } + if changed: + with pytest.raises(LiveSkillTreeDashboardError, match="content"): + validate_live_skilltree_dashboard(**arguments) + else: + result = validate_live_skilltree_dashboard(**arguments) + assert result["liveContent"]["checked"] is True and result["liveContent"]["complete"] is True + assert result["exportContentDigest"] == export_content_digest(exported) + assert result["dashboardInstanceId"] == "live-validation-fixture" + assert "isCorrect" not in json.dumps(result) and "fixture-password" not in json.dumps(result) diff --git a/tests/test_dashboard_mount.py b/tests/test_dashboard_mount.py new file mode 100644 index 0000000..1c3aa95 --- /dev/null +++ b/tests/test_dashboard_mount.py @@ -0,0 +1,82 @@ +"""Opt-in actual-browser acceptance; never substitutes a fake SDK or CLI result.""" + +from __future__ import annotations + +import json +import os +import shutil +import threading +from http.server import ThreadingHTTPServer +from pathlib import Path + +import pytest + +from src.dashboard.browser_validation import ( + BrowserDisplayError, + browser_receipt_complete, + display_probe, + validate_dashboard_browser, +) +from src.dashboard.config import make_dashboard_config +from src.dashboard.server import make_dashboard_handler + + +@pytest.fixture +def authorized_browser(): + url = os.environ.get("ASF_TEST_DASHBOARD_URL", "") + if not url: + pytest.skip("set ASF_TEST_DASHBOARD_URL only after live browser acceptance is authorized") + command = os.environ.get("ASF_TEST_BROWSER_COMMAND", "chrome-devtools-axi") + if shutil.which(command) is None: + pytest.skip(f"existing browser CLI unavailable: {command}") + return url, command + + +@pytest.mark.requires_browser +def test_actual_official_project_subject_and_quiz_are_mounted(authorized_browser) -> None: + url, command = authorized_browser + receipt = validate_dashboard_browser(url, browser_command=command, timeout=60, settle_seconds=1) + assert browser_receipt_complete(receipt) + quiz = next(view for view in receipt["views"] if view["kind"] == "quiz") + observed = display_probe(browser_command=command, timeout=10) + assert observed["state"] == "ready" and observed["route"] == quiz["route"] + assert observed["frameVisible"] is True and observed["busy"] is False + assert observed["projectId"] == receipt["projectId"] + assert observed["dashboardInstanceId"] == receipt["dashboardInstanceId"] + assert observed["exportContentDigest"] == receipt["exportContentDigest"] + + +@pytest.mark.requires_browser +def test_real_dashboard_fallback_cannot_pass_mounted_view_proof( + tmp_path: Path, httpserver, content_project, authorized_browser, +) -> None: + _authorized_url, command = authorized_browser + export = tmp_path / "output" / "data" / "skilltree_project.json" + export.parent.mkdir(parents=True) + export.write_text(json.dumps(content_project.to_export_dict()), encoding="utf-8") + httpserver.expect_request("/public/status").respond_with_json({"status": "OK"}) + config = make_dashboard_config( + project_root=tmp_path, skilltree_url=httpserver.url_for(""), + project_id="p", project_name="Project", host="127.0.0.1", port=8765, + skills_client_module_url="/assets/missing-official-module.js", + local_auth_enabled=False, + ) + server = ThreadingHTTPServer(("127.0.0.1", 0), make_dashboard_handler(config)) + thread = threading.Thread(target=server.serve_forever, daemon=True) + thread.start() + url = f"http://127.0.0.1:{server.server_address[1]}" + try: + with pytest.raises(BrowserDisplayError, match="fallback|did not reach ready"): + validate_dashboard_browser(url, browser_command=command, timeout=15, settle_seconds=0) + observed = display_probe(browser_command=command, timeout=5) + assert observed["state"] != "ready" + finally: + server.shutdown() + server.server_close() + thread.join(timeout=5) + + +@pytest.mark.parametrize("timeout", [0.0, -1.0, float("nan"), float("inf")]) +def test_invalid_browser_timeout_is_rejected_before_network_or_cli(timeout: float) -> None: + with pytest.raises(BrowserDisplayError, match="finite and positive"): + validate_dashboard_browser("not a URL", timeout=timeout) diff --git a/tests/test_dashboard_server.py b/tests/test_dashboard_server.py index 13b4f7c..6d9e922 100644 --- a/tests/test_dashboard_server.py +++ b/tests/test_dashboard_server.py @@ -640,3 +640,65 @@ def test_request_without_host_header_is_allowed(tmp_path: Path) -> None: status_line = raw.split(b"\r\n", 1)[0].decode("ascii") assert "200" in status_line assert b'"ok": true' in raw + + +def test_malformed_provider_token_is_a_secret_safe_service_error( + tmp_path: Path, httpserver: HTTPServer, +) -> None: + provider = httpserver.url_for("").rstrip("/") + httpserver.expect_request("/oauth/token", method="POST").respond_with_data( + '{"access_token": "private-provider-fragment"', content_type="application/json", + ) + state_dir = tmp_path / "state" + _write_local_auth_state(state_dir, skilltree_url=provider) + config = make_dashboard_config( + project_root=tmp_path, skilltree_url=provider, + local_auth_enabled=True, local_state_dir=state_dir, + ) + with _DashboardServer(config) as base_url: + try: + _request(f"{base_url}/api/skilltree/token") + except urllib.error.HTTPError as error: + status, raw = error.code, error.read() + else: + raise AssertionError("malformed provider token was accepted") + assert status == 503 + assert "private-provider-fragment" not in raw.decode("utf-8") + assert "trusted-secret" not in raw.decode("utf-8") + + +def test_status_requests_see_atomic_export_replacement_without_exposing_answer_keys( + tmp_path: Path, httpserver, content_project, +) -> None: + from copy import deepcopy + + from src.artifacts.json_io import write_json_artifact + from src.skilltree.content import export_content_digest + + exported = content_project.to_export_dict() + path = tmp_path / "output" / "data" / "skilltree_project.json" + path.parent.mkdir(parents=True, exist_ok=True) + write_json_artifact(path, exported) + httpserver.expect_request("/public/status").respond_with_json({"status": "OK"}) + config = make_dashboard_config( + project_root=tmp_path, project_id="p", project_name="Project", + skilltree_url=httpserver.url_for(""), + ) + with _DashboardServer(config) as url: + code, _headers, body = _request(url + "/api/status?defer_live=1") + first = json.loads(body) + assert code == 200 and first["project"]["available"] is True + assert first["project"]["exportContentDigest"] == export_content_digest(exported) + replacement = deepcopy(exported) + replacement["subjects"][0]["skills"][0]["description"] = "Replacement lesson" + replacement["quizzes"][0]["questions"][0]["answers"][0]["answer"] = "private-replacement-key-7ce8" + write_json_artifact(path, replacement) + code, _headers, body = _request(url + "/api/status?defer_live=1") + second = json.loads(body) + + assert code == 200 + assert first["instanceId"] == second["instanceId"] + assert second["project"]["exportContentDigest"] == export_content_digest(replacement) + assert first["project"]["exportContentDigest"] != second["project"]["exportContentDigest"] + assert "isCorrect" not in body.decode("utf-8") + assert "private-replacement-key-7ce8" not in body.decode("utf-8") diff --git a/tests/test_figures.py b/tests/test_figures.py index 47bc619..5f0fef3 100644 --- a/tests/test_figures.py +++ b/tests/test_figures.py @@ -605,13 +605,15 @@ def test_validation_surface_matrix_figure_written(tmp_path) -> None: def test_validation_intent_mismatch_matrix_figure_written(tmp_path) -> None: + from src.figures.plots.registry_io import planned_figure_registry_entries + path = plot_validation_intent_mismatch_matrix( tmp_path / "validation-intent.png", - figure_count=planned_figure_count( + figure_count=len(planned_figure_registry_entries( include_scholarship=True, include_claim_support=True, include_roadmap=True, - ), + )), include_scholarship=True, include_claim_support=True, include_roadmap=True, @@ -913,3 +915,19 @@ def test_validation_surface_matrix_is_derived_from_validator_files( # Against an empty root no validator files exist -> every cell is n/a (0). _, _, empty = validation_surface_matrix(tmp_path) assert empty.sum() == 0 + + +def test_registered_total_and_rendered_subset_keep_distinct_contracts() -> None: + from itertools import product + + from src.figures.plots.registry_io import planned_figure_registry_entries + from src.figures.registry_manifest import figure_registry_payload + + for scholarship, claims, roadmap in product((False, True), repeat=3): + flags = dict(include_scholarship=scholarship, include_claim_support=claims, include_roadmap=roadmap) + registered = figure_registry_payload(**flags) + rendered = planned_figure_registry_entries(**flags) + assert planned_figure_count(**flags) == len(registered) + assert {row["filename"] for row in rendered} == { + row["filename"] for row in registered if row.get("evidence_kind", "rendered") != "captured" + } diff --git a/tests/test_full_app_stack.py b/tests/test_full_app_stack.py index ae4110f..35d2422 100644 --- a/tests/test_full_app_stack.py +++ b/tests/test_full_app_stack.py @@ -8,6 +8,8 @@ import sys from pathlib import Path +import pytest + from src.integration import full_app_stack as stack REPO = Path(__file__).resolve().parents[1] @@ -20,7 +22,7 @@ def test_full_app_stack_builds_strict_commands() -> None: state_dir="state", learner_id="learner", skip_dependencies=True, - skip_seed_if_live_counts_ready=True, + skip_seed_if_live_content_ready=True, ) dashboard = stack.dashboard_command( python="python", @@ -57,7 +59,7 @@ def test_full_app_stack_builds_strict_commands() -> None: assert bootstrap[:2] == ["python", "scripts/bootstrap_local_skilltree.py"] assert "--learner-id" in bootstrap assert "--skip-dependencies" in bootstrap - assert "--skip-seed-if-live-counts-ready" in bootstrap + assert "--skip-seed-if-live-content-ready" in bootstrap assert dashboard[:2] == ["python", "scripts/serve_learner_dashboard.py"] assert "--local-auth" in dashboard assert reconcile[:2] == ["python", "scripts/reconcile_live_dependencies.py"] @@ -312,48 +314,58 @@ def test_full_app_stack_starts_keep_running_dashboard_with_log(tmp_path: Path) - stack.stop_dashboard_process(process, timeout=5) +def _browser_receipt() -> dict: + return { + "ok": True, "checked": True, "projectId": "p", + "dashboardInstanceId": "instance", "exportContentDigest": "a" * 64, + "views": [ + {"kind": "project", "route": "/", "state": "ready", "visibleContentChecked": True}, + {"kind": "subject", "route": "/subjects/s", "state": "ready", "visibleContentChecked": True}, + {"kind": "quiz", "route": "/subjects/s/skills/k/quizzes/q", "state": "ready", "visibleContentChecked": True}, + ], + } + + +def _live_receipt() -> dict: + return { + "ok": True, "dashboardInstanceId": "instance", "exportContentDigest": "a" * 64, + "counts": {"skills": 630}, + "artifacts": {"present": 30, "missing": 0}, + "courseMaterials": {"records": 630}, + "figureContracts": {"checked": True, "registered": 19}, + "liveAdminProject": {"checked": True, "skills": 630}, + "liveDependencies": {"checked": True, "complete": True, "expectedDependencies": 1199}, + "liveContent": { + "checked": True, "complete": True, "projectId": "p", + "exportContentDigest": "a" * 64, + }, + } + + def test_full_app_stack_proof_summary_shape() -> None: - pid_path = Path("/tmp/dashboard.pid") - log_path = Path("/tmp/dashboard.log") + pid_path, log_path = Path("/tmp/dashboard.pid"), Path("/tmp/dashboard.log") proof = stack.proof_summary( skilltree_url="http://localhost:8080", dashboard_url="http://127.0.0.1:8765", skilltree_status={"status": "OK"}, - dashboard_health={"ok": True}, - live_validation={ - "counts": {"skills": 630}, - "artifacts": {"present": 30, "missing": 0}, - "courseMaterials": {"records": 630}, - "figureContracts": {"checked": True, "registered": 19}, - "liveAdminProject": {"checked": True, "skills": 630}, - "liveDependencies": {"checked": True, "expectedDependencies": 1199}, - }, - browser_console_stdout="dashboard console gate OK", + dashboard_health={"ok": True, "instanceId": "instance"}, + live_validation=_live_receipt(), + browser_console_stdout=json.dumps(_browser_receipt()), raw_skilltree_login={"ok": True, "adminSurfaceVisible": True}, - kept_running=True, - dashboard_pid=12345, - dashboard_pid_path=pid_path, - dashboard_log_path=log_path, - dependency_reconciliation={"ok": True, "missingAfter": 0}, - app_ready=False, + kept_running=True, dashboard_pid=12345, + dashboard_pid_path=pid_path, dashboard_log_path=log_path, + dependency_reconciliation={"ok": True, "complete": True, "missingAfter": 0}, ) - - encoded = json.dumps(proof, sort_keys=True) assert proof["ok"] is True - assert proof["counts"] == {"skills": 630} assert proof["browserConsole"]["checked"] is True - assert proof["rawSkillTreeLogin"] == {"ok": True, "adminSurfaceVisible": True} - assert proof["figureContracts"] == {"checked": True, "registered": 19} - assert proof["liveDependencies"] == { - "checked": True, - "expectedDependencies": 1199, - } - assert proof["dependencyReconciliation"] == {"ok": True, "missingAfter": 0} + assert proof["liveContent"]["complete"] is True + assert proof["liveDependencies"]["complete"] is True + assert proof["runId"] == proof["dashboardInstanceId"] == "instance" + assert proof["exportContentDigest"] == proof["liveContent"]["exportContentDigest"] assert proof["mode"] == "full-parity" assert proof["dashboardPid"] == 12345 assert proof["dashboardPidPath"] == str(pid_path) assert proof["dashboardLogPath"] == str(log_path) - assert "dashboard console gate OK" in encoded def test_full_app_stack_proof_summary_does_not_certify_failed_subproof() -> None: @@ -383,7 +395,58 @@ def test_full_app_stack_writes_redacted_deterministic_proof(tmp_path: Path) -> N path = stack.write_proof_artifact(proof, tmp_path / "proof.json") payload = json.loads(path.read_text(encoding="utf-8")) - assert payload["proofVersion"] == 1 + assert payload["proofVersion"] == 2 assert payload["token"] == "[REDACTED]" assert "super-secret" not in path.read_text(encoding="utf-8") assert payload["counts"] == {"skills": 630} + + +@pytest.mark.parametrize("stdout", ["dashboard console gate OK", "{}", '{"ok": true, "checked": false}']) +def test_console_exit_text_or_unchecked_payload_cannot_certify_a_display(stdout: str) -> None: + with pytest.raises(stack.FullAppStackError): + stack.proof_summary( + skilltree_url="http://localhost:8080", dashboard_url="http://127.0.0.1:8765", + skilltree_status={"status": "OK"}, + dashboard_health={"ok": True, "instanceId": "instance"}, + live_validation=_live_receipt(), browser_console_stdout=stdout, + raw_skilltree_login={"ok": True}, kept_running=False, dashboard_pid=None, + ) + + +def test_proof_failure_stops_owned_dashboard_even_with_keep_running( + tmp_path: Path, httpserver, +) -> None: + import urllib.request + + httpserver.expect_request("/unrelated").respond_with_data("still running") + process = stack.start_dashboard([sys.executable, "-c", "import time; time.sleep(30)"]) + destination = tmp_path / "proof-is-a-directory" + destination.mkdir() + try: + with pytest.raises(OSError): + try: + stack.write_proof_artifact({"ok": True}, destination) + finally: + stack.finalize_dashboard_process( + process, state_dir=str(tmp_path), keep_running=True, successful=False, + ) + assert process.poll() is not None + assert not (tmp_path / "dashboard.pid").exists() + with urllib.request.urlopen(httpserver.url_for("/unrelated"), timeout=2) as response: + assert response.read() == b"still running" + finally: + stack.stop_dashboard_process(process) + + +def test_failed_pid_publication_does_not_leave_the_owned_process_alive(tmp_path: Path) -> None: + (tmp_path / "dashboard.pid").mkdir() + process = stack.start_dashboard([sys.executable, "-c", "import time; time.sleep(30)"]) + try: + with pytest.raises(OSError): + stack.finalize_dashboard_process( + process, state_dir=str(tmp_path), keep_running=True, successful=True, + ) + assert process.poll() is not None + assert (tmp_path / "dashboard.pid").is_dir() + finally: + stack.stop_dashboard_process(process) diff --git a/tests/test_generated_output_cli.py b/tests/test_generated_output_cli.py new file mode 100644 index 0000000..c9c16e9 --- /dev/null +++ b/tests/test_generated_output_cli.py @@ -0,0 +1,20 @@ +"""The CLI owns process exits while the importable pipeline raises package errors.""" + +from __future__ import annotations + +from pathlib import Path + +import pytest + +from scripts import run_kernels + + +def test_cli_reports_real_manuscript_rejection_without_replacing_the_builder(tmp_path: Path, monkeypatch) -> None: + manuscript = tmp_path / "manuscript" + manuscript.mkdir() + (manuscript / "00_fixture.md").write_text("The platform is NSA approved for all learners.\n", encoding="utf-8") + monkeypatch.setattr(run_kernels, "_ROOT", tmp_path) + with pytest.raises(SystemExit) as error: + run_kernels.main() + assert isinstance(error.value.code, str) + assert "endorsement-overreach" in error.value.code diff --git a/tests/test_gnn_conformance.py b/tests/test_gnn_conformance.py index 8adef9d..16b280b 100644 --- a/tests/test_gnn_conformance.py +++ b/tests/test_gnn_conformance.py @@ -114,10 +114,6 @@ def test_declared_dimensions_match_the_kernel() -> None: assert dimensions["o"] == (agent["num_observations"], 1) -def test_kernel_canonical_model_conforms() -> None: - checks = check_conformance(canonical_model()) - assert len(checks) == 6 - def test_mutated_model_is_rejected() -> None: model = copy.deepcopy(canonical_model()) @@ -126,6 +122,22 @@ def test_mutated_model_is_rejected() -> None: check_conformance(model) +@pytest.mark.parametrize( + ("field", "value"), + [ + ("A_matrix", [[1.0, 0.0]]), + ("B_matrix", [[1.0, 0.0], [0.0, 1.0]]), + ("C_vector", [1.0]), + ("D_vector", [1.0]), + ], +) +def test_invalid_declared_component_shapes_raise_conformance_error(field, value) -> None: + model = copy.deepcopy(canonical_model()) + model[field] = value + with pytest.raises(ConformanceError, match="declared probability/transition model"): + check_conformance(model) + + @pytest.mark.parametrize("schema", [None, "", 1, {}, "gnn-importer-pin/v2"]) def test_unsupported_pin_schema_is_rejected(tmp_path, schema) -> None: data = {**CANONICAL_PIN, "schema": schema} diff --git a/tests/test_kernels_distributions.py b/tests/test_kernels_distributions.py index 5ca67c5..96922f8 100644 --- a/tests/test_kernels_distributions.py +++ b/tests/test_kernels_distributions.py @@ -61,9 +61,14 @@ def test_higher_precision_sharpens_the_distribution(self) -> None: def test_rejects_non_positive_precision(self) -> None: """softmax() requires a positive inverse temperature.""" - with pytest.raises(ValueError, match="precision must be > 0"): + with pytest.raises(ValueError): softmax(np.array([1.0, 2.0], dtype=np.float64), precision=0.0) + def test_small_precision_preserves_mass_across_extreme_scores(self) -> None: + expected = np.exp([-2.0, 0.0]) + expected /= expected.sum() + assert_allclose(softmax([-1e308, 1e308], precision=1e-308), expected) + class TestLogStable: """Tests for log_stable().""" @@ -94,7 +99,7 @@ def test_dirichlet_expected_log_matches_digamma_identity(self) -> None: def test_dirichlet_helpers_reject_non_positive_concentrations(self) -> None: """Dirichlet expectations require strictly positive concentration parameters.""" alpha = np.array([1.0, 0.0, 2.0], dtype=np.float64) - with pytest.raises(ValueError, match="strictly positive"): + with pytest.raises(ValueError): dirichlet_expected(alpha) - with pytest.raises(ValueError, match="strictly positive"): + with pytest.raises(ValueError): dirichlet_expected_log(alpha) diff --git a/tests/test_manuscript_hydration.py b/tests/test_manuscript_hydration.py new file mode 100644 index 0000000..566e99d --- /dev/null +++ b/tests/test_manuscript_hydration.py @@ -0,0 +1,30 @@ +"""Unknown prose placeholders cannot silently become plausible manuscript text.""" + +from __future__ import annotations + +import pytest + +from scripts.z_generate_manuscript_variables import _render_markdown + + +def test_known_prose_variables_and_claims_resolve_to_the_supplied_value() -> None: + assert _render_markdown("Result {{value}} / {{claim:value}}.", {"value": "computed"}) == "Result computed / computed." + + +@pytest.mark.parametrize("text", ["Result {{misspelled}}.", "Result {{claim:missing}}."]) +def test_unknown_prose_placeholder_fails_instead_of_stripping_its_delimiters(text: str) -> None: + with pytest.raises(KeyError): + _render_markdown(text, {"value": "computed"}) + + +def test_code_examples_normalize_locally_without_admitting_unknown_adjacent_prose() -> None: + example = "Use `{{unknown}}` and:\n```text\n{{claim:missing}}\n```\n" + assert _render_markdown(example, {}) == "Use `unknown` and:\n```text\nclaim:missing\n```\n" + with pytest.raises(KeyError): + _render_markdown(example + "Prose {{unknown}}.", {}) + + +@pytest.mark.parametrize("text", ["Result {{missing token}}.", "Result {{ claim:value }}."]) +def test_unrecognized_prose_token_shape_remains_an_error(text: str) -> None: + with pytest.raises(ValueError, match="unresolved manuscript tokens"): + _render_markdown(text, {"value": "computed"}) diff --git a/tests/test_material_claim_fields.py b/tests/test_material_claim_fields.py new file mode 100644 index 0000000..c097ae2 --- /dev/null +++ b/tests/test_material_claim_fields.py @@ -0,0 +1,153 @@ +"""Learner fields share canonical claim admission and retain original claim IDs.""" + +from __future__ import annotations + +from copy import deepcopy +from dataclasses import replace + +import pytest + +from src.curriculum.graph import CurriculumGraph +from src.curriculum.materials import audit_course_material_index, build_course_material_index +from src.curriculum.model import AssessmentQuestion, Skill, Stratum, Subject +from src.provenance.binding import ProvenanceError, lint_graph, render_graph, validate_curriculum, validate_skill +from src.provenance.claims import ClaimRegistry + +TOKEN = "{{cl\u200baim:bo\u200bund}}" + + +def _registry() -> ClaimRegistry: + registry = ClaimRegistry() + registry.register("bound", "bound-value") + return registry + + +def _skill() -> Skill: + return Skill( + "a", "A", Stratum.MATH_FOUNDATIONS, "Objective", "Description", (), + (AssessmentQuestion("Question", ("one", "two"), (0,)),), + ) + + +def _field(skill: Skill, name: str, value: str) -> Skill: + if name == "assessment_prompt": + return replace(skill, assessments=(replace(skill.assessments[0], prompt=value),)) + if name == "assessment_answer": + return replace(skill, assessments=(replace(skill.assessments[0], answers=(value, "two")),)) + return replace(skill, **{name: value}) + + +@pytest.mark.parametrize("field", ["name", "objective", "description", "assessment_prompt", "assessment_answer"]) +@pytest.mark.parametrize("token", ["{{claim:missing}}", "{{ claim:bound }}", "{{claim:bad-id}}", "{{cl\u200baim:missing}}"]) +def test_standalone_skill_and_curriculum_validation_reject_every_bad_field(field: str, token: str) -> None: + skill = _field(_skill(), field, token) + with pytest.raises(ProvenanceError): + validate_skill(skill, _registry()) + with pytest.raises(ProvenanceError): + validate_curriculum([skill], _registry()) + + +@pytest.mark.parametrize("field", ["name", "objective", "description", "assessment_prompt", "assessment_answer"]) +def test_format_character_normalization_preserves_original_claim_identity(field: str) -> None: + skill = _field(_skill(), field, TOKEN) + assert skill.claim_tokens() == {"bound"} + validate_skill(skill, _registry()) + + +def _graph() -> CurriculumGraph: + first = replace(_skill(), name=f"A {TOKEN}", objective=f"Objective {TOKEN}", description=f"Description {TOKEN}") + second = replace(_skill(), skill_id="b", name=f"B {TOKEN}", prerequisites=("a",)) + return CurriculumGraph([ + Subject("sub_m", f"Subject {TOKEN}", Stratum.MATH_FOUNDATIONS, "Subject description", (first, second)) + ]) + + +def _sources(*, claims: bool = True) -> dict: + value = TOKEN if claims else "bound-value" + pointer = {"source_id": "fixture", "path": "lessons/intro.md", "title": f"Pointer {value}", "role": f"Role {value}"} + return { + "version": 1, + "sources": [{ + "source_id": "fixture", "repo_url": "https://github.com/fixture/materials", "commit": "a" * 40, + "title": f"Source {value}", "license_status": f"License {value}", "status_note": f"Status {value}", + "path_globs": ["lessons/*.md"], + }], + "stratum_mappings": {stratum.value: [dict(pointer)] for stratum in Stratum}, + "topic_mappings": [], + } + + +def test_material_names_references_context_and_source_labels_render_without_losing_claim_ids() -> None: + graph = _graph() + index = build_course_material_index(graph, _sources(), registry=_registry()) + first, second = index["skills"]["a"], index["skills"]["b"] + + assert first["name"] == "A bound-value" and first["subjectName"] == "Subject bound-value" + assert second["prerequisites"][0]["name"] == "A bound-value" + assert second["relatedSkills"][0]["subjectName"] == "Subject bound-value" + assert first["downstreamSkills"][0]["name"] == "B bound-value" + assert "A bound-value (a)" in "\n".join(section["body"] for section in second["sections"]) + assert "B bound-value (b)" in "\n".join(section["body"] for section in first["sections"]) + assert first["provenance"]["claimTokens"] == ["bound"] + assert first["externalPointers"][0]["title"] == "Pointer bound-value" + assert first["externalPointers"][0]["role"] == "Role bound-value" + assert index["sources"][0]["title"] == "Source bound-value" + assert index["sources"][0]["statusNote"] == "Status bound-value" + assert audit_course_material_index(index, graph).issues == () + + +def test_material_builder_without_registry_accepts_claim_free_input_but_refuses_raw_claims() -> None: + graph = _graph() + with pytest.raises(ProvenanceError): + build_course_material_index(graph, _sources(claims=False)) + rendered = render_graph(graph, _registry()) + index = build_course_material_index(rendered, _sources(claims=False)) + assert index["skills"]["a"]["name"] == "A bound-value" + + +@pytest.mark.parametrize("field", ["name", "subjectName", "prerequisite", "section", "source", "pointer"]) +@pytest.mark.parametrize("token", ["{{claim:missing}}", "{{ claim:bound }}", "{{cl\u200baim:missing}}"]) +def test_material_audit_catches_unrendered_tokens_outside_section_bodies(field: str, token: str) -> None: + graph = _graph() + index = deepcopy(build_course_material_index(graph, _sources(), registry=_registry())) + record = index["skills"]["b"] + if field in ("name", "subjectName"): + record[field] = token + elif field == "prerequisite": + record["prerequisites"][0]["name"] = token + elif field == "section": + record["sections"][0]["body"] = token + elif field == "source": + index["sources"][0]["title"] = token + else: + record["externalPointers"][0]["title"] = token + issues = audit_course_material_index(index, graph).issues + assert any(issue.code == "unresolved-claim-token" for issue in issues) + + +def test_subject_name_is_rendered_and_is_part_of_numeric_lint() -> None: + graph = _graph() + assert render_graph(graph, _registry()).subjects[0].name == "Subject bound-value" + malformed = CurriculumGraph([replace(graph.subjects[0], name="{{ claim:bound }}")]) + with pytest.raises(ProvenanceError): + render_graph(malformed, _registry()) + unbacked = CurriculumGraph([replace(graph.subjects[0], name="Result 2.75 nats")]) + violations = lint_graph(unbacked) + assert len(violations) == 1 and violations[0].value == "2.75" + + +def test_subject_names_that_collide_after_rendering_are_refused() -> None: + from src.curriculum.model import CurriculumError + + registry = ClaimRegistry() + registry.register("first", "same") + registry.register("second", "same") + first = _skill() + second = replace(first, skill_id="b", name="B", prerequisites=("a",)) + graph = CurriculumGraph([ + Subject("sub_a", "Course {{claim:first}}", first.stratum, "", (first,)), + Subject("sub_b", "Course {{claim:second}}", second.stratum, "", (second,)), + ]) + graph.validate() + with pytest.raises(CurriculumError, match="duplicate subject name"): + render_graph(graph, registry) diff --git a/tests/test_pdf_provenance.py b/tests/test_pdf_provenance.py new file mode 100644 index 0000000..afbcae5 --- /dev/null +++ b/tests/test_pdf_provenance.py @@ -0,0 +1,121 @@ +"""Receipt/freshness contracts use disposable real files, never historical outputs.""" + +from __future__ import annotations + +from copy import deepcopy + +import pytest + +from src.manuscript.pdf_receipt import ( + PdfReceiptError, + add_pdf_assets, + capture_pdf_inputs, + latex_asset_paths, + pdf_receipt_issues, + validate_pdf_inputs, + write_pdf_receipt, +) + + +def _receipt(project): + return write_pdf_receipt( + project["target"], root=project["root"], source_revision=project["revision"], inputs=project["inputs"], + ) + + +@pytest.mark.parametrize("relative", [ + "manuscript/config.yaml", "manuscript/preamble.md", "manuscript/references.bib", + "output/manuscript/01_fixture.md", "output/manuscript/assets/cover.pdf", + "output/manuscript/assets/atlas.pdf", "output/manuscript/assets/body.pdf", +]) +def test_changed_actual_input_invalidates_unchanged_pdf_and_cannot_refresh_receipt(pdf_project, relative: str) -> None: + receipt = _receipt(pdf_project) + old_receipt, old_pdf = receipt.read_bytes(), pdf_project["target"].read_bytes() + path = pdf_project["root"] / relative + path.write_bytes(path.read_bytes() + b"\nchanged input\n") + + assert pdf_receipt_issues(pdf_project["root"]) + with pytest.raises(PdfReceiptError, match="changed"): + _receipt(pdf_project) + assert receipt.read_bytes() == old_receipt + assert pdf_project["target"].read_bytes() == old_pdf + + +@pytest.mark.parametrize("change", ["add-section", "delete-section", "delete-image", "replace-pdf", "delete-pdf"]) +def test_changed_input_set_or_final_pdf_is_not_fresh(pdf_project, pdf_bytes, change: str) -> None: + receipt = _receipt(pdf_project) + old_receipt = receipt.read_bytes() + if change == "add-section": + (pdf_project["section"].parent / "02_added.md").write_text("# Added section\n") + elif change == "delete-section": + pdf_project["section"].unlink() + elif change == "delete-image": + pdf_project["assets"][1].unlink() + elif change == "replace-pdf": + pdf_project["target"].write_bytes(pdf_bytes("Different finalized PDF")) + else: + pdf_project["target"].unlink() + assert pdf_receipt_issues(pdf_project["root"]) + assert receipt.read_bytes() == old_receipt + + +@pytest.mark.parametrize("raw", [b"not JSON", b"[]", b'{"schema":"active-skillference-pdf/v1"}', b'{"schema":"active-skillference-pdf/v2"}']) +def test_legacy_or_incomplete_receipt_requires_real_rerender_without_relabeling(pdf_project, raw: bytes) -> None: + receipt = pdf_project["target"].with_suffix(".receipt.json") + receipt.write_bytes(raw) + assert pdf_receipt_issues(pdf_project["root"]) + assert receipt.read_bytes() == raw + + +def test_missing_receipt_is_not_created_by_read_only_freshness_audit(pdf_project) -> None: + receipt = pdf_project["target"].with_suffix(".receipt.json") + assert pdf_receipt_issues(pdf_project["root"]) + assert not receipt.exists() + + +@pytest.mark.parametrize("damage", ["missing-primary", "missing-section", "missing-image", "duplicate-path", "unsafe-path", "bad-digest", "missing-assets-list"]) +def test_incomplete_or_unsafe_input_manifest_is_rejected(pdf_project, damage: str) -> None: + manifest = deepcopy(pdf_project["inputs"]) + rows = manifest["files"] + if damage.startswith("missing-") and damage != "missing-assets-list": + kind = {"missing-primary": "config", "missing-section": "section", "missing-image": "asset"}[damage] + rows.remove(next(row for row in rows if row["kind"] == kind)) + elif damage == "duplicate-path": + rows.append(dict(rows[0])) + elif damage == "unsafe-path": + rows[0]["path"] = "../outside" + elif damage == "bad-digest": + rows[0]["sha256"] = "not-a-content-digest" + else: + manifest.pop("assets") + with pytest.raises(PdfReceiptError): + validate_pdf_inputs(pdf_project["root"], manifest) + + +def test_actual_latex_image_inputs_are_deduplicated_and_comments_do_not_create_inputs(pdf_project) -> None: + root = pdf_project["root"] + tex = root / "build.tex" + cover, _atlas, body = pdf_project["assets"] + tex.write_text( + "% \\includegraphics{missing-commented.pdf}\n" + f"\\includegraphics*[width=.5\\linewidth]{{{cover.as_posix()}}}\n" + f"\\includegraphics{{{cover.as_posix()}}}\n\\includegraphics{{{body.as_posix()}}}\n" + ) + paths = latex_asset_paths(tex, build_directory=root, root=root) + assert paths == sorted([cover.resolve(), body.resolve()]) + manifest = add_pdf_assets(root, capture_pdf_inputs(root, [pdf_project["section"]]), paths) + assert set(manifest["assets"]) == {path.relative_to(root).as_posix() for path in (cover, body)} + + +def test_nonliteral_latex_image_input_cannot_create_incomplete_provenance(pdf_project) -> None: + root = pdf_project["root"] + tex = root / "build.tex" + tex.write_text(r"\includegraphics{\chosenImage{}}") + with pytest.raises(PdfReceiptError, match="non-literal"): + latex_asset_paths(tex, build_directory=root, root=root) + + +def test_unrelated_repository_file_does_not_invalidate_actual_input_snapshot(pdf_project) -> None: + _receipt(pdf_project) + (pdf_project["root"] / "unrelated.txt").write_text("Not a PDF compiler input.") + assert pdf_receipt_issues(pdf_project["root"]) == [] diff --git a/tests/test_pilot_schema_boundaries.py b/tests/test_pilot_schema_boundaries.py new file mode 100644 index 0000000..938bd0d --- /dev/null +++ b/tests/test_pilot_schema_boundaries.py @@ -0,0 +1,48 @@ +"""A protocol needs linkable event schemas, not merely list-shaped fields.""" + +from __future__ import annotations + +import json +from pathlib import Path + +import pytest + +from src.pilot.contract import PilotContractError, validate_pilot_contract + + +@pytest.fixture +def pilot_payload() -> dict: + path = Path(__file__).resolve().parents[1] / "data" / "learner_pilot_contract.json" + return json.loads(path.read_text(encoding="utf-8")) + + +@pytest.mark.parametrize( + ("section", "field"), + [("participants", "requiredFields"), ("events", "allowedEventTypes"), + ("events", "requiredFields"), ("export", "fields")], +) +def test_schema_lists_cannot_be_empty(pilot_payload: dict, section: str, field: str) -> None: + pilot_payload[section][field] = [] + with pytest.raises(PilotContractError, match="must not be empty"): + validate_pilot_contract(pilot_payload) + + +@pytest.mark.parametrize( + ("section", "field", "missing"), + [("participants", "requiredFields", "participant_id_hash"), + ("events", "requiredFields", "event_timestamp"), + ("export", "fields", "participant_id_hash"), + ("export", "fields", "event_type")], +) +def test_schema_must_retain_linking_fields( + pilot_payload: dict, section: str, field: str, missing: str +) -> None: + pilot_payload[section][field].remove(missing) + with pytest.raises(PilotContractError, match=missing): + validate_pilot_contract(pilot_payload) + + +def test_linkage_does_not_require_demographics_or_secondary_outcomes(pilot_payload: dict) -> None: + pilot_payload["participants"]["requiredFields"] = ["participant_id_hash"] + pilot_payload["outcomes"]["secondary"] = [] + validate_pilot_contract(pilot_payload) diff --git a/tests/test_pipeline_generated_outputs.py b/tests/test_pipeline_generated_outputs.py index afb1df9..5f1b86b 100644 --- a/tests/test_pipeline_generated_outputs.py +++ b/tests/test_pipeline_generated_outputs.py @@ -1,5 +1,7 @@ from __future__ import annotations +import json +import shutil from pathlib import Path import pytest @@ -11,12 +13,19 @@ REPO = Path(__file__).resolve().parents[1] -# Rebuilds the whole generated artifact bundle; `./gates.sh fast` deselects it. -pytestmark = pytest.mark.slow -def test_build_generated_outputs_returns_existing_artifact_bundle() -> None: - bundle = build_generated_outputs(REPO) +@pytest.mark.slow +def test_build_generated_outputs_returns_existing_artifact_bundle(tmp_path: Path) -> None: + root = tmp_path / "project" + shutil.copytree( + REPO, root, + ignore=shutil.ignore_patterns( + ".git", ".venv", ".local", "__pycache__", ".pytest_cache", + ".mypy_cache", ".ruff_cache", "node_modules", + ), + ) + bundle = build_generated_outputs(root) assert isinstance(bundle, GeneratedOutputBundle) assert bundle.provenance_claims == 79 @@ -25,20 +34,33 @@ def test_build_generated_outputs_returns_existing_artifact_bundle() -> None: assert bundle.claim_rows == 74 assert bundle.frontier_sources == 45 assert bundle.material_records == 630 - assert (REPO / "output" / "data" / "claims.json") in bundle.paths + assert (root / "output" / "data" / "claims.json") in bundle.paths assert ( - REPO / "output" / "data" / "curriculum_diagnostic_gap_map.json" + root / "output" / "data" / "curriculum_diagnostic_gap_map.json" ) in bundle.paths - assert (REPO / "output" / "data" / "manuscript_variables.json") in bundle.paths + assert (root / "output" / "data" / "manuscript_variables.json") in bundle.paths assert ( - REPO / "output" / "data" / "scholarship_source_health_network.json" + root / "output" / "data" / "scholarship_source_health_network.json" ) in bundle.paths - assert (REPO / "output" / "data" / "figure_table_companions.json") in bundle.paths - assert (REPO / "output" / "data" / "course_material_index.json") in bundle.paths + assert (root / "output" / "data" / "figure_table_companions.json") in bundle.paths + assert (root / "output" / "data" / "course_material_index.json") in bundle.paths assert ( - REPO / "output" / "figures" / "active_skillference_cover.png" + root / "output" / "figures" / "active_skillference_cover.png" ) in bundle.paths assert ( - REPO / "output" / "figures" / "active_skillference_skill_atlas.png" + root / "output" / "figures" / "active_skillference_skill_atlas.png" ) in bundle.paths assert all(path.exists() for path in bundle.paths) + variables = json.loads((root / "output" / "data" / "manuscript_variables.json").read_text()) + registered = json.loads((root / "output" / "figures" / "figure_registry.json").read_text()) + assert int(variables["figure_count"]) == len(registered) == bundle.registered_figures + + +def test_importable_builder_raises_package_error_for_endorsement_overreach(tmp_path: Path) -> None: + manuscript = tmp_path / "manuscript" + manuscript.mkdir() + (manuscript / "00_fixture.md").write_text( + "The platform is NSA approved for all learners.", encoding="utf-8", + ) + with pytest.raises(ValueError, match="endorsement-overreach"): + build_generated_outputs(tmp_path) diff --git a/tests/test_predictive_coding_diagnostics.py b/tests/test_predictive_coding_diagnostics.py new file mode 100644 index 0000000..3e4f333 --- /dev/null +++ b/tests/test_predictive_coding_diagnostics.py @@ -0,0 +1,30 @@ +"""The update-threshold diagnostic does not certify distance to the fixed point.""" + +import numpy as np +import pytest + +from src.kernels.predictive_coding import gaussian_posterior_mean, predictive_coding_descent + + +def test_small_update_can_qualify_far_from_the_analytic_target() -> None: + result = predictive_coding_descent( + observation=10.0, prior_mean=0.0, prior_variance=1.0, obs_variance=1.0, + rate=1e-9, steps=3, tolerance=1e-6, + ) + target = gaussian_posterior_mean(10.0, 0.0, 1.0, 1.0) + assert target == pytest.approx(5.0) + assert result.steps_to_converge == 1 + assert abs(result.estimate - target) > 4.0 + assert result.trajectory.shape == (4,) + assert result.estimate == result.trajectory[-1] + assert np.all(np.diff(result.trajectory) > 0.0) + + +def test_no_qualifying_update_retains_steps_sentinel_and_full_trajectory() -> None: + result = predictive_coding_descent( + observation=10.0, prior_mean=0.0, prior_variance=1.0, obs_variance=1.0, + rate=0.1, steps=2, tolerance=1e-12, + ) + np.testing.assert_allclose(result.trajectory, [0.0, 1.0, 1.8]) + assert result.steps_to_converge == 2 + assert result.estimate == pytest.approx(1.8) diff --git a/tests/test_probability_support.py b/tests/test_probability_support.py new file mode 100644 index 0000000..860bf0d --- /dev/null +++ b/tests/test_probability_support.py @@ -0,0 +1,87 @@ +"""Exact support and magnitude-independent probability boundary contracts.""" + +from __future__ import annotations + +import numpy as np +import pytest + +from src.kernels.active_inference import ( + DiscreteActiveInferenceAgent, + expected_free_energy, + policy_posterior, +) +from src.kernels.distributions import ( + dirichlet_expected, + dirichlet_expected_log, + normalize, + softmax, +) +from src.kernels.inference import elbo, log_evidence, variational_free_energy + + +def test_impossible_observation_retains_exact_zero_support() -> None: + likelihood = np.eye(2) + assert log_evidence(likelihood, 1, [1.0, 0.0]) == -np.inf + assert variational_free_energy([0.0, 1.0], likelihood, 1, [1.0, 0.0]) == np.inf + assert elbo([0.0, 1.0], likelihood, 1, [1.0, 0.0]) == -np.inf + assert variational_free_energy([1.0, 0.0], likelihood, 0, [1.0, 0.0]) == 0.0 + + +def test_small_nonzero_evidence_is_not_epsilon_regularized() -> None: + mass = 1e-30 + expected = np.log(mass) + assert log_evidence(np.eye(2), 0, [mass, 1.0 - mass]) == pytest.approx(expected) + assert variational_free_energy( + [1.0, 0.0], np.eye(2), 0, [mass, 1.0 - mass] + ) == pytest.approx(-expected) + + +def test_extreme_masses_and_policy_scores_preserve_normalization() -> None: + np.testing.assert_allclose(normalize([1e308, 1e308]), [0.5, 0.5]) + np.testing.assert_allclose(dirichlet_expected([1e308, 1e308]), [0.5, 0.5]) + np.testing.assert_allclose( + dirichlet_expected_log([1e308, 1e308]), [-np.log(2.0)] * 2, + atol=1e-12, rtol=0.0, + ) + np.testing.assert_allclose(softmax([1e308, 1e308], precision=2.0), [0.5, 0.5]) + np.testing.assert_allclose(policy_posterior([-1e308, -1e308], gamma=2.0), [0.5, 0.5]) + np.testing.assert_allclose(softmax([-1e308, 1e308], precision=2.0), [0.0, 1.0]) + + +@pytest.mark.parametrize("bad", [np.nan, np.inf, -np.inf]) +@pytest.mark.parametrize("function", [normalize, dirichlet_expected, dirichlet_expected_log]) +def test_nonfinite_masses_are_rejected(function, bad: float) -> None: + with pytest.raises(ValueError): + function([bad, 1.0]) + + +@pytest.mark.parametrize("gamma", [-1.0, 0.0, np.inf, np.nan]) +def test_invalid_policy_precision_is_rejected(gamma: float) -> None: + with pytest.raises(ValueError): + policy_posterior([0.0, 1.0], gamma=gamma) + + +@pytest.mark.parametrize("belief", [[-1.0, 2.0], [0.2, 0.2], [np.nan, 1.0]]) +def test_planning_rejects_invalid_state_beliefs(belief: list[float]) -> None: + with pytest.raises(ValueError): + expected_free_energy(belief, np.full((2, 2), 0.5), [0.5, 0.5]) + + +def test_agent_rejects_invalid_transition_at_its_owning_boundary() -> None: + with pytest.raises(ValueError, match="B"): + DiscreteActiveInferenceAgent( + A=np.eye(2), B=np.array([[[-1.0], [-1.0]], [[2.0], [2.0]]]), + C=np.array([0.5, 0.5]), D=np.array([0.5, 0.5]), + ) + + +def test_prediction_rejects_out_of_range_actions_and_wrong_belief_dimension() -> None: + agent = DiscreteActiveInferenceAgent( + A=np.eye(2), B=np.eye(2)[:, :, None], + C=np.array([0.5, 0.5]), D=np.array([0.5, 0.5]), + ) + with pytest.raises(ValueError, match="action"): + agent.predict([0.5, 0.5], -1) + with pytest.raises(ValueError, match="dimension"): + agent.predict([1.0], 0) + np.testing.assert_allclose(agent.predict([0.25, 0.75], 0), [0.25, 0.75]) diff --git a/tests/test_recommendations.py b/tests/test_recommendations.py index 23ca073..1304388 100644 --- a/tests/test_recommendations.py +++ b/tests/test_recommendations.py @@ -9,6 +9,8 @@ def _project() -> dict[str, object]: return { + "projectId": "recommendations", + "name": "Recommendations", "subjects": [ { "subjectId": "sub_foundations", @@ -48,15 +50,15 @@ def _project() -> dict[str, object]: "quizzes": [ { "quizId": "math_sets_quiz", - "questions": [{"question": "What is a set?"}], + "questions": [{"question": "What is a set?", "answers": [{"answer": "A collection", "isCorrect": True}]}], }, { "quizId": "prob_events_quiz", - "questions": [{"question": "How do events relate to sets?"}], + "questions": [{"question": "How do events relate to sets?", "answers": [{"answer": "Events are sets", "isCorrect": True}]}], }, { "quizId": "bayes_update_quiz", - "questions": [{"question": "How does evidence update belief?"}], + "questions": [{"question": "How does evidence update belief?", "answers": [{"answer": "Bayesian conditioning", "isCorrect": True}]}], }, ], } @@ -134,25 +136,22 @@ def test_frontier_skill_whose_prerequisite_failed_is_remediation() -> None: assert "retake the official quiz" not in prob["reason"] -def test_cyclic_exports_fail_closed_to_zero_depths() -> None: - """A dependency cycle in a malformed export makes _depths fail closed (depth 0).""" - project = { - "subjects": [ - { - "subjectId": "s", - "name": "S", - "skills": [ - {"skillId": "a", "name": "A", "description": "a"}, - {"skillId": "b", "name": "B", "description": "b"}, - ], - } - ], - "dependencies": [ - {"skillId": "a", "dependsOnSkillId": "b"}, - {"skillId": "b", "dependsOnSkillId": "a"}, - ], - "quizzes": [], - } - result = recommend_next_skills(project) - assert result["available"] is True - assert all(row["depth"] == 0 for row in result["recommendations"]) +@pytest.mark.parametrize( + "dependency", + [ + {"skillId": "math_sets", "dependsOnSkillId": "prob_events"}, + {"skillId": "prob_events", "dependsOnSkillId": "missing"}, + {"skillId": "prob_events", "dependsOnSkillId": "prob_events"}, + {"skillId": "prob_events", "dependsOnSkillId": "math_sets"}, + "not an edge", + ], +) +def test_invalid_export_graph_cannot_produce_an_unsuccessful_retake(dependency) -> None: + project = _project() + project["dependencies"].append(dependency) + result = recommend_next_skills( + project, unsuccessful_skill_ids=["bayes_update"], + ) + assert result["available"] is False + assert result["recommendations"] == [] + assert result["error"] diff --git a/tests/test_release_readiness_inventory.py b/tests/test_release_readiness_inventory.py index 4bf8145..0e1e61a 100644 --- a/tests/test_release_readiness_inventory.py +++ b/tests/test_release_readiness_inventory.py @@ -280,10 +280,33 @@ def test_summarize_probe_distinguishes_reachable_and_blocked() -> None: assert _summarize_probe(silent) == "blocked (exit 3)" -def test_collect_blocker_probes_returns_string_summary(tmp_path: Path) -> None: - from src.release_readiness import BlockerProbeSummary, collect_blocker_probes +def test_public_probe_distinguishes_empty_populated_and_tags_only_local_repositories(tmp_path: Path) -> None: + from src.release_readiness import probe_public_repository - summary = collect_blocker_probes(tmp_path) - assert isinstance(summary, BlockerProbeSummary) - assert isinstance(summary.public_repository, str) and summary.public_repository - assert isinstance(summary.docker_daemon, str) and summary.docker_daemon + empty = tmp_path / "empty.git" + subprocess.run(["git", "init", "--bare", "--quiet", str(empty)], check=True) + result = probe_public_repository(tmp_path, str(empty)) + assert result.returncode == 0 + assert result.stdout.strip() == "" + + root = _repo_with_dirty_release_state(tmp_path) + populated = probe_public_repository(tmp_path, str(root)) + assert populated.returncode == 0 + assert any(line.split()[1].startswith("refs/heads/") for line in populated.stdout.splitlines()) + + subprocess.run(["git", "-C", str(root), "tag", "fixture_tag"], check=True) + branch = subprocess.run( + ["git", "-C", str(root), "symbolic-ref", "HEAD"], + check=True, capture_output=True, text=True, + ).stdout.strip() + subprocess.run(["git", "-C", str(root), "update-ref", "-d", branch], check=True) + tags_only = probe_public_repository(tmp_path, str(root)) + assert tags_only.returncode == 0 + assert {line.split()[1] for line in tags_only.stdout.splitlines()} == {"refs/tags/fixture_tag"} + + +def test_public_probe_missing_local_target_is_not_an_empty_success(tmp_path: Path) -> None: + from src.release_readiness import probe_public_repository + + result = probe_public_repository(tmp_path, str(tmp_path / "missing.git")) + assert result.returncode != 0 diff --git a/tests/test_render_manuscript.py b/tests/test_render_manuscript.py index 5f6e105..20e31af 100644 --- a/tests/test_render_manuscript.py +++ b/tests/test_render_manuscript.py @@ -6,49 +6,70 @@ import json import shutil import subprocess -from pathlib import Path - -from scripts.render_manuscript import PDF_NAME, ROOT, _write_pdf_receipt - - -def test_pdf_receipt_records_finalized_output(tmp_path: Path) -> None: - subprocess.run(["git", "init", "--quiet", str(tmp_path)], check=True) - subprocess.run( - [ - "git", - "-c", - "user.name=Receipt Test", - "-c", - "user.email=receipt@example.org", - "commit", - "--quiet", - "--allow-empty", - "-m", - "Source revision", - ], - cwd=tmp_path, - check=True, - ) - revision = subprocess.run( - ["git", "rev-parse", "HEAD"], - cwd=tmp_path, - capture_output=True, - text=True, - check=True, - ).stdout.strip() - target = tmp_path / "output" / "pdf" / PDF_NAME - target.parent.mkdir(parents=True) - shutil.copyfile(ROOT / "output" / "pdf" / PDF_NAME, target) + +import pytest + +from scripts import render_manuscript as renderer +from src.manuscript.pdf_receipt import capture_pdf_inputs, pdf_receipt_issues + + +def test_pdf_receipt_records_finalized_output_and_actual_inputs(pdf_project) -> None: + target, root = pdf_project["target"], pdf_project["root"] receipt = target.with_suffix(".receipt.json") receipt.write_text("outdated receipt", encoding="utf-8") - assert _write_pdf_receipt(target, root=tmp_path) == receipt + assert renderer._write_pdf_receipt(target, root=root, inputs=pdf_project["inputs"]) == receipt content = target.read_bytes() - assert json.loads(receipt.read_text(encoding="utf-8")) == { - "schema": "active-skillference-pdf/v1", - "output_path": f"output/pdf/{PDF_NAME}", - "size_bytes": len(content), - "sha256": hashlib.sha256(content).hexdigest(), - "source_revision": revision, + payload = json.loads(receipt.read_text(encoding="utf-8")) + assert payload["schema"] == "active-skillference-pdf/v2" + assert payload["output_path"] == f"output/pdf/{renderer.PDF_NAME}" + assert payload["size_bytes"] == len(content) + assert payload["sha256"] == hashlib.sha256(content).hexdigest() + assert payload["source_revision"] == pdf_project["revision"] + expected = { + "manuscript/config.yaml", "manuscript/preamble.md", "manuscript/references.bib", + "output/manuscript/01_fixture.md", "output/manuscript/assets/cover.pdf", + "output/manuscript/assets/atlas.pdf", "output/manuscript/assets/body.pdf", + } + rows = payload["inputManifest"]["files"] + assert {row["path"] for row in rows} == expected + for row in rows: + actual = (root / row["path"]).read_bytes() + assert row["size_bytes"] == len(actual) + assert row["sha256"] == hashlib.sha256(actual).hexdigest() + assert pdf_receipt_issues(root) == [] + + +@pytest.mark.slow +def test_real_render_places_configured_atlas_on_physical_page_two(pdf_project, monkeypatch) -> None: + required = ("pandoc", "pandoc-crossref", "xelatex", "bibtex", "pdftotext") + missing = [tool for tool in required if shutil.which(tool) is None] + if missing: + pytest.skip("existing render tools unavailable: " + ", ".join(missing)) + root = pdf_project["root"] + monkeypatch.setattr(renderer, "ROOT", root) + monkeypatch.setattr(renderer, "MANUSCRIPT", root / "manuscript") + monkeypatch.setattr(renderer, "OUTPUT_MANUSCRIPT", root / "output" / "manuscript") + monkeypatch.setattr(renderer, "OUTPUT_PDF", root / "output" / "pdf") + scratch = root / "build" + scratch.mkdir() + inputs = capture_pdf_inputs(root, [pdf_project["section"]]) + + target = renderer._render_pdf(scratch, renderer._config(), inputs=inputs) + + extracted = subprocess.run( + ["pdftotext", "-layout", str(target), "-"], + check=True, capture_output=True, text=True, + ).stdout.split("\f") + assert "FixtureTitlePage" in extracted[0] + assert "FixtureAtlasPage" in extracted[1] + assert "FixtureAtlasCaption" in extracted[1] + assert "AtlasGraphic" in extracted[1] + assert "FixtureBodySection" not in extracted[1] + assert "FixtureBodySection" in "\f".join(extracted[2:]) + receipt = json.loads(target.with_suffix(".receipt.json").read_text(encoding="utf-8")) + assert set(receipt["inputManifest"]["assets"]) == { + path.relative_to(root).as_posix() for path in pdf_project["assets"] } + assert pdf_receipt_issues(root) == [] diff --git a/tests/test_roadmap.py b/tests/test_roadmap.py index 870e0f5..bda4662 100644 --- a/tests/test_roadmap.py +++ b/tests/test_roadmap.py @@ -671,3 +671,40 @@ def _copy_minimal_source_surfaces(tmp_path: Path) -> None: (manuscript_dir / src.name).write_text( src.read_text(encoding="utf-8"), encoding="utf-8" ) + + +def test_blocked_task_retains_named_blocker_without_completion_evidence(tmp_path: Path) -> None: + payload = _valid_roadmap_payload() + task = payload["tasks"][1] + task.update( + status="blocked", evidence=[], + notes="Blocked until docker info reported: daemon unavailable.", + ) + _write_roadmap(tmp_path, payload) + + assert audit_roadmap(tmp_path) == [] + todo = (tmp_path / "TODO.md").read_text(encoding="utf-8") + details = "\n".join(render_todo_detail_pages(payload).values()) + assert "[!]" in todo and task["id"] in todo + assert "daemon unavailable." in details + + task.update(status="in_progress", notes="Implementation admitted.") + _write_roadmap(tmp_path, payload) + assert audit_roadmap(tmp_path) == [] + assert "[~]" in (tmp_path / "TODO.md").read_text(encoding="utf-8") + assert "daemon unavailable." not in "\n".join(render_todo_detail_pages(payload).values()) + + +def test_live_count_diagnostic_keeps_real_line_number_after_fenced_example(tmp_path: Path) -> None: + _copy_minimal_source_surfaces(tmp_path) + current = build_live_counts(REPO)["skill_count"].value + stale = current + 7 + (tmp_path / "README.md").write_text( + f"# Counts\n```text\nExample: {stale} skills.\n```\n\nCurrent: {stale} skills.\n", + encoding="utf-8", + ) + + issues = [issue for issue in audit_live_count_literals(tmp_path) if issue.key == "skill_count"] + + assert len(issues) == 1 + assert issues[0].line == 6 and issues[0].found == stale and issues[0].expected == current diff --git a/tests/test_run_entrypoint.py b/tests/test_run_entrypoint.py new file mode 100644 index 0000000..9df1b0e --- /dev/null +++ b/tests/test_run_entrypoint.py @@ -0,0 +1,61 @@ +"""Run the real entrypoint without fake uv/docker executables or repository writes.""" + +from __future__ import annotations + +import os +import shutil +import stat +import subprocess +from pathlib import Path + +import pytest + +REPO = Path(__file__).resolve().parents[1] + + +@pytest.mark.parametrize("argument,code", [("--help", 0), ("--unknown", 2), ("--check", 1), ("--offline", 1)]) +@pytest.mark.parametrize("existing", [False, True]) +def test_argument_and_prerequisite_exits_do_not_initialize_postgres_state( + tmp_path: Path, argument: str, code: int, existing: bool, +) -> None: + bash = shutil.which("bash", path=os.defpath) + if bash is None: + pytest.skip("existing bash runtime unavailable") + script = tmp_path / "run.sh" + shutil.copyfile(REPO / "run.sh", script) + programs = tmp_path / "programs" + programs.mkdir() + # Real system utilities make the old entropy path observable, but no uv or + # Docker binary is admitted: --check/--offline must stop at that prerequisite. + for name in ("dirname", "sed", "mkdir", "cat", "tr", "head", "chmod"): + source = shutil.which(name, path=os.defpath) + if source is None: + pytest.skip(f"existing system utility unavailable: {name}") + (programs / name).symlink_to(source) + state = tmp_path / "private-state" + password = state / "postgres_password" + before = None + if existing: + state.mkdir() + password.write_text("kept-synthetic-password\n", encoding="utf-8") + password.chmod(0o600) + before = (password.read_bytes(), password.stat().st_mtime_ns, stat.S_IMODE(password.stat().st_mode)) + env = { + key: value for key, value in os.environ.items() + if key not in {"ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD", "BASH_ENV", "ENV"} + and not key.startswith("BASH_FUNC_") + } + env.update(PATH=str(programs), ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR=str(state)) + + result = subprocess.run( + [bash, str(script), argument], cwd=tmp_path, env=env, + capture_output=True, text=True, timeout=5, check=False, + ) + + assert result.returncode == code, result.stdout + result.stderr + if argument in ("--check", "--offline"): + assert "uv is not installed" in result.stderr + if existing: + assert (password.read_bytes(), password.stat().st_mtime_ns, stat.S_IMODE(password.stat().st_mode)) == before + else: + assert not state.exists() diff --git a/tests/test_scholarship.py b/tests/test_scholarship.py index 60c3f08..d57e9c0 100644 --- a/tests/test_scholarship.py +++ b/tests/test_scholarship.py @@ -1386,3 +1386,25 @@ def test_role_gate_rejects_lane_name_padding() -> None: ) is True ) + + +@pytest.mark.parametrize("status,bucket", [(403, "botBlocks"), (404, "deadLinks")]) +def test_online_source_failure_appears_once_in_network_report(tmp_path: Path, httpserver, status: int, bucket: str) -> None: + httpserver.expect_request("/failed-source", method="HEAD").respond_with_data("", status=status) + _manuscript, matrix_path, bib_path = _write_fixture( + tmp_path, manuscript="Claim [@known].", bib="@article{known,\n}\n", + matrix=_health_matrix(httpserver.url_for("/failed-source")), + ) + health = audit_source_health(matrix_path, bib_path, online=True, timeout=5.0, allow_private=True) + scholarship = validate_scholarship( + REPO / "manuscript", REPO / "data" / "scholarship_sources.yaml", + REPO / "manuscript" / "references.bib", + ) + gaps = validate_scholarship_gaps( + scholarship, _real_claim_support(), REPO / "manuscript" / "references.bib", + ) + + report = scholarship_source_health_network(scholarship=scholarship, source_health=health, gap_report=gaps) + + assert report["summary"][bucket] == 1 + assert [row["key"] for row in report[bucket]] == ["known"] diff --git a/tests/test_screenshot_context.py b/tests/test_screenshot_context.py new file mode 100644 index 0000000..645bde5 --- /dev/null +++ b/tests/test_screenshot_context.py @@ -0,0 +1,146 @@ +"""Association-level fixtures are not browser proof; HTTP and proof files are real.""" + +from __future__ import annotations + +import hashlib +import json +from dataclasses import replace +from pathlib import Path + +import pytest +from pytest_httpserver import HTTPServer + +from src.screenshots.capture import ( + CaptureSettings, + ScreenshotCaptureError, + _capture_target, + _capture_url, + capture_context, + capture_screenshots, +) + + +def _case(httpserver, tmp_path: Path): + url = httpserver.url_for("").rstrip("/") + digest = hashlib.sha256(b"private-export-fixture").hexdigest() + instance = "capture-run-fixture" + status = { + "instanceId": instance, + "project": {"exportContentDigest": digest, "counts": {"skills": 1, "subjects": 1, "dependencies": 0, "quizzes": 1}}, + "config": {"project": {"id": "p"}, "skilltree": {"url": url}}, + "skilltree": {"available": True}, + "liveSkilltree": {"checked": True, "complete": True, "expectedDependencies": 0, "matchedDependencies": 0, + "content": {"checked": True, "complete": True, "exportContentDigest": digest}}, + } + httpserver.expect_request("/api/status", method="GET").respond_with_json(status) + browser = { + "state": "ready", "url": url, "projectId": "p", "route": "/", + "dashboardInstanceId": instance, "exportContentDigest": digest, + "frameUrl": url + "/skills-display/p", "frameVisible": True, "busy": False, + } + receipt = { + "checked": True, "ok": True, "projectId": "p", "dashboardInstanceId": instance, + "exportContentDigest": digest, + "views": [{"kind": kind, "route": route, "state": "ready", "visibleContentChecked": True} + for kind, route in (("project", "/"), ("subject", "/subjects/s"), ("quiz", "/subjects/s/skills/k/quizzes/q"))], + } + proof = { + "proofVersion": 2, "ok": True, "mode": "full-parity", "dashboardUrl": url, "skilltreeUrl": url, + "projectId": "p", "runId": instance, "dashboardInstanceId": instance, "exportContentDigest": digest, + "liveContent": {"checked": True, "complete": True, "exportContentDigest": digest}, + "liveDependencies": {"checked": True, "complete": True}, "browserConsole": receipt, + } + path = tmp_path / "explicit-proof.json" + settings = CaptureSettings(url, tmp_path / "captures", tmp_path, proof_path=path) + return settings, status, browser, proof + + +def test_explicit_matching_v2_proof_is_bound_to_real_capture_time_status(httpserver, tmp_path: Path) -> None: + settings, _status, browser, proof = _case(httpserver, tmp_path) + settings.proof_path.write_text(json.dumps(proof), encoding="utf-8") + raw = settings.proof_path.read_bytes() + context = capture_context(settings, browser_probe=browser) + association = context["proofAssociation"] + assert association["checked"] is True and association["mode"] == "full-parity" + assert association["runId"] == context["dashboardInstanceId"] + assert association["proofSha256"] == hashlib.sha256(raw).hexdigest() + assert context["exportContentDigest"] == proof["exportContentDigest"] + assert settings.proof_path.read_bytes() == raw + + +@pytest.mark.parametrize("damage", ["failed", "legacy", "url", "skilltree", "project", "digest", "instance", "content", "browser", "dependencies"]) +def test_unrelated_or_incomplete_proof_cannot_certify_current_capture(httpserver, tmp_path: Path, damage: str) -> None: + settings, _status, browser, proof = _case(httpserver, tmp_path) + if damage == "failed": + proof["ok"] = False + elif damage == "legacy": + proof["proofVersion"] = 1 + elif damage == "url": + proof["dashboardUrl"] += "/another-dashboard" + elif damage == "skilltree": + proof["skilltreeUrl"] = "http://127.0.0.1:1" + elif damage == "project": + proof["projectId"] = "other" + elif damage == "digest": + proof["exportContentDigest"] = "b" * 64 + elif damage == "instance": + proof["dashboardInstanceId"] = "different-run" + elif damage == "content": + proof["liveContent"]["complete"] = False + elif damage == "browser": + proof["browserConsole"]["checked"] = False + else: + proof["liveDependencies"]["complete"] = False + settings.proof_path.write_text(json.dumps(proof), encoding="utf-8") + before = settings.proof_path.read_bytes() + context = capture_context(settings, browser_probe=browser) + assert context["proofAssociation"]["checked"] is False + assert context["fullAppStackProofMode"] == "not associated" + assert settings.proof_path.read_bytes() == before + + +def test_missing_malformed_or_implicit_proof_is_unassociated_without_rewriting_it(httpserver, tmp_path: Path) -> None: + settings, _status, browser, proof = _case(httpserver, tmp_path) + assert capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] is False + settings.proof_path.write_text("not JSON", encoding="utf-8") + assert capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] is False + assert settings.proof_path.read_text(encoding="utf-8") == "not JSON" + settings.proof_path.write_text(json.dumps(proof), encoding="utf-8") + assert capture_context(replace(settings, proof_path=None), browser_probe=browser)["proofAssociation"]["checked"] is False + assert capture_context(settings)["proofAssociation"]["checked"] is False + + +@pytest.mark.parametrize("field,value", [("state", "fallback"), ("dashboardInstanceId", "old"), ("exportContentDigest", "c" * 64), ("projectId", "other"), ("url", "not a URL")]) +def test_browser_snapshot_must_match_current_status(httpserver, tmp_path: Path, field: str, value: str) -> None: + settings, _status, browser, _proof = _case(httpserver, tmp_path) + browser[field] = value + with pytest.raises(ScreenshotCaptureError, match="does not match"): + capture_context(settings, browser_probe=browser) + assert not settings.output_dir.exists() + + +@pytest.mark.parametrize("url", ["file:///tmp/not-a-dashboard", "https://example.invalid", "http://user:password@127.0.0.1", "http://127.0.0.1/?query=yes", "http://127.0.0.1/#fragment"]) +def test_capture_url_admission_is_pure_and_rejects_nonlocal_or_credential_bearing_targets(url: str) -> None: + with pytest.raises(ScreenshotCaptureError): + _capture_url(url) + + +def test_capture_target_comparison_normalizes_origin_and_default_port_without_merging_paths() -> None: + assert _capture_target("HTTP://LOCALHOST:80/") == _capture_target("http://localhost") + assert _capture_target("http://localhost/a") != _capture_target("http://localhost/b") + + +def test_off_origin_loopback_redirect_is_refused_before_browser_or_output_creation(httpserver, tmp_path: Path) -> None: + unrelated = HTTPServer(host="127.0.0.1", port=0) + unrelated.start() + try: + httpserver.expect_request("/", method="GET").respond_with_data( + "", status=302, headers={"Location": unrelated.url_for("/")}, + ) + settings = CaptureSettings(httpserver.url_for(""), tmp_path / "captures", tmp_path) + with pytest.raises(ScreenshotCaptureError, match="changed the requested origin"): + capture_screenshots(settings) + assert unrelated.log == [] + assert not settings.output_dir.exists() + finally: + unrelated.stop() diff --git a/tests/test_screenshots.py b/tests/test_screenshots.py index 74fcdf2..537f4a6 100644 --- a/tests/test_screenshots.py +++ b/tests/test_screenshots.py @@ -10,6 +10,7 @@ from __future__ import annotations import json +import hashlib import re import subprocess import sys @@ -173,7 +174,7 @@ def test_captured_images_are_tracked_and_carry_provenance() -> None: image = SCREENSHOT_DIR / view.filename assert image.is_file(), f"{view.filename} is not tracked" assert image.stat().st_size > 0 - assert recorded[view.filename]["sha256"] + assert recorded[view.filename]["sha256"] == hashlib.sha256(image.read_bytes()).hexdigest() def test_captured_images_are_readable_and_not_blank() -> None: diff --git a/tests/test_scripts_smoke.py b/tests/test_scripts_smoke.py index bbf90c9..a8c960e 100644 --- a/tests/test_scripts_smoke.py +++ b/tests/test_scripts_smoke.py @@ -5,6 +5,7 @@ import importlib.util import json import os +import shutil import subprocess import sys import urllib.error @@ -28,6 +29,27 @@ _SUBPROCESS_TIMEOUT = 300 +@pytest.fixture(scope="module", autouse=True) +def isolated_script_project(tmp_path_factory): + """Run real generators in disposable source, never in the audited checkout.""" + global REPO + source = REPO + project = tmp_path_factory.mktemp("script-project") + for name in ("data", "manuscript", "scripts", "src", "gnn", "tests"): + shutil.copytree( + source / name, + project / name, + ignore=shutil.ignore_patterns("__pycache__", "*.pyc"), + ) + for name in ("docker-compose.yml", "tasks.yaml", "pyproject.toml"): + shutil.copyfile(source / name, project / name) + REPO = project + try: + yield project + finally: + REPO = source + + def _run(script: str) -> subprocess.CompletedProcess[str]: return subprocess.run( [sys.executable, str(REPO / "scripts" / script)], @@ -70,20 +92,6 @@ def test_validate_docker_digests_helper_flags_tag_only_lines() -> None: ) -def test_local_skilltree_runtime_security_defaults() -> None: - compose = (REPO / "docker-compose.yml").read_text(encoding="utf-8") - run_skilltree = (REPO / "scripts" / "run_skilltree.sh").read_text(encoding="utf-8") - run_dashboard = (REPO / "scripts" / "run_learning_dashboard.sh").read_text( - encoding="utf-8" - ) - - assert "127.0.0.1:5432:5432" in compose - assert "127.0.0.1:8080:8080" in compose - assert "ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD:?" in compose - assert '--token "$SKILLTREE_TOKEN"' not in run_skilltree - assert '--token "$SKILLTREE_TOKEN"' not in run_dashboard - assert "POSTGRES_PASSWORD_FILE" in run_skilltree - assert "POSTGRES_PASSWORD_FILE" in run_dashboard def test_form_login_password_handoff_avoids_eval_argv_secret() -> None: @@ -477,8 +485,7 @@ def test_validate_live_skilltree_dashboard_script_accepts_healthy_payload( httpserver.expect_request("/api/status").respond_with_json( { "project": { - "projectId": "activeInference", - "name": "Active Skillference", + "project": {"id": "activeInference", "name": "Active Skillference"}, "counts": counts, }, "config": { @@ -635,8 +642,7 @@ def _serve_strict_live_dashboard_payload( httpserver.expect_request("/api/status").respond_with_json( { "project": { - "projectId": "activeInference", - "name": "Active Skillference", + "project": {"id": "activeInference", "name": "Active Skillference"}, "counts": counts, }, "config": { @@ -897,8 +903,7 @@ def test_validate_live_skilltree_dashboard_rejects_missing_live_dependency( httpserver.expect_request("/api/status").respond_with_json( { "project": { - "projectId": "activeInference", - "name": "Active Skillference", + "project": {"id": "activeInference", "name": "Active Skillference"}, "dependencies": [{"skillId": "leaf", "dependsOnSkillId": "root"}], "counts": { "subjects": 1, diff --git a/tests/test_skilltree_client.py b/tests/test_skilltree_client.py index 197016c..f524614 100644 --- a/tests/test_skilltree_client.py +++ b/tests/test_skilltree_client.py @@ -133,7 +133,7 @@ def test_create_quiz_does_not_duplicate_existing_questions( ).respond_with_json({"quizId": "bayes_quiz"}) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [{"id": 1}]}) + ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) assert client.create_quiz(quiz) == {"quizId": "bayes_quiz"} @@ -154,7 +154,7 @@ def test_create_quiz_appends_missing_question_suffix( ).respond_with_json({"quizId": "bayes_quiz"}) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [{"id": 1}]}) + ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/create-question", method="POST", diff --git a/tests/test_skilltree_content.py b/tests/test_skilltree_content.py new file mode 100644 index 0000000..dd65224 --- /dev/null +++ b/tests/test_skilltree_content.py @@ -0,0 +1,251 @@ +"""Real loopback regressions for identity/content parity and idempotent quiz seeding.""" + +from __future__ import annotations + +import copy +import json +from pathlib import Path + +import pytest +from pytest_httpserver import HTTPServer + +from src.dashboard.artifacts import load_skilltree_project +from src.dashboard.config import DashboardConfig +from src.dashboard.payload import live_skilltree_summary +from src.skilltree.client import SkillTreeClient +from src.skilltree.content import SkillTreeContentError, compare_live_content, export_content_digest +from src.skilltree.local_auth import ( + LocalSkillTreeState, + SkillTreeSessionClient, + ensure_local_skilltree_auth, + save_local_state, +) +from src.skilltree.local_auth_state import LocalSkillTreeAuthError +from src.skilltree.model import SkillTreeProject + + +def _compare(httpserver: HTTPServer, project: SkillTreeProject) -> dict: + client = SkillTreeClient(httpserver.url_for(""), project.project_id, token="synthetic-test-token") + return compare_live_content( + get=client._get, base_url=client.base_url, project=project.to_export_dict(), + ) + + +def test_matching_actual_content_is_digest_bound_and_answer_key_free( + httpserver: HTTPServer, content_project, skilltree_admin_content, +) -> None: + skilltree_admin_content(httpserver, content_project) + result = _compare(httpserver, content_project) + assert result["checked"] is True and result["complete"] is True + assert result["projectId"] == "p" + assert result["exportContentDigest"] == export_content_digest(content_project.to_export_dict()) + assert "isCorrect" not in json.dumps(result) + assert "synthetic-test-token" not in json.dumps(result) + + +@pytest.mark.parametrize("change", ["answer", "lesson", "extra_question"]) +def test_same_counts_cannot_hide_changed_content( + httpserver: HTTPServer, content_project, skilltree_admin_content, change: str, +) -> None: + live = copy.deepcopy(content_project.to_export_dict()) + if change == "answer": + live["quizzes"][0]["questions"][0]["answers"][0]["isCorrect"] = False + elif change == "lesson": + live["subjects"][0]["skills"][0]["description"] = "Different lesson" + else: + live["quizzes"][0]["questions"].append(live["quizzes"][0]["questions"][0]) + skilltree_admin_content(httpserver, live) + result = _compare(httpserver, content_project) + assert result["checked"] is True and result["complete"] is False + assert result["changedResources"] + assert result["subjects"] == 1 and result["skills"] == 1 + + +def test_count_only_admin_response_is_unverifiable(httpserver: HTTPServer, content_project) -> None: + httpserver.expect_request("/admin/projects/p", method="GET").respond_with_json( + {"numSkills": 1, "numSubjects": 1} + ) + with pytest.raises(SkillTreeContentError, match="unverifiable"): + _compare(httpserver, content_project) + + +@pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) +def test_existing_second_question_does_not_suppress_missing_first( + httpserver: HTTPServer, content_project, client_type, +) -> None: + quiz = content_project.quizzes[0] + httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) + httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( + {"questions": [quiz.questions[1].to_dict()]} + ) + httpserver.expect_request( + "/admin/quiz-definitions/q/create-question", method="POST", + json=quiz.questions[0].to_admin_create_payload(), + ).respond_with_json({"id": 1}) + client = client_type(httpserver.url_for(""), "p") + client.create_quiz(quiz) + posted = [request.get_json() for request, _ in httpserver.log + if request.path.endswith("/create-question")] + assert posted == [quiz.questions[0].to_admin_create_payload()] + httpserver.check_assertions() + + +@pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) +def test_extra_question_occurrence_is_refused_without_any_question_write( + httpserver: HTTPServer, content_project, client_type, +) -> None: + quiz = content_project.quizzes[0] + questions = [question.to_dict() for question in quiz.questions] + httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) + httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( + {"questions": questions + [questions[0]]} + ) + with pytest.raises(SkillTreeContentError, match="extra or changed"): + client_type(httpserver.url_for(""), "p").create_quiz(quiz) + assert not any(request.path.endswith("/create-question") for request, _ in httpserver.log) + + +def test_live_summary_never_sends_stored_credentials_to_second_origin( + tmp_path: Path, httpserver: HTTPServer, content_project, +) -> None: + original = httpserver.url_for("").rstrip("/") + second = HTTPServer(host="127.0.0.1", port=0) + second.start() + try: + state_dir = tmp_path / "private" + save_local_state(state_dir, LocalSkillTreeState( + admin_username="admin@localhost.invalid", admin_password="stored-secret", + learner_id="learner", project_id="p", skilltree_url=original, + client_secret="trusted-secret", + )) + config = DashboardConfig( + project_root=tmp_path, project_id="p", project_name="Project", + skilltree_url=second.url_for(""), local_auth_enabled=True, local_state_dir=state_dir, + ) + exported = content_project.to_export_dict() + public = load_skilltree_project(tmp_path, export_payload=exported) + result = live_skilltree_summary(config, public, export_payload=exported) + assert result["checked"] is False + assert "does not match" in result["reason"] + assert second.log == [] + assert httpserver.log == [] + assert "stored-secret" not in json.dumps(result) + finally: + second.stop() + + +def test_bootstrap_target_mismatch_is_refused_before_any_network( + tmp_path: Path, httpserver: HTTPServer, content_project, +) -> None: + with pytest.raises(LocalSkillTreeAuthError, match="bootstrap target"): + ensure_local_skilltree_auth( + base_url=httpserver.url_for(""), project_id="different", + project=content_project, state_dir=tmp_path, + ) + assert httpserver.log == [] + + +@pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) +def test_missing_duplicate_question_occurrence_is_created_once(httpserver, content_project, client_type) -> None: + quiz = content_project.quizzes[0] + first = copy.deepcopy(quiz.questions[0]) + quiz.questions[:] = [first, copy.deepcopy(first)] + httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) + httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( + {"questions": [first.to_dict()]} + ) + httpserver.expect_request( + "/admin/quiz-definitions/q/create-question", method="POST", + json=first.to_admin_create_payload(), + ).respond_with_json({"id": 2}) + + client_type(httpserver.url_for(""), "p").create_quiz(quiz) + + posted = [request.get_json() for request, _response in httpserver.log if request.path.endswith("/create-question")] + assert posted == [first.to_admin_create_payload()] + httpserver.check_assertions() + + +@pytest.mark.parametrize("changed", ["lesson", "answer"]) +def test_existing_changed_content_is_refused_before_any_seed_write( + httpserver, content_project, skilltree_admin_content, changed: str, +) -> None: + from src.skilltree.local_auth import _seed_project_if_requested + + live = copy.deepcopy(content_project.to_export_dict()) + if changed == "lesson": + live["subjects"][0]["skills"][0]["description"] = "Different lesson" + else: + live["quizzes"][0]["questions"][0]["answers"][0]["isCorrect"] = False + skilltree_admin_content(httpserver, live) + client = SkillTreeSessionClient(httpserver.url_for(""), "p") + with pytest.raises(LocalSkillTreeAuthError, match="not overwritten"): + _seed_project_if_requested( + client=client, project=content_project, seed_project=True, seed_dependencies=False, + skip_seed_if_live_content_ready=True, verify_seed=True, + ) + assert httpserver.log and all(request.method == "GET" for request, _response in httpserver.log) + + +@pytest.mark.parametrize( + ("settings", "complete"), + [ + ([], True), + ([{"setting": "quizUserRole", "value": "ROLE_QUIZ_ADMIN"}], True), + ([{"setting": "quizLength", "value": "-1"}], True), + ([{"setting": "quizPassingReq", "value": "-1"}], True), + ([{"setting": "quizPassingReq", "value": "0"}], True), + ([{"setting": "quizPassingReq", "value": "2"}], True), + ([{"setting": "quizPassingReq", "value": "1"}], False), + ([{"setting": "quizLength", "value": "1"}], False), + ], +) +def test_native_quiz_settings_preserve_full_question_completion( + httpserver, content_project, skilltree_admin_content, settings, complete, +) -> None: + skilltree_admin_content( + httpserver, content_project, quiz_settings={"q": settings}, + ) + result = _compare(httpserver, content_project) + assert result["complete"] is complete + assert result["changedResources"] == ([] if complete else ["quiz:q"]) + assert result["missingResources"] == [] + httpserver.check_assertions() + + +def test_export_percentage_is_compared_with_native_correct_question_count( + httpserver, content_project, skilltree_admin_content, +) -> None: + from dataclasses import replace + + content_project.quizzes[0] = replace(content_project.quizzes[0], passing_percent=50) + skilltree_admin_content( + httpserver, content_project, + quiz_settings={"q": [{"setting": "quizPassingReq", "value": "1"}]}, + ) + assert _compare(httpserver, content_project)["complete"] is True + + +@pytest.mark.parametrize( + "settings", + [ + [{"setting": "quizPassingReq", "value": True}], + [{"setting": "quizPassingReq", "value": "NaN"}], + [{"setting": "quizPassingReq", "value": "3"}], + [{"setting": "quizLength", "value": "0"}], + [{"setting": "quizLength", "value": "-2"}], + [ + {"setting": "quizPassingReq", "value": "2"}, + {"setting": "quizPassingReq", "value": "1"}, + ], + ], +) +def test_unverifiable_native_quiz_settings_fail_closed( + httpserver, content_project, skilltree_admin_content, settings, +) -> None: + skilltree_admin_content( + httpserver, content_project, quiz_settings={"q": settings}, + ) + with pytest.raises(SkillTreeContentError): + _compare(httpserver, content_project) + assert all(request.method == "GET" for request, _response in httpserver.log) diff --git a/tests/test_skilltree_local_auth.py b/tests/test_skilltree_local_auth.py index 6ccaf05..47d24c5 100644 --- a/tests/test_skilltree_local_auth.py +++ b/tests/test_skilltree_local_auth.py @@ -686,6 +686,9 @@ def test_admin_create_methods_hit_expected_paths(httpserver: HTTPServer) -> None httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz", method="POST" ).respond_with_json({}) + httpserver.expect_request( + "/admin/quiz-definitions/m_sets_quiz/questions", method="GET" + ).respond_with_json({"questions": []}) client.create_project("Active Inference") client.create_subject(subject) @@ -745,7 +748,7 @@ def test_admin_create_quiz_appends_missing_question_suffix( ).respond_with_json({}) httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [{"id": 1}]}) + ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz/create-question", method="POST", @@ -763,7 +766,7 @@ def test_admin_create_quiz_appends_missing_question_suffix( def _bootstrap_httpserver_expectations( - httpserver: HTTPServer, *, needs_bootstrap: bool + httpserver: HTTPServer, skilltree_admin_content, *, needs_bootstrap: bool ) -> None: httpserver.expect_request("/public/config", method="GET").respond_with_json( {"needToBootstrap": needs_bootstrap} @@ -788,11 +791,7 @@ def _bootstrap_httpserver_expectations( httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz", method="POST" ).respond_with_json({}) - # Post-seed verification (verify_seed_landed) reads the live skill count; the - # tiny project defines exactly 1 skill, so report 1 to confirm the seed landed. - httpserver.expect_request( - "/admin/projects/activeInference", method="GET" - ).respond_with_json({"numSkills": 1, "numSubjects": 1}) + skilltree_admin_content(httpserver, _tiny_project(), initially_missing=True) httpserver.expect_request( "/admin/projects/activeInference/clientSecret", method="GET" ).respond_with_data("bootstrap-secret", content_type="text/plain") @@ -823,14 +822,25 @@ def _tiny_project() -> SkillTreeProject: ) ], dependencies=[], - quizzes=[Quiz(quiz_id="m_sets_quiz", name="Sets Quiz")], + quizzes=[Quiz( + quiz_id="m_sets_quiz", name="Sets Quiz", + questions=[QuizQuestion( + question="Which collection is a set?", + answers=[ + {"answer": "A collection of distinct elements", "isCorrect": True}, + {"answer": "An ordered list requiring repetition", "isCorrect": False}, + ], + )], + )], ) def test_ensure_local_skilltree_auth_full_first_bootstrap( - tmp_path: Path, httpserver: HTTPServer + tmp_path: Path, httpserver: HTTPServer, skilltree_admin_content, ) -> None: - _bootstrap_httpserver_expectations(httpserver, needs_bootstrap=True) + _bootstrap_httpserver_expectations( + httpserver, skilltree_admin_content, needs_bootstrap=True, + ) base = httpserver.url_for("").rstrip("/") result = ensure_local_skilltree_auth( @@ -854,6 +864,13 @@ def test_ensure_local_skilltree_auth_full_first_bootstrap( assert stored is not None assert stored.client_secret == "bootstrap-secret" assert stored.skilltree_url == base + question_writes = [ + request.get_json() for request, _response in httpserver.log + if request.method == "POST" and request.path.endswith("/create-question") + ] + assert question_writes == [ + _tiny_project().quizzes[0].questions[0].to_admin_create_payload() + ] def test_ensure_local_skilltree_auth_reuses_existing_state( @@ -929,8 +946,8 @@ def test_ensure_local_skilltree_auth_rejects_state_url_mismatch_before_login( httpserver.check_assertions() -def test_ensure_local_skilltree_auth_skips_seed_when_live_counts_ready( - tmp_path: Path, httpserver: HTTPServer +def test_ensure_local_skilltree_auth_skips_seed_when_live_content_matches( + tmp_path: Path, httpserver: HTTPServer, skilltree_admin_content, ) -> None: base = httpserver.url_for("").rstrip("/") save_local_state( @@ -953,9 +970,7 @@ def test_ensure_local_skilltree_auth_skips_seed_when_live_counts_ready( content_type="text/html", ) httpserver.expect_request("/performLogin", method="POST").respond_with_json({}) - httpserver.expect_request( - "/admin/projects/activeInference", method="GET" - ).respond_with_json({"numSkills": 1, "numSubjects": 1}) + skilltree_admin_content(httpserver, _tiny_project()) httpserver.expect_request( "/admin/projects/activeInference/clientSecret", method="GET" ).respond_with_data("trusted-secret", content_type="text/plain") @@ -966,7 +981,7 @@ def test_ensure_local_skilltree_auth_skips_seed_when_live_counts_ready( state_dir=tmp_path, project=_tiny_project(), seed_project=True, - skip_seed_if_live_counts_ready=True, + skip_seed_if_live_content_ready=True, verify_token=False, ) @@ -1283,10 +1298,10 @@ def test_live_skill_count_rejects_missing_num_skills(httpserver: HTTPServer) -> raise AssertionError("missing numSkills did not raise") -def test_verify_seed_landed_passes_when_live_matches(httpserver: HTTPServer) -> None: - httpserver.expect_request("/admin/projects/p", method="GET").respond_with_json( - {"numSkills": 2, "numSubjects": 1} - ) +def test_verify_seed_landed_passes_when_live_matches( + httpserver: HTTPServer, skilltree_admin_content, +) -> None: + skilltree_admin_content(httpserver, _two_skill_project()) client = SkillTreeSessionClient(httpserver.url_for("/"), "p") # Exactly-equal live count must not raise. verify_seed_landed(client, _two_skill_project()) diff --git a/tests/test_skilltree_seeder.py b/tests/test_skilltree_seeder.py index 2e8fb99..e8572cd 100644 --- a/tests/test_skilltree_seeder.py +++ b/tests/test_skilltree_seeder.py @@ -68,7 +68,8 @@ def test_seed_returns_correct_summary(httpserver: HTTPServer) -> None: def handler(request): # type: ignore[no-untyped-def] order.append(request.path) - return Response("{}", content_type="application/json", status=200) + body = '{"questions": []}' if request.path.endswith("/questions") else "{}" + return Response(body, content_type="application/json", status=200) httpserver.expect_request(re.compile(r"^/.*$")).respond_with_handler(handler) @@ -91,7 +92,8 @@ def test_seed_issues_dependencies_after_all_skills(httpserver: HTTPServer) -> No def handler(request): # type: ignore[no-untyped-def] order.append(request.path) - return Response("{}", content_type="application/json", status=200) + body = '{"questions": []}' if request.path.endswith("/questions") else "{}" + return Response(body, content_type="application/json", status=200) httpserver.expect_request(re.compile(r"^/.*$")).respond_with_handler(handler) @@ -128,7 +130,8 @@ def test_seed_full_path_ordering_exact(httpserver: HTTPServer) -> None: def handler(request): # type: ignore[no-untyped-def] order.append(f"{request.method} {request.path}") - return Response("{}", content_type="application/json", status=200) + body = '{"questions": []}' if request.path.endswith("/questions") else "{}" + return Response(body, content_type="application/json", status=200) httpserver.expect_request(re.compile(r"^/.*$")).respond_with_handler(handler) @@ -159,7 +162,8 @@ def test_seed_can_defer_dependencies_for_dedicated_reconciler( def handler(request): # type: ignore[no-untyped-def] order.append(f"{request.method} {request.path}") - return Response("{}", content_type="application/json", status=200) + body = '{"questions": []}' if request.path.endswith("/questions") else "{}" + return Response(body, content_type="application/json", status=200) httpserver.expect_request(re.compile(r"^/.*$")).respond_with_handler(handler) client = SkillTreeClient( diff --git a/tests/test_skilltree_transport_boundaries.py b/tests/test_skilltree_transport_boundaries.py new file mode 100644 index 0000000..329b94a --- /dev/null +++ b/tests/test_skilltree_transport_boundaries.py @@ -0,0 +1,89 @@ +"""Malformed JSON and real socket stalls stay inside the SkillTree error contract.""" + +from __future__ import annotations + +import threading +from concurrent.futures import ThreadPoolExecutor + +import pytest +from pytest_httpserver import HTTPServer +from werkzeug.wrappers import Response + +from src.skilltree.client import SkillTreeClient, SkillTreeClientError +from src.skilltree.local_auth import ( + LocalSkillTreeAuthError, + SkillTreeSessionClient, + exchange_trusted_client_token, +) + + +@pytest.mark.parametrize("method", ["GET", "POST"]) +def test_bearer_read_timeout_is_bounded_and_domain_typed(httpserver: HTTPServer, method: str) -> None: + release = threading.Event() + def stalled_body(): + yield b"{" + release.wait() + yield b"}" + httpserver.expect_request("/slow", method=method).respond_with_handler( + lambda request: Response(stalled_body(), content_type="application/json") + ) + client = SkillTreeClient(httpserver.url_for(""), "p", token="private-token", timeout=0.05) + operation = client._get if method == "GET" else client._post + with ThreadPoolExecutor(max_workers=1) as pool: + future = pool.submit(operation, "/slow") + try: + with pytest.raises(SkillTreeClientError, match="timed out") as error: + future.result(timeout=5) + assert "private-token" not in str(error.value) + finally: + release.set() + + +@pytest.mark.parametrize("timeout", [0.0, -1.0, float("inf"), float("nan")]) +def test_bearer_timeout_must_be_finite_and_positive(timeout: float) -> None: + with pytest.raises(ValueError, match="finite and positive"): + SkillTreeClient("http://127.0.0.1:1", "p", timeout=timeout) + + +def test_session_malformed_json_is_domain_typed_and_secret_safe(httpserver: HTTPServer) -> None: + httpserver.expect_request("/broken", method="GET").respond_with_data( + '{"secret": "private-response-fragment"', content_type="application/json", + ) + client = SkillTreeSessionClient(httpserver.url_for(""), "p") + with pytest.raises(SkillTreeClientError, match="malformed JSON") as error: + client.request_json("GET", "/broken") + assert "private-response-fragment" not in str(error.value) + + +def test_token_malformed_json_is_auth_domain_error(httpserver: HTTPServer) -> None: + httpserver.expect_request("/oauth/token", method="POST").respond_with_data( + '{"access_token": "private-fragment"', content_type="application/json", + ) + with pytest.raises(LocalSkillTreeAuthError, match="malformed JSON") as error: + exchange_trusted_client_token( + base_url=httpserver.url_for(""), project_id="p", + client_secret="private-secret", learner_id="learner", + ) + assert "private-fragment" not in str(error.value) + assert "private-secret" not in str(error.value) + + +def test_token_socket_timeout_is_domain_typed(httpserver: HTTPServer) -> None: + release = threading.Event() + def stalled_body(): + yield b"{" + release.wait() + yield b"}" + httpserver.expect_request("/oauth/token", method="POST").respond_with_handler( + lambda request: Response(stalled_body(), content_type="application/json") + ) + with ThreadPoolExecutor(max_workers=1) as pool: + future = pool.submit( + exchange_trusted_client_token, base_url=httpserver.url_for(""), project_id="p", + client_secret="synthetic-secret", learner_id="learner", timeout=0.05, + ) + try: + with pytest.raises(SkillTreeClientError, match="timed out"): + future.result(timeout=5) + finally: + release.set() diff --git a/tests/test_zenodo.py b/tests/test_zenodo.py new file mode 100644 index 0000000..d0c14c9 --- /dev/null +++ b/tests/test_zenodo.py @@ -0,0 +1,132 @@ +"""Publication API contracts over loopback, with synthetic credentials only.""" + +from __future__ import annotations + +import json +import os +import urllib.request +from pathlib import Path + +import pytest +from werkzeug.wrappers import Response + +from src.publication import zenodo + +TOKEN = "loopback-test-token" +DOI = "10.5072/fixture.7" + + +@pytest.fixture +def api(httpserver, monkeypatch): + monkeypatch.setattr(zenodo, "ZENODO_API", httpserver.url_for("/").rstrip("/")) + return httpserver + + +def _draft(identifier: int, title: str = "Wanted") -> dict: + return { + "id": identifier, "title": title, "state": "unsubmitted", "submitted": False, + "metadata": {"prereserve_doi": {"doi": DOI}}, + "links": {"html": "http://localhost/draft"}, + } + + +@pytest.fixture +def observed_upload(monkeypatch): + streams = [] + + class ObserveRequest(urllib.request.BaseHandler): + def http_request(self, request): + if request.get_method() == "PUT": + stream = request.data + assert hasattr(stream, "read") and not stream.closed + assert not isinstance(stream, (bytes, bytearray)) + assert stream.tell() == 0 + assert request.get_header("Content-length") == str(os.fstat(stream.fileno()).st_size) + streams.append(stream) + return request + + https_request = http_request + + # A request processor observes the real urllib transfer; no response is faked. + monkeypatch.setattr(urllib.request, "_opener", urllib.request.build_opener(ObserveRequest())) + return streams + + +def _upload_bucket(api) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json( + {"links": {"bucket": api.url_for("/bucket/fixture")}} + ) + + +def test_upload_streams_open_file_with_exact_length_and_closes_it(api, tmp_path: Path, observed_upload) -> None: + content = b"actual streamed archive bytes\n" * 32768 + path = tmp_path / "artifact.zip" + path.write_bytes(content) + _upload_bucket(api) + + def receive(request): + assert request.headers["Authorization"] == f"Bearer {TOKEN}" + assert request.headers["Content-Length"] == str(len(content)) + assert request.get_data() == content + return Response(json.dumps({"key": path.name}), content_type="application/json") + + api.expect_request("/bucket/fixture/artifact.zip", method="PUT").respond_with_handler(receive) + assert zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=path) == path.name + api.check_assertions() + assert len(observed_upload) == 1 and observed_upload[0].closed + + +@pytest.mark.parametrize("status,body", [(500, "failed"), (200, "not JSON"), (200, "{}")]) +def test_upload_failure_closes_stream_and_never_invents_a_confirmed_filename( + api, tmp_path: Path, observed_upload, status: int, body: str, +) -> None: + path = tmp_path / "artifact.zip" + path.write_bytes(b"archive") + _upload_bucket(api) + api.expect_request("/bucket/fixture/artifact.zip", method="PUT").respond_with_data(body, status=status) + with pytest.raises(zenodo.ZenodoError): + zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=path) + assert len(observed_upload) == 1 and observed_upload[0].closed + + +def test_missing_upload_bucket_refuses_before_opening_or_sending_a_file(api, tmp_path: Path) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json({"links": {}}) + with pytest.raises(zenodo.ZenodoError, match="no upload bucket"): + zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=tmp_path / "absent.zip") + assert [request.method for request, _response in api.log] == ["GET"] + + +def test_metadata_preserves_declared_identity_but_excludes_non_api_creator_fields(tmp_path: Path) -> None: + path = tmp_path / ".zenodo.json" + path.write_text(json.dumps({ + "title": "Fixture", "description": "Declared description", "license": "MIT", + "version": "", "creators": [{"name": "Fixture Author", "affiliation": "Lab", "email": "private@example.org"}], + })) + metadata = zenodo.zenodo_metadata(path) + assert metadata["title"] == "Fixture" and metadata["license"] == "mit" + assert metadata["creators"] == [{"name": "Fixture Author", "affiliation": "Lab"}] + assert "version" not in metadata and "private@example.org" not in json.dumps(metadata) + + +@pytest.mark.parametrize("payload", [{}, {"id": 7}, {**_draft(7), "submitted": True}]) +def test_reservation_rejects_missing_identity_or_doi_and_already_published_result(api, payload) -> None: + api.expect_request("/deposit/depositions", method="POST").respond_with_json(payload) + with pytest.raises(zenodo.ZenodoError): + zenodo.reserve_doi(token=TOKEN, metadata={"title": "Fixture"}) + + +def test_doi_mismatch_sends_no_publication_post(api) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json(_draft(7)) + with pytest.raises(zenodo.ZenodoError, match="refusing to publish"): + zenodo.publish_deposit(token=TOKEN, deposition_id=7, expected_doi="10.5072/other") + assert [request.method for request, _response in api.log] == ["GET"] + + +def test_explicit_matching_doi_allows_only_the_requested_loopback_publication(api) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json(_draft(7)) + api.expect_request("/deposit/depositions/7/actions/publish", method="POST").respond_with_json( + {"doi": DOI, "state": "done", "submitted": True, "links": {"record_html": api.url_for("/record/7")}} + ) + result = zenodo.publish_deposit(token=TOKEN, deposition_id=7, expected_doi=DOI) + assert result["doi"] == DOI and result["submitted"] and result["state"] == "done" + assert [request.method for request, _response in api.log] == ["GET", "POST"] From db829597a8d4ff97b7f8f44c489491a9db505904 Mon Sep 17 00:00:00 2001 From: Daniel Ari Friedman <6911384+docxology@users.noreply.github.com> Date: Mon, 16 Nov 2026 13:27:21 -0800 Subject: [PATCH 3/3] Resolve final software gate diagnostics without weakening proof boundaries - Preserve the audited checkout while providing the isolated validator its required primary HTML fixture. - Make proof JSON type guards explicit and preserve native malformed-response findings and strict action-type rejection. - Apply reported formatting and import fixes only to selected software paths; retain the uv, GNN, per-module, real-stack and owner holds. --- scripts/bootstrap_local_skilltree.py | 8 +- scripts/capture_dashboard_screenshots.py | 6 +- scripts/render_manuscript.py | 75 +++++-- scripts/serve_learner_dashboard.py | 4 +- scripts/validate_dashboard_console.py | 26 ++- scripts/validate_rendered_freshness.py | 1 + src/artifacts/json_io.py | 12 +- src/curriculum/graph.py | 16 +- src/curriculum/materials/builders.py | 15 +- src/curriculum/materials/github_sources.py | 9 +- src/curriculum/materials/helpers.py | 24 +- src/curriculum/materials/validation.py | 7 +- src/curriculum/model.py | 4 +- src/dashboard/artifacts.py | 10 +- src/dashboard/browser_validation.py | 186 ++++++++++++---- src/dashboard/live_validation.py | 13 +- src/dashboard/payload.py | 5 +- src/dashboard/server.py | 7 +- src/figures/plots/registry_io.py | 12 +- src/integration/full_app_stack.py | 56 +++-- src/kernels/active_inference.py | 10 +- src/kernels/inference.py | 14 +- src/manuscript/pdf_receipt.py | 123 ++++++++--- src/provenance/binding.py | 2 - src/publication/zenodo.py | 13 +- src/release_readiness.py | 11 +- src/scholarship/source_health_report.py | 2 - src/screenshots/capture.py | 217 ++++++++++++++----- src/skilltree/client.py | 10 +- src/skilltree/content.py | 120 +++++++--- tests/conftest.py | 180 ++++++++++----- tests/test_atomic_json_publication.py | 25 ++- tests/test_course_materials.py | 51 ++++- tests/test_curriculum_graph.py | 19 +- tests/test_dashboard_console.py | 1 + tests/test_dashboard_live_validation.py | 54 +++-- tests/test_dashboard_mount.py | 33 ++- tests/test_dashboard_server.py | 37 +++- tests/test_figures.py | 22 +- tests/test_full_app_stack.py | 95 ++++++-- tests/test_generated_output_cli.py | 8 +- tests/test_gnn_conformance.py | 5 +- tests/test_manuscript_hydration.py | 30 ++- tests/test_material_claim_fields.py | 147 ++++++++++--- tests/test_pdf_provenance.py | 96 ++++++-- tests/test_pilot_schema_boundaries.py | 26 ++- tests/test_pipeline_generated_outputs.py | 37 +++- tests/test_predictive_coding_diagnostics.py | 23 +- tests/test_probability_support.py | 38 +++- tests/test_recommendations.py | 26 ++- tests/test_release_readiness_inventory.py | 21 +- tests/test_render_manuscript.py | 27 ++- tests/test_roadmap.py | 27 ++- tests/test_run_entrypoint.py | 33 ++- tests/test_scholarship.py | 27 ++- tests/test_screenshot_context.py | 173 ++++++++++++--- tests/test_screenshots.py | 7 +- tests/test_scripts_smoke.py | 4 +- tests/test_skilltree_client.py | 8 +- tests/test_skilltree_content.py | 177 ++++++++++----- tests/test_skilltree_local_auth.py | 51 +++-- tests/test_skilltree_transport_boundaries.py | 36 ++- tests/test_zenodo.py | 111 +++++++--- 63 files changed, 2000 insertions(+), 673 deletions(-) diff --git a/scripts/bootstrap_local_skilltree.py b/scripts/bootstrap_local_skilltree.py index 4f63bed..4feb901 100755 --- a/scripts/bootstrap_local_skilltree.py +++ b/scripts/bootstrap_local_skilltree.py @@ -43,9 +43,7 @@ PROJECT_NAME = SKILLTREE_PROJECT_NAME -def build_project( - project_id: str = PROJECT_ID, project_name: str = PROJECT_NAME -): # type: ignore[no-untyped-def] +def build_project(project_id: str = PROJECT_ID, project_name: str = PROJECT_NAME): # type: ignore[no-untyped-def] """Build the provenance-rendered SkillTree project (no network).""" graph = build_curriculum() registry = build_claim_registry() @@ -175,7 +173,9 @@ def main() -> int: ) return 0 - project = None if args.no_seed else build_project(args.project_id, args.project_name) + project = ( + None if args.no_seed else build_project(args.project_id, args.project_name) + ) try: result = ensure_local_skilltree_auth( base_url=args.base_url, diff --git a/scripts/capture_dashboard_screenshots.py b/scripts/capture_dashboard_screenshots.py index fb83378..a92b0de 100755 --- a/scripts/capture_dashboard_screenshots.py +++ b/scripts/capture_dashboard_screenshots.py @@ -37,7 +37,11 @@ def parse_args(argv: list[str] | None = None) -> argparse.Namespace: help="tracked directory receiving the captured PNGs and provenance record", ) parser.add_argument("--browser-command", default=DEFAULT_BROWSER_COMMAND) - parser.add_argument("--proof-path", type=Path, help="Explicit current v2 full-stack proof; older/unmatched proofs remain unassociated.") + parser.add_argument( + "--proof-path", + type=Path, + help="Explicit current v2 full-stack proof; older/unmatched proofs remain unassociated.", + ) parser.add_argument("--viewport-width", type=int, default=VIEWPORT_WIDTH) parser.add_argument("--viewport-height", type=int, default=VIEWPORT_HEIGHT) return parser.parse_args(argv) diff --git a/scripts/render_manuscript.py b/scripts/render_manuscript.py index 9879df0..116c0da 100755 --- a/scripts/render_manuscript.py +++ b/scripts/render_manuscript.py @@ -51,10 +51,16 @@ sys.path.insert(0, str(ROOT)) from src.manuscript.pdf_receipt import ( - PDF_NAME, add_pdf_assets, capture_pdf_inputs, hydrated_section_files, - latex_asset_paths, validate_pdf_inputs, write_pdf_receipt, + PDF_NAME, + add_pdf_assets, + capture_pdf_inputs, + hydrated_section_files, + latex_asset_paths, + validate_pdf_inputs, + write_pdf_receipt, ) + def _require(tool: str) -> str: found = shutil.which(tool) if found is None: @@ -177,21 +183,39 @@ def _page_two_visual(config: dict[str, object]) -> str: visual = front["page_two_visual"] if not isinstance(visual, dict): raise SystemExit("front_matter.page_two_visual must be a mapping") - image_name, title, caption = (visual.get(key) for key in ("image", "title", "caption")) - if not all(isinstance(value, str) and value.strip() for value in (image_name, title, caption)): - raise SystemExit("front_matter.page_two_visual requires image, title and caption") + image_name, title, caption = ( + visual.get(key) for key in ("image", "title", "caption") + ) + if not all( + isinstance(value, str) and value.strip() + for value in (image_name, title, caption) + ): + raise SystemExit( + "front_matter.page_two_visual requires image, title and caption" + ) image = (OUTPUT_MANUSCRIPT / str(image_name)).resolve() if not image.is_file(): - raise SystemExit(f"front_matter.page_two_visual.image does not resolve to a file: {image}") - return "\n".join([ - "```{=latex}", r"\thispagestyle{empty}", r"\begin{center}", - r"{\Large\bfseries " + _latex_escape(str(title)) + r"\par}", - r"\vspace{0.6em}", - r"\includegraphics[width=\textwidth,height=0.75\textheight,keepaspectratio]{" - + image.as_posix() + "}", - r"\par\vspace{0.6em}", _latex_escape(str(caption)), - r"\end{center}", r"\clearpage", "```", "", - ]) + raise SystemExit( + f"front_matter.page_two_visual.image does not resolve to a file: {image}" + ) + return "\n".join( + [ + "```{=latex}", + r"\thispagestyle{empty}", + r"\begin{center}", + r"{\Large\bfseries " + _latex_escape(str(title)) + r"\par}", + r"\vspace{0.6em}", + r"\includegraphics[width=\textwidth,height=0.75\textheight,keepaspectratio]{" + + image.as_posix() + + "}", + r"\par\vspace{0.6em}", + _latex_escape(str(caption)), + r"\end{center}", + r"\clearpage", + "```", + "", + ] + ) def _section_files() -> list[Path]: @@ -206,7 +230,8 @@ def _assemble(config: dict[str, object], destination: Path) -> list[Path]: "`uv run python scripts/z_generate_manuscript_variables.py`" ) parts = [ - _titlepage(config), _page_two_visual(config), + _titlepage(config), + _page_two_visual(config), "```{=latex}\n\\setcounter{tocdepth}{2}\n\\tableofcontents\n\\clearpage\n```\n", ] for index, path in enumerate(sections): @@ -277,12 +302,18 @@ def _write_pdf_receipt(target: Path, *, root: Path, inputs: dict[str, object]) - text=True, ).stdout.strip() return write_pdf_receipt( - target, root=root, source_revision=revision, inputs=inputs, + target, + root=root, + source_revision=revision, + inputs=inputs, ) def _render_pdf( - scratch: Path, config: dict[str, object], *, inputs: dict[str, object], + scratch: Path, + config: dict[str, object], + *, + inputs: dict[str, object], ) -> Path: pandoc = _require("pandoc") crossref = _require("pandoc-crossref") @@ -323,7 +354,9 @@ def _render_pdf( what="pandoc", ) inputs = add_pdf_assets( - ROOT, inputs, latex_asset_paths(tex, build_directory=scratch, root=ROOT), + ROOT, + inputs, + latex_asset_paths(tex, build_directory=scratch, root=ROOT), ) latex = [xelatex, "-interaction=nonstopmode", "-halt-on-error", tex.name] @@ -341,7 +374,9 @@ def _render_pdf( target = OUTPUT_PDF / PDF_NAME temporary: Path | None = None try: - with tempfile.NamedTemporaryFile(dir=OUTPUT_PDF, prefix=f".{PDF_NAME}.", delete=False) as stream: + with tempfile.NamedTemporaryFile( + dir=OUTPUT_PDF, prefix=f".{PDF_NAME}.", delete=False + ) as stream: temporary = Path(stream.name) with produced.open("rb") as source_stream: shutil.copyfileobj(source_stream, stream) diff --git a/scripts/serve_learner_dashboard.py b/scripts/serve_learner_dashboard.py index 55a540c..0d34e63 100755 --- a/scripts/serve_learner_dashboard.py +++ b/scripts/serve_learner_dashboard.py @@ -34,7 +34,9 @@ def main() -> int: parser.add_argument("--skilltree-url", default="http://localhost:8080") parser.add_argument("--host", default="127.0.0.1") parser.add_argument("--port", type=int, default=8765) - parser.add_argument("--instance-id", default="", help="Public identity for an owned proof run.") + parser.add_argument( + "--instance-id", default="", help="Public identity for an owned proof run." + ) parser.add_argument( "--authenticator", default=os.environ.get("SKILLTREE_AUTHENTICATOR", ""), diff --git a/scripts/validate_dashboard_console.py b/scripts/validate_dashboard_console.py index e4c49c3..7b7ed7b 100755 --- a/scripts/validate_dashboard_console.py +++ b/scripts/validate_dashboard_console.py @@ -12,20 +12,22 @@ import argparse import json -import sys -from pathlib import Path import shutil +import sys import urllib.error import urllib.request +from pathlib import Path from urllib.parse import urlparse - _ALLOWED_URL_SCHEMES = frozenset({"http", "https"}) _ROOT = Path(__file__).resolve().parents[1] sys.path.insert(0, str(_ROOT)) -from src.dashboard.browser_validation import BrowserDisplayError, validate_dashboard_browser +from src.dashboard.browser_validation import ( + BrowserDisplayError, + validate_dashboard_browser, +) def _dashboard_reachable(url: str, timeout: float) -> bool: @@ -39,16 +41,14 @@ def _dashboard_reachable(url: str, timeout: float) -> bool: return False - - - - def main() -> int: parser = argparse.ArgumentParser(description=__doc__) parser.add_argument("--dashboard-url", default="http://127.0.0.1:8765/") parser.add_argument("--settle-seconds", type=float, default=5.0) parser.add_argument("--timeout", type=float, default=45.0) - parser.add_argument("--json", action="store_true", help="Emit the structured browser proof.") + parser.add_argument( + "--json", action="store_true", help="Emit the structured browser proof." + ) parser.add_argument( "--require-browser", action="store_true", @@ -80,7 +80,9 @@ def main() -> int: try: summary = validate_dashboard_browser( - args.dashboard_url, timeout=args.timeout, settle_seconds=args.settle_seconds, + args.dashboard_url, + timeout=args.timeout, + settle_seconds=args.settle_seconds, ) except BrowserDisplayError as exc: if args.json: @@ -91,7 +93,9 @@ def main() -> int: if args.json: print(json.dumps(summary, sort_keys=True)) else: - print("dashboard console gate OK (official project, subject and quiz views ready)") + print( + "dashboard console gate OK (official project, subject and quiz views ready)" + ) return 0 diff --git a/scripts/validate_rendered_freshness.py b/scripts/validate_rendered_freshness.py index d9f507a..90afe60 100755 --- a/scripts/validate_rendered_freshness.py +++ b/scripts/validate_rendered_freshness.py @@ -23,6 +23,7 @@ sys.path.insert(0, str(ROOT / "scripts")) import z_generate_manuscript_variables as zg + from src.manuscript.pdf_receipt import pdf_receipt_issues ROADMAP_OPEN_RE = re.compile( diff --git a/src/artifacts/json_io.py b/src/artifacts/json_io.py index 2fb8be8..f664ad7 100644 --- a/src/artifacts/json_io.py +++ b/src/artifacts/json_io.py @@ -33,14 +33,20 @@ def write_json_artifact( ensure_ascii: bool = True, ) -> Path: path = Path(path) - content = json.dumps(payload, indent=2, sort_keys=True, ensure_ascii=ensure_ascii) + "\n" + content = ( + json.dumps(payload, indent=2, sort_keys=True, ensure_ascii=ensure_ascii) + "\n" + ) path.parent.mkdir(parents=True, exist_ok=True) mode = stat.S_IMODE(path.stat().st_mode) if path.exists() else None temporary: Path | None = None try: with tempfile.NamedTemporaryFile( - mode="w", encoding="utf-8", dir=path.parent, - prefix=f".{path.name}.", suffix=".tmp", delete=False, + mode="w", + encoding="utf-8", + dir=path.parent, + prefix=f".{path.name}.", + suffix=".tmp", + delete=False, ) as stream: temporary = Path(stream.name) stream.write(content) diff --git a/src/curriculum/graph.py b/src/curriculum/graph.py index cdb25f1..957320c 100644 --- a/src/curriculum/graph.py +++ b/src/curriculum/graph.py @@ -295,7 +295,8 @@ def reachable_from_foundations(self) -> set[str]: for prereq in skill.prerequisites: forward[prereq].append(skill.skill_id) roots = [ - skill.skill_id for skill in skills + skill.skill_id + for skill in skills if skill.is_root and skill.stratum.is_foundational ] seen: set[str] = set() @@ -341,21 +342,26 @@ def stats(self) -> dict[str, object]: levels: dict[str, int] = {} core_levels: dict[str, int] = {} core_ids = { - skill.skill_id for skill in ordered + skill.skill_id + for skill in ordered if skill.skill_id.split("_")[0] not in _SCAFFOLD_SKILL_PREFIXES } skills_per_stratum = {stratum.value: 0 for stratum in ALL_STRATA} for skill in ordered: levels[skill.skill_id] = ( 1 + max(levels[prereq] for prereq in skill.prerequisites) - if skill.prerequisites else 0 + if skill.prerequisites + else 0 ) skills_per_stratum[skill.stratum.value] += 1 if skill.skill_id in core_ids: - core_prereqs = [prereq for prereq in skill.prerequisites if prereq in core_ids] + core_prereqs = [ + prereq for prereq in skill.prerequisites if prereq in core_ids + ] core_levels[skill.skill_id] = ( 1 + max(core_levels[prereq] for prereq in core_prereqs) - if core_prereqs else 0 + if core_prereqs + else 0 ) strata = {subject.stratum for subject in self.subjects} return { diff --git a/src/curriculum/materials/builders.py b/src/curriculum/materials/builders.py index abd7036..7ed0542 100644 --- a/src/curriculum/materials/builders.py +++ b/src/curriculum/materials/builders.py @@ -56,12 +56,17 @@ def build_course_material_index( subject = subject_by_skill[skill.skill_id] pointers = _pointers_for_skill(skill, source_payload, source_index, registry) prerequisites = [ - _skill_ref(skill_by_id[prereq], subject_by_skill[prereq], levels[prereq], registry) + _skill_ref( + skill_by_id[prereq], subject_by_skill[prereq], levels[prereq], registry + ) for prereq in skill.prerequisites ] dependents = [ _skill_ref( - skill_by_id[dependent], subject_by_skill[dependent], levels[dependent], registry, + skill_by_id[dependent], + subject_by_skill[dependent], + levels[dependent], + registry, ) for dependent in downstream.get(skill.skill_id, ()) ] @@ -88,7 +93,11 @@ def build_course_material_index( ), "prerequisites": prerequisites, "relatedSkills": _related_skills( - skill, skill_by_id, subject_by_skill, levels, registry, + skill, + skill_by_id, + subject_by_skill, + levels, + registry, ), "downstreamSkills": dependents, "assessmentSummary": _assessment_summary(skill), diff --git a/src/curriculum/materials/github_sources.py b/src/curriculum/materials/github_sources.py index dbbd62e..3fad5fc 100644 --- a/src/curriculum/materials/github_sources.py +++ b/src/curriculum/materials/github_sources.py @@ -12,12 +12,9 @@ from src.net_guard import require_http_url from .helpers import _iter_configured_pointers, _required_string - from .models import CourseMaterialIssue - - def _validate_online_trees( source_index: dict[str, dict[str, Any]], payload: dict[str, Any], @@ -34,7 +31,7 @@ def _validate_online_trees( continue try: tree_paths = _fetch_github_tree_paths(parts[0], parts[1], commit, timeout) - except (OSError, urllib.error.URLError, ValueError) as exc: + except (OSError, urllib.error.URLError, ValueError, TypeError) as exc: issues.append( CourseMaterialIssue( "github-tree", @@ -91,7 +88,7 @@ def _github_tree_payload( with urllib.request.urlopen(request, timeout=timeout) as response: payload = json.loads(response.read().decode("utf-8")) if not isinstance(payload, dict): - raise ValueError("GitHub response was not a JSON object") + raise TypeError("GitHub response was not a JSON object") return payload except urllib.error.HTTPError as exc: last_exc = exc @@ -130,7 +127,7 @@ def _fetch_github_tree_paths( payload = _github_tree_payload(request, timeout=timeout) tree = payload.get("tree") if isinstance(payload, dict) else None if not isinstance(tree, list): - raise ValueError("GitHub response did not include a tree") + raise TypeError("GitHub response did not include a tree") return { str(item.get("path")) for item in tree diff --git a/src/curriculum/materials/helpers.py b/src/curriculum/materials/helpers.py index cff6fad..d8db8f4 100644 --- a/src/curriculum/materials/helpers.py +++ b/src/curriculum/materials/helpers.py @@ -255,13 +255,17 @@ def _pointers_for_skill( repo_url = _required_string(source, "repo_url").rstrip("/") merged[key] = { "sourceId": source_id, - "sourceTitle": _render_text(_required_string(source, "title"), registry), + "sourceTitle": _render_text( + _required_string(source, "title"), registry + ), "title": _render_text(_required_string(pointer, "title"), registry), "role": _render_text(role, registry), "path": path, "commit": commit, "url": f"{repo_url}/blob/{commit}/{path}", - "licenseStatus": _render_text(_required_string(source, "license_status"), registry), + "licenseStatus": _render_text( + _required_string(source, "license_status"), registry + ), } scope_sets[key].add(match_scope) term_sets[key].update(matched_terms) @@ -390,7 +394,10 @@ def _related_skills( def _skill_ref( - skill: Skill, subject: Subject, depth: int, registry: ClaimRegistry, + skill: Skill, + subject: Subject, + depth: int, + registry: ClaimRegistry, ) -> dict[str, object]: return { "skillId": skill.skill_id, @@ -416,7 +423,8 @@ def _downstream_map(graph: CurriculumGraph) -> dict[str, tuple[str, ...]]: def _source_summaries( - source_index: dict[str, dict[str, Any]], registry: ClaimRegistry, + source_index: dict[str, dict[str, Any]], + registry: ClaimRegistry, ) -> list[dict[str, str]]: rows: list[dict[str, str]] = [] for source_id in sorted(source_index): @@ -427,8 +435,12 @@ def _source_summaries( "title": _render_text(_required_string(source, "title"), registry), "repoUrl": _required_string(source, "repo_url"), "commit": _required_string(source, "commit"), - "licenseStatus": _render_text(_required_string(source, "license_status"), registry), - "statusNote": _render_text(_required_string(source, "status_note"), registry), + "licenseStatus": _render_text( + _required_string(source, "license_status"), registry + ), + "statusNote": _render_text( + _required_string(source, "status_note"), registry + ), } ) return rows diff --git a/src/curriculum/materials/validation.py b/src/curriculum/materials/validation.py index dacfd8b..8aebbd7 100644 --- a/src/curriculum/materials/validation.py +++ b/src/curriculum/materials/validation.py @@ -229,14 +229,17 @@ def audit_course_material_index( except ProvenanceError: issues.append( CourseMaterialIssue( - "unresolved-claim-token", location, + "unresolved-claim-token", + location, "learner-facing material contains an unresolved or malformed claim token", ) ) elif isinstance(value, dict): pending.extend((f"{location}.{key}", child) for key, child in value.items()) elif isinstance(value, (list, tuple)): - pending.extend((f"{location}[{idx}]", child) for idx, child in enumerate(value)) + pending.extend( + (f"{location}[{idx}]", child) for idx, child in enumerate(value) + ) normalized = normalize_course_material_index(index) if not normalized.skills_is_mapping: issues.append( diff --git a/src/curriculum/model.py b/src/curriculum/model.py index 7ccd368..11cc9d5 100644 --- a/src/curriculum/model.py +++ b/src/curriculum/model.py @@ -30,7 +30,9 @@ def _strip_format_chars(text: str) -> str: Format characters can split an otherwise identical token or result number. Rendering, validation, linting and original-token provenance share this policy. """ - return "".join(character for character in text if unicodedata.category(character) != "Cf") + return "".join( + character for character in text if unicodedata.category(character) != "Cf" + ) SelfReportingType = Literal["HonorSystem", "Approval", "Quiz"] diff --git a/src/dashboard/artifacts.py b/src/dashboard/artifacts.py index 4a590cf..70e29b8 100644 --- a/src/dashboard/artifacts.py +++ b/src/dashboard/artifacts.py @@ -12,12 +12,12 @@ DashboardArtifact, ) from src.dashboard.figure_metadata import load_figure_metadata as _load_figure_metadata +from src.skilltree.content import export_content_digest from src.skilltree.summary import ( SkillTreeSummaryError, summarize_export_payload, validate_export_payload, ) -from src.skilltree.content import export_content_digest class DashboardArtifactError(ValueError): @@ -94,11 +94,15 @@ def load_redacted_skilltree_export( def load_skilltree_project( - project_root: Path, *, export_payload: dict[str, Any] | None = None, + project_root: Path, + *, + export_payload: dict[str, Any] | None = None, ) -> dict[str, Any]: """Load the exported SkillTree project, summarized for dashboard use.""" path = project_root / "output" / "data" / "skilltree_project.json" - payload, error = (export_payload, None) if export_payload is not None else read_json_file(path) + payload, error = ( + (export_payload, None) if export_payload is not None else read_json_file(path) + ) if payload is None: return { "available": False, diff --git a/src/dashboard/browser_validation.py b/src/dashboard/browser_validation.py index e9bf79d..8fa2eaf 100644 --- a/src/dashboard/browser_validation.py +++ b/src/dashboard/browser_validation.py @@ -47,24 +47,37 @@ class BrowserDisplayError(RuntimeError): def unexpected_errors(errors: list[str]) -> list[str]: - return [error for error in errors if not any(pattern.search(error) for pattern in _ALLOWED_ERROR_PATTERNS)] + return [ + error + for error in errors + if not any(pattern.search(error) for pattern in _ALLOWED_ERROR_PATTERNS) + ] -def browser_output(arguments: list[str], *, browser_command: str, timeout: float) -> str: +def browser_output( + arguments: list[str], *, browser_command: str, timeout: float +) -> str: try: result = run_browser_command([browser_command, *arguments], timeout=timeout) except SkillTreeFormLoginError as exc: raise BrowserDisplayError(str(exc)) from exc - if re.search(r"(?im)^\s*(?:code:\s*(?:TIMEOUT|ERROR)|isError:\s*true)\s*$", result.stdout): + if re.search( + r"(?im)^\s*(?:code:\s*(?:TIMEOUT|ERROR)|isError:\s*true)\s*$", result.stdout + ): raise BrowserDisplayError("browser command returned an explicit tool failure") return result.stdout def display_probe( - *, browser_command: str, timeout: float, activate: bool = False, + *, + browser_command: str, + timeout: float, + activate: bool = False, ) -> dict[str, Any]: script = f"() => ({DISPLAY_PROBE_JS})({'true' if activate else 'false'})" - stdout = browser_output(["eval", script], browser_command=browser_command, timeout=timeout) + stdout = browser_output( + ["eval", script], browser_command=browser_command, timeout=timeout + ) try: value = parse_axi_result(stdout) except SkillTreeFormLoginError as exc: @@ -75,32 +88,56 @@ def display_probe( def wait_for_official_display( - *, browser_command: str, dashboard_url: str, skilltree_url: str, - project_id: str, route: str, timeout: float = 45.0, - dashboard_instance_id: str | None = None, export_digest: str | None = None, + *, + browser_command: str, + dashboard_url: str, + skilltree_url: str, + project_id: str, + route: str, + timeout: float = 45.0, + dashboard_instance_id: str | None = None, + export_digest: str | None = None, ) -> dict[str, Any]: if not math.isfinite(timeout) or timeout <= 0: - raise BrowserDisplayError("browser readiness timeout must be finite and positive") + raise BrowserDisplayError( + "browser readiness timeout must be finite and positive" + ) deadline = time.monotonic() + timeout while True: remaining = deadline - time.monotonic() if remaining <= 0: - raise BrowserDisplayError(f"official SkillTree display did not reach ready route {route!r}") + raise BrowserDisplayError( + f"official SkillTree display did not reach ready route {route!r}" + ) probe = display_probe( - browser_command=browser_command, timeout=min(remaining, 30.0), activate=True, + browser_command=browser_command, + timeout=min(remaining, 30.0), + activate=True, ) if _origin(str(probe.get("url", ""))) != _origin(dashboard_url): - raise BrowserDisplayError("browser navigated away from the requested dashboard origin") + raise BrowserDisplayError( + "browser navigated away from the requested dashboard origin" + ) if probe.get("state") == "fallback": - raise BrowserDisplayError("official SkillTree display is in fallback, not mounted") + raise BrowserDisplayError( + "official SkillTree display is in fallback, not mounted" + ) if probe.get("state") == "ready" and probe.get("route") == route: if ( probe.get("projectId") != project_id or _origin(str(probe.get("frameUrl", ""))) != _origin(skilltree_url) - or (dashboard_instance_id is not None and probe.get("dashboardInstanceId") != dashboard_instance_id) - or (export_digest is not None and probe.get("exportContentDigest") != export_digest) + or ( + dashboard_instance_id is not None + and probe.get("dashboardInstanceId") != dashboard_instance_id + ) + or ( + export_digest is not None + and probe.get("exportContentDigest") != export_digest + ) ): - raise BrowserDisplayError("official frame identity or export association is invalid") + raise BrowserDisplayError( + "official frame identity or export association is invalid" + ) if probe.get("busy") is False and probe.get("frameVisible") is True: return probe time.sleep(min(0.25, max(0.0, deadline - time.monotonic()))) @@ -109,7 +146,11 @@ def wait_for_official_display( def _view_contracts(status: Mapping[str, Any]) -> tuple[str, str, list[dict[str, str]]]: config = status.get("config", {}) project = status.get("project", {}) - if not isinstance(config, Mapping) or not isinstance(project, Mapping) or not project.get("available"): + if ( + not isinstance(config, Mapping) + or not isinstance(project, Mapping) + or not project.get("available") + ): raise BrowserDisplayError("dashboard has no available exported project") identity = config.get("project", {}) if not isinstance(identity, Mapping) or not isinstance(identity.get("id"), str): @@ -118,7 +159,11 @@ def _view_contracts(status: Mapping[str, Any]) -> tuple[str, str, list[dict[str, skilltree = config.get("skilltree", {}) if not isinstance(skilltree, Mapping) or not isinstance(skilltree.get("url"), str): raise BrowserDisplayError("SkillTree target is unavailable") - quizzes = {quiz["quizId"]: quiz for quiz in project.get("quizzes", []) if isinstance(quiz, Mapping) and isinstance(quiz.get("quizId"), str)} + quizzes = { + quiz["quizId"]: quiz + for quiz in project.get("quizzes", []) + if isinstance(quiz, Mapping) and isinstance(quiz.get("quizId"), str) + } for subject in project.get("subjects", []): if not isinstance(subject, Mapping): continue @@ -127,19 +172,36 @@ def _view_contracts(status: Mapping[str, Any]) -> tuple[str, str, list[dict[str, continue quiz = quizzes[skill["quizId"]] views = [ - {"kind": "project", "route": "/", "visibleText": str(identity.get("name", ""))}, - {"kind": "subject", "route": f"/subjects/{subject['subjectId']}", "visibleText": str(subject.get("name", ""))}, - {"kind": "quiz", "route": f"/subjects/{subject['subjectId']}/skills/{skill['skillId']}/quizzes/{quiz['quizId']}", "visibleText": str(quiz.get("name", ""))}, + { + "kind": "project", + "route": "/", + "visibleText": str(identity.get("name", "")), + }, + { + "kind": "subject", + "route": f"/subjects/{subject['subjectId']}", + "visibleText": str(subject.get("name", "")), + }, + { + "kind": "quiz", + "route": f"/subjects/{subject['subjectId']}/skills/{skill['skillId']}/quizzes/{quiz['quizId']}", + "visibleText": str(quiz.get("name", "")), + }, ] if any(not view["visibleText"].strip() for view in views): raise BrowserDisplayError("official view contract lacks a display name") return project_id, skilltree["url"], views - raise BrowserDisplayError("export has no quiz-gated skill for the required official quiz proof") + raise BrowserDisplayError( + "export has no quiz-gated skill for the required official quiz proof" + ) def validate_dashboard_browser( - dashboard_url: str, *, browser_command: str = "chrome-devtools-axi", - timeout: float = 45.0, settle_seconds: float = 5.0, + dashboard_url: str, + *, + browser_command: str = "chrome-devtools-axi", + timeout: float = 45.0, + settle_seconds: float = 5.0, ) -> dict[str, Any]: if not math.isfinite(timeout) or timeout <= 0: raise BrowserDisplayError("timeout must be finite and positive") @@ -154,37 +216,68 @@ def validate_dashboard_browser( project_id, skilltree_url, views = _view_contracts(status) instance_id = status.get("instanceId") digest = status.get("project", {}).get("exportContentDigest") - if not isinstance(instance_id, str) or not instance_id or not isinstance(digest, str) or len(digest) != 64: + if ( + not isinstance(instance_id, str) + or not instance_id + or not isinstance(digest, str) + or len(digest) != 64 + ): raise BrowserDisplayError("dashboard instance/export identity is missing") allowed_count = 0 results = [] for view in views: query = dict(parse_qsl(parsed.query)) query["skillsClientDisplayPath"] = view["route"] - url = urlunsplit((parsed.scheme, parsed.netloc, parsed.path or "/", urlencode(query), "")) + url = urlunsplit( + (parsed.scheme, parsed.netloc, parsed.path or "/", urlencode(query), "") + ) browser_output(["open", url], browser_command=browser_command, timeout=60.0) probe = wait_for_official_display( - browser_command=browser_command, dashboard_url=dashboard_url, - skilltree_url=skilltree_url, project_id=project_id, route=view["route"], - timeout=timeout, dashboard_instance_id=instance_id, export_digest=digest, + browser_command=browser_command, + dashboard_url=dashboard_url, + skilltree_url=skilltree_url, + project_id=project_id, + route=view["route"], + timeout=timeout, + dashboard_instance_id=instance_id, + export_digest=digest, ) if settle_seconds: time.sleep(settle_seconds) - snapshot = browser_output(["snapshot"], browser_command=browser_command, timeout=30.0) + snapshot = browser_output( + ["snapshot"], browser_command=browser_command, timeout=30.0 + ) visible = " ".join(html.unescape(snapshot).split()).casefold() expected = " ".join(html.unescape(view["visibleText"]).split()).casefold() if expected not in visible: - raise BrowserDisplayError(f"official {view['kind']} view lacks its expected visible content") - console = browser_output(["console", "--type", "error"], browser_command=browser_command, timeout=30.0) + raise BrowserDisplayError( + f"official {view['kind']} view lacks its expected visible content" + ) + console = browser_output( + ["console", "--type", "error"], + browser_command=browser_command, + timeout=30.0, + ) errors = [line.strip() for line in console.splitlines() if "[error]" in line] if unexpected_errors(errors): raise BrowserDisplayError("dashboard has unexpected browser console errors") allowed_count += len(errors) - results.append({"kind": view["kind"], "route": probe["route"], "state": "ready", "visibleContentChecked": True}) + results.append( + { + "kind": view["kind"], + "route": probe["route"], + "state": "ready", + "visibleContentChecked": True, + } + ) return { - "ok": True, "checked": True, "projectId": project_id, - "dashboardInstanceId": instance_id, "exportContentDigest": digest, - "views": results, "allowedConsoleErrors": allowed_count, + "ok": True, + "checked": True, + "projectId": project_id, + "dashboardInstanceId": instance_id, + "exportContentDigest": digest, + "views": results, + "allowedConsoleErrors": allowed_count, "evidenceKind": "real official SDK route, visible frame and browser accessibility content", } @@ -198,11 +291,17 @@ def _origin(url: str) -> tuple[str, str, int]: def browser_receipt_complete(payload: object) -> bool: """Require all actual official view observations; an empty console is insufficient.""" - if not isinstance(payload, Mapping) or payload.get("ok") is not True or payload.get("checked") is not True: + if ( + not isinstance(payload, Mapping) + or payload.get("ok") is not True + or payload.get("checked") is not True + ): return False if ( - not isinstance(payload.get("projectId"), str) or not payload["projectId"] - or not isinstance(payload.get("dashboardInstanceId"), str) or not payload["dashboardInstanceId"] + not isinstance(payload.get("projectId"), str) + or not payload["projectId"] + or not isinstance(payload.get("dashboardInstanceId"), str) + or not payload["dashboardInstanceId"] or not isinstance(payload.get("exportContentDigest"), str) or len(payload["exportContentDigest"]) != 64 ): @@ -216,9 +315,12 @@ def browser_receipt_complete(payload: object) -> bool: return False kind, route = view.get("kind"), view.get("route") if ( - not isinstance(kind, str) or kind not in {"project", "subject", "quiz"} - or kind in kinds or not isinstance(route, str) - or view.get("state") != "ready" or view.get("visibleContentChecked") is not True + not isinstance(kind, str) + or kind not in {"project", "subject", "quiz"} + or kind in kinds + or not isinstance(route, str) + or view.get("state") != "ready" + or view.get("visibleContentChecked") is not True ): return False if kind == "project" and route != "/": diff --git a/src/dashboard/live_validation.py b/src/dashboard/live_validation.py index a866c09..d73fd70 100644 --- a/src/dashboard/live_validation.py +++ b/src/dashboard/live_validation.py @@ -17,8 +17,8 @@ ) from src.net_guard import NonHttpUrlError, require_http_url from src.project_identity import SKILLTREE_PROJECT_ID, SKILLTREE_PROJECT_NAME -from src.skilltree.summary import SkillTreeSummaryError, summarize_export_payload from src.skilltree.content import compare_live_content, export_content_digest +from src.skilltree.summary import SkillTreeSummaryError, summarize_export_payload class LiveSkillTreeDashboardError(RuntimeError): @@ -77,7 +77,9 @@ def _load_exported_project(project_path: Path) -> dict[str, Any]: def _count_exported_project(project_path: Path) -> dict[str, int]: - return summarize_export_payload(_load_exported_project(project_path)).project_counts() + return summarize_export_payload( + _load_exported_project(project_path) + ).project_counts() def _require(condition: bool, message: str) -> None: @@ -318,7 +320,9 @@ def validate_live_skilltree_dashboard( f"dashboard counts {counts!r} != export {exported_counts!r}", ) public_project = project.get("project", {}) - _require(isinstance(public_project, dict), "dashboard project identity is not an object") + _require( + isinstance(public_project, dict), "dashboard project identity is not an object" + ) project_id = public_project.get("id") project_name = public_project.get("name") _require( @@ -454,7 +458,8 @@ def validate_live_skilltree_dashboard( or ( live_content.get("checked") is True and live_content.get("complete") is True - and project.get("exportContentDigest") == export_content_digest(exported_project) + and project.get("exportContentDigest") + == export_content_digest(exported_project) ), "live SkillTree content or dashboard export digest does not match the current export", ) diff --git a/src/dashboard/payload.py b/src/dashboard/payload.py index 3ec816a..6ca76af 100644 --- a/src/dashboard/payload.py +++ b/src/dashboard/payload.py @@ -16,6 +16,7 @@ from src.dashboard.config import DashboardConfig, DashboardPayload from src.dashboard.graph_analysis import analyze_project_graph from src.dashboard.status import check_skilltree_status +from src.skilltree.content import compare_live_content from src.skilltree.deployment_matrix import skilltree_deployment_matrix from src.skilltree.local_auth import ( DEFAULT_LOCAL_STATE_DIR, @@ -24,7 +25,6 @@ load_local_state, ) from src.skilltree.local_auth_state import _require_same_skilltree_origin -from src.skilltree.content import compare_live_content DEFERRED_LIVE_PARITY_REASON = ( "live dependency parity was deferred for first paint; " @@ -142,7 +142,8 @@ def live_skilltree_summary( if state is None: raise LocalSkillTreeAuthError("local SkillTree auth state is missing") _require_same_skilltree_origin( - base_url=config.skilltree_url, state_url=state.skilltree_url, + base_url=config.skilltree_url, + state_url=state.skilltree_url, ) if state.project_id != config.project_id: raise LocalSkillTreeAuthError( diff --git a/src/dashboard/server.py b/src/dashboard/server.py index c18ce61..2c8125d 100644 --- a/src/dashboard/server.py +++ b/src/dashboard/server.py @@ -41,8 +41,11 @@ def dashboard_recommendations(config: DashboardConfig) -> dict[str, Any]: ) if project is None: return { - "available": False, "error": error, - "recommendations": [], "personalized": False, "signalSource": "none", + "available": False, + "error": error, + "recommendations": [], + "personalized": False, + "signalSource": "none", } return recommend_next_skills(project) diff --git a/src/figures/plots/registry_io.py b/src/figures/plots/registry_io.py index 6f02b62..f11f47c 100644 --- a/src/figures/plots/registry_io.py +++ b/src/figures/plots/registry_io.py @@ -99,11 +99,13 @@ def render_all( include_scholarship = scholarship_audit is not None include_claim_support = claim_support_audit is not None include_roadmap = roadmap is not None - planned_count = len(planned_figure_registry_entries( - include_scholarship=include_scholarship, - include_claim_support=include_claim_support, - include_roadmap=include_roadmap, - )) + planned_count = len( + planned_figure_registry_entries( + include_scholarship=include_scholarship, + include_claim_support=include_claim_support, + include_roadmap=include_roadmap, + ) + ) figures = [ plot_curriculum_dag(graph, out_dir / "curriculum_dag.png"), plot_dependency_strata_flow(graph, out_dir / "dependency_strata_flow.png"), diff --git a/src/integration/full_app_stack.py b/src/integration/full_app_stack.py index 51b5b42..01ae40f 100644 --- a/src/integration/full_app_stack.py +++ b/src/integration/full_app_stack.py @@ -460,7 +460,9 @@ def wait_for_dashboard( try: health = _fetch_json(health_url, timeout=_REQUEST_TIMEOUT_SECONDS) if instance_id is not None and health.get("instanceId") != instance_id: - raise FullAppStackError("health endpoint belongs to another dashboard instance") + raise FullAppStackError( + "health endpoint belongs to another dashboard instance" + ) return health except ( FullAppStackError, @@ -498,9 +500,13 @@ def proof_summary( try: candidate = json.loads(browser_console_stdout) except json.JSONDecodeError as exc: - raise FullAppStackError("browser validator did not emit a structured proof") from exc + raise FullAppStackError( + "browser validator did not emit a structured proof" + ) from exc if not browser_receipt_complete(candidate): - raise FullAppStackError("browser validator did not prove all official learner views") + raise FullAppStackError( + "browser validator did not prove all official learner views" + ) browser = candidate content = live_validation.get("liveContent", {}) dependencies = live_validation.get("liveDependencies", {}) @@ -511,21 +517,29 @@ def proof_summary( ok = ( live_validation.get("ok") is True and dashboard_health.get("ok") is True - and isinstance(instance_id, str) and bool(instance_id) + and isinstance(instance_id, str) + and bool(instance_id) and live_validation.get("dashboardInstanceId") == instance_id - and isinstance(digest, str) and len(digest) == 64 - and content.get("checked") is True and content.get("complete") is True + and isinstance(digest, str) + and len(digest) == 64 + and content.get("checked") is True + and content.get("complete") is True and content.get("exportContentDigest") == digest ) if not app_ready: - ok = ok and dependencies.get("checked") is True and dependencies.get("complete") is True + ok = ( + ok + and dependencies.get("checked") is True + and dependencies.get("complete") is True + ) if raw_skilltree_login is not None: ok = ok and raw_skilltree_login.get("ok") is True if dependency_reconciliation is not None: ok = ok and dependency_reconciliation.get("complete") is True if browser_console_stdout is not None: ok = ( - ok and browser.get("dashboardInstanceId") == instance_id + ok + and browser.get("dashboardInstanceId") == instance_id and browser.get("exportContentDigest") == digest and browser.get("projectId") == content.get("projectId") ) @@ -549,12 +563,17 @@ def proof_summary( "dependencyReconciliation": dependency_reconciliation or {"checked": False}, "mode": "app-ready" if app_ready else "full-parity", "browserConsole": browser, - "rawSkillTreeLogin": raw_skilltree_login if raw_skilltree_login is not None + "rawSkillTreeLogin": raw_skilltree_login + if raw_skilltree_login is not None else {"checked": False, "reason": "skipped by --skip-browser"}, "keptRunning": kept_running, "dashboardPid": dashboard_pid if kept_running else None, - "dashboardPidPath": str(dashboard_pid_path) if kept_running and dashboard_pid_path else None, - "dashboardLogPath": str(dashboard_log_path) if kept_running and dashboard_log_path else None, + "dashboardPidPath": str(dashboard_pid_path) + if kept_running and dashboard_pid_path + else None, + "dashboardLogPath": str(dashboard_log_path) + if kept_running and dashboard_log_path + else None, } @@ -754,19 +773,26 @@ def validate_full_app_stack(args: argparse.Namespace) -> dict[str, Any]: return proof finally: finalize_dashboard_process( - dashboard, state_dir=state_dir, keep_running=args.keep_running, + dashboard, + state_dir=state_dir, + keep_running=args.keep_running, successful=proof_written, ) def finalize_dashboard_process( - process: subprocess.Popen[str], *, state_dir: str, - keep_running: bool, successful: bool, + process: subprocess.Popen[str], + *, + state_dir: str, + keep_running: bool, + successful: bool, ) -> None: """Keep an owned process only after successful proof publication.""" if successful and keep_running: if process.poll() is not None: - raise FullAppStackError("owned dashboard exited before keep-running finalization") + raise FullAppStackError( + "owned dashboard exited before keep-running finalization" + ) try: record_dashboard_process(state_dir=state_dir, pid=process.pid) except BaseException: diff --git a/src/kernels/active_inference.py b/src/kernels/active_inference.py index eca1754..03099e1 100644 --- a/src/kernels/active_inference.py +++ b/src/kernels/active_inference.py @@ -96,7 +96,9 @@ def __post_init__(self) -> None: or np.any(self.B < 0.0) or not np.allclose(self.B.sum(axis=0), 1.0, atol=1e-9, rtol=0.0) ): - raise ValueError("B must contain finite non-negative column-stochastic state transitions") + raise ValueError( + "B must contain finite non-negative column-stochastic state transitions" + ) if normalize(self.C).shape != (self.A.shape[0],): raise ValueError("C length must match the number of observations") @@ -111,8 +113,10 @@ def predict(self, qs: ArrayLike, action: int) -> FloatArray: state_probs = _as_probability_vector(qs, name="qs") if state_probs.shape != self.D.shape: raise ValueError("qs length must match the model state dimension") - if not isinstance(action, (int, np.integer)) or isinstance(action, (bool, np.bool_)): - raise ValueError("action must be an integer index") + if not isinstance(action, (int, np.integer)) or isinstance( + action, (bool, np.bool_) + ): + raise TypeError("action must be an integer index") if action < 0 or action >= self.B.shape[2]: raise ValueError("action index out of range") return self.B[:, :, action] @ state_probs diff --git a/src/kernels/inference.py b/src/kernels/inference.py index 160d30d..ca03fd3 100644 --- a/src/kernels/inference.py +++ b/src/kernels/inference.py @@ -44,7 +44,9 @@ def _validate_model(prior: FloatArray, A: FloatArray, obs: int) -> None: if obs < 0 or obs >= A.shape[0]: raise ValueError("obs index out of range for A") if not np.all(np.isfinite(A)) or np.any(A < 0.0): - raise ValueError("A must be finite and non-negative (it is a likelihood P(o|s))") + raise ValueError( + "A must be finite and non-negative (it is a likelihood P(o|s))" + ) if not np.allclose(np.sum(A, axis=0), 1.0, atol=_PROB_TOL, rtol=0.0): raise ValueError("each column of A (P(o|s) for fixed state s) must sum to 1") @@ -105,7 +107,11 @@ def variational_free_energy( mask = q_probs > 0.0 if np.any(prior_probs[mask] == 0.0) or np.any(likelihood[obs, mask] == 0.0): return float("inf") - terms = np.log(q_probs[mask]) - np.log(prior_probs[mask]) - np.log(likelihood[obs, mask]) + terms = ( + np.log(q_probs[mask]) + - np.log(prior_probs[mask]) + - np.log(likelihood[obs, mask]) + ) return float(np.dot(q_probs[mask], terms)) @@ -130,7 +136,9 @@ def log_evidence(A: ArrayLike, obs: int, prior: ArrayLike) -> float: support = (likelihood[obs, :] > 0.0) & (prior_probs > 0.0) if not np.any(support): return float("-inf") - return float(logsumexp(np.log(likelihood[obs, support]) + np.log(prior_probs[support]))) + return float( + logsumexp(np.log(likelihood[obs, support]) + np.log(prior_probs[support])) + ) def belief_propagation_chain( diff --git a/src/manuscript/pdf_receipt.py b/src/manuscript/pdf_receipt.py index 3d9b633..d24ea4e 100644 --- a/src/manuscript/pdf_receipt.py +++ b/src/manuscript/pdf_receipt.py @@ -16,7 +16,9 @@ from src.artifacts.json_io import write_json_artifact -SECTION_EXCLUDE = frozenset({"config.yaml", "preamble.md", "README.md", "AGENTS.md", "SYNTAX.md"}) +SECTION_EXCLUDE = frozenset( + {"config.yaml", "preamble.md", "README.md", "AGENTS.md", "SYNTAX.md"} +) PDF_NAME = "Active_Skillference_combined.pdf" _PRIMARY_INPUTS = { "manuscript/config.yaml": "config", @@ -31,14 +33,20 @@ class PdfReceiptError(ValueError): def hydrated_section_files(directory: Path) -> list[Path]: - return [path for path in sorted(directory.glob("*.md")) if path.name not in SECTION_EXCLUDE] + return [ + path + for path in sorted(directory.glob("*.md")) + if path.name not in SECTION_EXCLUDE + ] def _relative(path: Path, root: Path) -> str: try: return path.resolve().relative_to(root.resolve()).as_posix() except ValueError as exc: - raise PdfReceiptError("PDF inputs/output must remain inside the project root") from exc + raise PdfReceiptError( + "PDF inputs/output must remain inside the project root" + ) from exc def _record(path: Path, root: Path, kind: str) -> dict[str, Any]: @@ -69,7 +77,9 @@ def latex_asset_paths(tex: Path, *, build_directory: Path, root: Path) -> list[P text = re.sub(r"(? list[P path = build_directory / path _relative(path, root) if not path.is_file(): - raise PdfReceiptError(f"LaTeX image input is missing: {_relative(path, root)}") + raise PdfReceiptError( + f"LaTeX image input is missing: {_relative(path, root)}" + ) paths.add(path.resolve()) return sorted(paths) def add_pdf_assets( - root: Path, manifest: Mapping[str, Any], assets: Iterable[Path], + root: Path, + manifest: Mapping[str, Any], + assets: Iterable[Path], ) -> dict[str, Any]: validate_pdf_inputs(root, manifest) rows = {row["path"]: dict(row) for row in manifest["files"]} @@ -103,16 +117,23 @@ def validate_pdf_inputs(root: Path, manifest: Mapping[str, Any]) -> None: sections, rows = manifest.get("sections"), manifest.get("files") assets = manifest.get("assets") if ( - not isinstance(sections, list) or not sections + not isinstance(sections, list) + or not sections or any(not isinstance(path, str) for path in sections) or len(set(sections)) != len(sections) - or not isinstance(rows, list) or not rows + or not isinstance(rows, list) + or not rows or not isinstance(assets, list) or any(not isinstance(path, str) for path in assets) or len(set(assets)) != len(assets) ): - raise PdfReceiptError("PDF input manifest is missing its complete section/file set") - expected_sections = [_relative(path, root) for path in hydrated_section_files(root / "output" / "manuscript")] + raise PdfReceiptError( + "PDF input manifest is missing its complete section/file set" + ) + expected_sections = [ + _relative(path, root) + for path in hydrated_section_files(root / "output" / "manuscript") + ] if sections != expected_sections: raise PdfReceiptError("hydrated PDF section set/order changed") by_path: dict[str, Mapping[str, Any]] = {} @@ -121,42 +142,75 @@ def validate_pdf_inputs(root: Path, manifest: Mapping[str, Any]) -> None: raise PdfReceiptError("PDF input manifest contains a malformed file row") name = row["path"] path = Path(name) - if path.is_absolute() or ".." in path.parts or _relative(root / path, root) != name or name in by_path: + if ( + path.is_absolute() + or ".." in path.parts + or _relative(root / path, root) != name + or name in by_path + ): raise PdfReceiptError("PDF input manifest has an unsafe or duplicate path") if ( - not isinstance(row.get("sha256"), str) or not re.fullmatch(r"[0-9a-f]{64}", row["sha256"]) - or type(row.get("size_bytes")) is not int or row["size_bytes"] < 0 - or row.get("kind") not in ("config", "preamble", "bibliography", "section", "asset") + not isinstance(row.get("sha256"), str) + or not re.fullmatch(r"[0-9a-f]{64}", row["sha256"]) + or type(row.get("size_bytes")) is not int + or row["size_bytes"] < 0 + or row.get("kind") + not in ("config", "preamble", "bibliography", "section", "asset") ): raise PdfReceiptError(f"PDF input digest row is malformed: {name}") current = _record(root / path, root, row["kind"]) - if current["sha256"] != row["sha256"] or current["size_bytes"] != row["size_bytes"]: + if ( + current["sha256"] != row["sha256"] + or current["size_bytes"] != row["size_bytes"] + ): raise PdfReceiptError(f"PDF input changed: {name}") by_path[name] = row - if any(name not in by_path or by_path[name]["kind"] != kind for name, kind in _PRIMARY_INPUTS.items()): - raise PdfReceiptError("PDF input manifest is missing configuration/preamble/bibliography") - if {name for name, row in by_path.items() if row["kind"] == "section"} != set(sections): - raise PdfReceiptError("PDF input manifest does not cover the complete hydrated section set") + if any( + name not in by_path or by_path[name]["kind"] != kind + for name, kind in _PRIMARY_INPUTS.items() + ): + raise PdfReceiptError( + "PDF input manifest is missing configuration/preamble/bibliography" + ) + if {name for name, row in by_path.items() if row["kind"] == "section"} != set( + sections + ): + raise PdfReceiptError( + "PDF input manifest does not cover the complete hydrated section set" + ) if {name for name, row in by_path.items() if row["kind"] == "asset"} != set(assets): - raise PdfReceiptError("PDF input manifest does not cover its complete image set") + raise PdfReceiptError( + "PDF input manifest does not cover its complete image set" + ) def write_pdf_receipt( - target: Path, *, root: Path, source_revision: str, inputs: Mapping[str, Any], + target: Path, + *, + root: Path, + source_revision: str, + inputs: Mapping[str, Any], ) -> Path: validate_pdf_inputs(root, inputs) if not re.fullmatch(r"[0-9a-f]{40}", source_revision): - raise PdfReceiptError("PDF source revision must be the observed full Git commit") + raise PdfReceiptError( + "PDF source revision must be the observed full Git commit" + ) output = _record(target, root, "pdf") if output["size_bytes"] <= 0: raise PdfReceiptError("finalized PDF is empty") receipt = target.with_suffix(".receipt.json") - return write_json_artifact(receipt, { - "schema": "active-skillference-pdf/v2", - "output_path": output["path"], "size_bytes": output["size_bytes"], - "sha256": output["sha256"], "source_revision": source_revision, - "inputManifest": dict(inputs), - }) + return write_json_artifact( + receipt, + { + "schema": "active-skillference-pdf/v2", + "output_path": output["path"], + "size_bytes": output["size_bytes"], + "sha256": output["sha256"], + "source_revision": source_revision, + "inputManifest": dict(inputs), + }, + ) def pdf_receipt_issues(root: Path, target: Path | None = None) -> list[str]: @@ -164,8 +218,13 @@ def pdf_receipt_issues(root: Path, target: Path | None = None) -> list[str]: receipt = target.with_suffix(".receipt.json") try: payload = json.loads(receipt.read_text(encoding="utf-8")) - if not isinstance(payload, dict) or payload.get("schema") != "active-skillference-pdf/v2": - raise PdfReceiptError("missing/legacy PDF input provenance; a genuine re-render is required") + if ( + not isinstance(payload, dict) + or payload.get("schema") != "active-skillference-pdf/v2" + ): + raise PdfReceiptError( + "missing/legacy PDF input provenance; a genuine re-render is required" + ) manifest = payload.get("inputManifest") if not isinstance(manifest, Mapping): raise PdfReceiptError("PDF receipt has no complete input manifest") @@ -176,7 +235,9 @@ def pdf_receipt_issues(root: Path, target: Path | None = None) -> list[str]: or payload.get("sha256") != output["sha256"] or payload.get("size_bytes") != output["size_bytes"] ): - raise PdfReceiptError("finalized PDF bytes do not match the recorded render") + raise PdfReceiptError( + "finalized PDF bytes do not match the recorded render" + ) revision = payload.get("source_revision") if not isinstance(revision, str) or not re.fullmatch(r"[0-9a-f]{40}", revision): raise PdfReceiptError("PDF receipt has no observed source revision") diff --git a/src/provenance/binding.py b/src/provenance/binding.py index a08b10e..f3aa201 100644 --- a/src/provenance/binding.py +++ b/src/provenance/binding.py @@ -65,8 +65,6 @@ class UnbackedNumber: context: str - - def render(text: str, registry: ClaimRegistry) -> str: """Replace every ``{{claim:NAME}}`` with its rendered value. diff --git a/src/publication/zenodo.py b/src/publication/zenodo.py index 9a6e673..26e04c8 100644 --- a/src/publication/zenodo.py +++ b/src/publication/zenodo.py @@ -163,12 +163,15 @@ def upload_deposit_file(*, token: str, deposition_id: int, path: Path) -> str: links = deposition.get("links", {}) if isinstance(deposition, dict) else {} bucket = links.get("bucket") if isinstance(links, dict) else None if not isinstance(bucket, str) or not bucket: - raise ZenodoError(f"deposition {deposition_id} exposes no upload bucket; it may already be published") + raise ZenodoError( + f"deposition {deposition_id} exposes no upload bucket; it may already be published" + ) try: with path.open("rb") as stream: request = urllib.request.Request( f"{bucket}/{urllib.parse.quote(path.name)}", - data=stream, method="PUT", + data=stream, + method="PUT", headers={ "Authorization": f"Bearer {token}", "Content-Type": "application/octet-stream", @@ -182,7 +185,11 @@ def upload_deposit_file(*, token: str, deposition_id: int, path: Path) -> str: raise ZenodoError(f"Zenodo file upload failed ({exc.code}): {detail}") from exc except (OSError, ValueError) as exc: raise ZenodoError(f"Zenodo file upload failed: {exc}") from exc - if not isinstance(payload, dict) or not isinstance(payload.get("key"), str) or not payload["key"].strip(): + if ( + not isinstance(payload, dict) + or not isinstance(payload.get("key"), str) + or not payload["key"].strip() + ): raise ZenodoError("Zenodo upload returned no confirmed filename") return payload["key"] diff --git a/src/release_readiness.py b/src/release_readiness.py index 77ffe91..7950067 100644 --- a/src/release_readiness.py +++ b/src/release_readiness.py @@ -116,11 +116,14 @@ def build_release_readiness_snapshot(root: Path) -> ReleaseReadinessSnapshot: def probe_public_repository( - root: Path, repository_url: str = PUBLIC_REPOSITORY_URL, + root: Path, + repository_url: str = PUBLIC_REPOSITORY_URL, ) -> subprocess.CompletedProcess[str]: """Observe all advertised refs without fetching or mutating either repository.""" return _run_command( - ["git", "ls-remote", "--refs", repository_url], cwd=root, timeout=20, + ["git", "ls-remote", "--refs", repository_url], + cwd=root, + timeout=20, ) @@ -165,7 +168,9 @@ def render_release_readiness_inventory( elif probes.public_repository.startswith("reachable but empty"): public_observation = "The repository is reachable but advertises no refs." elif probes.public_repository.startswith("not refreshed"): - public_observation = "No fresh public-repository observation was made by this validation run." + public_observation = ( + "No fresh public-repository observation was made by this validation run." + ) else: public_observation = "The public-repository probe failed or was unavailable." public_release_line = ( diff --git a/src/scholarship/source_health_report.py b/src/scholarship/source_health_report.py index 9f74053..1609452 100644 --- a/src/scholarship/source_health_report.py +++ b/src/scholarship/source_health_report.py @@ -100,8 +100,6 @@ def _issue_rows( ] - - def _boundary_source_row(row: dict[str, Any]) -> dict[str, object]: return { "area": str(row.get("area", "")), diff --git a/src/screenshots/capture.py b/src/screenshots/capture.py index dbdabc0..1f5ae79 100644 --- a/src/screenshots/capture.py +++ b/src/screenshots/capture.py @@ -23,11 +23,13 @@ from typing import Any from src.artifacts.json_io import write_json_artifact -from src.net_guard import NonHttpUrlError, is_loopback_host, require_http_url from src.dashboard.browser_validation import ( - BrowserDisplayError, browser_receipt_complete, display_probe, wait_for_official_display, + BrowserDisplayError, + browser_receipt_complete, + display_probe, + wait_for_official_display, ) -from src.skilltree.local_auth_state import LocalSkillTreeAuthError, _skilltree_origin +from src.net_guard import NonHttpUrlError, is_loopback_host, require_http_url from src.screenshots.manifest import ( SCREENSHOT_PROVENANCE_FILENAME, VIEWPORT_HEIGHT, @@ -35,6 +37,7 @@ ScreenshotView, screenshot_views, ) +from src.skilltree.local_auth_state import LocalSkillTreeAuthError, _skilltree_origin DEFAULT_BROWSER_COMMAND = "chrome-devtools-axi" #: The browser automation server writes only inside this root on this platform. @@ -121,15 +124,26 @@ def _capture_url(url: str) -> str: parsed = urllib.parse.urlsplit(url) except ValueError as exc: raise ScreenshotCaptureError("capture URL is invalid") from exc - if parsed.username is not None or parsed.password is not None or parsed.query or parsed.fragment: - raise ScreenshotCaptureError("capture requires a credential-free dashboard base URL without query or fragment") + if ( + parsed.username is not None + or parsed.password is not None + or parsed.query + or parsed.fragment + ): + raise ScreenshotCaptureError( + "capture requires a credential-free dashboard base URL without query or fragment" + ) try: require_http_url(url, context="screenshot capture") port = parsed.port except (NonHttpUrlError, ValueError) as exc: - raise ScreenshotCaptureError("capture requires an absolute loopback HTTP(S) URL") from exc + raise ScreenshotCaptureError( + "capture requires an absolute loopback HTTP(S) URL" + ) from exc if not parsed.hostname or not is_loopback_host(parsed.hostname): - raise ScreenshotCaptureError("capture requires an absolute loopback HTTP(S) URL") + raise ScreenshotCaptureError( + "capture requires an absolute loopback HTTP(S) URL" + ) if port is not None and port <= 0: raise ScreenshotCaptureError("capture URL port is invalid") return url.rstrip("/") @@ -137,17 +151,26 @@ def _capture_url(url: str) -> str: def _capture_target(url: str) -> tuple[tuple[str, str, int], str]: normalized = _capture_url(url) - return _skilltree_origin(normalized), urllib.parse.urlsplit(normalized).path.rstrip("/") or "/" + return _skilltree_origin(normalized), urllib.parse.urlsplit(normalized).path.rstrip( + "/" + ) or "/" class _LoopbackCaptureRedirect(urllib.request.HTTPRedirectHandler): def redirect_request( - self, req: urllib.request.Request, fp: Any, code: int, msg: str, - headers: Any, newurl: str, + self, + req: urllib.request.Request, + fp: Any, + code: int, + msg: str, + headers: Any, + newurl: str, ) -> urllib.request.Request | None: _capture_url(newurl) if _skilltree_origin(req.full_url) != _skilltree_origin(newurl): - raise ScreenshotCaptureError("capture redirect changed the requested origin") + raise ScreenshotCaptureError( + "capture redirect changed the requested origin" + ) return super().redirect_request(req, fp, code, msg, headers, newurl) @@ -168,26 +191,45 @@ def _fetch_json(url: str, *, timeout: float) -> dict[str, Any]: def capture_context( - settings: CaptureSettings, *, browser_probe: dict[str, Any] | None = None, + settings: CaptureSettings, + *, + browser_probe: dict[str, Any] | None = None, ) -> dict[str, Any]: """Observe current status and associate only an explicit, matching successful proof.""" dashboard_url = _capture_url(settings.dashboard_url) status = _fetch_json(f"{dashboard_url}/api/status", timeout=_STATUS_TIMEOUT_SECONDS) - project, config, live = status.get("project"), status.get("config"), status.get("liveSkilltree") + project, config, live = ( + status.get("project"), + status.get("config"), + status.get("liveSkilltree"), + ) service = status.get("skilltree") - if not all(isinstance(value, dict) for value in (project, config, live, service)): - raise ScreenshotCaptureError("dashboard project/config/live state must contain objects") + if not ( + isinstance(project, dict) + and isinstance(config, dict) + and isinstance(live, dict) + and isinstance(service, dict) + ): + raise ScreenshotCaptureError( + "dashboard project/config/live state must contain objects" + ) project_counts = project.get("counts", {}) public_project, skilltree = config.get("project", {}), config.get("skilltree", {}) - if not all(isinstance(value, dict) for value in (project_counts, public_project, skilltree)): - raise ScreenshotCaptureError("dashboard project counts/identity/target must contain objects") + if not all( + isinstance(value, dict) for value in (project_counts, public_project, skilltree) + ): + raise ScreenshotCaptureError( + "dashboard project counts/identity/target must contain objects" + ) instance_id = status.get("instanceId") digest = project.get("exportContentDigest") project_id = public_project.get("id") skilltree_url = str(skilltree.get("url", "")) if browser_probe is not None: try: - browser_matches = _capture_target(str(browser_probe.get("url", ""))) == _capture_target(dashboard_url) + browser_matches = _capture_target( + str(browser_probe.get("url", "")) + ) == _capture_target(dashboard_url) except (ScreenshotCaptureError, LocalSkillTreeAuthError, ValueError): browser_matches = False if ( @@ -197,10 +239,16 @@ def capture_context( or browser_probe.get("projectId") != project_id or not browser_matches ): - raise ScreenshotCaptureError("browser capture state does not match current dashboard content") + raise ScreenshotCaptureError( + "browser capture state does not match current dashboard content" + ) association = _proof_association( - settings, instance_id=instance_id, digest=digest, project_id=project_id, - skilltree_url=skilltree_url, browser_probe=browser_probe, + settings, + instance_id=instance_id, + digest=digest, + project_id=project_id, + skilltree_url=skilltree_url, + browser_probe=browser_probe, ) content = live.get("content", {}) return { @@ -210,18 +258,32 @@ def capture_context( "projectId": project_id, "skilltreeUrl": skilltree_url, "skilltreeAvailable": bool(service.get("available")), - "viewport": {"width": settings.viewport_width, "height": settings.viewport_height}, - "exportedCounts": {key: project_counts.get(key, 0) for key in ("skills", "subjects", "dependencies", "quizzes")}, + "viewport": { + "width": settings.viewport_width, + "height": settings.viewport_height, + }, + "exportedCounts": { + key: project_counts.get(key, 0) + for key in ("skills", "subjects", "dependencies", "quizzes") + }, "liveDependencies": { - "checked": live.get("checked") is True, "matched": live.get("matchedDependencies"), - "expected": live.get("expectedDependencies"), "complete": live.get("complete") is True, + "checked": live.get("checked") is True, + "matched": live.get("matchedDependencies"), + "expected": live.get("expectedDependencies"), + "complete": live.get("complete") is True, }, "liveContent": { - key: content.get(key) for key in ("checked", "complete", "exportContentDigest") - } if isinstance(content, dict) else {"checked": False}, + key: content.get(key) + for key in ("checked", "complete", "exportContentDigest") + } + if isinstance(content, dict) + else {"checked": False}, "officialDisplay": { - key: browser_probe.get(key) for key in ("state", "route", "frameUrl", "frameVisible") - } if browser_probe is not None else {"state": "unobserved"}, + key: browser_probe.get(key) + for key in ("state", "route", "frameUrl", "frameVisible") + } + if browser_probe is not None + else {"state": "unobserved"}, "proofAssociation": association, "fullAppStackProofMode": association.get("mode", "not associated"), "evidenceKind": "captured browser recording of a live system; pixels are not reproducible byte-for-byte and are not a deterministic render", @@ -229,8 +291,13 @@ def capture_context( def _proof_association( - settings: CaptureSettings, *, instance_id: Any, digest: Any, project_id: Any, - skilltree_url: str, browser_probe: dict[str, Any] | None, + settings: CaptureSettings, + *, + instance_id: Any, + digest: Any, + project_id: Any, + skilltree_url: str, + browser_probe: dict[str, Any] | None, ) -> dict[str, Any]: if settings.proof_path is None: return {"checked": False, "reason": "no explicit proof path"} @@ -241,29 +308,42 @@ def _proof_association( raw = path.read_bytes() proof = json.loads(raw) except (OSError, ValueError) as exc: - return {"checked": False, "reason": f"proof is unreadable: {type(exc).__name__}"} + return { + "checked": False, + "reason": f"proof is unreadable: {type(exc).__name__}", + } if not isinstance(proof, dict) or proof.get("proofVersion") != 2: - return {"checked": False, "reason": "historical/unsupported proof schema is unassociated"} + return { + "checked": False, + "reason": "historical/unsupported proof schema is unassociated", + } content = proof.get("liveContent", {}) browser = proof.get("browserConsole", {}) try: - same_targets = ( - _capture_target(str(proof.get("dashboardUrl", ""))) == _capture_target(settings.dashboard_url) - and _skilltree_origin(str(proof.get("skilltreeUrl", ""))) == _skilltree_origin(skilltree_url) - ) + same_targets = _capture_target( + str(proof.get("dashboardUrl", "")) + ) == _capture_target(settings.dashboard_url) and _skilltree_origin( + str(proof.get("skilltreeUrl", "")) + ) == _skilltree_origin(skilltree_url) except (ScreenshotCaptureError, LocalSkillTreeAuthError, ValueError): same_targets = False complete = ( - proof.get("ok") is True and same_targets + proof.get("ok") is True + and same_targets and proof.get("mode") in ("app-ready", "full-parity") - and isinstance(instance_id, str) and bool(instance_id) - and isinstance(digest, str) and len(digest) == 64 + and isinstance(instance_id, str) + and bool(instance_id) + and isinstance(digest, str) + and len(digest) == 64 and proof.get("runId") == proof.get("dashboardInstanceId") == instance_id and proof.get("exportContentDigest") == digest and proof.get("projectId") == project_id - and isinstance(content, dict) and content.get("checked") is True - and content.get("complete") is True and content.get("exportContentDigest") == digest - and browser_probe is not None and browser_receipt_complete(browser) + and isinstance(content, dict) + and content.get("checked") is True + and content.get("complete") is True + and content.get("exportContentDigest") == digest + and browser_probe is not None + and browser_receipt_complete(browser) and browser.get("dashboardInstanceId") == instance_id and browser.get("exportContentDigest") == digest and browser.get("projectId") == project_id @@ -271,14 +351,22 @@ def _proof_association( if proof.get("mode") == "full-parity": dependencies = proof.get("liveDependencies", {}) complete = ( - complete and isinstance(dependencies, dict) - and dependencies.get("checked") is True and dependencies.get("complete") is True + complete + and isinstance(dependencies, dict) + and dependencies.get("checked") is True + and dependencies.get("complete") is True ) if not complete: - return {"checked": False, "reason": "proof does not match this successful current browser/content/instance state"} + return { + "checked": False, + "reason": "proof does not match this successful current browser/content/instance state", + } return { - "checked": True, "mode": proof["mode"], "runId": proof["runId"], - "proofSha256": hashlib.sha256(raw).hexdigest(), "proofPath": str(path), + "checked": True, + "mode": proof["mode"], + "runId": proof["runId"], + "proofSha256": hashlib.sha256(raw).hexdigest(), + "proofPath": str(path), } @@ -308,8 +396,19 @@ def _capture_one( after = display_probe(browser_command=settings.browser_command, timeout=30.0) except BrowserDisplayError as exc: raise ScreenshotCaptureError(str(exc)) from exc - if any(after.get(key) != probe.get(key) for key in ("state", "route", "url", "dashboardInstanceId", "exportContentDigest")): - raise ScreenshotCaptureError("browser state changed while the screenshot was being recorded") + if any( + after.get(key) != probe.get(key) + for key in ( + "state", + "route", + "url", + "dashboardInstanceId", + "exportContentDigest", + ) + ): + raise ScreenshotCaptureError( + "browser state changed while the screenshot was being recorded" + ) target = settings.output_dir / view.filename target.write_bytes(staged.read_bytes()) staged.unlink() @@ -328,7 +427,9 @@ def _capture_one( def capture_screenshots(settings: CaptureSettings) -> dict[str, Any]: """Capture every manifest view and write the registry plus provenance record.""" dashboard_url = _capture_url(settings.dashboard_url) - with _capture_opener().open(dashboard_url, timeout=_STATUS_TIMEOUT_SECONDS) as response: + with _capture_opener().open( + dashboard_url, timeout=_STATUS_TIMEOUT_SECONDS + ) as response: _capture_url(response.geturl()) restarted = ensure_browser_ready(browser_command=settings.browser_command) _run( @@ -340,18 +441,24 @@ def capture_screenshots(settings: CaptureSettings) -> dict[str, Any]: ] ) _run([settings.browser_command, "open", settings.dashboard_url]) - initial = _fetch_json(f"{dashboard_url}/api/status", timeout=_STATUS_TIMEOUT_SECONDS) + initial = _fetch_json( + f"{dashboard_url}/api/status", timeout=_STATUS_TIMEOUT_SECONDS + ) try: probe = wait_for_official_display( - browser_command=settings.browser_command, dashboard_url=dashboard_url, + browser_command=settings.browser_command, + dashboard_url=dashboard_url, skilltree_url=str(initial["config"]["skilltree"]["url"]), - project_id=str(initial["config"]["project"]["id"]), route="/", + project_id=str(initial["config"]["project"]["id"]), + route="/", dashboard_instance_id=str(initial["instanceId"]), export_digest=str(initial["project"]["exportContentDigest"]), timeout=_STATUS_TIMEOUT_SECONDS, ) except (BrowserDisplayError, KeyError, TypeError) as exc: - raise ScreenshotCaptureError("official display did not provide capture-ready identity") from exc + raise ScreenshotCaptureError( + "official display did not provide capture-ready identity" + ) from exc context = capture_context(settings, browser_probe=probe) settings.output_dir.mkdir(parents=True, exist_ok=True) diff --git a/src/skilltree/client.py b/src/skilltree/client.py index 876c908..b21b21b 100644 --- a/src/skilltree/client.py +++ b/src/skilltree/client.py @@ -29,6 +29,7 @@ subject_create_path, subject_payload, ) +from src.skilltree.content import missing_quiz_questions from src.skilltree.model import ( Quiz, QuizQuestion, @@ -36,7 +37,6 @@ SkillTreeSkill, SkillTreeSubject, ) -from src.skilltree.content import missing_quiz_questions from src.skilltree.transport import ( NoCrossOriginBearerRedirect, SkillTreeClientError, @@ -51,8 +51,12 @@ class SkillTreeClient: """Thin admin/report client for a single SkillTree project.""" def __init__( - self, base_url: str, project_id: str, token: str | None = None, - *, timeout: float = 30.0, + self, + base_url: str, + project_id: str, + token: str | None = None, + *, + timeout: float = 30.0, ) -> None: if not base_url.strip(): raise ValueError("base_url must be non-empty") diff --git a/src/skilltree/content.py b/src/skilltree/content.py index b2c5a62..c7d2e6a 100644 --- a/src/skilltree/content.py +++ b/src/skilltree/content.py @@ -52,7 +52,9 @@ def _rows(payload: Any, key: str, resource: str) -> list[Mapping[str, Any]]: payload = payload[key] elif "data" in payload: return _rows(payload["data"], key, resource) - if not isinstance(payload, list) or any(not isinstance(row, Mapping) for row in payload): + if not isinstance(payload, list) or any( + not isinstance(row, Mapping) for row in payload + ): raise SkillTreeContentError(f"{resource} returned no verifiable {key} list") return payload @@ -62,12 +64,16 @@ def _question_key(question: Mapping[str, Any]) -> str: question_type = question.get("questionType") answers = question.get("answers") if ( - not isinstance(text, str) or not text.strip() + not isinstance(text, str) + or not text.strip() or not isinstance(question_type, str) or question_type not in {"SingleChoice", "MultipleChoice", "TextInput"} - or not isinstance(answers, list) or not answers + or not isinstance(answers, list) + or not answers ): - raise SkillTreeContentError("quiz question response lacks verifiable semantic content") + raise SkillTreeContentError( + "quiz question response lacks verifiable semantic content" + ) normalized_answers = [] for answer in answers: if ( @@ -75,16 +81,20 @@ def _question_key(question: Mapping[str, Any]) -> str: or not isinstance(answer.get("answer"), str) or not isinstance(answer.get("isCorrect"), bool) ): - raise SkillTreeContentError("quiz answer response lacks verifiable correctness") + raise SkillTreeContentError( + "quiz answer response lacks verifiable correctness" + ) normalized_answers.append((answer["answer"], answer["isCorrect"])) return json.dumps( [text, question_type, sorted(normalized_answers)], - ensure_ascii=False, separators=(",", ":"), + ensure_ascii=False, + separators=(",", ":"), ) def missing_quiz_questions( - payload: Any, expected: Sequence[QuizQuestion], + payload: Any, + expected: Sequence[QuizQuestion], ) -> list[QuizQuestion]: """Return missing question occurrences; refuse extras or changed answer keys.""" current = Counter(_question_key(row) for row in _rows(payload, "questions", "quiz")) @@ -103,7 +113,9 @@ def missing_quiz_questions( return missing -def _index(rows: list[Mapping[str, Any]], field: str, resource: str) -> dict[str, Mapping[str, Any]]: +def _index( + rows: list[Mapping[str, Any]], field: str, resource: str +) -> dict[str, Mapping[str, Any]]: result = {} for row in rows: identifier = row.get(field) @@ -126,19 +138,29 @@ def integer(name: str) -> int: try: return int(value) except ValueError as exc: - raise SkillTreeContentError(f"quiz setting {name} is not an integer") from exc + raise SkillTreeContentError( + f"quiz setting {name} is not an integer" + ) from exc length = integer("quizLength") passing = integer("quizPassingReq") if length == -1: length = question_count - if length <= 0 or length > question_count or passing < -1 or passing > question_count: + if ( + length <= 0 + or length > question_count + or passing < -1 + or passing > question_count + ): raise SkillTreeContentError("quiz length/pass-count settings are unverifiable") return length, passing if passing > 0 else length def compare_live_content( - *, get: Callable[[str], Any], base_url: str, project: Mapping[str, Any], + *, + get: Callable[[str], Any], + base_url: str, + project: Mapping[str, Any], ) -> dict[str, Any]: """Read actual admin content and return a browser-safe comparison receipt. @@ -151,44 +173,75 @@ def compare_live_content( project_id = str(project["projectId"]) root = project_create_path(project_id) live_project = _object(get(root), "project") - if live_project.get("projectId") != project_id or not isinstance(live_project.get("name"), str): + if live_project.get("projectId") != project_id or not isinstance( + live_project.get("name"), str + ): raise SkillTreeContentError("live project identity/name is unverifiable") missing: list[str] = [] extra: list[str] = [] changed: list[str] = [] if live_project["name"] != project["name"]: changed.append(f"project:{project_id}") - live_subjects = _index(_rows(get(f"{root}/subjects"), "subjects", "project"), "subjectId", "project") - expected_subjects = {subject["subjectId"]: subject for subject in project["subjects"]} - missing.extend(f"subject:{identifier}" for identifier in expected_subjects.keys() - live_subjects.keys()) - extra.extend(f"subject:{identifier}" for identifier in live_subjects.keys() - expected_subjects.keys()) + live_subjects = _index( + _rows(get(f"{root}/subjects"), "subjects", "project"), "subjectId", "project" + ) + expected_subjects = { + subject["subjectId"]: subject for subject in project["subjects"] + } + missing.extend( + f"subject:{identifier}" + for identifier in expected_subjects.keys() - live_subjects.keys() + ) + extra.extend( + f"subject:{identifier}" + for identifier in live_subjects.keys() - expected_subjects.keys() + ) live_skill_count = 0 for subject_id, subject in expected_subjects.items(): if subject_id not in live_subjects: continue subject_path = subject_create_path(project_id, subject_id) live_subject = _object(get(subject_path), "subject") - fields = ("subjectId", "name", "description") + fields: tuple[str, ...] = ("subjectId", "name", "description") if any(field not in live_subject for field in fields): raise SkillTreeContentError(f"subject:{subject_id} content is unverifiable") if any(live_subject[field] != subject.get(field, "") for field in fields): changed.append(f"subject:{subject_id}") - live_skills = _index(_rows(get(f"{subject_path}/skills"), "skills", "subject"), "skillId", "subject") + live_skills = _index( + _rows(get(f"{subject_path}/skills"), "skills", "subject"), + "skillId", + "subject", + ) wanted_skills = {skill["skillId"]: skill for skill in subject["skills"]} live_skill_count += len(live_skills) - missing.extend(f"skill:{identifier}" for identifier in wanted_skills.keys() - live_skills.keys()) - extra.extend(f"skill:{identifier}" for identifier in live_skills.keys() - wanted_skills.keys()) + missing.extend( + f"skill:{identifier}" + for identifier in wanted_skills.keys() - live_skills.keys() + ) + extra.extend( + f"skill:{identifier}" + for identifier in live_skills.keys() - wanted_skills.keys() + ) for skill_id, skill in wanted_skills.items(): if skill_id not in live_skills: continue - live_skill = _object(get(skill_create_path(project_id, subject_id, skill_id)), "skill") - fields = ("skillId", "name", "description", "pointIncrement", "numPerformToCompletion", "helpUrl", "selfReportingType") + live_skill = _object( + get(skill_create_path(project_id, subject_id, skill_id)), "skill" + ) + fields = ( + "skillId", + "name", + "description", + "pointIncrement", + "numPerformToCompletion", + "helpUrl", + "selfReportingType", + ) if any(field not in live_skill for field in fields): raise SkillTreeContentError(f"skill:{skill_id} content is unverifiable") - if ( - any(live_skill[field] != skill.get(field) for field in fields) - or live_skill.get("quizId", "") != skill.get("quizId", "") - ): + if any( + live_skill[field] != skill.get(field) for field in fields + ) or live_skill.get("quizId", "") != skill.get("quizId", ""): changed.append(f"skill:{skill_id}") for quiz in project["quizzes"]: quiz_id = quiz["quizId"] @@ -207,18 +260,25 @@ def compare_live_content( question_count = len(quiz["questions"]) percent = quiz.get("passingPercent") if type(percent) is not int or not 1 <= percent <= 100: - raise SkillTreeContentError(f"quiz:{quiz_id} export pass percentage is invalid") + raise SkillTreeContentError( + f"quiz:{quiz_id} export pass percentage is invalid" + ) # Normalize defaults for the intended completed quiz. Missing questions # still make the receipt incomplete below and can be seeded without # misclassifying native default settings as an existing-content change. length, required = _quiz_completion_contract( - get(f"{quiz_definition_path(quiz_id)}/settings"), question_count, + get(f"{quiz_definition_path(quiz_id)}/settings"), + question_count, ) if (length, required) != ( - question_count, (question_count * percent + 99) // 100, + question_count, + (question_count * percent + 99) // 100, ): changed.append(f"quiz:{quiz_id}") - current = Counter(_question_key(row) for row in _rows(get(quiz_questions_path(quiz_id)), "questions", "quiz")) + current = Counter( + _question_key(row) + for row in _rows(get(quiz_questions_path(quiz_id)), "questions", "quiz") + ) wanted = Counter(_question_key(row) for row in quiz["questions"]) if current - wanted: changed.append(f"quiz-questions:{quiz_id}") diff --git a/tests/conftest.py b/tests/conftest.py index 4bd7b0b..e24f2d5 100644 --- a/tests/conftest.py +++ b/tests/conftest.py @@ -19,14 +19,17 @@ sys.path.insert(0, _path) - @pytest.fixture def skilltree_admin_content(): """Serve actual model content over loopback, including private answer keys.""" + def serve(httpserver, project, *, initially_missing=False, quiz_settings=None): - payload = project.to_export_dict() if hasattr(project, "to_export_dict") else project + payload = ( + project.to_export_dict() if hasattr(project, "to_export_dict") else project + ) root = f"/admin/projects/{payload['projectId']}" import json + from werkzeug.wrappers import Response root_requests = 0 @@ -35,76 +38,116 @@ def project_response(request): nonlocal root_requests root_requests += 1 missing = initially_missing and root_requests == 1 - body = {"error": "not found"} if missing else { - "projectId": payload["projectId"], "name": payload["name"], - "numSubjects": len(payload["subjects"]), - "numSkills": sum(len(subject["skills"]) for subject in payload["subjects"]), - } - return Response(json.dumps(body), status=404 if missing else 200, content_type="application/json") - - httpserver.expect_request(root, method="GET").respond_with_handler(project_response) + body = ( + {"error": "not found"} + if missing + else { + "projectId": payload["projectId"], + "name": payload["name"], + "numSubjects": len(payload["subjects"]), + "numSkills": sum( + len(subject["skills"]) for subject in payload["subjects"] + ), + } + ) + return Response( + json.dumps(body), + status=404 if missing else 200, + content_type="application/json", + ) + + httpserver.expect_request(root, method="GET").respond_with_handler( + project_response + ) httpserver.expect_request(f"{root}/subjects", method="GET").respond_with_json( payload["subjects"] ) for subject in payload["subjects"]: subject_path = f"{root}/subjects/{subject['subjectId']}" - httpserver.expect_request(subject_path, method="GET").respond_with_json(subject) - httpserver.expect_request(f"{subject_path}/skills", method="GET").respond_with_json( - subject["skills"] + httpserver.expect_request(subject_path, method="GET").respond_with_json( + subject ) + httpserver.expect_request( + f"{subject_path}/skills", method="GET" + ).respond_with_json(subject["skills"]) for skill in subject["skills"]: httpserver.expect_request( - f"{subject_path}/skills/{skill['skillId']}", method="GET", + f"{subject_path}/skills/{skill['skillId']}", + method="GET", ).respond_with_json(skill) for quiz in payload["quizzes"]: quiz_path = f"/admin/quiz-definitions/{quiz['quizId']}" - httpserver.expect_request(quiz_path, method="GET").respond_with_json({ - key: quiz[key] for key in ("quizId", "name", "type") - }) - httpserver.expect_request(f"{quiz_path}/settings", method="GET").respond_with_json( - (quiz_settings or {}).get(quiz["quizId"], []) + httpserver.expect_request(quiz_path, method="GET").respond_with_json( + {key: quiz[key] for key in ("quizId", "name", "type")} ) + httpserver.expect_request( + f"{quiz_path}/settings", method="GET" + ).respond_with_json((quiz_settings or {}).get(quiz["quizId"], [])) questions = [] if initially_missing else list(quiz["questions"]) def read_questions(_request, rows=questions): return Response( - json.dumps({"questions": rows}), content_type="application/json", + json.dumps({"questions": rows}), + content_type="application/json", ) def create_question(request, rows=questions): rows.append(request.get_json()) return Response("{}", content_type="application/json") - httpserver.expect_request(f"{quiz_path}/questions", method="GET").respond_with_handler( - read_questions - ) + httpserver.expect_request( + f"{quiz_path}/questions", method="GET" + ).respond_with_handler(read_questions) if initially_missing: httpserver.expect_request( - f"{quiz_path}/create-question", method="POST", + f"{quiz_path}/create-question", + method="POST", ).respond_with_handler(create_question) + return serve @pytest.fixture def content_project(): - from src.skilltree.model import Quiz, QuizQuestion, SkillTreeProject, SkillTreeSkill, SkillTreeSubject + from src.skilltree.model import ( + Quiz, + QuizQuestion, + SkillTreeProject, + SkillTreeSkill, + SkillTreeSubject, + ) quiz = Quiz( - quiz_id="q", name="Quiz", + quiz_id="q", + name="Quiz", questions=[ - QuizQuestion(question="First?", answers=[{"answer": "one", "isCorrect": True}]), - QuizQuestion(question="Second?", answers=[{"answer": "two", "isCorrect": True}]), + QuizQuestion( + question="First?", answers=[{"answer": "one", "isCorrect": True}] + ), + QuizQuestion( + question="Second?", answers=[{"answer": "two", "isCorrect": True}] + ), ], ) return SkillTreeProject( - project_id="p", name="Project", - subjects=[SkillTreeSubject( - subject_id="s", name="Subject", description="Subject content", - skills=[SkillTreeSkill( - skill_id="k", name="Skill", description="Lesson content", - self_reporting_type="Quiz", quiz_id="q", - )], - )], + project_id="p", + name="Project", + subjects=[ + SkillTreeSubject( + subject_id="s", + name="Subject", + description="Subject content", + skills=[ + SkillTreeSkill( + skill_id="k", + name="Skill", + description="Lesson content", + self_reporting_type="Quiz", + quiz_id="q", + ) + ], + ) + ], quizzes=[quiz], ) @@ -112,6 +155,7 @@ def content_project(): @pytest.fixture def pdf_bytes(): """Build a structurally valid one-page PDF, not a copied historical output.""" + def make(text="Fixture image"): escaped = text.replace("\\", "\\\\").replace("(", "\\(").replace(")", "\\)") stream = f"BT /F1 18 Tf 20 120 Td ({escaped}) Tj ET\n".encode("ascii") @@ -120,7 +164,11 @@ def make(text="Fixture image"): b"<< /Type /Pages /Kids [3 0 R] /Count 1 >>", b"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 300 200] /Resources << /Font << /F1 4 0 R >> >> /Contents 5 0 R >>", b"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>", - b"<< /Length " + str(len(stream)).encode("ascii") + b" >>\nstream\n" + stream + b"endstream", + b"<< /Length " + + str(len(stream)).encode("ascii") + + b" >>\nstream\n" + + stream + + b"endstream", ] content = bytearray(b"%PDF-1.4\n%\xe2\xe3\xcf\xd3\n") offsets = [] @@ -128,13 +176,18 @@ def make(text="Fixture image"): offsets.append(len(content)) content.extend(f"{index} 0 obj\n".encode("ascii") + body + b"\nendobj\n") xref = len(content) - content.extend(f"xref\n0 {len(objects) + 1}\n0000000000 65535 f \n".encode("ascii")) + content.extend( + f"xref\n0 {len(objects) + 1}\n0000000000 65535 f \n".encode("ascii") + ) for offset in offsets: content.extend(f"{offset:010d} 00000 n \n".encode("ascii")) content.extend( - f"trailer\n<< /Root 1 0 R /Size {len(objects) + 1} >>\nstartxref\n{xref}\n%%EOF\n".encode("ascii") + f"trailer\n<< /Root 1 0 R /Size {len(objects) + 1} >>\nstartxref\n{xref}\n%%EOF\n".encode( + "ascii" + ) ) return bytes(content) + return make @@ -154,7 +207,9 @@ def pdf_project(tmp_path, pdf_bytes): assets_dir = hydrated / "assets" assets_dir.mkdir() assets = [assets_dir / name for name in ("cover.pdf", "atlas.pdf", "body.pdf")] - for path, label in zip(assets, ("CoverGraphic", "AtlasGraphic", "BodyGraphic"), strict=True): + for path, label in zip( + assets, ("CoverGraphic", "AtlasGraphic", "BodyGraphic"), strict=True + ): path.write_bytes(pdf_bytes(label)) (manuscript / "config.yaml").write_text( "paper:\n title: FixtureTitlePage\n" @@ -164,7 +219,9 @@ def pdf_project(tmp_path, pdf_bytes): " title: FixtureAtlasPage\n caption: FixtureAtlasCaption\n", encoding="utf-8", ) - (manuscript / "preamble.md").write_text("```latex\n\\usepackage{graphicx}\n```\n", encoding="utf-8") + (manuscript / "preamble.md").write_text( + "```latex\n\\usepackage{graphicx}\n```\n", encoding="utf-8" + ) (manuscript / "references.bib").write_text( "@book{fixture,\n title={Fixture Publication},\n author={Example, Test},\n" " year={2020},\n publisher={Fixture Press}\n}\n", @@ -178,24 +235,47 @@ def pdf_project(tmp_path, pdf_bytes): target = root / "output" / "pdf" / PDF_NAME target.parent.mkdir() target.write_bytes(pdf_bytes("Finalized fixture PDF")) - git_env = {key: value for key, value in os.environ.items() if not key.startswith("GIT_")} + git_env = { + key: value for key, value in os.environ.items() if not key.startswith("GIT_") + } git_env.update(GIT_CONFIG_NOSYSTEM="1", GIT_CONFIG_GLOBAL=os.devnull) subprocess.run(["git", "init", "--quiet", str(root)], env=git_env, check=True) subprocess.run(["git", "add", "."], cwd=root, env=git_env, check=True) subprocess.run( [ - "git", "-c", "user.name=Receipt Test", "-c", "user.email=receipt@example.org", - "-c", "commit.gpgsign=false", "-c", "core.hooksPath=/dev/null", - "commit", "--quiet", "-m", "Disposable publication source", + "git", + "-c", + "user.name=Receipt Test", + "-c", + "user.email=receipt@example.org", + "-c", + "commit.gpgsign=false", + "-c", + "core.hooksPath=/dev/null", + "commit", + "--quiet", + "-m", + "Disposable publication source", ], - cwd=root, env=git_env, check=True, + cwd=root, + env=git_env, + check=True, ) revision = subprocess.run( - ["git", "rev-parse", "HEAD"], cwd=root, env=git_env, - check=True, capture_output=True, text=True, + ["git", "rev-parse", "HEAD"], + cwd=root, + env=git_env, + check=True, + capture_output=True, + text=True, ).stdout.strip() inputs = add_pdf_assets(root, capture_pdf_inputs(root, [section]), assets) return { - "root": root, "section": section, "assets": assets, "target": target, - "revision": revision, "inputs": inputs, "git_env": git_env, + "root": root, + "section": section, + "assets": assets, + "target": target, + "revision": revision, + "inputs": inputs, + "git_env": git_env, } diff --git a/tests/test_atomic_json_publication.py b/tests/test_atomic_json_publication.py index 5799fff..7003998 100644 --- a/tests/test_atomic_json_publication.py +++ b/tests/test_atomic_json_publication.py @@ -4,9 +4,9 @@ import json import os -from pathlib import Path import subprocess import sys +from pathlib import Path import pytest @@ -14,7 +14,9 @@ @pytest.mark.skipif(os.name != "posix", reason="RLIMIT_FSIZE is a POSIX I/O boundary") -def test_failed_write_preserves_existing_bytes_and_cleans_temporary(tmp_path: Path) -> None: +def test_failed_write_preserves_existing_bytes_and_cleans_temporary( + tmp_path: Path, +) -> None: artifact = tmp_path / "artifact.json" original = b'{"generation": "previous"}\n' artifact.write_bytes(original) @@ -38,20 +40,29 @@ def test_failed_write_preserves_existing_bytes_and_cleans_temporary(tmp_path: Pa result = subprocess.run( [sys.executable, "-B", "-c", program, str(artifact)], cwd=Path(__file__).resolve().parents[1], - capture_output=True, text=True, timeout=15, check=False, + capture_output=True, + text=True, + timeout=15, + check=False, ) assert result.returncode == 0, result.stdout + result.stderr assert artifact.read_bytes() == original assert sorted(path.name for path in tmp_path.iterdir()) == [artifact.name] -def test_atomic_replacement_preserves_content_format_and_existing_mode(tmp_path: Path) -> None: +def test_atomic_replacement_preserves_content_format_and_existing_mode( + tmp_path: Path, +) -> None: artifact = tmp_path / "artifact.json" artifact.write_text('{"old": true}\n', encoding="utf-8") artifact.chmod(0o640) write_json_artifact(artifact, {"z": "λ", "a": [1, 2]}, ensure_ascii=False) - assert artifact.read_text(encoding="utf-8") == json.dumps( - {"z": "λ", "a": [1, 2]}, indent=2, sort_keys=True, ensure_ascii=False - ) + "\n" + assert ( + artifact.read_text(encoding="utf-8") + == json.dumps( + {"z": "λ", "a": [1, 2]}, indent=2, sort_keys=True, ensure_ascii=False + ) + + "\n" + ) assert artifact.stat().st_mode & 0o777 == 0o640 assert sorted(path.name for path in tmp_path.iterdir()) == [artifact.name] diff --git a/tests/test_course_materials.py b/tests/test_course_materials.py index 52841b3..40b806e 100644 --- a/tests/test_course_materials.py +++ b/tests/test_course_materials.py @@ -368,9 +368,13 @@ def test_github_tree_fetch_retries_transient_5xx(httpserver) -> None: from src.curriculum.materials.github_sources import _github_tree_payload httpserver.expect_ordered_request("/tree").respond_with_json({}, status=503) - httpserver.expect_ordered_request("/tree").respond_with_json({"tree": [{"path": "a.md", "type": "blob"}]}) + httpserver.expect_ordered_request("/tree").respond_with_json( + {"tree": [{"path": "a.md", "type": "blob"}]} + ) result = _github_tree_payload( - urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, + urllib.request.Request(httpserver.url_for("/tree")), + timeout=2, + base_delay=0, ) assert result == {"tree": [{"path": "a.md", "type": "blob"}]} assert len(httpserver.log) == 2 @@ -385,7 +389,9 @@ def test_github_tree_fetch_raises_immediately_on_permanent_4xx(httpserver) -> No httpserver.expect_ordered_request("/tree").respond_with_json({}, status=404) with pytest.raises(urllib.error.HTTPError) as error: _github_tree_payload( - urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, + urllib.request.Request(httpserver.url_for("/tree")), + timeout=2, + base_delay=0, ) assert error.value.code == 404 assert len(httpserver.log) == 1 @@ -404,11 +410,16 @@ def test_github_tree_fetch_raises_connection_error_from_real_loopback() -> None: url = f"http://127.0.0.1:{endpoint.getsockname()[1]}/tree" with pytest.raises(urllib.error.URLError): _github_tree_payload( - urllib.request.Request(url), timeout=0.2, attempts=2, base_delay=0, + urllib.request.Request(url), + timeout=0.2, + attempts=2, + base_delay=0, ) -def test_github_tree_fetch_raises_final_error_after_exhausting_attempts(httpserver) -> None: +def test_github_tree_fetch_raises_final_error_after_exhausting_attempts( + httpserver, +) -> None: import urllib.error import urllib.request @@ -418,7 +429,9 @@ def test_github_tree_fetch_raises_final_error_after_exhausting_attempts(httpserv httpserver.expect_ordered_request("/tree").respond_with_json({}, status=503) with pytest.raises(urllib.error.HTTPError) as error: _github_tree_payload( - urllib.request.Request(httpserver.url_for("/tree")), timeout=2, base_delay=0, + urllib.request.Request(httpserver.url_for("/tree")), + timeout=2, + base_delay=0, ) assert error.value.code == 503 assert len(httpserver.log) == 3 @@ -426,7 +439,9 @@ def test_github_tree_fetch_raises_final_error_after_exhausting_attempts(httpserv @pytest.mark.parametrize("malformed", [[], None, {"tree": None}]) def test_malformed_successful_github_response_becomes_one_source_issue( - httpserver, monkeypatch, malformed, + httpserver, + monkeypatch, + malformed, ) -> None: httpserver.expect_request( "/repos/testorg/testrepo/git/trees/0123456789abcdef0123456789abcdef01234567" @@ -440,7 +455,9 @@ def test_malformed_successful_github_response_becomes_one_source_issue( @pytest.mark.parametrize("include_second", [False, True]) def test_offline_and_online_pointer_owners_agree_on_whitespace_and_all_list_entries( - httpserver, monkeypatch, include_second: bool, + httpserver, + monkeypatch, + include_second: bool, ) -> None: from src.curriculum.model import ALL_STRATA @@ -453,8 +470,18 @@ def test_offline_and_online_pointer_owners_agree_on_whitespace_and_all_list_entr source["path_globs"] = ["*.md"] payload["stratum_mappings"] = { stratum.value: [ - {"source_id": f" {source_id} ", "path": " first.md ", "title": " First ", "role": " Context "}, - {"source_id": f" {source_id} ", "path": " second.md ", "title": " Second ", "role": " Context "}, + { + "source_id": f" {source_id} ", + "path": " first.md ", + "title": " First ", + "role": " Context ", + }, + { + "source_id": f" {source_id} ", + "path": " second.md ", + "title": " Second ", + "role": " Context ", + }, ] for stratum in ALL_STRATA } @@ -464,7 +491,9 @@ def test_offline_and_online_pointer_owners_agree_on_whitespace_and_all_list_entr paths = ["first.md", "second.md"] if include_second else ["first.md"] httpserver.expect_request( "/repos/testorg/testrepo/git/trees/0123456789abcdef0123456789abcdef01234567" - ).respond_with_json({"tree": [{"path": path, "type": "blob"} for path in paths], "truncated": False}) + ).respond_with_json( + {"tree": [{"path": path, "type": "blob"} for path in paths], "truncated": False} + ) issues = validate_material_sources(payload, online=True, timeout=2) diff --git a/tests/test_curriculum_graph.py b/tests/test_curriculum_graph.py index 73fd68f..10331d8 100644 --- a/tests/test_curriculum_graph.py +++ b/tests/test_curriculum_graph.py @@ -224,14 +224,21 @@ def test_diamond_closure_deduplicates_shared_ancestors_and_refuses_cycles() -> N from src.curriculum.quality import transitive_prerequisites graph = _graph( - _skill("a", _M, ()), _skill("b", _M, ("a",)), - _skill("c", _M, ("a",)), _skill("d", _M, ("b", "c")), + _skill("a", _M, ()), + _skill("b", _M, ("a",)), + _skill("c", _M, ("a",)), + _skill("d", _M, ("b", "c")), ) assert transitive_prerequisites(graph) == { - "a": set(), "b": {"a"}, "c": {"a"}, "d": {"a", "b", "c"}, + "a": set(), + "b": {"a"}, + "c": {"a"}, + "d": {"a", "b", "c"}, } with pytest.raises(CurriculumError): - transitive_prerequisites(_graph(_skill("a", _M, ("b",)), _skill("b", _M, ("a",)))) + transitive_prerequisites( + _graph(_skill("a", _M, ("b",)), _skill("b", _M, ("a",))) + ) with pytest.raises(CurriculumError): transitive_prerequisites(_graph(_skill("a", _M, ("missing",)))) @@ -239,7 +246,9 @@ def test_diamond_closure_deduplicates_shared_ancestors_and_refuses_cycles() -> N def test_call_scoped_snapshots_do_not_cache_mutable_subject_membership() -> None: graph = _graph(_skill("a", _M, ())) before = graph.skills - graph.subjects.append(Subject("extra", "Extra", _M, "d", (_skill("b", _M, ("a",)),))) + graph.subjects.append( + Subject("extra", "Extra", _M, "d", (_skill("b", _M, ("a",)),)) + ) graph.validate() assert [skill.skill_id for skill in before] == ["a"] assert [skill.skill_id for skill in graph.skills] == ["a", "b"] diff --git a/tests/test_dashboard_console.py b/tests/test_dashboard_console.py index e171552..ea3ae66 100644 --- a/tests/test_dashboard_console.py +++ b/tests/test_dashboard_console.py @@ -12,6 +12,7 @@ sys.path.insert(0, str(REPO / "scripts")) import validate_dashboard_console as gate + from src.dashboard.browser_validation import unexpected_errors _INCEPTION = ( diff --git a/tests/test_dashboard_live_validation.py b/tests/test_dashboard_live_validation.py index 4d13992..62bfcb8 100644 --- a/tests/test_dashboard_live_validation.py +++ b/tests/test_dashboard_live_validation.py @@ -59,7 +59,9 @@ def _serve_dashboard_payloads( httpserver.expect_request("/api/status").respond_with_json( { "instanceId": "live-validation-fixture", - "project": project_summary if project_summary is not None else { + "project": project_summary + if project_summary is not None + else { "project": {"id": SKILLTREE_PROJECT_ID, "name": SKILLTREE_PROJECT_NAME}, "dependencies": [], "counts": project_counts, @@ -181,12 +183,18 @@ def test_fetch_json_rejects_non_http_scheme() -> None: @pytest.mark.parametrize("changed", [False, True]) def test_strict_live_validator_binds_actual_admin_content_not_just_matching_counts( - tmp_path: Path, httpserver, content_project, skilltree_admin_content, changed: bool, + tmp_path: Path, + httpserver, + content_project, + skilltree_admin_content, + changed: bool, ) -> None: exported = content_project.to_export_dict() exported.update(projectId=SKILLTREE_PROJECT_ID, name=SKILLTREE_PROJECT_NAME) exported["subjects"][0]["skills"][0]["skillId"] = "live_validation_01" - exported["quizzes"][0]["questions"][0]["answers"].append({"answer": "other", "isCorrect": False}) + exported["quizzes"][0]["questions"][0]["answers"].append( + {"answer": "other", "isCorrect": False} + ) _write_project_export(tmp_path, exported) public = load_skilltree_project(tmp_path, export_payload=exported) base_url = _serve_dashboard_payloads(httpserver, project_summary=public) @@ -195,27 +203,45 @@ def test_strict_live_validator_binds_actual_admin_content_not_just_matching_coun answers = live["quizzes"][0]["questions"][0]["answers"] answers[0]["isCorrect"], answers[1]["isCorrect"] = False, True skilltree_admin_content(httpserver, live) - httpserver.expect_request("/", method="GET").respond_with_data("Session", content_type="text/html") + httpserver.expect_request("/", method="GET").respond_with_data( + "Session", content_type="text/html" + ) httpserver.expect_request("/performLogin", method="POST").respond_with_json({}) httpserver.expect_request( - f"/admin/projects/{SKILLTREE_PROJECT_ID}/dependency/graph", method="GET", + f"/admin/projects/{SKILLTREE_PROJECT_ID}/dependency/graph", + method="GET", ).respond_with_json({"nodes": [], "edges": []}) state_dir = tmp_path / "private-state" - save_local_state(state_dir, LocalSkillTreeState( - admin_username="fixture-admin", admin_password="fixture-password", - learner_id="fixture-learner", project_id=SKILLTREE_PROJECT_ID, - skilltree_url=base_url, client_secret="fixture-client-secret", - )) + save_local_state( + state_dir, + LocalSkillTreeState( + admin_username="fixture-admin", + admin_password="fixture-password", + learner_id="fixture-learner", + project_id=SKILLTREE_PROJECT_ID, + skilltree_url=base_url, + client_secret="fixture-client-secret", + ), + ) arguments = { - "dashboard_url": base_url, "skilltree_url": base_url, "project_root": tmp_path, - "local_state_dir": state_dir, "require_token": False, "require_live_admin": True, + "dashboard_url": base_url, + "skilltree_url": base_url, + "project_root": tmp_path, + "local_state_dir": state_dir, + "require_token": False, + "require_live_admin": True, } if changed: with pytest.raises(LiveSkillTreeDashboardError, match="content"): validate_live_skilltree_dashboard(**arguments) else: result = validate_live_skilltree_dashboard(**arguments) - assert result["liveContent"]["checked"] is True and result["liveContent"]["complete"] is True + assert ( + result["liveContent"]["checked"] is True + and result["liveContent"]["complete"] is True + ) assert result["exportContentDigest"] == export_content_digest(exported) assert result["dashboardInstanceId"] == "live-validation-fixture" - assert "isCorrect" not in json.dumps(result) and "fixture-password" not in json.dumps(result) + assert "isCorrect" not in json.dumps( + result + ) and "fixture-password" not in json.dumps(result) diff --git a/tests/test_dashboard_mount.py b/tests/test_dashboard_mount.py index 1c3aa95..dbcc7f5 100644 --- a/tests/test_dashboard_mount.py +++ b/tests/test_dashboard_mount.py @@ -25,7 +25,9 @@ def authorized_browser(): url = os.environ.get("ASF_TEST_DASHBOARD_URL", "") if not url: - pytest.skip("set ASF_TEST_DASHBOARD_URL only after live browser acceptance is authorized") + pytest.skip( + "set ASF_TEST_DASHBOARD_URL only after live browser acceptance is authorized" + ) command = os.environ.get("ASF_TEST_BROWSER_COMMAND", "chrome-devtools-axi") if shutil.which(command) is None: pytest.skip(f"existing browser CLI unavailable: {command}") @@ -33,9 +35,13 @@ def authorized_browser(): @pytest.mark.requires_browser -def test_actual_official_project_subject_and_quiz_are_mounted(authorized_browser) -> None: +def test_actual_official_project_subject_and_quiz_are_mounted( + authorized_browser, +) -> None: url, command = authorized_browser - receipt = validate_dashboard_browser(url, browser_command=command, timeout=60, settle_seconds=1) + receipt = validate_dashboard_browser( + url, browser_command=command, timeout=60, settle_seconds=1 + ) assert browser_receipt_complete(receipt) quiz = next(view for view in receipt["views"] if view["kind"] == "quiz") observed = display_probe(browser_command=command, timeout=10) @@ -48,7 +54,10 @@ def test_actual_official_project_subject_and_quiz_are_mounted(authorized_browser @pytest.mark.requires_browser def test_real_dashboard_fallback_cannot_pass_mounted_view_proof( - tmp_path: Path, httpserver, content_project, authorized_browser, + tmp_path: Path, + httpserver, + content_project, + authorized_browser, ) -> None: _authorized_url, command = authorized_browser export = tmp_path / "output" / "data" / "skilltree_project.json" @@ -56,8 +65,12 @@ def test_real_dashboard_fallback_cannot_pass_mounted_view_proof( export.write_text(json.dumps(content_project.to_export_dict()), encoding="utf-8") httpserver.expect_request("/public/status").respond_with_json({"status": "OK"}) config = make_dashboard_config( - project_root=tmp_path, skilltree_url=httpserver.url_for(""), - project_id="p", project_name="Project", host="127.0.0.1", port=8765, + project_root=tmp_path, + skilltree_url=httpserver.url_for(""), + project_id="p", + project_name="Project", + host="127.0.0.1", + port=8765, skills_client_module_url="/assets/missing-official-module.js", local_auth_enabled=False, ) @@ -67,7 +80,9 @@ def test_real_dashboard_fallback_cannot_pass_mounted_view_proof( url = f"http://127.0.0.1:{server.server_address[1]}" try: with pytest.raises(BrowserDisplayError, match="fallback|did not reach ready"): - validate_dashboard_browser(url, browser_command=command, timeout=15, settle_seconds=0) + validate_dashboard_browser( + url, browser_command=command, timeout=15, settle_seconds=0 + ) observed = display_probe(browser_command=command, timeout=5) assert observed["state"] != "ready" finally: @@ -77,6 +92,8 @@ def test_real_dashboard_fallback_cannot_pass_mounted_view_proof( @pytest.mark.parametrize("timeout", [0.0, -1.0, float("nan"), float("inf")]) -def test_invalid_browser_timeout_is_rejected_before_network_or_cli(timeout: float) -> None: +def test_invalid_browser_timeout_is_rejected_before_network_or_cli( + timeout: float, +) -> None: with pytest.raises(BrowserDisplayError, match="finite and positive"): validate_dashboard_browser("not a URL", timeout=timeout) diff --git a/tests/test_dashboard_server.py b/tests/test_dashboard_server.py index 6d9e922..2f06f76 100644 --- a/tests/test_dashboard_server.py +++ b/tests/test_dashboard_server.py @@ -643,17 +643,21 @@ def test_request_without_host_header_is_allowed(tmp_path: Path) -> None: def test_malformed_provider_token_is_a_secret_safe_service_error( - tmp_path: Path, httpserver: HTTPServer, + tmp_path: Path, + httpserver: HTTPServer, ) -> None: provider = httpserver.url_for("").rstrip("/") httpserver.expect_request("/oauth/token", method="POST").respond_with_data( - '{"access_token": "private-provider-fragment"', content_type="application/json", + '{"access_token": "private-provider-fragment"', + content_type="application/json", ) state_dir = tmp_path / "state" _write_local_auth_state(state_dir, skilltree_url=provider) config = make_dashboard_config( - project_root=tmp_path, skilltree_url=provider, - local_auth_enabled=True, local_state_dir=state_dir, + project_root=tmp_path, + skilltree_url=provider, + local_auth_enabled=True, + local_state_dir=state_dir, ) with _DashboardServer(config) as base_url: try: @@ -668,7 +672,9 @@ def test_malformed_provider_token_is_a_secret_safe_service_error( def test_status_requests_see_atomic_export_replacement_without_exposing_answer_keys( - tmp_path: Path, httpserver, content_project, + tmp_path: Path, + httpserver, + content_project, ) -> None: from copy import deepcopy @@ -681,24 +687,35 @@ def test_status_requests_see_atomic_export_replacement_without_exposing_answer_k write_json_artifact(path, exported) httpserver.expect_request("/public/status").respond_with_json({"status": "OK"}) config = make_dashboard_config( - project_root=tmp_path, project_id="p", project_name="Project", + project_root=tmp_path, + project_id="p", + project_name="Project", skilltree_url=httpserver.url_for(""), ) with _DashboardServer(config) as url: code, _headers, body = _request(url + "/api/status?defer_live=1") first = json.loads(body) assert code == 200 and first["project"]["available"] is True - assert first["project"]["exportContentDigest"] == export_content_digest(exported) + assert first["project"]["exportContentDigest"] == export_content_digest( + exported + ) replacement = deepcopy(exported) replacement["subjects"][0]["skills"][0]["description"] = "Replacement lesson" - replacement["quizzes"][0]["questions"][0]["answers"][0]["answer"] = "private-replacement-key-7ce8" + replacement["quizzes"][0]["questions"][0]["answers"][0]["answer"] = ( + "private-replacement-key-7ce8" + ) write_json_artifact(path, replacement) code, _headers, body = _request(url + "/api/status?defer_live=1") second = json.loads(body) assert code == 200 assert first["instanceId"] == second["instanceId"] - assert second["project"]["exportContentDigest"] == export_content_digest(replacement) - assert first["project"]["exportContentDigest"] != second["project"]["exportContentDigest"] + assert second["project"]["exportContentDigest"] == export_content_digest( + replacement + ) + assert ( + first["project"]["exportContentDigest"] + != second["project"]["exportContentDigest"] + ) assert "isCorrect" not in body.decode("utf-8") assert "private-replacement-key-7ce8" not in body.decode("utf-8") diff --git a/tests/test_figures.py b/tests/test_figures.py index 5f0fef3..287f206 100644 --- a/tests/test_figures.py +++ b/tests/test_figures.py @@ -609,11 +609,13 @@ def test_validation_intent_mismatch_matrix_figure_written(tmp_path) -> None: path = plot_validation_intent_mismatch_matrix( tmp_path / "validation-intent.png", - figure_count=len(planned_figure_registry_entries( - include_scholarship=True, - include_claim_support=True, - include_roadmap=True, - )), + figure_count=len( + planned_figure_registry_entries( + include_scholarship=True, + include_claim_support=True, + include_roadmap=True, + ) + ), include_scholarship=True, include_claim_support=True, include_roadmap=True, @@ -924,10 +926,16 @@ def test_registered_total_and_rendered_subset_keep_distinct_contracts() -> None: from src.figures.registry_manifest import figure_registry_payload for scholarship, claims, roadmap in product((False, True), repeat=3): - flags = dict(include_scholarship=scholarship, include_claim_support=claims, include_roadmap=roadmap) + flags = { + "include_scholarship": scholarship, + "include_claim_support": claims, + "include_roadmap": roadmap, + } registered = figure_registry_payload(**flags) rendered = planned_figure_registry_entries(**flags) assert planned_figure_count(**flags) == len(registered) assert {row["filename"] for row in rendered} == { - row["filename"] for row in registered if row.get("evidence_kind", "rendered") != "captured" + row["filename"] + for row in registered + if row.get("evidence_kind", "rendered") != "captured" } diff --git a/tests/test_full_app_stack.py b/tests/test_full_app_stack.py index 35d2422..ee2e710 100644 --- a/tests/test_full_app_stack.py +++ b/tests/test_full_app_stack.py @@ -316,27 +316,53 @@ def test_full_app_stack_starts_keep_running_dashboard_with_log(tmp_path: Path) - def _browser_receipt() -> dict: return { - "ok": True, "checked": True, "projectId": "p", - "dashboardInstanceId": "instance", "exportContentDigest": "a" * 64, + "ok": True, + "checked": True, + "projectId": "p", + "dashboardInstanceId": "instance", + "exportContentDigest": "a" * 64, "views": [ - {"kind": "project", "route": "/", "state": "ready", "visibleContentChecked": True}, - {"kind": "subject", "route": "/subjects/s", "state": "ready", "visibleContentChecked": True}, - {"kind": "quiz", "route": "/subjects/s/skills/k/quizzes/q", "state": "ready", "visibleContentChecked": True}, + { + "kind": "project", + "route": "/", + "state": "ready", + "visibleContentChecked": True, + }, + { + "kind": "subject", + "route": "/subjects/s", + "state": "ready", + "visibleContentChecked": True, + }, + { + "kind": "quiz", + "route": "/subjects/s/skills/k/quizzes/q", + "state": "ready", + "visibleContentChecked": True, + }, ], } def _live_receipt() -> dict: return { - "ok": True, "dashboardInstanceId": "instance", "exportContentDigest": "a" * 64, + "ok": True, + "dashboardInstanceId": "instance", + "exportContentDigest": "a" * 64, "counts": {"skills": 630}, "artifacts": {"present": 30, "missing": 0}, "courseMaterials": {"records": 630}, "figureContracts": {"checked": True, "registered": 19}, "liveAdminProject": {"checked": True, "skills": 630}, - "liveDependencies": {"checked": True, "complete": True, "expectedDependencies": 1199}, + "liveDependencies": { + "checked": True, + "complete": True, + "expectedDependencies": 1199, + }, "liveContent": { - "checked": True, "complete": True, "projectId": "p", + "checked": True, + "complete": True, + "projectId": "p", "exportContentDigest": "a" * 64, }, } @@ -352,8 +378,10 @@ def test_full_app_stack_proof_summary_shape() -> None: live_validation=_live_receipt(), browser_console_stdout=json.dumps(_browser_receipt()), raw_skilltree_login={"ok": True, "adminSurfaceVisible": True}, - kept_running=True, dashboard_pid=12345, - dashboard_pid_path=pid_path, dashboard_log_path=log_path, + kept_running=True, + dashboard_pid=12345, + dashboard_pid_path=pid_path, + dashboard_log_path=log_path, dependency_reconciliation={"ok": True, "complete": True, "missingAfter": 0}, ) assert proof["ok"] is True @@ -401,25 +429,36 @@ def test_full_app_stack_writes_redacted_deterministic_proof(tmp_path: Path) -> N assert payload["counts"] == {"skills": 630} -@pytest.mark.parametrize("stdout", ["dashboard console gate OK", "{}", '{"ok": true, "checked": false}']) -def test_console_exit_text_or_unchecked_payload_cannot_certify_a_display(stdout: str) -> None: +@pytest.mark.parametrize( + "stdout", ["dashboard console gate OK", "{}", '{"ok": true, "checked": false}'] +) +def test_console_exit_text_or_unchecked_payload_cannot_certify_a_display( + stdout: str, +) -> None: with pytest.raises(stack.FullAppStackError): stack.proof_summary( - skilltree_url="http://localhost:8080", dashboard_url="http://127.0.0.1:8765", + skilltree_url="http://localhost:8080", + dashboard_url="http://127.0.0.1:8765", skilltree_status={"status": "OK"}, dashboard_health={"ok": True, "instanceId": "instance"}, - live_validation=_live_receipt(), browser_console_stdout=stdout, - raw_skilltree_login={"ok": True}, kept_running=False, dashboard_pid=None, + live_validation=_live_receipt(), + browser_console_stdout=stdout, + raw_skilltree_login={"ok": True}, + kept_running=False, + dashboard_pid=None, ) def test_proof_failure_stops_owned_dashboard_even_with_keep_running( - tmp_path: Path, httpserver, + tmp_path: Path, + httpserver, ) -> None: import urllib.request httpserver.expect_request("/unrelated").respond_with_data("still running") - process = stack.start_dashboard([sys.executable, "-c", "import time; time.sleep(30)"]) + process = stack.start_dashboard( + [sys.executable, "-c", "import time; time.sleep(30)"] + ) destination = tmp_path / "proof-is-a-directory" destination.mkdir() try: @@ -428,23 +467,35 @@ def test_proof_failure_stops_owned_dashboard_even_with_keep_running( stack.write_proof_artifact({"ok": True}, destination) finally: stack.finalize_dashboard_process( - process, state_dir=str(tmp_path), keep_running=True, successful=False, + process, + state_dir=str(tmp_path), + keep_running=True, + successful=False, ) assert process.poll() is not None assert not (tmp_path / "dashboard.pid").exists() - with urllib.request.urlopen(httpserver.url_for("/unrelated"), timeout=2) as response: + with urllib.request.urlopen( + httpserver.url_for("/unrelated"), timeout=2 + ) as response: assert response.read() == b"still running" finally: stack.stop_dashboard_process(process) -def test_failed_pid_publication_does_not_leave_the_owned_process_alive(tmp_path: Path) -> None: +def test_failed_pid_publication_does_not_leave_the_owned_process_alive( + tmp_path: Path, +) -> None: (tmp_path / "dashboard.pid").mkdir() - process = stack.start_dashboard([sys.executable, "-c", "import time; time.sleep(30)"]) + process = stack.start_dashboard( + [sys.executable, "-c", "import time; time.sleep(30)"] + ) try: with pytest.raises(OSError): stack.finalize_dashboard_process( - process, state_dir=str(tmp_path), keep_running=True, successful=True, + process, + state_dir=str(tmp_path), + keep_running=True, + successful=True, ) assert process.poll() is not None assert (tmp_path / "dashboard.pid").is_dir() diff --git a/tests/test_generated_output_cli.py b/tests/test_generated_output_cli.py index c9c16e9..26a5d8a 100644 --- a/tests/test_generated_output_cli.py +++ b/tests/test_generated_output_cli.py @@ -9,10 +9,14 @@ from scripts import run_kernels -def test_cli_reports_real_manuscript_rejection_without_replacing_the_builder(tmp_path: Path, monkeypatch) -> None: +def test_cli_reports_real_manuscript_rejection_without_replacing_the_builder( + tmp_path: Path, monkeypatch +) -> None: manuscript = tmp_path / "manuscript" manuscript.mkdir() - (manuscript / "00_fixture.md").write_text("The platform is NSA approved for all learners.\n", encoding="utf-8") + (manuscript / "00_fixture.md").write_text( + "The platform is NSA approved for all learners.\n", encoding="utf-8" + ) monkeypatch.setattr(run_kernels, "_ROOT", tmp_path) with pytest.raises(SystemExit) as error: run_kernels.main() diff --git a/tests/test_gnn_conformance.py b/tests/test_gnn_conformance.py index 16b280b..5e1e9c4 100644 --- a/tests/test_gnn_conformance.py +++ b/tests/test_gnn_conformance.py @@ -114,7 +114,6 @@ def test_declared_dimensions_match_the_kernel() -> None: assert dimensions["o"] == (agent["num_observations"], 1) - def test_mutated_model_is_rejected() -> None: model = copy.deepcopy(canonical_model()) model["B_matrix"][0][0][1] = 0.0 # break the commit transition @@ -131,7 +130,9 @@ def test_mutated_model_is_rejected() -> None: ("D_vector", [1.0]), ], ) -def test_invalid_declared_component_shapes_raise_conformance_error(field, value) -> None: +def test_invalid_declared_component_shapes_raise_conformance_error( + field, value +) -> None: model = copy.deepcopy(canonical_model()) model[field] = value with pytest.raises(ConformanceError, match="declared probability/transition model"): diff --git a/tests/test_manuscript_hydration.py b/tests/test_manuscript_hydration.py index 566e99d..70e7e1f 100644 --- a/tests/test_manuscript_hydration.py +++ b/tests/test_manuscript_hydration.py @@ -8,23 +8,37 @@ def test_known_prose_variables_and_claims_resolve_to_the_supplied_value() -> None: - assert _render_markdown("Result {{value}} / {{claim:value}}.", {"value": "computed"}) == "Result computed / computed." - - -@pytest.mark.parametrize("text", ["Result {{misspelled}}.", "Result {{claim:missing}}."]) -def test_unknown_prose_placeholder_fails_instead_of_stripping_its_delimiters(text: str) -> None: + assert ( + _render_markdown("Result {{value}} / {{claim:value}}.", {"value": "computed"}) + == "Result computed / computed." + ) + + +@pytest.mark.parametrize( + "text", ["Result {{misspelled}}.", "Result {{claim:missing}}."] +) +def test_unknown_prose_placeholder_fails_instead_of_stripping_its_delimiters( + text: str, +) -> None: with pytest.raises(KeyError): _render_markdown(text, {"value": "computed"}) -def test_code_examples_normalize_locally_without_admitting_unknown_adjacent_prose() -> None: +def test_code_examples_normalize_locally_without_admitting_unknown_adjacent_prose() -> ( + None +): example = "Use `{{unknown}}` and:\n```text\n{{claim:missing}}\n```\n" - assert _render_markdown(example, {}) == "Use `unknown` and:\n```text\nclaim:missing\n```\n" + assert ( + _render_markdown(example, {}) + == "Use `unknown` and:\n```text\nclaim:missing\n```\n" + ) with pytest.raises(KeyError): _render_markdown(example + "Prose {{unknown}}.", {}) -@pytest.mark.parametrize("text", ["Result {{missing token}}.", "Result {{ claim:value }}."]) +@pytest.mark.parametrize( + "text", ["Result {{missing token}}.", "Result {{ claim:value }}."] +) def test_unrecognized_prose_token_shape_remains_an_error(text: str) -> None: with pytest.raises(ValueError, match="unresolved manuscript tokens"): _render_markdown(text, {"value": "computed"}) diff --git a/tests/test_material_claim_fields.py b/tests/test_material_claim_fields.py index c097ae2..2a08db6 100644 --- a/tests/test_material_claim_fields.py +++ b/tests/test_material_claim_fields.py @@ -8,9 +8,18 @@ import pytest from src.curriculum.graph import CurriculumGraph -from src.curriculum.materials import audit_course_material_index, build_course_material_index +from src.curriculum.materials import ( + audit_course_material_index, + build_course_material_index, +) from src.curriculum.model import AssessmentQuestion, Skill, Stratum, Subject -from src.provenance.binding import ProvenanceError, lint_graph, render_graph, validate_curriculum, validate_skill +from src.provenance.binding import ( + ProvenanceError, + lint_graph, + render_graph, + validate_curriculum, + validate_skill, +) from src.provenance.claims import ClaimRegistry TOKEN = "{{cl\u200baim:bo\u200bund}}" @@ -24,22 +33,44 @@ def _registry() -> ClaimRegistry: def _skill() -> Skill: return Skill( - "a", "A", Stratum.MATH_FOUNDATIONS, "Objective", "Description", (), + "a", + "A", + Stratum.MATH_FOUNDATIONS, + "Objective", + "Description", + (), (AssessmentQuestion("Question", ("one", "two"), (0,)),), ) def _field(skill: Skill, name: str, value: str) -> Skill: if name == "assessment_prompt": - return replace(skill, assessments=(replace(skill.assessments[0], prompt=value),)) + return replace( + skill, assessments=(replace(skill.assessments[0], prompt=value),) + ) if name == "assessment_answer": - return replace(skill, assessments=(replace(skill.assessments[0], answers=(value, "two")),)) + return replace( + skill, assessments=(replace(skill.assessments[0], answers=(value, "two")),) + ) return replace(skill, **{name: value}) -@pytest.mark.parametrize("field", ["name", "objective", "description", "assessment_prompt", "assessment_answer"]) -@pytest.mark.parametrize("token", ["{{claim:missing}}", "{{ claim:bound }}", "{{claim:bad-id}}", "{{cl\u200baim:missing}}"]) -def test_standalone_skill_and_curriculum_validation_reject_every_bad_field(field: str, token: str) -> None: +@pytest.mark.parametrize( + "field", + ["name", "objective", "description", "assessment_prompt", "assessment_answer"], +) +@pytest.mark.parametrize( + "token", + [ + "{{claim:missing}}", + "{{ claim:bound }}", + "{{claim:bad-id}}", + "{{cl\u200baim:missing}}", + ], +) +def test_standalone_skill_and_curriculum_validation_reject_every_bad_field( + field: str, token: str +) -> None: skill = _field(_skill(), field, token) with pytest.raises(ProvenanceError): validate_skill(skill, _registry()) @@ -47,47 +78,85 @@ def test_standalone_skill_and_curriculum_validation_reject_every_bad_field(field validate_curriculum([skill], _registry()) -@pytest.mark.parametrize("field", ["name", "objective", "description", "assessment_prompt", "assessment_answer"]) -def test_format_character_normalization_preserves_original_claim_identity(field: str) -> None: +@pytest.mark.parametrize( + "field", + ["name", "objective", "description", "assessment_prompt", "assessment_answer"], +) +def test_format_character_normalization_preserves_original_claim_identity( + field: str, +) -> None: skill = _field(_skill(), field, TOKEN) assert skill.claim_tokens() == {"bound"} validate_skill(skill, _registry()) def _graph() -> CurriculumGraph: - first = replace(_skill(), name=f"A {TOKEN}", objective=f"Objective {TOKEN}", description=f"Description {TOKEN}") + first = replace( + _skill(), + name=f"A {TOKEN}", + objective=f"Objective {TOKEN}", + description=f"Description {TOKEN}", + ) second = replace(_skill(), skill_id="b", name=f"B {TOKEN}", prerequisites=("a",)) - return CurriculumGraph([ - Subject("sub_m", f"Subject {TOKEN}", Stratum.MATH_FOUNDATIONS, "Subject description", (first, second)) - ]) + return CurriculumGraph( + [ + Subject( + "sub_m", + f"Subject {TOKEN}", + Stratum.MATH_FOUNDATIONS, + "Subject description", + (first, second), + ) + ] + ) def _sources(*, claims: bool = True) -> dict: value = TOKEN if claims else "bound-value" - pointer = {"source_id": "fixture", "path": "lessons/intro.md", "title": f"Pointer {value}", "role": f"Role {value}"} + pointer = { + "source_id": "fixture", + "path": "lessons/intro.md", + "title": f"Pointer {value}", + "role": f"Role {value}", + } return { "version": 1, - "sources": [{ - "source_id": "fixture", "repo_url": "https://github.com/fixture/materials", "commit": "a" * 40, - "title": f"Source {value}", "license_status": f"License {value}", "status_note": f"Status {value}", - "path_globs": ["lessons/*.md"], - }], + "sources": [ + { + "source_id": "fixture", + "repo_url": "https://github.com/fixture/materials", + "commit": "a" * 40, + "title": f"Source {value}", + "license_status": f"License {value}", + "status_note": f"Status {value}", + "path_globs": ["lessons/*.md"], + } + ], "stratum_mappings": {stratum.value: [dict(pointer)] for stratum in Stratum}, "topic_mappings": [], } -def test_material_names_references_context_and_source_labels_render_without_losing_claim_ids() -> None: +def test_material_names_references_context_and_source_labels_render_without_losing_claim_ids() -> ( + None +): graph = _graph() index = build_course_material_index(graph, _sources(), registry=_registry()) first, second = index["skills"]["a"], index["skills"]["b"] - assert first["name"] == "A bound-value" and first["subjectName"] == "Subject bound-value" + assert ( + first["name"] == "A bound-value" + and first["subjectName"] == "Subject bound-value" + ) assert second["prerequisites"][0]["name"] == "A bound-value" assert second["relatedSkills"][0]["subjectName"] == "Subject bound-value" assert first["downstreamSkills"][0]["name"] == "B bound-value" - assert "A bound-value (a)" in "\n".join(section["body"] for section in second["sections"]) - assert "B bound-value (b)" in "\n".join(section["body"] for section in first["sections"]) + assert "A bound-value (a)" in "\n".join( + section["body"] for section in second["sections"] + ) + assert "B bound-value (b)" in "\n".join( + section["body"] for section in first["sections"] + ) assert first["provenance"]["claimTokens"] == ["bound"] assert first["externalPointers"][0]["title"] == "Pointer bound-value" assert first["externalPointers"][0]["role"] == "Role bound-value" @@ -96,7 +165,9 @@ def test_material_names_references_context_and_source_labels_render_without_losi assert audit_course_material_index(index, graph).issues == () -def test_material_builder_without_registry_accepts_claim_free_input_but_refuses_raw_claims() -> None: +def test_material_builder_without_registry_accepts_claim_free_input_but_refuses_raw_claims() -> ( + None +): graph = _graph() with pytest.raises(ProvenanceError): build_course_material_index(graph, _sources(claims=False)) @@ -105,11 +176,19 @@ def test_material_builder_without_registry_accepts_claim_free_input_but_refuses_ assert index["skills"]["a"]["name"] == "A bound-value" -@pytest.mark.parametrize("field", ["name", "subjectName", "prerequisite", "section", "source", "pointer"]) -@pytest.mark.parametrize("token", ["{{claim:missing}}", "{{ claim:bound }}", "{{cl\u200baim:missing}}"]) -def test_material_audit_catches_unrendered_tokens_outside_section_bodies(field: str, token: str) -> None: +@pytest.mark.parametrize( + "field", ["name", "subjectName", "prerequisite", "section", "source", "pointer"] +) +@pytest.mark.parametrize( + "token", ["{{claim:missing}}", "{{ claim:bound }}", "{{cl\u200baim:missing}}"] +) +def test_material_audit_catches_unrendered_tokens_outside_section_bodies( + field: str, token: str +) -> None: graph = _graph() - index = deepcopy(build_course_material_index(graph, _sources(), registry=_registry())) + index = deepcopy( + build_course_material_index(graph, _sources(), registry=_registry()) + ) record = index["skills"]["b"] if field in ("name", "subjectName"): record[field] = token @@ -144,10 +223,12 @@ def test_subject_names_that_collide_after_rendering_are_refused() -> None: registry.register("second", "same") first = _skill() second = replace(first, skill_id="b", name="B", prerequisites=("a",)) - graph = CurriculumGraph([ - Subject("sub_a", "Course {{claim:first}}", first.stratum, "", (first,)), - Subject("sub_b", "Course {{claim:second}}", second.stratum, "", (second,)), - ]) + graph = CurriculumGraph( + [ + Subject("sub_a", "Course {{claim:first}}", first.stratum, "", (first,)), + Subject("sub_b", "Course {{claim:second}}", second.stratum, "", (second,)), + ] + ) graph.validate() with pytest.raises(CurriculumError, match="duplicate subject name"): render_graph(graph, registry) diff --git a/tests/test_pdf_provenance.py b/tests/test_pdf_provenance.py index afbcae5..b3b0da4 100644 --- a/tests/test_pdf_provenance.py +++ b/tests/test_pdf_provenance.py @@ -19,16 +19,28 @@ def _receipt(project): return write_pdf_receipt( - project["target"], root=project["root"], source_revision=project["revision"], inputs=project["inputs"], + project["target"], + root=project["root"], + source_revision=project["revision"], + inputs=project["inputs"], ) -@pytest.mark.parametrize("relative", [ - "manuscript/config.yaml", "manuscript/preamble.md", "manuscript/references.bib", - "output/manuscript/01_fixture.md", "output/manuscript/assets/cover.pdf", - "output/manuscript/assets/atlas.pdf", "output/manuscript/assets/body.pdf", -]) -def test_changed_actual_input_invalidates_unchanged_pdf_and_cannot_refresh_receipt(pdf_project, relative: str) -> None: +@pytest.mark.parametrize( + "relative", + [ + "manuscript/config.yaml", + "manuscript/preamble.md", + "manuscript/references.bib", + "output/manuscript/01_fixture.md", + "output/manuscript/assets/cover.pdf", + "output/manuscript/assets/atlas.pdf", + "output/manuscript/assets/body.pdf", + ], +) +def test_changed_actual_input_invalidates_unchanged_pdf_and_cannot_refresh_receipt( + pdf_project, relative: str +) -> None: receipt = _receipt(pdf_project) old_receipt, old_pdf = receipt.read_bytes(), pdf_project["target"].read_bytes() path = pdf_project["root"] / relative @@ -41,8 +53,13 @@ def test_changed_actual_input_invalidates_unchanged_pdf_and_cannot_refresh_recei assert pdf_project["target"].read_bytes() == old_pdf -@pytest.mark.parametrize("change", ["add-section", "delete-section", "delete-image", "replace-pdf", "delete-pdf"]) -def test_changed_input_set_or_final_pdf_is_not_fresh(pdf_project, pdf_bytes, change: str) -> None: +@pytest.mark.parametrize( + "change", + ["add-section", "delete-section", "delete-image", "replace-pdf", "delete-pdf"], +) +def test_changed_input_set_or_final_pdf_is_not_fresh( + pdf_project, pdf_bytes, change: str +) -> None: receipt = _receipt(pdf_project) old_receipt = receipt.read_bytes() if change == "add-section": @@ -59,26 +76,55 @@ def test_changed_input_set_or_final_pdf_is_not_fresh(pdf_project, pdf_bytes, cha assert receipt.read_bytes() == old_receipt -@pytest.mark.parametrize("raw", [b"not JSON", b"[]", b'{"schema":"active-skillference-pdf/v1"}', b'{"schema":"active-skillference-pdf/v2"}']) -def test_legacy_or_incomplete_receipt_requires_real_rerender_without_relabeling(pdf_project, raw: bytes) -> None: +@pytest.mark.parametrize( + "raw", + [ + b"not JSON", + b"[]", + b'{"schema":"active-skillference-pdf/v1"}', + b'{"schema":"active-skillference-pdf/v2"}', + ], +) +def test_legacy_or_incomplete_receipt_requires_real_rerender_without_relabeling( + pdf_project, raw: bytes +) -> None: receipt = pdf_project["target"].with_suffix(".receipt.json") receipt.write_bytes(raw) assert pdf_receipt_issues(pdf_project["root"]) assert receipt.read_bytes() == raw -def test_missing_receipt_is_not_created_by_read_only_freshness_audit(pdf_project) -> None: +def test_missing_receipt_is_not_created_by_read_only_freshness_audit( + pdf_project, +) -> None: receipt = pdf_project["target"].with_suffix(".receipt.json") assert pdf_receipt_issues(pdf_project["root"]) assert not receipt.exists() -@pytest.mark.parametrize("damage", ["missing-primary", "missing-section", "missing-image", "duplicate-path", "unsafe-path", "bad-digest", "missing-assets-list"]) -def test_incomplete_or_unsafe_input_manifest_is_rejected(pdf_project, damage: str) -> None: +@pytest.mark.parametrize( + "damage", + [ + "missing-primary", + "missing-section", + "missing-image", + "duplicate-path", + "unsafe-path", + "bad-digest", + "missing-assets-list", + ], +) +def test_incomplete_or_unsafe_input_manifest_is_rejected( + pdf_project, damage: str +) -> None: manifest = deepcopy(pdf_project["inputs"]) rows = manifest["files"] if damage.startswith("missing-") and damage != "missing-assets-list": - kind = {"missing-primary": "config", "missing-section": "section", "missing-image": "asset"}[damage] + kind = { + "missing-primary": "config", + "missing-section": "section", + "missing-image": "asset", + }[damage] rows.remove(next(row for row in rows if row["kind"] == kind)) elif damage == "duplicate-path": rows.append(dict(rows[0])) @@ -92,7 +138,9 @@ def test_incomplete_or_unsafe_input_manifest_is_rejected(pdf_project, damage: st validate_pdf_inputs(pdf_project["root"], manifest) -def test_actual_latex_image_inputs_are_deduplicated_and_comments_do_not_create_inputs(pdf_project) -> None: +def test_actual_latex_image_inputs_are_deduplicated_and_comments_do_not_create_inputs( + pdf_project, +) -> None: root = pdf_project["root"] tex = root / "build.tex" cover, _atlas, body = pdf_project["assets"] @@ -103,11 +151,17 @@ def test_actual_latex_image_inputs_are_deduplicated_and_comments_do_not_create_i ) paths = latex_asset_paths(tex, build_directory=root, root=root) assert paths == sorted([cover.resolve(), body.resolve()]) - manifest = add_pdf_assets(root, capture_pdf_inputs(root, [pdf_project["section"]]), paths) - assert set(manifest["assets"]) == {path.relative_to(root).as_posix() for path in (cover, body)} + manifest = add_pdf_assets( + root, capture_pdf_inputs(root, [pdf_project["section"]]), paths + ) + assert set(manifest["assets"]) == { + path.relative_to(root).as_posix() for path in (cover, body) + } -def test_nonliteral_latex_image_input_cannot_create_incomplete_provenance(pdf_project) -> None: +def test_nonliteral_latex_image_input_cannot_create_incomplete_provenance( + pdf_project, +) -> None: root = pdf_project["root"] tex = root / "build.tex" tex.write_text(r"\includegraphics{\chosenImage{}}") @@ -115,7 +169,9 @@ def test_nonliteral_latex_image_input_cannot_create_incomplete_provenance(pdf_pr latex_asset_paths(tex, build_directory=root, root=root) -def test_unrelated_repository_file_does_not_invalidate_actual_input_snapshot(pdf_project) -> None: +def test_unrelated_repository_file_does_not_invalidate_actual_input_snapshot( + pdf_project, +) -> None: _receipt(pdf_project) (pdf_project["root"] / "unrelated.txt").write_text("Not a PDF compiler input.") assert pdf_receipt_issues(pdf_project["root"]) == [] diff --git a/tests/test_pilot_schema_boundaries.py b/tests/test_pilot_schema_boundaries.py index 938bd0d..c33c04d 100644 --- a/tests/test_pilot_schema_boundaries.py +++ b/tests/test_pilot_schema_boundaries.py @@ -18,10 +18,16 @@ def pilot_payload() -> dict: @pytest.mark.parametrize( ("section", "field"), - [("participants", "requiredFields"), ("events", "allowedEventTypes"), - ("events", "requiredFields"), ("export", "fields")], + [ + ("participants", "requiredFields"), + ("events", "allowedEventTypes"), + ("events", "requiredFields"), + ("export", "fields"), + ], ) -def test_schema_lists_cannot_be_empty(pilot_payload: dict, section: str, field: str) -> None: +def test_schema_lists_cannot_be_empty( + pilot_payload: dict, section: str, field: str +) -> None: pilot_payload[section][field] = [] with pytest.raises(PilotContractError, match="must not be empty"): validate_pilot_contract(pilot_payload) @@ -29,10 +35,12 @@ def test_schema_lists_cannot_be_empty(pilot_payload: dict, section: str, field: @pytest.mark.parametrize( ("section", "field", "missing"), - [("participants", "requiredFields", "participant_id_hash"), - ("events", "requiredFields", "event_timestamp"), - ("export", "fields", "participant_id_hash"), - ("export", "fields", "event_type")], + [ + ("participants", "requiredFields", "participant_id_hash"), + ("events", "requiredFields", "event_timestamp"), + ("export", "fields", "participant_id_hash"), + ("export", "fields", "event_type"), + ], ) def test_schema_must_retain_linking_fields( pilot_payload: dict, section: str, field: str, missing: str @@ -42,7 +50,9 @@ def test_schema_must_retain_linking_fields( validate_pilot_contract(pilot_payload) -def test_linkage_does_not_require_demographics_or_secondary_outcomes(pilot_payload: dict) -> None: +def test_linkage_does_not_require_demographics_or_secondary_outcomes( + pilot_payload: dict, +) -> None: pilot_payload["participants"]["requiredFields"] = ["participant_id_hash"] pilot_payload["outcomes"]["secondary"] = [] validate_pilot_contract(pilot_payload) diff --git a/tests/test_pipeline_generated_outputs.py b/tests/test_pipeline_generated_outputs.py index 5f1b86b..ad1e8c7 100644 --- a/tests/test_pipeline_generated_outputs.py +++ b/tests/test_pipeline_generated_outputs.py @@ -14,15 +14,23 @@ REPO = Path(__file__).resolve().parents[1] - @pytest.mark.slow -def test_build_generated_outputs_returns_existing_artifact_bundle(tmp_path: Path) -> None: +def test_build_generated_outputs_returns_existing_artifact_bundle( + tmp_path: Path, +) -> None: root = tmp_path / "project" shutil.copytree( - REPO, root, + REPO, + root, ignore=shutil.ignore_patterns( - ".git", ".venv", ".local", "__pycache__", ".pytest_cache", - ".mypy_cache", ".ruff_cache", "node_modules", + ".git", + ".venv", + ".local", + "__pycache__", + ".pytest_cache", + ".mypy_cache", + ".ruff_cache", + "node_modules", ), ) bundle = build_generated_outputs(root) @@ -51,16 +59,25 @@ def test_build_generated_outputs_returns_existing_artifact_bundle(tmp_path: Path root / "output" / "figures" / "active_skillference_skill_atlas.png" ) in bundle.paths assert all(path.exists() for path in bundle.paths) - variables = json.loads((root / "output" / "data" / "manuscript_variables.json").read_text()) - registered = json.loads((root / "output" / "figures" / "figure_registry.json").read_text()) - assert int(variables["figure_count"]) == len(registered) == bundle.registered_figures + variables = json.loads( + (root / "output" / "data" / "manuscript_variables.json").read_text() + ) + registered = json.loads( + (root / "output" / "figures" / "figure_registry.json").read_text() + ) + assert ( + int(variables["figure_count"]) == len(registered) == bundle.registered_figures + ) -def test_importable_builder_raises_package_error_for_endorsement_overreach(tmp_path: Path) -> None: +def test_importable_builder_raises_package_error_for_endorsement_overreach( + tmp_path: Path, +) -> None: manuscript = tmp_path / "manuscript" manuscript.mkdir() (manuscript / "00_fixture.md").write_text( - "The platform is NSA approved for all learners.", encoding="utf-8", + "The platform is NSA approved for all learners.", + encoding="utf-8", ) with pytest.raises(ValueError, match="endorsement-overreach"): build_generated_outputs(tmp_path) diff --git a/tests/test_predictive_coding_diagnostics.py b/tests/test_predictive_coding_diagnostics.py index 3e4f333..87cf1c4 100644 --- a/tests/test_predictive_coding_diagnostics.py +++ b/tests/test_predictive_coding_diagnostics.py @@ -3,13 +3,21 @@ import numpy as np import pytest -from src.kernels.predictive_coding import gaussian_posterior_mean, predictive_coding_descent +from src.kernels.predictive_coding import ( + gaussian_posterior_mean, + predictive_coding_descent, +) def test_small_update_can_qualify_far_from_the_analytic_target() -> None: result = predictive_coding_descent( - observation=10.0, prior_mean=0.0, prior_variance=1.0, obs_variance=1.0, - rate=1e-9, steps=3, tolerance=1e-6, + observation=10.0, + prior_mean=0.0, + prior_variance=1.0, + obs_variance=1.0, + rate=1e-9, + steps=3, + tolerance=1e-6, ) target = gaussian_posterior_mean(10.0, 0.0, 1.0, 1.0) assert target == pytest.approx(5.0) @@ -22,8 +30,13 @@ def test_small_update_can_qualify_far_from_the_analytic_target() -> None: def test_no_qualifying_update_retains_steps_sentinel_and_full_trajectory() -> None: result = predictive_coding_descent( - observation=10.0, prior_mean=0.0, prior_variance=1.0, obs_variance=1.0, - rate=0.1, steps=2, tolerance=1e-12, + observation=10.0, + prior_mean=0.0, + prior_variance=1.0, + obs_variance=1.0, + rate=0.1, + steps=2, + tolerance=1e-12, ) np.testing.assert_allclose(result.trajectory, [0.0, 1.0, 1.8]) assert result.steps_to_converge == 2 diff --git a/tests/test_probability_support.py b/tests/test_probability_support.py index 860bf0d..fda0c51 100644 --- a/tests/test_probability_support.py +++ b/tests/test_probability_support.py @@ -40,16 +40,22 @@ def test_extreme_masses_and_policy_scores_preserve_normalization() -> None: np.testing.assert_allclose(normalize([1e308, 1e308]), [0.5, 0.5]) np.testing.assert_allclose(dirichlet_expected([1e308, 1e308]), [0.5, 0.5]) np.testing.assert_allclose( - dirichlet_expected_log([1e308, 1e308]), [-np.log(2.0)] * 2, - atol=1e-12, rtol=0.0, + dirichlet_expected_log([1e308, 1e308]), + [-np.log(2.0)] * 2, + atol=1e-12, + rtol=0.0, ) np.testing.assert_allclose(softmax([1e308, 1e308], precision=2.0), [0.5, 0.5]) - np.testing.assert_allclose(policy_posterior([-1e308, -1e308], gamma=2.0), [0.5, 0.5]) + np.testing.assert_allclose( + policy_posterior([-1e308, -1e308], gamma=2.0), [0.5, 0.5] + ) np.testing.assert_allclose(softmax([-1e308, 1e308], precision=2.0), [0.0, 1.0]) @pytest.mark.parametrize("bad", [np.nan, np.inf, -np.inf]) -@pytest.mark.parametrize("function", [normalize, dirichlet_expected, dirichlet_expected_log]) +@pytest.mark.parametrize( + "function", [normalize, dirichlet_expected, dirichlet_expected_log] +) def test_nonfinite_masses_are_rejected(function, bad: float) -> None: with pytest.raises(ValueError): function([bad, 1.0]) @@ -70,18 +76,34 @@ def test_planning_rejects_invalid_state_beliefs(belief: list[float]) -> None: def test_agent_rejects_invalid_transition_at_its_owning_boundary() -> None: with pytest.raises(ValueError, match="B"): DiscreteActiveInferenceAgent( - A=np.eye(2), B=np.array([[[-1.0], [-1.0]], [[2.0], [2.0]]]), - C=np.array([0.5, 0.5]), D=np.array([0.5, 0.5]), + A=np.eye(2), + B=np.array([[[-1.0], [-1.0]], [[2.0], [2.0]]]), + C=np.array([0.5, 0.5]), + D=np.array([0.5, 0.5]), ) def test_prediction_rejects_out_of_range_actions_and_wrong_belief_dimension() -> None: agent = DiscreteActiveInferenceAgent( - A=np.eye(2), B=np.eye(2)[:, :, None], - C=np.array([0.5, 0.5]), D=np.array([0.5, 0.5]), + A=np.eye(2), + B=np.eye(2)[:, :, None], + C=np.array([0.5, 0.5]), + D=np.array([0.5, 0.5]), ) with pytest.raises(ValueError, match="action"): agent.predict([0.5, 0.5], -1) with pytest.raises(ValueError, match="dimension"): agent.predict([1.0], 0) np.testing.assert_allclose(agent.predict([0.25, 0.75], 0), [0.25, 0.75]) + + +@pytest.mark.parametrize("action", [True, np.bool_(False), 0.5, "0", None]) +def test_prediction_rejects_noninteger_action_types(action) -> None: + agent = DiscreteActiveInferenceAgent( + A=np.eye(2), + B=np.eye(2)[:, :, None], + C=np.array([0.5, 0.5]), + D=np.array([0.5, 0.5]), + ) + with pytest.raises(TypeError, match="action"): + agent.predict([0.5, 0.5], action) diff --git a/tests/test_recommendations.py b/tests/test_recommendations.py index 1304388..043252e 100644 --- a/tests/test_recommendations.py +++ b/tests/test_recommendations.py @@ -50,15 +50,32 @@ def _project() -> dict[str, object]: "quizzes": [ { "quizId": "math_sets_quiz", - "questions": [{"question": "What is a set?", "answers": [{"answer": "A collection", "isCorrect": True}]}], + "questions": [ + { + "question": "What is a set?", + "answers": [{"answer": "A collection", "isCorrect": True}], + } + ], }, { "quizId": "prob_events_quiz", - "questions": [{"question": "How do events relate to sets?", "answers": [{"answer": "Events are sets", "isCorrect": True}]}], + "questions": [ + { + "question": "How do events relate to sets?", + "answers": [{"answer": "Events are sets", "isCorrect": True}], + } + ], }, { "quizId": "bayes_update_quiz", - "questions": [{"question": "How does evidence update belief?", "answers": [{"answer": "Bayesian conditioning", "isCorrect": True}]}], + "questions": [ + { + "question": "How does evidence update belief?", + "answers": [ + {"answer": "Bayesian conditioning", "isCorrect": True} + ], + } + ], }, ], } @@ -150,7 +167,8 @@ def test_invalid_export_graph_cannot_produce_an_unsuccessful_retake(dependency) project = _project() project["dependencies"].append(dependency) result = recommend_next_skills( - project, unsuccessful_skill_ids=["bayes_update"], + project, + unsuccessful_skill_ids=["bayes_update"], ) assert result["available"] is False assert result["recommendations"] == [] diff --git a/tests/test_release_readiness_inventory.py b/tests/test_release_readiness_inventory.py index 0e1e61a..193d9c4 100644 --- a/tests/test_release_readiness_inventory.py +++ b/tests/test_release_readiness_inventory.py @@ -280,7 +280,9 @@ def test_summarize_probe_distinguishes_reachable_and_blocked() -> None: assert _summarize_probe(silent) == "blocked (exit 3)" -def test_public_probe_distinguishes_empty_populated_and_tags_only_local_repositories(tmp_path: Path) -> None: +def test_public_probe_distinguishes_empty_populated_and_tags_only_local_repositories( + tmp_path: Path, +) -> None: from src.release_readiness import probe_public_repository empty = tmp_path / "empty.git" @@ -292,20 +294,29 @@ def test_public_probe_distinguishes_empty_populated_and_tags_only_local_reposito root = _repo_with_dirty_release_state(tmp_path) populated = probe_public_repository(tmp_path, str(root)) assert populated.returncode == 0 - assert any(line.split()[1].startswith("refs/heads/") for line in populated.stdout.splitlines()) + assert any( + line.split()[1].startswith("refs/heads/") + for line in populated.stdout.splitlines() + ) subprocess.run(["git", "-C", str(root), "tag", "fixture_tag"], check=True) branch = subprocess.run( ["git", "-C", str(root), "symbolic-ref", "HEAD"], - check=True, capture_output=True, text=True, + check=True, + capture_output=True, + text=True, ).stdout.strip() subprocess.run(["git", "-C", str(root), "update-ref", "-d", branch], check=True) tags_only = probe_public_repository(tmp_path, str(root)) assert tags_only.returncode == 0 - assert {line.split()[1] for line in tags_only.stdout.splitlines()} == {"refs/tags/fixture_tag"} + assert {line.split()[1] for line in tags_only.stdout.splitlines()} == { + "refs/tags/fixture_tag" + } -def test_public_probe_missing_local_target_is_not_an_empty_success(tmp_path: Path) -> None: +def test_public_probe_missing_local_target_is_not_an_empty_success( + tmp_path: Path, +) -> None: from src.release_readiness import probe_public_repository result = probe_public_repository(tmp_path, str(tmp_path / "missing.git")) diff --git a/tests/test_render_manuscript.py b/tests/test_render_manuscript.py index 20e31af..7addfdc 100644 --- a/tests/test_render_manuscript.py +++ b/tests/test_render_manuscript.py @@ -18,7 +18,10 @@ def test_pdf_receipt_records_finalized_output_and_actual_inputs(pdf_project) -> receipt = target.with_suffix(".receipt.json") receipt.write_text("outdated receipt", encoding="utf-8") - assert renderer._write_pdf_receipt(target, root=root, inputs=pdf_project["inputs"]) == receipt + assert ( + renderer._write_pdf_receipt(target, root=root, inputs=pdf_project["inputs"]) + == receipt + ) content = target.read_bytes() payload = json.loads(receipt.read_text(encoding="utf-8")) @@ -28,9 +31,13 @@ def test_pdf_receipt_records_finalized_output_and_actual_inputs(pdf_project) -> assert payload["sha256"] == hashlib.sha256(content).hexdigest() assert payload["source_revision"] == pdf_project["revision"] expected = { - "manuscript/config.yaml", "manuscript/preamble.md", "manuscript/references.bib", - "output/manuscript/01_fixture.md", "output/manuscript/assets/cover.pdf", - "output/manuscript/assets/atlas.pdf", "output/manuscript/assets/body.pdf", + "manuscript/config.yaml", + "manuscript/preamble.md", + "manuscript/references.bib", + "output/manuscript/01_fixture.md", + "output/manuscript/assets/cover.pdf", + "output/manuscript/assets/atlas.pdf", + "output/manuscript/assets/body.pdf", } rows = payload["inputManifest"]["files"] assert {row["path"] for row in rows} == expected @@ -42,7 +49,9 @@ def test_pdf_receipt_records_finalized_output_and_actual_inputs(pdf_project) -> @pytest.mark.slow -def test_real_render_places_configured_atlas_on_physical_page_two(pdf_project, monkeypatch) -> None: +def test_real_render_places_configured_atlas_on_physical_page_two( + pdf_project, monkeypatch +) -> None: required = ("pandoc", "pandoc-crossref", "xelatex", "bibtex", "pdftotext") missing = [tool for tool in required if shutil.which(tool) is None] if missing: @@ -60,7 +69,9 @@ def test_real_render_places_configured_atlas_on_physical_page_two(pdf_project, m extracted = subprocess.run( ["pdftotext", "-layout", str(target), "-"], - check=True, capture_output=True, text=True, + check=True, + capture_output=True, + text=True, ).stdout.split("\f") assert "FixtureTitlePage" in extracted[0] assert "FixtureAtlasPage" in extracted[1] @@ -68,7 +79,9 @@ def test_real_render_places_configured_atlas_on_physical_page_two(pdf_project, m assert "AtlasGraphic" in extracted[1] assert "FixtureBodySection" not in extracted[1] assert "FixtureBodySection" in "\f".join(extracted[2:]) - receipt = json.loads(target.with_suffix(".receipt.json").read_text(encoding="utf-8")) + receipt = json.loads( + target.with_suffix(".receipt.json").read_text(encoding="utf-8") + ) assert set(receipt["inputManifest"]["assets"]) == { path.relative_to(root).as_posix() for path in pdf_project["assets"] } diff --git a/tests/test_roadmap.py b/tests/test_roadmap.py index bda4662..165ca22 100644 --- a/tests/test_roadmap.py +++ b/tests/test_roadmap.py @@ -673,11 +673,14 @@ def _copy_minimal_source_surfaces(tmp_path: Path) -> None: ) -def test_blocked_task_retains_named_blocker_without_completion_evidence(tmp_path: Path) -> None: +def test_blocked_task_retains_named_blocker_without_completion_evidence( + tmp_path: Path, +) -> None: payload = _valid_roadmap_payload() task = payload["tasks"][1] task.update( - status="blocked", evidence=[], + status="blocked", + evidence=[], notes="Blocked until docker info reported: daemon unavailable.", ) _write_roadmap(tmp_path, payload) @@ -692,10 +695,14 @@ def test_blocked_task_retains_named_blocker_without_completion_evidence(tmp_path _write_roadmap(tmp_path, payload) assert audit_roadmap(tmp_path) == [] assert "[~]" in (tmp_path / "TODO.md").read_text(encoding="utf-8") - assert "daemon unavailable." not in "\n".join(render_todo_detail_pages(payload).values()) + assert "daemon unavailable." not in "\n".join( + render_todo_detail_pages(payload).values() + ) -def test_live_count_diagnostic_keeps_real_line_number_after_fenced_example(tmp_path: Path) -> None: +def test_live_count_diagnostic_keeps_real_line_number_after_fenced_example( + tmp_path: Path, +) -> None: _copy_minimal_source_surfaces(tmp_path) current = build_live_counts(REPO)["skill_count"].value stale = current + 7 @@ -704,7 +711,15 @@ def test_live_count_diagnostic_keeps_real_line_number_after_fenced_example(tmp_p encoding="utf-8", ) - issues = [issue for issue in audit_live_count_literals(tmp_path) if issue.key == "skill_count"] + issues = [ + issue + for issue in audit_live_count_literals(tmp_path) + if issue.key == "skill_count" + ] assert len(issues) == 1 - assert issues[0].line == 6 and issues[0].found == stale and issues[0].expected == current + assert ( + issues[0].line == 6 + and issues[0].found == stale + and issues[0].expected == current + ) diff --git a/tests/test_run_entrypoint.py b/tests/test_run_entrypoint.py index 9df1b0e..35b747a 100644 --- a/tests/test_run_entrypoint.py +++ b/tests/test_run_entrypoint.py @@ -13,10 +13,15 @@ REPO = Path(__file__).resolve().parents[1] -@pytest.mark.parametrize("argument,code", [("--help", 0), ("--unknown", 2), ("--check", 1), ("--offline", 1)]) +@pytest.mark.parametrize( + "argument,code", [("--help", 0), ("--unknown", 2), ("--check", 1), ("--offline", 1)] +) @pytest.mark.parametrize("existing", [False, True]) def test_argument_and_prerequisite_exits_do_not_initialize_postgres_state( - tmp_path: Path, argument: str, code: int, existing: bool, + tmp_path: Path, + argument: str, + code: int, + existing: bool, ) -> None: bash = shutil.which("bash", path=os.defpath) if bash is None: @@ -39,23 +44,37 @@ def test_argument_and_prerequisite_exits_do_not_initialize_postgres_state( state.mkdir() password.write_text("kept-synthetic-password\n", encoding="utf-8") password.chmod(0o600) - before = (password.read_bytes(), password.stat().st_mtime_ns, stat.S_IMODE(password.stat().st_mode)) + before = ( + password.read_bytes(), + password.stat().st_mtime_ns, + stat.S_IMODE(password.stat().st_mode), + ) env = { - key: value for key, value in os.environ.items() + key: value + for key, value in os.environ.items() if key not in {"ACTIVE_SKILLFERENCE_POSTGRES_PASSWORD", "BASH_ENV", "ENV"} and not key.startswith("BASH_FUNC_") } env.update(PATH=str(programs), ACTIVE_SKILLFERENCE_LOCAL_STATE_DIR=str(state)) result = subprocess.run( - [bash, str(script), argument], cwd=tmp_path, env=env, - capture_output=True, text=True, timeout=5, check=False, + [bash, str(script), argument], + cwd=tmp_path, + env=env, + capture_output=True, + text=True, + timeout=5, + check=False, ) assert result.returncode == code, result.stdout + result.stderr if argument in ("--check", "--offline"): assert "uv is not installed" in result.stderr if existing: - assert (password.read_bytes(), password.stat().st_mtime_ns, stat.S_IMODE(password.stat().st_mode)) == before + assert ( + password.read_bytes(), + password.stat().st_mtime_ns, + stat.S_IMODE(password.stat().st_mode), + ) == before else: assert not state.exists() diff --git a/tests/test_scholarship.py b/tests/test_scholarship.py index d57e9c0..94c7c77 100644 --- a/tests/test_scholarship.py +++ b/tests/test_scholarship.py @@ -1389,22 +1389,35 @@ def test_role_gate_rejects_lane_name_padding() -> None: @pytest.mark.parametrize("status,bucket", [(403, "botBlocks"), (404, "deadLinks")]) -def test_online_source_failure_appears_once_in_network_report(tmp_path: Path, httpserver, status: int, bucket: str) -> None: - httpserver.expect_request("/failed-source", method="HEAD").respond_with_data("", status=status) +def test_online_source_failure_appears_once_in_network_report( + tmp_path: Path, httpserver, status: int, bucket: str +) -> None: + httpserver.expect_request("/failed-source", method="HEAD").respond_with_data( + "", status=status + ) _manuscript, matrix_path, bib_path = _write_fixture( - tmp_path, manuscript="Claim [@known].", bib="@article{known,\n}\n", + tmp_path, + manuscript="Claim [@known].", + bib="@article{known,\n}\n", matrix=_health_matrix(httpserver.url_for("/failed-source")), ) - health = audit_source_health(matrix_path, bib_path, online=True, timeout=5.0, allow_private=True) + health = audit_source_health( + matrix_path, bib_path, online=True, timeout=5.0, allow_private=True + ) scholarship = validate_scholarship( - REPO / "manuscript", REPO / "data" / "scholarship_sources.yaml", + REPO / "manuscript", + REPO / "data" / "scholarship_sources.yaml", REPO / "manuscript" / "references.bib", ) gaps = validate_scholarship_gaps( - scholarship, _real_claim_support(), REPO / "manuscript" / "references.bib", + scholarship, + _real_claim_support(), + REPO / "manuscript" / "references.bib", ) - report = scholarship_source_health_network(scholarship=scholarship, source_health=health, gap_report=gaps) + report = scholarship_source_health_network( + scholarship=scholarship, source_health=health, gap_report=gaps + ) assert report["summary"][bucket] == 1 assert [row["key"] for row in report[bucket]] == ["known"] diff --git a/tests/test_screenshot_context.py b/tests/test_screenshot_context.py index 645bde5..30591c9 100644 --- a/tests/test_screenshot_context.py +++ b/tests/test_screenshot_context.py @@ -26,36 +26,82 @@ def _case(httpserver, tmp_path: Path): instance = "capture-run-fixture" status = { "instanceId": instance, - "project": {"exportContentDigest": digest, "counts": {"skills": 1, "subjects": 1, "dependencies": 0, "quizzes": 1}}, + "project": { + "exportContentDigest": digest, + "counts": {"skills": 1, "subjects": 1, "dependencies": 0, "quizzes": 1}, + }, "config": {"project": {"id": "p"}, "skilltree": {"url": url}}, "skilltree": {"available": True}, - "liveSkilltree": {"checked": True, "complete": True, "expectedDependencies": 0, "matchedDependencies": 0, - "content": {"checked": True, "complete": True, "exportContentDigest": digest}}, + "liveSkilltree": { + "checked": True, + "complete": True, + "expectedDependencies": 0, + "matchedDependencies": 0, + "content": { + "checked": True, + "complete": True, + "exportContentDigest": digest, + }, + }, } httpserver.expect_request("/api/status", method="GET").respond_with_json(status) browser = { - "state": "ready", "url": url, "projectId": "p", "route": "/", - "dashboardInstanceId": instance, "exportContentDigest": digest, - "frameUrl": url + "/skills-display/p", "frameVisible": True, "busy": False, + "state": "ready", + "url": url, + "projectId": "p", + "route": "/", + "dashboardInstanceId": instance, + "exportContentDigest": digest, + "frameUrl": url + "/skills-display/p", + "frameVisible": True, + "busy": False, } receipt = { - "checked": True, "ok": True, "projectId": "p", "dashboardInstanceId": instance, + "checked": True, + "ok": True, + "projectId": "p", + "dashboardInstanceId": instance, "exportContentDigest": digest, - "views": [{"kind": kind, "route": route, "state": "ready", "visibleContentChecked": True} - for kind, route in (("project", "/"), ("subject", "/subjects/s"), ("quiz", "/subjects/s/skills/k/quizzes/q"))], + "views": [ + { + "kind": kind, + "route": route, + "state": "ready", + "visibleContentChecked": True, + } + for kind, route in ( + ("project", "/"), + ("subject", "/subjects/s"), + ("quiz", "/subjects/s/skills/k/quizzes/q"), + ) + ], } proof = { - "proofVersion": 2, "ok": True, "mode": "full-parity", "dashboardUrl": url, "skilltreeUrl": url, - "projectId": "p", "runId": instance, "dashboardInstanceId": instance, "exportContentDigest": digest, - "liveContent": {"checked": True, "complete": True, "exportContentDigest": digest}, - "liveDependencies": {"checked": True, "complete": True}, "browserConsole": receipt, + "proofVersion": 2, + "ok": True, + "mode": "full-parity", + "dashboardUrl": url, + "skilltreeUrl": url, + "projectId": "p", + "runId": instance, + "dashboardInstanceId": instance, + "exportContentDigest": digest, + "liveContent": { + "checked": True, + "complete": True, + "exportContentDigest": digest, + }, + "liveDependencies": {"checked": True, "complete": True}, + "browserConsole": receipt, } path = tmp_path / "explicit-proof.json" settings = CaptureSettings(url, tmp_path / "captures", tmp_path, proof_path=path) return settings, status, browser, proof -def test_explicit_matching_v2_proof_is_bound_to_real_capture_time_status(httpserver, tmp_path: Path) -> None: +def test_explicit_matching_v2_proof_is_bound_to_real_capture_time_status( + httpserver, tmp_path: Path +) -> None: settings, _status, browser, proof = _case(httpserver, tmp_path) settings.proof_path.write_text(json.dumps(proof), encoding="utf-8") raw = settings.proof_path.read_bytes() @@ -68,8 +114,24 @@ def test_explicit_matching_v2_proof_is_bound_to_real_capture_time_status(httpser assert settings.proof_path.read_bytes() == raw -@pytest.mark.parametrize("damage", ["failed", "legacy", "url", "skilltree", "project", "digest", "instance", "content", "browser", "dependencies"]) -def test_unrelated_or_incomplete_proof_cannot_certify_current_capture(httpserver, tmp_path: Path, damage: str) -> None: +@pytest.mark.parametrize( + "damage", + [ + "failed", + "legacy", + "url", + "skilltree", + "project", + "digest", + "instance", + "content", + "browser", + "dependencies", + ], +) +def test_unrelated_or_incomplete_proof_cannot_certify_current_capture( + httpserver, tmp_path: Path, damage: str +) -> None: settings, _status, browser, proof = _case(httpserver, tmp_path) if damage == "failed": proof["ok"] = False @@ -99,19 +161,43 @@ def test_unrelated_or_incomplete_proof_cannot_certify_current_capture(httpserver assert settings.proof_path.read_bytes() == before -def test_missing_malformed_or_implicit_proof_is_unassociated_without_rewriting_it(httpserver, tmp_path: Path) -> None: +def test_missing_malformed_or_implicit_proof_is_unassociated_without_rewriting_it( + httpserver, tmp_path: Path +) -> None: settings, _status, browser, proof = _case(httpserver, tmp_path) - assert capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] is False + assert ( + capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] + is False + ) settings.proof_path.write_text("not JSON", encoding="utf-8") - assert capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] is False + assert ( + capture_context(settings, browser_probe=browser)["proofAssociation"]["checked"] + is False + ) assert settings.proof_path.read_text(encoding="utf-8") == "not JSON" settings.proof_path.write_text(json.dumps(proof), encoding="utf-8") - assert capture_context(replace(settings, proof_path=None), browser_probe=browser)["proofAssociation"]["checked"] is False + assert ( + capture_context(replace(settings, proof_path=None), browser_probe=browser)[ + "proofAssociation" + ]["checked"] + is False + ) assert capture_context(settings)["proofAssociation"]["checked"] is False -@pytest.mark.parametrize("field,value", [("state", "fallback"), ("dashboardInstanceId", "old"), ("exportContentDigest", "c" * 64), ("projectId", "other"), ("url", "not a URL")]) -def test_browser_snapshot_must_match_current_status(httpserver, tmp_path: Path, field: str, value: str) -> None: +@pytest.mark.parametrize( + "field,value", + [ + ("state", "fallback"), + ("dashboardInstanceId", "old"), + ("exportContentDigest", "c" * 64), + ("projectId", "other"), + ("url", "not a URL"), + ], +) +def test_browser_snapshot_must_match_current_status( + httpserver, tmp_path: Path, field: str, value: str +) -> None: settings, _status, browser, _proof = _case(httpserver, tmp_path) browser[field] = value with pytest.raises(ScreenshotCaptureError, match="does not match"): @@ -119,26 +205,51 @@ def test_browser_snapshot_must_match_current_status(httpserver, tmp_path: Path, assert not settings.output_dir.exists() -@pytest.mark.parametrize("url", ["file:///tmp/not-a-dashboard", "https://example.invalid", "http://user:password@127.0.0.1", "http://127.0.0.1/?query=yes", "http://127.0.0.1/#fragment"]) -def test_capture_url_admission_is_pure_and_rejects_nonlocal_or_credential_bearing_targets(url: str) -> None: +@pytest.mark.parametrize( + "url", + [ + "file:///tmp/not-a-dashboard", + "https://example.invalid", + "http://user:password@127.0.0.1", + "http://127.0.0.1/?query=yes", + "http://127.0.0.1/#fragment", + ], +) +def test_capture_url_admission_is_pure_and_rejects_nonlocal_or_credential_bearing_targets( + url: str, +) -> None: with pytest.raises(ScreenshotCaptureError): _capture_url(url) -def test_capture_target_comparison_normalizes_origin_and_default_port_without_merging_paths() -> None: - assert _capture_target("HTTP://LOCALHOST:80/") == _capture_target("http://localhost") - assert _capture_target("http://localhost/a") != _capture_target("http://localhost/b") +def test_capture_target_comparison_normalizes_origin_and_default_port_without_merging_paths() -> ( + None +): + assert _capture_target("HTTP://LOCALHOST:80/") == _capture_target( + "http://localhost" + ) + assert _capture_target("http://localhost/a") != _capture_target( + "http://localhost/b" + ) -def test_off_origin_loopback_redirect_is_refused_before_browser_or_output_creation(httpserver, tmp_path: Path) -> None: +def test_off_origin_loopback_redirect_is_refused_before_browser_or_output_creation( + httpserver, tmp_path: Path +) -> None: unrelated = HTTPServer(host="127.0.0.1", port=0) unrelated.start() try: httpserver.expect_request("/", method="GET").respond_with_data( - "", status=302, headers={"Location": unrelated.url_for("/")}, + "", + status=302, + headers={"Location": unrelated.url_for("/")}, + ) + settings = CaptureSettings( + httpserver.url_for(""), tmp_path / "captures", tmp_path ) - settings = CaptureSettings(httpserver.url_for(""), tmp_path / "captures", tmp_path) - with pytest.raises(ScreenshotCaptureError, match="changed the requested origin"): + with pytest.raises( + ScreenshotCaptureError, match="changed the requested origin" + ): capture_screenshots(settings) assert unrelated.log == [] assert not settings.output_dir.exists() diff --git a/tests/test_screenshots.py b/tests/test_screenshots.py index 537f4a6..726673d 100644 --- a/tests/test_screenshots.py +++ b/tests/test_screenshots.py @@ -9,8 +9,8 @@ from __future__ import annotations -import json import hashlib +import json import re import subprocess import sys @@ -174,7 +174,10 @@ def test_captured_images_are_tracked_and_carry_provenance() -> None: image = SCREENSHOT_DIR / view.filename assert image.is_file(), f"{view.filename} is not tracked" assert image.stat().st_size > 0 - assert recorded[view.filename]["sha256"] == hashlib.sha256(image.read_bytes()).hexdigest() + assert ( + recorded[view.filename]["sha256"] + == hashlib.sha256(image.read_bytes()).hexdigest() + ) def test_captured_images_are_readable_and_not_blank() -> None: diff --git a/tests/test_scripts_smoke.py b/tests/test_scripts_smoke.py index a8c960e..50c0567 100644 --- a/tests/test_scripts_smoke.py +++ b/tests/test_scripts_smoke.py @@ -92,8 +92,6 @@ def test_validate_docker_digests_helper_flags_tag_only_lines() -> None: ) - - def test_form_login_password_handoff_avoids_eval_argv_secret() -> None: module = _load_script_module("validate_skilltree_form_login.py") @@ -290,7 +288,7 @@ def test_validate_american_english_script_runs() -> None: def test_validate_web_math_script_runs_with_rendered_html() -> None: web_dir = REPO / "output" / "web" web_dir.mkdir(parents=True, exist_ok=True) - (web_dir / "math_smoke.html").write_text( + (web_dir / "index.html").write_text( """ diff --git a/tests/test_skilltree_client.py b/tests/test_skilltree_client.py index f524614..0545c0f 100644 --- a/tests/test_skilltree_client.py +++ b/tests/test_skilltree_client.py @@ -133,7 +133,9 @@ def test_create_quiz_does_not_duplicate_existing_questions( ).respond_with_json({"quizId": "bayes_quiz"}) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) + ).respond_with_json( + {"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]} + ) assert client.create_quiz(quiz) == {"quizId": "bayes_quiz"} @@ -154,7 +156,9 @@ def test_create_quiz_appends_missing_question_suffix( ).respond_with_json({"quizId": "bayes_quiz"}) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) + ).respond_with_json( + {"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]} + ) httpserver.expect_request( "/admin/quiz-definitions/bayes_quiz/create-question", method="POST", diff --git a/tests/test_skilltree_content.py b/tests/test_skilltree_content.py index dd65224..7bebf6d 100644 --- a/tests/test_skilltree_content.py +++ b/tests/test_skilltree_content.py @@ -13,7 +13,11 @@ from src.dashboard.config import DashboardConfig from src.dashboard.payload import live_skilltree_summary from src.skilltree.client import SkillTreeClient -from src.skilltree.content import SkillTreeContentError, compare_live_content, export_content_digest +from src.skilltree.content import ( + SkillTreeContentError, + compare_live_content, + export_content_digest, +) from src.skilltree.local_auth import ( LocalSkillTreeState, SkillTreeSessionClient, @@ -25,27 +29,38 @@ def _compare(httpserver: HTTPServer, project: SkillTreeProject) -> dict: - client = SkillTreeClient(httpserver.url_for(""), project.project_id, token="synthetic-test-token") + client = SkillTreeClient( + httpserver.url_for(""), project.project_id, token="synthetic-test-token" + ) return compare_live_content( - get=client._get, base_url=client.base_url, project=project.to_export_dict(), + get=client._get, + base_url=client.base_url, + project=project.to_export_dict(), ) def test_matching_actual_content_is_digest_bound_and_answer_key_free( - httpserver: HTTPServer, content_project, skilltree_admin_content, + httpserver: HTTPServer, + content_project, + skilltree_admin_content, ) -> None: skilltree_admin_content(httpserver, content_project) result = _compare(httpserver, content_project) assert result["checked"] is True and result["complete"] is True assert result["projectId"] == "p" - assert result["exportContentDigest"] == export_content_digest(content_project.to_export_dict()) + assert result["exportContentDigest"] == export_content_digest( + content_project.to_export_dict() + ) assert "isCorrect" not in json.dumps(result) assert "synthetic-test-token" not in json.dumps(result) @pytest.mark.parametrize("change", ["answer", "lesson", "extra_question"]) def test_same_counts_cannot_hide_changed_content( - httpserver: HTTPServer, content_project, skilltree_admin_content, change: str, + httpserver: HTTPServer, + content_project, + skilltree_admin_content, + change: str, ) -> None: live = copy.deepcopy(content_project.to_export_dict()) if change == "answer": @@ -61,7 +76,9 @@ def test_same_counts_cannot_hide_changed_content( assert result["subjects"] == 1 and result["skills"] == 1 -def test_count_only_admin_response_is_unverifiable(httpserver: HTTPServer, content_project) -> None: +def test_count_only_admin_response_is_unverifiable( + httpserver: HTTPServer, content_project +) -> None: httpserver.expect_request("/admin/projects/p", method="GET").respond_with_json( {"numSkills": 1, "numSubjects": 1} ) @@ -71,56 +88,82 @@ def test_count_only_admin_response_is_unverifiable(httpserver: HTTPServer, conte @pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) def test_existing_second_question_does_not_suppress_missing_first( - httpserver: HTTPServer, content_project, client_type, + httpserver: HTTPServer, + content_project, + client_type, ) -> None: quiz = content_project.quizzes[0] - httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) - httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( - {"questions": [quiz.questions[1].to_dict()]} - ) httpserver.expect_request( - "/admin/quiz-definitions/q/create-question", method="POST", + "/admin/quiz-definitions/q", method="POST" + ).respond_with_json({}) + httpserver.expect_request( + "/admin/quiz-definitions/q/questions", method="GET" + ).respond_with_json({"questions": [quiz.questions[1].to_dict()]}) + httpserver.expect_request( + "/admin/quiz-definitions/q/create-question", + method="POST", json=quiz.questions[0].to_admin_create_payload(), ).respond_with_json({"id": 1}) client = client_type(httpserver.url_for(""), "p") client.create_quiz(quiz) - posted = [request.get_json() for request, _ in httpserver.log - if request.path.endswith("/create-question")] + posted = [ + request.get_json() + for request, _ in httpserver.log + if request.path.endswith("/create-question") + ] assert posted == [quiz.questions[0].to_admin_create_payload()] httpserver.check_assertions() @pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) def test_extra_question_occurrence_is_refused_without_any_question_write( - httpserver: HTTPServer, content_project, client_type, + httpserver: HTTPServer, + content_project, + client_type, ) -> None: quiz = content_project.quizzes[0] questions = [question.to_dict() for question in quiz.questions] - httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) - httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( - {"questions": questions + [questions[0]]} - ) + httpserver.expect_request( + "/admin/quiz-definitions/q", method="POST" + ).respond_with_json({}) + httpserver.expect_request( + "/admin/quiz-definitions/q/questions", method="GET" + ).respond_with_json({"questions": questions + [questions[0]]}) with pytest.raises(SkillTreeContentError, match="extra or changed"): client_type(httpserver.url_for(""), "p").create_quiz(quiz) - assert not any(request.path.endswith("/create-question") for request, _ in httpserver.log) + assert not any( + request.path.endswith("/create-question") for request, _ in httpserver.log + ) def test_live_summary_never_sends_stored_credentials_to_second_origin( - tmp_path: Path, httpserver: HTTPServer, content_project, + tmp_path: Path, + httpserver: HTTPServer, + content_project, ) -> None: original = httpserver.url_for("").rstrip("/") second = HTTPServer(host="127.0.0.1", port=0) second.start() try: state_dir = tmp_path / "private" - save_local_state(state_dir, LocalSkillTreeState( - admin_username="admin@localhost.invalid", admin_password="stored-secret", - learner_id="learner", project_id="p", skilltree_url=original, - client_secret="trusted-secret", - )) + save_local_state( + state_dir, + LocalSkillTreeState( + admin_username="admin@localhost.invalid", + admin_password="stored-secret", + learner_id="learner", + project_id="p", + skilltree_url=original, + client_secret="trusted-secret", + ), + ) config = DashboardConfig( - project_root=tmp_path, project_id="p", project_name="Project", - skilltree_url=second.url_for(""), local_auth_enabled=True, local_state_dir=state_dir, + project_root=tmp_path, + project_id="p", + project_name="Project", + skilltree_url=second.url_for(""), + local_auth_enabled=True, + local_state_dir=state_dir, ) exported = content_project.to_export_dict() public = load_skilltree_project(tmp_path, export_payload=exported) @@ -135,40 +178,56 @@ def test_live_summary_never_sends_stored_credentials_to_second_origin( def test_bootstrap_target_mismatch_is_refused_before_any_network( - tmp_path: Path, httpserver: HTTPServer, content_project, + tmp_path: Path, + httpserver: HTTPServer, + content_project, ) -> None: with pytest.raises(LocalSkillTreeAuthError, match="bootstrap target"): ensure_local_skilltree_auth( - base_url=httpserver.url_for(""), project_id="different", - project=content_project, state_dir=tmp_path, + base_url=httpserver.url_for(""), + project_id="different", + project=content_project, + state_dir=tmp_path, ) assert httpserver.log == [] @pytest.mark.parametrize("client_type", [SkillTreeClient, SkillTreeSessionClient]) -def test_missing_duplicate_question_occurrence_is_created_once(httpserver, content_project, client_type) -> None: +def test_missing_duplicate_question_occurrence_is_created_once( + httpserver, content_project, client_type +) -> None: quiz = content_project.quizzes[0] first = copy.deepcopy(quiz.questions[0]) quiz.questions[:] = [first, copy.deepcopy(first)] - httpserver.expect_request("/admin/quiz-definitions/q", method="POST").respond_with_json({}) - httpserver.expect_request("/admin/quiz-definitions/q/questions", method="GET").respond_with_json( - {"questions": [first.to_dict()]} - ) httpserver.expect_request( - "/admin/quiz-definitions/q/create-question", method="POST", + "/admin/quiz-definitions/q", method="POST" + ).respond_with_json({}) + httpserver.expect_request( + "/admin/quiz-definitions/q/questions", method="GET" + ).respond_with_json({"questions": [first.to_dict()]}) + httpserver.expect_request( + "/admin/quiz-definitions/q/create-question", + method="POST", json=first.to_admin_create_payload(), ).respond_with_json({"id": 2}) client_type(httpserver.url_for(""), "p").create_quiz(quiz) - posted = [request.get_json() for request, _response in httpserver.log if request.path.endswith("/create-question")] + posted = [ + request.get_json() + for request, _response in httpserver.log + if request.path.endswith("/create-question") + ] assert posted == [first.to_admin_create_payload()] httpserver.check_assertions() @pytest.mark.parametrize("changed", ["lesson", "answer"]) def test_existing_changed_content_is_refused_before_any_seed_write( - httpserver, content_project, skilltree_admin_content, changed: str, + httpserver, + content_project, + skilltree_admin_content, + changed: str, ) -> None: from src.skilltree.local_auth import _seed_project_if_requested @@ -181,10 +240,16 @@ def test_existing_changed_content_is_refused_before_any_seed_write( client = SkillTreeSessionClient(httpserver.url_for(""), "p") with pytest.raises(LocalSkillTreeAuthError, match="not overwritten"): _seed_project_if_requested( - client=client, project=content_project, seed_project=True, seed_dependencies=False, - skip_seed_if_live_content_ready=True, verify_seed=True, + client=client, + project=content_project, + seed_project=True, + seed_dependencies=False, + skip_seed_if_live_content_ready=True, + verify_seed=True, ) - assert httpserver.log and all(request.method == "GET" for request, _response in httpserver.log) + assert httpserver.log and all( + request.method == "GET" for request, _response in httpserver.log + ) @pytest.mark.parametrize( @@ -201,10 +266,16 @@ def test_existing_changed_content_is_refused_before_any_seed_write( ], ) def test_native_quiz_settings_preserve_full_question_completion( - httpserver, content_project, skilltree_admin_content, settings, complete, + httpserver, + content_project, + skilltree_admin_content, + settings, + complete, ) -> None: skilltree_admin_content( - httpserver, content_project, quiz_settings={"q": settings}, + httpserver, + content_project, + quiz_settings={"q": settings}, ) result = _compare(httpserver, content_project) assert result["complete"] is complete @@ -214,13 +285,16 @@ def test_native_quiz_settings_preserve_full_question_completion( def test_export_percentage_is_compared_with_native_correct_question_count( - httpserver, content_project, skilltree_admin_content, + httpserver, + content_project, + skilltree_admin_content, ) -> None: from dataclasses import replace content_project.quizzes[0] = replace(content_project.quizzes[0], passing_percent=50) skilltree_admin_content( - httpserver, content_project, + httpserver, + content_project, quiz_settings={"q": [{"setting": "quizPassingReq", "value": "1"}]}, ) assert _compare(httpserver, content_project)["complete"] is True @@ -241,10 +315,15 @@ def test_export_percentage_is_compared_with_native_correct_question_count( ], ) def test_unverifiable_native_quiz_settings_fail_closed( - httpserver, content_project, skilltree_admin_content, settings, + httpserver, + content_project, + skilltree_admin_content, + settings, ) -> None: skilltree_admin_content( - httpserver, content_project, quiz_settings={"q": settings}, + httpserver, + content_project, + quiz_settings={"q": settings}, ) with pytest.raises(SkillTreeContentError): _compare(httpserver, content_project) diff --git a/tests/test_skilltree_local_auth.py b/tests/test_skilltree_local_auth.py index 47d24c5..1a33c5a 100644 --- a/tests/test_skilltree_local_auth.py +++ b/tests/test_skilltree_local_auth.py @@ -748,7 +748,9 @@ def test_admin_create_quiz_appends_missing_question_suffix( ).respond_with_json({}) httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz/questions", method="GET" - ).respond_with_json({"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]}) + ).respond_with_json( + {"quizType": "Quiz", "questions": [quiz.questions[0].to_dict()]} + ) httpserver.expect_request( "/admin/quiz-definitions/m_sets_quiz/create-question", method="POST", @@ -822,24 +824,39 @@ def _tiny_project() -> SkillTreeProject: ) ], dependencies=[], - quizzes=[Quiz( - quiz_id="m_sets_quiz", name="Sets Quiz", - questions=[QuizQuestion( - question="Which collection is a set?", - answers=[ - {"answer": "A collection of distinct elements", "isCorrect": True}, - {"answer": "An ordered list requiring repetition", "isCorrect": False}, + quizzes=[ + Quiz( + quiz_id="m_sets_quiz", + name="Sets Quiz", + questions=[ + QuizQuestion( + question="Which collection is a set?", + answers=[ + { + "answer": "A collection of distinct elements", + "isCorrect": True, + }, + { + "answer": "An ordered list requiring repetition", + "isCorrect": False, + }, + ], + ) ], - )], - )], + ) + ], ) def test_ensure_local_skilltree_auth_full_first_bootstrap( - tmp_path: Path, httpserver: HTTPServer, skilltree_admin_content, + tmp_path: Path, + httpserver: HTTPServer, + skilltree_admin_content, ) -> None: _bootstrap_httpserver_expectations( - httpserver, skilltree_admin_content, needs_bootstrap=True, + httpserver, + skilltree_admin_content, + needs_bootstrap=True, ) base = httpserver.url_for("").rstrip("/") @@ -865,7 +882,8 @@ def test_ensure_local_skilltree_auth_full_first_bootstrap( assert stored.client_secret == "bootstrap-secret" assert stored.skilltree_url == base question_writes = [ - request.get_json() for request, _response in httpserver.log + request.get_json() + for request, _response in httpserver.log if request.method == "POST" and request.path.endswith("/create-question") ] assert question_writes == [ @@ -947,7 +965,9 @@ def test_ensure_local_skilltree_auth_rejects_state_url_mismatch_before_login( def test_ensure_local_skilltree_auth_skips_seed_when_live_content_matches( - tmp_path: Path, httpserver: HTTPServer, skilltree_admin_content, + tmp_path: Path, + httpserver: HTTPServer, + skilltree_admin_content, ) -> None: base = httpserver.url_for("").rstrip("/") save_local_state( @@ -1299,7 +1319,8 @@ def test_live_skill_count_rejects_missing_num_skills(httpserver: HTTPServer) -> def test_verify_seed_landed_passes_when_live_matches( - httpserver: HTTPServer, skilltree_admin_content, + httpserver: HTTPServer, + skilltree_admin_content, ) -> None: skilltree_admin_content(httpserver, _two_skill_project()) client = SkillTreeSessionClient(httpserver.url_for("/"), "p") diff --git a/tests/test_skilltree_transport_boundaries.py b/tests/test_skilltree_transport_boundaries.py index 329b94a..daa21f5 100644 --- a/tests/test_skilltree_transport_boundaries.py +++ b/tests/test_skilltree_transport_boundaries.py @@ -18,16 +18,22 @@ @pytest.mark.parametrize("method", ["GET", "POST"]) -def test_bearer_read_timeout_is_bounded_and_domain_typed(httpserver: HTTPServer, method: str) -> None: +def test_bearer_read_timeout_is_bounded_and_domain_typed( + httpserver: HTTPServer, method: str +) -> None: release = threading.Event() + def stalled_body(): yield b"{" release.wait() yield b"}" + httpserver.expect_request("/slow", method=method).respond_with_handler( lambda request: Response(stalled_body(), content_type="application/json") ) - client = SkillTreeClient(httpserver.url_for(""), "p", token="private-token", timeout=0.05) + client = SkillTreeClient( + httpserver.url_for(""), "p", token="private-token", timeout=0.05 + ) operation = client._get if method == "GET" else client._post with ThreadPoolExecutor(max_workers=1) as pool: future = pool.submit(operation, "/slow") @@ -45,9 +51,12 @@ def test_bearer_timeout_must_be_finite_and_positive(timeout: float) -> None: SkillTreeClient("http://127.0.0.1:1", "p", timeout=timeout) -def test_session_malformed_json_is_domain_typed_and_secret_safe(httpserver: HTTPServer) -> None: +def test_session_malformed_json_is_domain_typed_and_secret_safe( + httpserver: HTTPServer, +) -> None: httpserver.expect_request("/broken", method="GET").respond_with_data( - '{"secret": "private-response-fragment"', content_type="application/json", + '{"secret": "private-response-fragment"', + content_type="application/json", ) client = SkillTreeSessionClient(httpserver.url_for(""), "p") with pytest.raises(SkillTreeClientError, match="malformed JSON") as error: @@ -57,12 +66,15 @@ def test_session_malformed_json_is_domain_typed_and_secret_safe(httpserver: HTTP def test_token_malformed_json_is_auth_domain_error(httpserver: HTTPServer) -> None: httpserver.expect_request("/oauth/token", method="POST").respond_with_data( - '{"access_token": "private-fragment"', content_type="application/json", + '{"access_token": "private-fragment"', + content_type="application/json", ) with pytest.raises(LocalSkillTreeAuthError, match="malformed JSON") as error: exchange_trusted_client_token( - base_url=httpserver.url_for(""), project_id="p", - client_secret="private-secret", learner_id="learner", + base_url=httpserver.url_for(""), + project_id="p", + client_secret="private-secret", + learner_id="learner", ) assert "private-fragment" not in str(error.value) assert "private-secret" not in str(error.value) @@ -70,17 +82,23 @@ def test_token_malformed_json_is_auth_domain_error(httpserver: HTTPServer) -> No def test_token_socket_timeout_is_domain_typed(httpserver: HTTPServer) -> None: release = threading.Event() + def stalled_body(): yield b"{" release.wait() yield b"}" + httpserver.expect_request("/oauth/token", method="POST").respond_with_handler( lambda request: Response(stalled_body(), content_type="application/json") ) with ThreadPoolExecutor(max_workers=1) as pool: future = pool.submit( - exchange_trusted_client_token, base_url=httpserver.url_for(""), project_id="p", - client_secret="synthetic-secret", learner_id="learner", timeout=0.05, + exchange_trusted_client_token, + base_url=httpserver.url_for(""), + project_id="p", + client_secret="synthetic-secret", + learner_id="learner", + timeout=0.05, ) try: with pytest.raises(SkillTreeClientError, match="timed out"): diff --git a/tests/test_zenodo.py b/tests/test_zenodo.py index d0c14c9..611efe4 100644 --- a/tests/test_zenodo.py +++ b/tests/test_zenodo.py @@ -24,7 +24,10 @@ def api(httpserver, monkeypatch): def _draft(identifier: int, title: str = "Wanted") -> dict: return { - "id": identifier, "title": title, "state": "unsubmitted", "submitted": False, + "id": identifier, + "title": title, + "state": "unsubmitted", + "submitted": False, "metadata": {"prereserve_doi": {"doi": DOI}}, "links": {"html": "http://localhost/draft"}, } @@ -41,14 +44,18 @@ def http_request(self, request): assert hasattr(stream, "read") and not stream.closed assert not isinstance(stream, (bytes, bytearray)) assert stream.tell() == 0 - assert request.get_header("Content-length") == str(os.fstat(stream.fileno()).st_size) + assert request.get_header("Content-length") == str( + os.fstat(stream.fileno()).st_size + ) streams.append(stream) return request https_request = http_request # A request processor observes the real urllib transfer; no response is faked. - monkeypatch.setattr(urllib.request, "_opener", urllib.request.build_opener(ObserveRequest())) + monkeypatch.setattr( + urllib.request, "_opener", urllib.request.build_opener(ObserveRequest()) + ) return streams @@ -58,7 +65,9 @@ def _upload_bucket(api) -> None: ) -def test_upload_streams_open_file_with_exact_length_and_closes_it(api, tmp_path: Path, observed_upload) -> None: +def test_upload_streams_open_file_with_exact_length_and_closes_it( + api, tmp_path: Path, observed_upload +) -> None: content = b"actual streamed archive bytes\n" * 32768 path = tmp_path / "artifact.zip" path.write_bytes(content) @@ -70,62 +79,114 @@ def receive(request): assert request.get_data() == content return Response(json.dumps({"key": path.name}), content_type="application/json") - api.expect_request("/bucket/fixture/artifact.zip", method="PUT").respond_with_handler(receive) - assert zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=path) == path.name + api.expect_request( + "/bucket/fixture/artifact.zip", method="PUT" + ).respond_with_handler(receive) + assert ( + zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=path) == path.name + ) api.check_assertions() assert len(observed_upload) == 1 and observed_upload[0].closed -@pytest.mark.parametrize("status,body", [(500, "failed"), (200, "not JSON"), (200, "{}")]) +@pytest.mark.parametrize( + "status,body", [(500, "failed"), (200, "not JSON"), (200, "{}")] +) def test_upload_failure_closes_stream_and_never_invents_a_confirmed_filename( - api, tmp_path: Path, observed_upload, status: int, body: str, + api, + tmp_path: Path, + observed_upload, + status: int, + body: str, ) -> None: path = tmp_path / "artifact.zip" path.write_bytes(b"archive") _upload_bucket(api) - api.expect_request("/bucket/fixture/artifact.zip", method="PUT").respond_with_data(body, status=status) + api.expect_request("/bucket/fixture/artifact.zip", method="PUT").respond_with_data( + body, status=status + ) with pytest.raises(zenodo.ZenodoError): zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=path) assert len(observed_upload) == 1 and observed_upload[0].closed -def test_missing_upload_bucket_refuses_before_opening_or_sending_a_file(api, tmp_path: Path) -> None: - api.expect_request("/deposit/depositions/7", method="GET").respond_with_json({"links": {}}) +def test_missing_upload_bucket_refuses_before_opening_or_sending_a_file( + api, tmp_path: Path +) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json( + {"links": {}} + ) with pytest.raises(zenodo.ZenodoError, match="no upload bucket"): - zenodo.upload_deposit_file(token=TOKEN, deposition_id=7, path=tmp_path / "absent.zip") + zenodo.upload_deposit_file( + token=TOKEN, deposition_id=7, path=tmp_path / "absent.zip" + ) assert [request.method for request, _response in api.log] == ["GET"] -def test_metadata_preserves_declared_identity_but_excludes_non_api_creator_fields(tmp_path: Path) -> None: +def test_metadata_preserves_declared_identity_but_excludes_non_api_creator_fields( + tmp_path: Path, +) -> None: path = tmp_path / ".zenodo.json" - path.write_text(json.dumps({ - "title": "Fixture", "description": "Declared description", "license": "MIT", - "version": "", "creators": [{"name": "Fixture Author", "affiliation": "Lab", "email": "private@example.org"}], - })) + path.write_text( + json.dumps( + { + "title": "Fixture", + "description": "Declared description", + "license": "MIT", + "version": "", + "creators": [ + { + "name": "Fixture Author", + "affiliation": "Lab", + "email": "private@example.org", + } + ], + } + ) + ) metadata = zenodo.zenodo_metadata(path) assert metadata["title"] == "Fixture" and metadata["license"] == "mit" assert metadata["creators"] == [{"name": "Fixture Author", "affiliation": "Lab"}] - assert "version" not in metadata and "private@example.org" not in json.dumps(metadata) + assert "version" not in metadata and "private@example.org" not in json.dumps( + metadata + ) @pytest.mark.parametrize("payload", [{}, {"id": 7}, {**_draft(7), "submitted": True}]) -def test_reservation_rejects_missing_identity_or_doi_and_already_published_result(api, payload) -> None: +def test_reservation_rejects_missing_identity_or_doi_and_already_published_result( + api, payload +) -> None: api.expect_request("/deposit/depositions", method="POST").respond_with_json(payload) with pytest.raises(zenodo.ZenodoError): zenodo.reserve_doi(token=TOKEN, metadata={"title": "Fixture"}) def test_doi_mismatch_sends_no_publication_post(api) -> None: - api.expect_request("/deposit/depositions/7", method="GET").respond_with_json(_draft(7)) + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json( + _draft(7) + ) with pytest.raises(zenodo.ZenodoError, match="refusing to publish"): - zenodo.publish_deposit(token=TOKEN, deposition_id=7, expected_doi="10.5072/other") + zenodo.publish_deposit( + token=TOKEN, deposition_id=7, expected_doi="10.5072/other" + ) assert [request.method for request, _response in api.log] == ["GET"] -def test_explicit_matching_doi_allows_only_the_requested_loopback_publication(api) -> None: - api.expect_request("/deposit/depositions/7", method="GET").respond_with_json(_draft(7)) - api.expect_request("/deposit/depositions/7/actions/publish", method="POST").respond_with_json( - {"doi": DOI, "state": "done", "submitted": True, "links": {"record_html": api.url_for("/record/7")}} +def test_explicit_matching_doi_allows_only_the_requested_loopback_publication( + api, +) -> None: + api.expect_request("/deposit/depositions/7", method="GET").respond_with_json( + _draft(7) + ) + api.expect_request( + "/deposit/depositions/7/actions/publish", method="POST" + ).respond_with_json( + { + "doi": DOI, + "state": "done", + "submitted": True, + "links": {"record_html": api.url_for("/record/7")}, + } ) result = zenodo.publish_deposit(token=TOKEN, deposition_id=7, expected_doi=DOI) assert result["doi"] == DOI and result["submitted"] and result["state"] == "done"