Repository navigation
Release 2026-10-27: Arena Studio v0.99 - #279
Merged
Merged
Conversation
--runlog was parsed but never read: it only relabelled the reconstructed step edges as "commanded". Now, with a run log: - /arena/stimulus markers sit at each trialParams' logged send time (labelled commanded); qc step windows use the controller's first displayed frame after it. - Clock: the run log's PC clock. Each TIFF is anchored by one point, its first DO1 pulse = the controller's first displayed frame for that color (controller t_us -> PC ms by a fit of the executed trialParams against their send times). The ScanImage header epoch is no longer used for placement: on 2026-10-16 it ran 531-671 ms ahead of the first frame, different per grab. The per-file correction, display latency and an end check (last DO1 vs allOff + latency) go to /run/provenance. - /arena/display comes from the controller's displayed-frame records (observed) instead of .pat at nominal fps, when the log carries them. - /run/events carries every ScanImage connector command and reply, plus display_start events. Without --runlog the output is unchanged, except that it no longer claims "commanded". tools/check_tiff2mcap.py checks a --runlog MCAP against its log. Bergamo 2026-10-16 run 2: markers at send times (max |delta| 0.000 ms), 13/13 commands + replies, 1909 frames + 4 allOff, end checks +1.3 to +2.8 ms, qc d255 jump -4.6 / +20.3 / +3.5 ms from display_start (violet/green/yellow; blue below the drift, skipped). tests/test_tiff2mcap_runlog.py covers the run-log side on a synthetic log (19 checks). Rebased over the ros2-poc/ → rig/ rename (2026-10-26). Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…eep gateway, composition, viz (LAB-338)
Extraction step of docs/development/rig-software-repos-and-runners.md §3.3 / §7 / §9. The shared
infrastructure under rig/ moved with history to reiserlab/rig-nodes (package set 0.2.1:
rig_interfaces + rig_nodes) and is installed here from that repository's channel.
Removed (moved): rig/interfaces, rig/channel, rig/config (the Zenoh configs ship inside rig_nodes),
rig/nodes/{_common, fictrac_udp_adapter, kinefly_udp_adapter, kinefly_sim, bias_node, bias_sim,
bias_cmd, scanimage_node, scanimage_cmd, recorder, mcap_index, graph_dump, build_interfaces}.py,
rig/tests/si_json_mock.py, rig/sizing, the benches, three rig/docs notes, fictrac-bridge/fictrac_sim.py
(its tests → rig-nodes; the closed-loop-report half stays as tests/test-closed-loop-report.py),
.github/workflows/rig-interfaces.yml (→ rig-nodes package-set.yml).
Rewired: rig.py composes with rig_nodes.supervise and starts nodes as `python -m rig_nodes.<module>`;
bridge_node imports rig_nodes._common / recorder / mcap_index and records the package-set versions
(hello_ack + /arena/session `packages`); mcap_convert imports the CDR decoder from rig_nodes.cdr;
_rows re-exports the FicTrac grammar from rig_nodes.fictrac; foxglove_json is the arena `enrich`
wrapper; the harness drives the installed nodes; root `pixi run sim` delegates to rig/.
rig/pixi.toml: channel raw.githubusercontent.com/reiserlab/rig-nodes/v0.2.1/channel, rig_nodes ==0.2.1,
ros2-distro-mutex 0.19 + Python 3.12 pinned, macOS 14 floor; environments default / typed (+
rig_interfaces, same solve group) / viz / dev (sibling ../../rig-nodes path deps). pixi.lock is
re-generated against the tagged channel in the next commit.
Verified on osx-arm64 against the installed packages (local index of the CI-equivalent build):
harness 59/59 (String) and 62/62 (typed), test-viz, root suite.
Docs: record §9 completion note, CLAUDE.md boundaries, orientation skill, rig/README banner,
WINDOWS-TEST 1b/7b, fictrac-bridge/README.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…x 0.19 + Python 3.12, all three platforms) Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…, PR #274) Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…ng path (fixes `pixi install` on rig PCs) `cd rig && pixi install` failed on any machine without a `../../rig-nodes` checkout (ww18, 2026-10-27): pixi validates every environment's source PATHS before installing ANY environment, so the `dev` feature's path dependencies broke the default install. `pixi install --locked` reported "lock-file not up-to-date", `pixi install -e default` "source path … could not be found". The `dev` feature now uses git source dependencies on reiserlab/rig-nodes `main` (`subdirectory = src/rig_nodes` / `src/rig_interfaces`), pinned to one commit in pixi.lock (75b9986 = the v0.2.1 release commit without the channel payload). Git sources are fetched only when `dev` itself is installed. Verified on the Mac with the sibling hidden and the network blocked (dead proxy, pixi git cache removed, environments removed first): `pixi install --locked` and `pixi install --locked -e typed` succeed and leave pixi.lock byte-identical. Lock diff is confined to the dev block (the two source lines per platform) and to the source packages' build records (uv 0.13.5 → 0.13.6). Dev Zenoh config URIs: with a git source the package is a real (non-editable) install, so the `$CONDA_PREFIX/…/site-packages/rig_nodes/config/*.json5` URIs resolve in `dev` too (they did not with the editable path build — LAB-338 note). Also found: a source build of the same version as the release lands in pixi's package cache under the release's filename and shadows a later channel install on that machine (my default env held the editable 25-file stub while conda-meta showed the channel URL and sha). After purging that entry: default 59/59, typed 62/62 harness checks against the genuine channel packages. The manifest comment, CLAUDE.md, rig/README.md, WINDOWS-TEST.md and the record (§7 item 6, §9) say how to point `dev` at a local checkout without committing it and name the cache gotcha. LAB-338, LAB-303. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Since Studio v0.91 / bridge 3.4 every startClosedLoop config carries start_frame, and the bridge sets offset = (start_frame mod frames) x deg_per_frame for that message. The logged browser config still says offset: 0 and the bridge's "[cfg] applied" line goes to stderr only, so the report predicted every start_frame epoch from frame 0 and flagged 100 % idx mismatch (ww18 2026-10-27, bench_closed_loop_2x10.yaml). The report now replays the config echoes in order the way bridge.py's handler does (apply_config): offset from the message, overridden by start_frame using the same message's pitch and frame count, cleared by the next message's offset (one-shot). Logs without start_frame are unchanged. The table shows "frames @start" and the legend names it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
…rig-nodes (extraction, LAB-338)
Contributor
Author
|
✅ Validated ww18 ( Mac (osx-arm64), candidate SHA Known, not blocking: a plain |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Release candidate
2026-10-27rc1 · candidate commit458c6758fe· built on maina8ef775a96Live on Next: https://reiserlab.github.io/webDisplayTools/next/
ca9f5c431f0c33668729174966e6c2Promotion checklist
pixi run releasechecks)Validated 458c675 …(pixi run tiers -- validate --rig … --by …). Any change to a constituent PR = rebuild (rc+1) = validate again.releaselists recent rig runs — advisory)fictrac-bridge/changed, rig PCsgit pull+ restart the bridgePromote with
pixi run release(merges with a merge commit — never squash — and tagsrelease-2026-10-27). Manifest:releases/2026-10-27.json.🤖 Generated with Claude Code release tooling (
scripts/tiers/candidate.py)