Repository navigation
Prove line-ending normalization and the header/streaming boundaries - #15
Conversation
…position Add safe compatible Agda source, a full Cubical consumer, Lean/Rocq hand re-proofs, finite octet correspondence, audited assumptions and rejecting controls. Qualify normalization lens laws and separate checked mathematics from the unimplemented runtime.
|
Important Draft PR not reviewedDraft PRs are not automatically reviewed by default.
To automatically review draft PRs, update your CodeRabbit configuration: reviews:
auto_review:
drafts: trueThanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. |
|
🟢 clean · commit Advisory: 11 row(s), worst 🔴 high (0 critical, 9 high in grid scope) — reported, never grading the change. Unique graded findings across gating and advisory rows: 36 (36 reporter observations; 20 without sufficient identity to corroborate). Row counts remain per reporter; unique totals do not change gating.
Hypatia scan: 1 finding(s)
panic-attack assail: 18 finding(s)
pons asinorum: 9 finding(s)
echidnabot: not_applicable — template default: enrol once the repository carries proofs and an ECHIDNA engine is configured
rhodibot (RSR compliance): 11 finding(s)
panicbot: not_applicable — panic-attack is already graded by the verdict grid's own panic-attack row, from static-analysis-gate.yml evidence; enrolling panicbot would report one scanner twice and give two rows that can disagree
sustainabot: not_applicable — reserved fleet slot; no adapter exists (gitbot-fleet bots/sustainabot/README.adoc)
glambot (presentation quality): 87 finding(s)
seambot: not_applicable — template default: enrol once a seam register exists at spec/seams/seam-register.json
finishingbot (release readiness): 176 finding(s)
cipherbot: not_applicable — template default: enrol once the repository implements cryptography
accessibilitybot: not_applicable — template default: enrol once the repository serves HTML
Scope: static findings reported by each row's tool at the named commit. The table grades only findings in files this change touches, so the colour answers "does this change introduce anything?", not "what does the whole repository still carry?". Standing findings are counted in each accordion and in report.json; the baseline filter remains the place where acknowledged debt is recorded. The lane state is the worst GATING row; advisory rows are listed and counted but never set it. Purple, red, yellow and blue grade the worst GRADED finding present; orange is evidence missing, malformed or a scanner that did not run; grey is an explicit scope exclusion with its reason. No row is a security certification. report.json, badge.json and the raw findings are in the verdict-grid artifact on the linked run. |
|
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b9256fd609
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| check-line-endings: | ||
| bash scripts/check-line-endings.sh |
There was a problem hiding this comment.
Wire the strict line-ending gate into CI
This recipe is currently only a manual entry point. I checked .github/workflows, the ci recipe, and build/just/proofs.just; none invokes check-line-endings, while proof-check-all only scans the existing prover manifests and therefore never checks verification/proofs/cubical/SholesCubical.agda or runs the new assumption audits and rejection controls. A later regression in those checks can consequently merge without any automated failure, so this recipe should be added to the CI dependency chain or an equivalent workflow step.
Useful? React with 👍 / 👎.
| These are source proofs, not a verified Rust implementation. The initial | ||
| checked work is dated 2026-11-16. Run `bash scripts/check-line-endings.sh` |
There was a problem hiding this comment.
Add these proofs to the canonical proof census
The repository's verification/proofs/README.adoc procedure requires every new obligation to be recorded in docs/status/PROOF-NEEDS.adoc and every completed proof and recheck leg in docs/status/PROOF-STATUS.adoc, but this commit leaves the project-specific needs table empty and the status page reporting 0% proven with “No proofs completed yet.” Readers and tooling that use those canonical status files therefore receive the opposite result from this new checked-work claim, so both census files need entries for the Agda statements and their Lean/Rocq legs.
Useful? React with 👍 / 👎.



Sholes described line-ending proofs at paths that were absent. This adds the canonical token model and checked laws for LF/CR/CRLF normalization, then proves the specific header and chunk-boundary results needed by Atoma and Proven.
verification/proofs/SHOLES.adoc. These are not automatically exported Agda objects.Validation at signed commit
b9256fd6090af9199d7307e2381f7a5e9c5cad44(local files matched the committed tree):--safe --without-K --cubical-compatible --double-check --warning=error.--safe --cubical --double-check --warning=error.coqchk; every named lemma audited without additional axioms.just check-line-endings(orbash scripts/check-line-endings.sh); the runner emits version, source-hash and checker logs.Consumer evidence: at Proven
c13619f9a67a1b9e8234e22e22dae0940d2480cb, actual RefC-compiledhasCRLFreturns true for lone CR/LF and embedded CR, while the compiled Zig export returns false. This proof explains why normalization cannot repair that pair-only detector; it does not claim that the runtime mismatch is fixed.Open boundaries: automatic witness correspondence, executable/compiler refinement, real classifier soundness, buffer/ownership proof and matching Lean4Lean replay. The installed Lean4Lean binary rejected object headers, so that leg is explicitly not passed. No PWG grade or verified runtime claim is made.
The proof sources use the canonical
verification/proofs/tree. The root-shape and AsciiDoc failures introduced during preparation were repaired without weakening gates. Local commit signature checks reportGfor the final head. Companion consumer design and executable diagnostic: hyperpolymath/atoma#26.