NRCL-99 Match MLB propositions and expose settlement differences - #33
Conversation
📚 Documentation Status✅ Code changes detected
This comment is automatically generated by the documentation workflow. |
|
Linked to Plane Work Item(s) References This comment was auto-generated by Plane |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5c61d8ae28
ℹ️ About Codex in GitHub
Codex has been enabled to automatically 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 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review Review repaired head ded734c. Addressed all three findings from the previous official review: fixture-scope evidence now requires fixture-prefixed native event/market/outcome identities and canonical event identity when known; source URLs validate DNS/IPv6 hostname and numeric port; text fields reject lone surrogates before UTF-8 transport. Added targeted regression cases and documented the boundary. No wire fields or existing StrategySpec/contracts schemas changed. Exact-head validation passed: 366 tests passed, 6 skipped; ruff; explicit changed-file Black; mypy across 79 source files; docs/examples validators; docs-site typecheck/build; seven-case offline demo. Repository-wide Black still selects no files due its pre-existing include pattern, so changed files were checked explicitly. Novig fixture provenance was already the exact MLB Winner Series PDF path with temporary query parameters omitted, bound to SHA-256 a60394032c519301b91a452c998ef4e77c5153ff92e4cee29d6d3d949b30b7dc. It does not attach that digest to the contract-directory HTML. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: ded734c6cb
ℹ️ About Codex in GitHub
Codex has been enabled to automatically 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 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review Review head 391b7f5. The latest official finding is fixed: bracketed source authorities always use IPv6Address rather than DNS validation after urlsplit. Added the exact https://[v1.foo]/rules rejection regression. No schema or adapter interface changes. Exact-head checks passed: 367 tests passed, 6 skipped; Ruff, explicit changed-file Black, mypy79 files, docs/examples validators, docs-site typecheck/build, seven-case offline demo. PR body updated. Merge remains separately gated. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 391b7f583b
ℹ️ About Codex in GitHub
Codex has been enabled to automatically 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 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review Review head a7ee8f0. The latest official URL finding is fixed by forbidding IPv6 zone identifiers in rule provenance URLs. Added literal and percent-encoded scope regressions; ordinary DNS and unscoped IPv6 remain supported. No schema/interface changes. Exact-head validation passed: 369 tests passed, 6 skipped; Ruff, explicit changed-file Black, mypy (79 files), docs/examples validators, docs-site typecheck/build, and seven-case offline demo. Body updated. Merge remains separately gated. |
|
Codex Review: Didn't find any major issues. Keep them coming! Reviewed commit: ℹ️ About Codex in GitHubCodex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback". |
1 participant
Sports markets with the same teams can settle differently. Add a standalone versioned contract that retains native venue IDs and canonical game/outcome identity, then compares the sporting proposition separately from reviewed settlement terms. Missing identities or incomplete rule coverage return unknown; doubleheaders and partial games stay distinct. Existing StrategySpec and neural.contracts v1 remain unchanged.
Seven synthetic scenarios cover ordinary winner, forfeit, postponement, doubleheader, partial-game exclusion and missing identity/rules. Real rule references retain source content digests but deliberately do not certify a real equivalent market pair. Fixture rules require fixture-prefixed native event/market/outcome identities and canonical event identity when known. Strict wire validation accepts DNS and unscoped IPv6 source authorities, rejects IPv6 zone identifiers and text that cannot encode as UTF-8.
Validation at a7ee8f0: 369 tests passed, 6 skipped; Ruff, explicit changed-file Black, mypy (79 files), documentation/example validators, docs-site typecheck/build and offline demo passed. Existing repository Black include pattern selects no files; changed Python files were checked explicitly.
Plane: https://app.plane.so/intelligent-intellectual-property/projects/7ffb1939-48c5-4a7a-a4be-6cd283fc8274/issues/730de44e-7f65-4c6b-95f7-3501f4e00d71
No live access, orders, deployment or new dependencies. Reviewed term values remain caller-supplied assertions. Merge awaits explicit approval.