Repository navigation
Validate Markdown links and headings from parsed documents. - #139
Conversation
Codex Review SummaryThis 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. |
|
The account paying for this security review has reached its Codex usage limits. The payer can check the Codex usage dashboard. For personal accounts, using credits requires enabling “Use credits for security reviews” in Code review settings. If you do not manage the paying account, contact this repository's admins. |
|
@codex review |
![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: 1ec740a042
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".
Line scans missed rendered links and invented headings inside comments or code. Share the GFM and footnote parser across link, heading and PR section checks, retaining each committed document for repeated links. Resolve repository-root paths and URL escapes once, accept source-file line fragments, and derive slugs from rendered heading text. Require visible prose alongside math-only or ruby fallback section content.
1ec740a to
4921dcb
Compare
|
@codex review |
|
Codex Review: Didn't find any major issues. 👍 Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
The PR checker uses one retained GitHub-flavored Markdown document for each committed file’s links and heading anchors. Linked images, nested or wrapped link text, parentheses in destinations, footnotes and long lines follow the parser’s structure. Comments, code spans and fenced or indented code do not become links or headings.
Repository-root paths resolve within the committed tree. Destinations use the renderer’s Markdown punctuation and entity decoding before URL escapes are decoded once. Links to the repository root are valid. Non-Markdown files accept GitHub line fragments such as
#L10-L20; the checker validates their syntax without claiming those lines exist. Malformed local URLs and unsupported source-file fragments produce attributable findings.Why
The old regular expressions truncated destinations, skipped rendered links and stopped silently at a long line. A separate line scan counted headings hidden in comments or code and dropped inline-code text from slugs. Links and Markdown headings now come from the same parsed tree, and heading slugs use rendered text, including code, emphasis, entities and link text. The parser configuration also matches the PR section checker’s GFM and footnote handling.
The required PR sections still need visible explanatory content. Ruby fallback text and math-only text do not satisfy that requirement; ordinary prose alongside a formula does. Math is recognized across inline formatting, including GitHub’s dollar/backtick delimiters and fenced math. Ordinary literal code and explicit dollar-sign spans remain visible. The command, flags and finding severities are unchanged. External network availability is not checked, and explicit HTML anchor support is unchanged.
Documentation
The contribution guide describes parsed link and heading checks, root paths, escape handling, source-file fragments and visible prose requirements. Regression cases use committed temporary repositories and independently assert destinations, anchors and reported failures.