test: update jinja rule fixtures #1925
Conversation
![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 suggestions.
Reply with @codex fix comments to fix any unresolved comments.
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, or 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 fix this CI failure" or "@codex address that feedback".
| fn find_raw_at_src_idx(segment: &ErasedSegment, src_idx: usize) -> Option<ErasedSegment> { | ||
| if segment.segments().is_empty() { | ||
| return None; | ||
| } | ||
| for seg in segment.segments() { | ||
| if let Some(pos_marker) = seg.get_position_marker() { | ||
| let src_slice = pos_marker.source_slice.clone(); | ||
| if src_slice.end <= src_idx { | ||
| continue; | ||
| } | ||
| if seg.segments().is_empty() { | ||
| return Some(seg.clone()); |
There was a problem hiding this comment.
[P1] Anchor Jinja fixes to the wrong segment when no AST node exists
The helper used by JJ01 to locate the segment for a raw Jinja slice only checks source_slice.end <= src_idx before returning a leaf. When a Jinja tag does not produce any SQL tokens (e.g. control tags like {% if %} or {% endif %}), there is no segment whose source_slice covers that tag. find_raw_at_src_idx therefore returns the first segment that follows the tag rather than failing, which means the subsequent LintFix uses the templated slice of an unrelated SQL token. Applying the fix will either modify the wrong templated region or fail due to mismatched templated/source slices. The search should ensure src_slice.start <= src_idx < src_slice.end and return None otherwise so that fixes target the correct segment.
Useful? React with 👍 / 👎.
Benchmark for 09d60e2Click to view benchmark
|
benfdking
force-pushed
the
codex/add-jinja-whitespace-rule-jj01
branch
from
ebebd4a to
63d8f16
Compare
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
This PR is being reviewed by Cursor Bugbot
Details
You are on the Bugbot Free tier. On this plan, Bugbot will review limited PRs each billing cycle.
To receive Bugbot reviews on all of your PRs, please upgrade to Bugbot Pro by visiting the Cursor dashboard. Your first 14 days will be free!
| main[pos + inner.len()..].to_string(), | ||
| post.to_string(), | ||
| ) | ||
| } |
There was a problem hiding this comment.
Bug: Jinja Tag Parsing Fails on Short Strings
The get_whitespace_ends function assumes a minimum length for Jinja tags when performing slicing, but its input validation only checks for opening and closing braces. This can cause panics for short strings like "{}" or "{}}" due to invalid slice indices.
Benchmark for 0cbca49Click to view benchmark
|
|
Looks like there are a few issues preventing this PR from being merged!
If you'd like me to help, just leave a comment, like Feel free to include any additional details that might help me get this PR into a better state. You can manage your notification settings |
No description provided.