Repository navigation
Add Python metrics utilities and dependency tooling - #6
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 automated review suggestions for this pull request.
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".
| satisfies = result.holds | ||
| if max_ratio is not None and result.violation_ratio > max_ratio: | ||
| satisfies = False | ||
| if max_violations is not None and result.violating_rows > max_violations: | ||
| satisfies = False |
There was a problem hiding this comment.
The dependency_summary helper always seeds satisfies with result.holds, so any dependency that contains even a single violation is reported as False before thresholds are considered. Because violation_ratio and violating_rows are both zero whenever result.holds is True, the max_ratio and max_violations comparisons can never flip a result from False to True; the parameters are effectively ignored. As written, callers cannot use max_ratio or max_violations to permit small numbers of violations, which is what the docstring implies. This makes the threshold arguments useless and prevents summarizing “almost satisfied” dependencies.
Useful? React with 👍 / 👎.
Summary
Testing
https://chatgpt.com/codex/tasks/task_e_68fade175cfc8320983331f681b8bc45