You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have checked there are no open bugs referencing the same bug or problem
Description
On a PR that updated a dependency version in our repo (Pydantic v1 to v2) and migrated the codebase to the new version's syntax, the initial "Summary by Sourcery" generated by the sourcery-ai bot was unusually verbose and contained a bunch of points unrelated to the changes in the PR.
See some examples highlighted below:
Not sure if the extra info was somehow being extracted from Pydantic's own release notes, or somewhere else?
Upon re-generating the summary using @sourcery-ai summary, the second attempt was much shorter and closer to what I expected:
Summary by Sourcery
Upgrade the project to Pydantic v2, updating query models to use the new syntax and methods. This includes changes to type annotations, validation logic, and dependency updates.
Enhancements:
Improve validation of the is_control parameter by converting valid string values to boolean using BeforeValidator.
Use model_validator instead of root_validator for model validation.
Replace constr with Annotated and StringConstraints for type hinting and validation.
Build:
Upgrade Pydantic to version 2.10.6 and add required dependencies, including annotated-types, pydantic-settings, pydantic_core, and python-dotenv.
Update the typing extensions to 4.12.2.
Tests:
Add tests to ensure that valid is_control values are correctly parsed as booleans.
Any advice on how to prevent this from happening in the future would be appreciated!
Side question: is there any way to tell Sourcery to replace existing PR summaries/reviewer's guides/etc. that it has generated when the bot is re-triggered manually, instead of always adding a new comment or text block on top of what's already there? This could prevent PRs from becoming unnecessarily long and verbose. Can open a separate issue for this if needed.
Code snippet that reproduces issue
Debug Information
We are using the sourcery-ai GitHub app.
IDE Version:
Sourcery Version:
Operating system and Version:
The text was updated successfully, but these errors were encountered:
Checklist
Description
On a PR that updated a dependency version in our repo (Pydantic v1 to v2) and migrated the codebase to the new version's syntax, the initial "Summary by Sourcery" generated by the sourcery-ai bot was unusually verbose and contained a bunch of points unrelated to the changes in the PR.
See some examples highlighted below:

Not sure if the extra info was somehow being extracted from Pydantic's own release notes, or somewhere else?
Upon re-generating the summary using
@sourcery-ai summary
, the second attempt was much shorter and closer to what I expected:Any advice on how to prevent this from happening in the future would be appreciated!
Side question: is there any way to tell Sourcery to replace existing PR summaries/reviewer's guides/etc. that it has generated when the bot is re-triggered manually, instead of always adding a new comment or text block on top of what's already there? This could prevent PRs from becoming unnecessarily long and verbose. Can open a separate issue for this if needed.
Code snippet that reproduces issue
Debug Information
We are using the sourcery-ai GitHub app.
IDE Version:
Sourcery Version:
Operating system and Version:
The text was updated successfully, but these errors were encountered: