Pierre Sassoulas

Thank you for the article @Privat33r-dev > Not all scenarios will be clear-cut for which pattern to use; use your best judgment to choose between these two styles. For a...

Hey @DervishD thank you opening an issue. Are you able to provide a dockerfile based on python:3.11 to reproduce ? For example: [3.11.6-bookworm](https://github.com/docker-library/python/blob/35d09c044857f7aef2bf24791027f3e3fe2c34dd/3.11/bookworm/Dockerfile) Or if this is windows specific: [3.11.6-windowsservercore-ltsc2022](https://github.com/docker-library/python/blob/35d09c044857f7aef2bf24791027f3e3fe2c34dd/3.11/windows/windowsservercore-ltsc2022/Dockerfile)...

Sorry, I should have said python 3.12, my mistake. The way to do that would be to install docker and then you can create an almost empty docker file ```dockerfile...

Thank you for the tip and all your effort to reproduce @DervishD, I can reproduce with: ``a.py`` ```python import dataclasses @dataclasses.dataclass class SomeDataClass(): ... ``` Command: ``pylint a.py --allow-global-unused-variables=no``

I think the ``*`` and dot are inverted in ``*.``, so it's "a ``*`` then one character", instead of "any characters 0 to n times" (``.*``). Maybe we also need...

Does this need reviews or do you need to polish some things still @rhyn0 ? I would be glad to have this in 3.2.0 😄

What is your version of pydantic ?

Feel free to close again Marc, I just wanted to make sure the discussion was not lost. (Type annotation taken into account issue : #4813 )

I think we decided at some point that explicitely failing is better than silentely failing the inference and having a poorer result without any explanation. At least here it's possible...

Thought about it some more, I think we need to transform all crashes into fatal level messages. There's no reason to prevent the remainder of the analysis to take place...