Suppress SyntaxWarnings when parsing modules

[cdce8p]

Description

Following the conclusion in pylint-dev/pylint#9322, ignore SyntaxWarnings when parsing modules with ast.parse.

[jacobtylerwalls]

What do you think about ignoring SyntaxWarning only if SyntaxWarning is missing from the warning filters registry, as it is by default? That would allow users to exercise control over the warnings.

[cdce8p]

What do you think about ignoring SyntaxWarning only if SyntaxWarning is missing from the warning filters registry, as it is by default? That would allow users to exercise control over the warnings.

Not sure it's worth the effort. For Home Assistant the solution should be start linting all requirements as well.
What I don't like about it currently is that modifying the warnings filter for each file is quite a performance hit. Maybe it's better to do that globally?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (e26f097) 92.75% compared to head (591c848) 92.75%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2386   +/-   ##
=======================================
  Coverage   92.75%   92.75%           
=======================================
  Files          94       94           
  Lines       11066    11070    +4     
=======================================
+ Hits        10264    10268    +4     
  Misses        802      802           

Flag	Coverage Δ
linux	92.56% <100.00%> (+<0.01%)	⬆️
pypy	90.88% <75.00%> (-0.01%)	⬇️
windows	92.34% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
astroid/builder.py	94.22% <100.00%> (+0.10%)	⬆️

[cdce8p]

It doesn't seem to make a difference performance wise. The first observation was probably a mistake. Never the less, I do like this version a bit more if I'm honest.

The PR would be ready IMO.

[jacobtylerwalls]

Great, thank you!

Next step would be to update pylint-dev/pylint@2034874 as part of the astroid 3.1 upgrade.

@Pierre-Sassoulas do you want to move forward on cutting astroid 3.1? I think leaving the numpy workaround in #2375 is alright for now.

[cdce8p]

@Pierre-Sassoulas do you want to move forward on cutting astroid 3.1?

I think a new release would make sense. This, together with the fix for the ValueError, is one of the last blockers before I can upgrade Home Assistant to use 3.12 for the pylint check.

[Pierre-Sassoulas]

Sure we were planning on releasing with Jacob. I was trying to find the time to check numpy pyi consistency against our astroid numpy's brain. (I don't remember the issue and I'm on mobile). It Can probably be dealt with later.