Bump actions/setup-python from 2.3.2 to 3

[dependabot]

Bumps actions/setup-python from 2.3.2 to 3.

Release notes

Sourced from actions/setup-python's releases.

v3.0.0

What's Changed

• Update default runtime to node16 (actions/setup-python#340)
• Update package-lock.json file version to 2, @types/node to 16.11.25 and typescript to 4.2.3 (actions/setup-python#341)
• Remove legacy pypy2 and pypy3 keywords (actions/setup-python#342)

Breaking Changes

With the update to Node 16, all scripts will now be run with Node 16 rather than Node 12.

This new major release removes support of legacy pypy2 and pypy3 keywords. Please use more specific and flexible syntax to specify a PyPy version:

jobs:
  build:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        python-version:
        - 'pypy-2.7' # the latest available version of PyPy that supports Python 2.7
        - 'pypy-3.8' # the latest available version of PyPy that supports Python 3.8
        - 'pypy-3.8-v7.3.8' # Python 3.8 and PyPy 7.3.8
    steps:
    - uses: actions/checkout@v2
    - uses: actions/setup-python@v3
      with:
        python-version: ${{ matrix.python-version }}

See more usage examples in the documentation

Commits

• 0ebf233 Remove legacy PyPy input (#342)
• 665cd78 Update lockfileversion (#341)
• 93cb78f Update to node16 (#340)
• See full diff in compare view

You can trigger a rebase of this PR by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[Pierre-Sassoulas]

The "clean way" ™️ to do that is probably to not take the line/col in the functional test for python < 3.8 or pypy. At some point we'll want to have a modern pypy, right ?

[DanielNoord]

I'm looking into this as we speak.

3.6 isn't even developed anymore and the branch has been removed from the main git repo of PyPy, see repo.
So we will probably want to update to 3.7 or 3.8. I'd prefer 3.8, but I'll have a look what needs to be changed to allow that.

[DanielNoord]

Is there a way to invalidate the pip cache? We need pytest to run with 3.7 instead of the 3.8 that is currently installed in the cache.

Supporting PyPy 3.8 will take some more time. Not sure what exactly they are doing in ast, but they break an important assumption we're making in astroid. Instead of setting end_lineno to None if not available/known they seem to give all nodes an end_lineno of -1. This is breaking a number of if statements both in astroid and pylint that guard against looking at end_lineno if it is not correct.

[Pierre-Sassoulas]

Supporting PyPy 3.8 will take some more time. Not sure what exactly they are doing in ast, but they break an important assumption we're making in astroid. Instead of setting end_lineno to None if not available/known they seem to give all nodes an end_lineno of -1. This is breaking a number of if statements both in astroid and pylint that guard against looking at end_lineno if it is not correct.

I see two ways to fix that: We can stop checking the line/col for pypy like we already for for <= 3.7 python interpreter, or we can maybe handle -1 and None the same way in astroid ?

[DanielNoord]

I see two ways to fix that: We can stop checking the line/col for pypy like we already for for <= 3.7 python interpreter, or we can maybe handle -1 and None the same way in astroid ?

Yeah, probably something like that. I don't have the time right now, but I'd like to see why PyPy doesn't provide the ast like cpython does. They seem to copy the stdblib for ast. I didn't find an issue about it in their tracker, but we probably need to have another look.
I'm not sure if the parser is in the "spec" of 3.8, but it feels like PyPy should do something with it. The fact that they seem to be returning -1 everywhere is at least different than cpython.

[coveralls]

Pull Request Test Coverage Report for Build 1916527175

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 93.997%

---

Totals
Change from base Build 1905718907:	0.0%
Covered Lines:	14937
Relevant Lines:	15891

---

💛 - Coveralls

[cdce8p]

Is there a way to invalidate the pip cache?

Pip itself isn't cached between runs, just the created venvs.

[DanielNoord]

😡

Apparently it takes 3+ min to clone our git repository for GitHub Actions which is now timing out all the set-up actions. I'll re-run.

[cdce8p]

Looks good, would you like to update it in astroid too?

[DanielNoord]

Looks good, would you like to update it in astroid too?

I'll have a look this evening!