SET-170: move some functions to utility

[ckrishk]

Moving function to utility file

[github-actions]

Test Results

0 tests   0 ✅  0s ⏱️
0 suites  0 💤
0 files    0 ❌

Results for commit 9e72971.

♻️ This comment has been updated with latest results.

[jmarshall]

Test action fails with:

ruff-format..............................................................Failed
- hook id: ruff-format
- files were modified by this hook

3 files reformatted, 40 files left unchanged
[…]
mypy.....................................................................Failed
[…]
cpg_infra/driver.py:71: error: Only concrete class can be given where
"type[CloudInfraBase]" is expected  [type-abstract]
        c.name(): c

I was able to reproduce this locally by setting up pre-commit. Here is what I think is going on here:

The original code on main is

    c.name(): c  # type: ignore
    for c in CloudInfraBase.__subclasses__()

Apparently the types here are too mysterious for mypy, hence the type: ignore annotation.

Your branch has reformatted that onto one line:

    c.name(): c for c in CloudInfraBase.__subclasses__()  # type: ignore

However for whatever reason ruff-format doesn't like that and has reformatted it back again:

    c.name(): c
    for c in CloudInfraBase.__subclasses__()  # type: ignore

Handily the later pre-commit hook apparently sees that uncommitted code rather than what's in your PR! And now the annotation is on the wrong line so is ineffective.

[jmarshall]

I feel like you've got a case of duelling linters and autoformatters here. For example, your PR changes all the '…' strings to "…" — and the pre-commit ruff-format hook is changing them all back again. So both these strings and the formatting of the c.name(): c for c in … line were perhaps altered by something you've got set up locally, and it probably would benefit from being configured to agree more with the linters the code is already formatted for…

[ckrishk]

Awesome thanks!! will give the linter change a try. I do remember the file was saved without IDE changes. I will try removing it to see if that works.

…

On Thu, 3 Oct 2024 at 10:58, John Marshall ***@***.***> wrote: I feel like you've got a case of duelling linters and autoformatters here. For example, your PR changes all the '…' strings to "…" — and the pre-commit ruff-format hook is changing them all back again. So both these strings and the formatting of the c.name(): c for c in … line were perhaps altered by something you've got set up locally, and it probably would benefit from being configured to agree more with the linters the code is already formatted for… — Reply to this email directly, view it on GitHub <#270 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKBWZDMU774Q2MOLQN5Z2PTZZSJAZAVCNFSM6AAAAABPBP6DIKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGOJQGI4TKMZXGI> . You are receiving this because you authored the thread.Message ID: ***@***.***>