[ruff] Unnecessary cast to int (RUF046)

[InSyncWithFoo]

Summary

Resolves #11412.

Test Plan

cargo nextest run and cargo insta test.

[InSyncWithFoo]

Maybe this should be expanded to cover those suggested in #14292 as well, in which case the rule should be renamed to unnecessary-cast-to-int?

[github-actions]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

ℹ️ ecosystem check detected linter changes. (+15 -0 violations, +0 -0 fixes in 5 projects; 50 projects unchanged)

RasaHQ/rasa (+1 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview

+ rasa/shared/core/training_data/loading.py:86:15: RUF046 Value being casted is already an integer

apache/superset (+2 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview --select ALL

+ superset/migrations/versions/2018-07-22_11-59_bebcf3fed1fe_convert_dashboard_v1_positions.py:197:40: RUF046 Value being casted is already an integer
+ superset/migrations/versions/2018-07-22_11-59_bebcf3fed1fe_convert_dashboard_v1_positions.py:199:29: RUF046 Value being casted is already an integer

bokeh/bokeh (+5 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview --select ALL

+ examples/server/app/spectrogram/main.py:107:13: RUF046 Value being casted is already an integer
+ src/bokeh/colors/color.py:338:60: RUF046 Value being casted is already an integer
+ src/bokeh/palettes.py:1530:27: RUF046 Value being casted is already an integer
+ src/bokeh/palettes.py:1568:10: RUF046 Value being casted is already an integer
+ src/bokeh/palettes.py:1569:10: RUF046 Value being casted is already an integer

indico/indico (+1 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview

+ indico/modules/rb/models/reservation_occurrences.py:223:24: RUF046 Value being casted is already an integer

astropy/astropy (+6 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --ignore RUF9 --output-format concise --preview

+ astropy/io/fits/hdu/compressed/tests/test_compressed.py:375:26: RUF046 Value being casted is already an integer
+ astropy/io/fits/hdu/compressed/utils.py:102:13: RUF046 Value being casted is already an integer
+ astropy/io/fits/tests/test_image.py:1003:26: RUF046 Value being casted is already an integer
+ astropy/io/votable/validator/html.py:246:14: RUF046 Value being casted is already an integer
+ astropy/modeling/_fitting_parallel.py:194:22: RUF046 Value being casted is already an integer
+ astropy/utils/console.py:365:21: RUF046 Value being casted is already an integer

Changes by rule (1 rules affected)

code	total	+ violation	- violation	+ fix	- fix
RUF046	15	15	0	0	0

Formatter (stable)

✅ ecosystem check detected no format changes.

Formatter (preview)

✅ ecosystem check detected no format changes.

[sbrugman]

+1 unnecessary cast to int. This is more future proof: when type inference is available unnecessary int casts could be detected for a much wider class of functions.

Consider expanding this rule to other builtin functions such as “len”. Edit: indeed as suggested in the other issue.

[MichaReiser]

I like your proposal to use the more generic name unnecessary-cast-to-int.

[MichaReiser]

Thanks.

The special casing where ndigitis=0 is incorrect. We should not flag that usage.

[NeilGirdhar]

when type inference is available unnecessary int casts could be detected for a much wider class of functions.

Just FYI, this may not be possible because Boolean values are (unfortunately) integers:

def f(x: int):
  assert isinstance(x, int)
  print(x, int(x))  # If you remove `int` here, you change things.
f(True)

You could detect unnecessary casts, but you have to check that it's not used in any way that would matter, which is a lot trickier.

[InSyncWithFoo]

@NeilGirdhar I made sure to take this into account: All functions listed there either delegate to their corresponding dunder methods or return a strict instance of int. If you can find one that doesn't, please let me know.

[NeilGirdhar]

@NeilGirdhar I made sure to take this into account: All functions listed there either delegate to their corresponding dunder methods or return a strict instance of int. If you can find one that doesn't, please let me know.

Yes, I wasn't talking about this excellent PR 😄 . I was just responding to the comment that suggested that type inference can be used to discover unnecessary integer casts.

[MichaReiser]

This looks good to me. I pushed a few smaller nits, updated the documentation, and marked the fix for round(2) as safe. Let me know if that looks good to you and I'll merge

[InSyncWithFoo]

@MichaReiser Looks good to me too. Let's get this merged.

[DanielYang59]

I completely missed the party, but I'm amazed by how fast you respond, couldn't appreciate more