[IbrahimKhan12]

Summary

Fixes #23281.

RUF069 currently treats division as a floating-point expression whenever either operand is syntactically numeric. This works for built-in numeric division, but Python allows / to be overloaded. As a result, the rule reports false positives for Decimal and custom numeric classes whose division returns an exact, non-float value.

This change removes the division-specific heuristic. Division now follows the same logic as other binary expressions: Ruff reports the comparison only when it can identify a float or complex value in the expression.

value / 2 == expected         # No diagnostic when the type of value is unknown.
value / 2.0 == expected       # RUF069 because 2.0 is a known float.
Decimal("1") / 2 == expected  # No diagnostic for overloaded division.

Without type inference, Ruff cannot determine whether value / 2 uses built-in integer division or an overloaded operator. This change therefore favors false negatives over false positives when the operand types are unknown. Explicit float and complex values remain detected. Future type inference could recover built-in integer-division cases without reintroducing false positives for overloaded operators.

The regression fixture covers direct and reverse Decimal division, marks unknown-type division as accepted, and verifies that a float nested inside division still triggers the rule.

Test Plan

• INSTA_FORCE_PASS=1 INSTA_UPDATE=always MDTEST_UPDATE_SNAPSHOTS=1 cargo nextest run -p ruff_linter ruf069
• INSTA_FORCE_PASS=1 INSTA_UPDATE=always MDTEST_UPDATE_SNAPSHOTS=1 cargo nextest run (9,203 passed, 44 skipped)
• cargo clippy --workspace --all-targets --all-features -- -D warnings
• cargo dev generate-all
• Ran prek for all changed files
• Verified the Decimal reproduction with the Ruff CLI

[ntBre]

Thanks! This looks good, just a couple of very small nits.

[astral-sh-bot]

ruff-ecosystem results

Linter (stable)

✅ ecosystem check detected no linter changes.

Linter (preview)

ℹ️ ecosystem check detected linter changes. (+6 -51 violations, +0 -0 fixes in 8 projects; 51 projects unchanged)

PlasmaPy/PlasmaPy (+0 -10 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- src/plasmapy/formulary/quantum.py:123:10: float-equality-comparison Unreliable floating point equality comparison `0 * u.m / u.s == V`
- tests/formulary/test_fusion.py:326:16: float-equality-comparison Unreliable floating point equality comparison `sigma.unit == u.m**3 / u.s`
- tests/formulary/test_fusion.py:430:16: float-equality-comparison Unreliable floating point equality comparison `sv.unit == u.m**3 / u.s`
- tests/formulary/test_fusion.py:451:16: float-equality-comparison Unreliable floating point equality comparison `sv.unit == u.m**3 / u.s`
- tests/formulary/test_laser.py:89:12: float-equality-comparison Unreliable floating point equality comparison `result.unit == u.Watt / u.m**2`
- tests/formulary/test_misc.py:131:12: float-equality-comparison Unreliable floating point equality comparison `(Bohm_diffusion(T_e, B)).unit == u.m**2 / u.s`
- tests/plasma/sources/test_plasma3d.py:48:12: float-equality-comparison Unreliable floating point equality comparison `test_plasma.density.si.unit == u.kg / u.m**3`
- tests/plasma/sources/test_plasma3d.py:51:12: float-equality-comparison Unreliable floating point equality comparison `test_plasma.momentum.si.unit == u.kg / (u.m**2 * u.s)`
- tests/plasma/sources/test_plasmablob.py:51:12: float-equality-comparison Unreliable floating point equality comparison `test_plasma.density.si.unit == u.kg / u.m**3`
- tests/plasma/sources/test_plasmablob.py:54:12: float-equality-comparison Unreliable floating point equality comparison `test_plasma.momentum.si.unit == u.kg / (u.m**2 * u.s)`

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

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- tests/unit/test_services_fees.py:61:12: float-equality-comparison Unreliable floating point equality comparison `fee / 1000 == 199`

milvus-io/pymilvus (+0 -1 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- examples/orm_deprecated/bulk_import/example_bulkwriter_with_nullable.py:300:16: float-equality-comparison Unreliable floating point equality comparison `len(b[0]) == DIM/8`

pandas-dev/pandas (+0 -12 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- pandas/core/indexes/range.py:1022:30: float-equality-comparison Unreliable floating point equality comparison `abs(start_s - start_o) == step_s / 2`
- pandas/core/indexes/range.py:1023:30: float-equality-comparison Unreliable floating point equality comparison `abs(end_s - end_o) == step_s / 2`
- pandas/core/indexes/range.py:1237:41: float-equality-comparison Unreliable floating point equality comparison `item == self[0] + self.step / 2`
- pandas/tests/scalar/timedelta/test_arithmetic.py:799:16: float-equality-comparison Unreliable floating point equality comparison `result == td._value / (86400 * 10**9)`
- pandas/tests/scalar/timedelta/test_arithmetic.py:801:16: float-equality-comparison Unreliable floating point equality comparison `result == td._value / 10**9`
- pandas/tests/scalar/timedelta/test_timedelta.py:139:16: float-equality-comparison Unreliable floating point equality comparison `res._value == td._value / 2`
- pandas/tests/scalar/timedelta/test_timedelta.py:143:16: float-equality-comparison Unreliable floating point equality comparison `res._value == td._value / 2`
- pandas/tests/scalar/timestamp/methods/test_to_pydatetime.py:70:13: float-equality-comparison Unreliable floating point equality comparison `pd.Timestamp(pydt_max).as_unit("ns")._value / 1000 == pd.Timestamp.max._value / 1000`
- pandas/tests/scalar/timestamp/methods/test_to_pydatetime.py:84:13: float-equality-comparison Unreliable floating point equality comparison `pd.Timestamp(pydt_min + tdus).as_unit("ns")._value / 1000 == pd.Timestamp.min._value / 1000`
- pandas/tests/tseries/offsets/test_ticks.py:255:12: float-equality-comparison Unreliable floating point equality comparison `off / 2 == cls(5)`
... 2 additional changes omitted for project

prefecthq/prefect (+0 -1 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- tests/events/server/test_events_counts.py:525:12: float-equality-comparison Unreliable floating point equality comparison `len(counts) == (until - since).total_seconds() / 1800 + 1`

indico/indico (+2 -0 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

+ indico/modules/designer/util.py:108:22: RUF069 Unreliable floating point equality comparison `frm_size[0] == float(tpl.data['width']) / PIXELS_CM`
+ indico/modules/designer/util.py:109:22: RUF069 Unreliable floating point equality comparison `frm_size[1] == float(tpl.data['height']) / PIXELS_CM`

astropy/astropy (+0 -25 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

- astropy/coordinates/tests/test_frames_with_velocity.py:218:12: float-equality-comparison Unreliable floating point equality comparison `icrs.pm_ra == 10 * u.mas / u.yr`
- astropy/coordinates/tests/test_frames_with_velocity.py:226:12: float-equality-comparison Unreliable floating point equality comparison `icrs.pm_ra == 10 * u.mas / u.yr`
- astropy/coordinates/tests/test_frames_with_velocity.py:246:12: float-equality-comparison Unreliable floating point equality comparison `icrs.v_x == 1 * u.km / u.s`
- astropy/coordinates/tests/test_frames_with_velocity.py:247:12: float-equality-comparison Unreliable floating point equality comparison `icrs.v_y == 2 * u.km / u.s`
- astropy/coordinates/tests/test_frames_with_velocity.py:248:12: float-equality-comparison Unreliable floating point equality comparison `icrs.v_z == 3 * u.km / u.s`
- astropy/coordinates/tests/test_representation.py:126:16: float-equality-comparison Unreliable floating point equality comparison `newrep.differentials["s"].d_distance == -9 * u.km / u.s`
- astropy/cosmology/_src/tests/flrw/test_base.py:192:16: float-equality-comparison Unreliable floating point equality comparison `cosmo.critical_density0.unit == u.g / u.cm**3`
- astropy/cosmology/_src/tests/flrw/test_lambdacdm.py:633:12: float-equality-comparison Unreliable floating point equality comparison `cosmo.critical_density(1.0).unit == u.g / u.cm**3`
- astropy/cosmology/_src/tests/test_units.py:57:13: float-equality-comparison Unreliable floating point equality comparison `3 * u.km / cu.redshift**3 == 3 * u.km`
- astropy/io/fits/util.py:245:12: float-equality-comparison Unreliable floating point equality comparison `ns.dtype.itemsize != s.dtype.itemsize / 4`
- astropy/io/fits/util.py:278:12: float-equality-comparison Unreliable floating point equality comparison `ns.dtype.itemsize / 4 != s.dtype.itemsize`
- astropy/modeling/tests/test_physical_models.py:69:12: float-equality-comparison Unreliable floating point equality comparison `b(1.0 * u.micron).unit == u.erg / (u.cm**2 * u.s * u.Hz * u.sr)`
- astropy/nddata/ccddata.py:77:16: float-equality-comparison Unreliable floating point equality comparison `unit == 1 / (parent_unit**2)`
- astropy/time/tests/test_methods.py:771:20: float-equality-comparison Unreliable floating point equality comparison `t.mean() == t1 + epsilon / 2`
- astropy/units/tests/test_format.py:967:12: float-equality-comparison Unreliable floating point equality comparison `x == number * (u.erg / u.s / u.cm**2 / u.Angstrom)`
- astropy/units/tests/test_format.py:971:12: float-equality-comparison Unreliable floating point equality comparison `x == number * (u.erg / u.s / u.cm**2 / u.Angstrom)`
- astropy/units/tests/test_quantity.py:608:17: float-equality-comparison Unreliable floating point equality comparison `1 / (u.cm * u.cm) == 1 * u.cm**-2`
- astropy/units/tests/test_quantity.py:855:12: float-equality-comparison Unreliable floating point equality comparison `q.si.unit == 1 / u.m`
- astropy/units/tests/test_quantity.py:869:12: float-equality-comparison Unreliable floating point equality comparison `q.cgs.unit == 1 / u.cm`
- astropy/units/tests/test_quantity_array_methods.py:434:16: float-equality-comparison Unreliable floating point equality comparison `q1[1] == 1000 * u.m / u.km`
- astropy/units/tests/test_quantity_array_methods.py:436:16: float-equality-comparison Unreliable floating point equality comparison `q1[1] == 1 * u.m / u.km`
... 4 additional changes omitted for project

home-assistant/core (+4 -1 violations, +0 -0 fixes)

ruff check --no-cache --exit-zero --no-fix --output-format concise --preview

+ tests/components/vizio/test_media_player.py:135:9: float-equality-comparison Unreliable floating point equality comparison `attr["volume_level"] == float(int(MAX_VOLUME[vizio_device_class] / 2))
+ tests/components/vizio/test_media_player.py:258:17: float-equality-comparison Unreliable floating point equality comparison `attr["volume_level"] == float(int(MAX_VOLUME[MediaPlayerDeviceClass.TV] / 2))
+ tests/components/vizio/test_media_player.py:353:17: float-equality-comparison Unreliable floating point equality comparison `attr["volume_level"] == float(volume) / MAX_VOLUME[MediaPlayerDeviceClass.TV]`
+ tests/components/vizio/test_media_player.py:384:17: float-equality-comparison Unreliable floating point equality comparison `attr["volume_level"] == float(volume) / MAX_VOLUME[MediaPlayerDeviceClass.TV]`
- tests/util/test_unit_conversion.py:1358:12: float-equality-comparison Unreliable floating point equality comparison `converter.get_unit_floored_log_ratio(to_unit, from_unit) == 1 / ratio`

Changes by rule (2 rules affected)

code	total	+ violation	- violation	+ fix	- fix
float-equality-comparison	55	4	51	0	0
RUF069	2	2	0	0	0

[IbrahimKhan12]

nits addressed!

[ntBre]

Thank you!