logging-f-string falsely passes for Logger instances

[jmelahman]

logging-f-string

• A minimal code snippet that reproduces the bug.

import logging

logging.info(f"{__name__}")

_LOGGER = logging.getLogger()
_LOGGER.info(f"{__name__}")

---

• The command you invoked (e.g., ruff /path/to/file.py --fix), ideally including the --isolated flag.

$ ruff check --isolated --select G004 repro.py
repro.py:3:14: G004 Logging statement uses f-string
Found 1 error.

while pylint's and flake8 will both error for the latter,

$ flake8 repro.py 
repro.py:3:14: G004 Logging statement uses f-string
repro.py:6:14: G004 Logging statement uses f-string

$ pylint repro.py 
************* Module repro
repro.py:3:0: W1203: Use lazy % formatting in logging functions (logging-fstring-interpolation)
repro.py:6:0: W1203: Use lazy % formatting in logging functions (logging-fstring-interpolation)

-----------------------------------
Your code has been rated at 2.50/10
 

---

• The current Ruff settings (any relevant sections from your pyproject.toml).

---

• The current Ruff version (ruff --version).

$ ruff --version
ruff 0.0.282

[charliermarsh]

Unfortunately we don't have the ability to know definitely whether a given variable is a logger object from the standard library. Neither does Flake8 or Pylint, so it's all a matter of heuristics that trade off false negatives vs. false positives. It looks like _LOGGER doesn't match our heuristic. (Anything ending with logger does, but it's not case insensitive.)