False positive for list comprehension #11482
Description
Ruff states that a list comprehension should be used, but the (unsafe) fix changes the syntax.
I have a script to make a POST request against an API and one of the inputs is a list of parameters to append onto the end of the query.
parser.add_argument(
"--params",
nargs=2,
action="append",
metavar=("KEY", "VALUE"),
help="Parameters to add to API call.",
)
These are then urlencoded and appended to the query.
if params:
url = f"{url}?{urllib_parse.urlencode(params)}"
If you just pass the params that come out of argparse, then urllib doesn't like that it is a list of lists and raises TypeError: not a valid non-string sequence or mapping object.
So, I convert to a list of tuples before passing the params in.
param_tuples = [(k, v) for k, v in args.params] if args.params else None
This is the line that Ruff is complaining about, but the suggested fix (list(args.params)) doesn't convert to a list of tuples, it leaves it as a list of lists.
It's subtle, but I do need to convert from lists to tuples! I know it is an unsafe fix, but thought this edge case was worth flagging.
Keywords: "list comprehension", tuple, argparse
Ruff version: 0.4.2
Command: ruff check query_api.py --config pyproject.toml --unsafe-fixes --fix
pyproject.toml
[tool.ruff] line-length = 100[tool.ruff.lint]
exclude = ["docs/", "build/"]
ignore = [
"B028", # allow quoting values in strings - see https://peps.python.org/pep-0498/#s-r-and-a-are-redundant
"COM812", # if we add trailing commas everywhere we also need to rerun black which will reformat them again
"D105", # missing docstring in magic method
"D107", # missing docstring in init
"D401", # first line of docstring should be imperative
"E501", # use Bugbear's B950 instead to allow lines to go slightly over the limit
"E203", # allow whitespace around : until https://github.com/PyCQA/pycodestyle/issues/373 is fixed
"RET503", # allow missing return None when other return values present
"RET504", # allow assignment before return
"TD002", # missing author in TODO comments
"TD003", # missing issue reference in TODO comments
"TD004", # missing expiry version in TODO comments
]
select = [
"A", # flake8-builtins
"B", # flake8-bugbear
"C4", # flake8-comprehensions
"COM", # flake8-commas
"D", # pydocstyle
"E", # error
"ERA", # eradicate
"F", # pyflakes
"G", # flake8-format-logging
"I", # isort
"N", # pep8-naming
"RET", # flake8-return
"TD", # flake8-todos
"T20", # flake8-print
"W", # warning,
]
[tool.ruff.lint.flake8-pytest-style]
fixture-parentheses = false
[tool.ruff.lint.pep8-naming]
ignore-names = [
"assertGeomsAll2D",
"assertHistoryEqual",
"assertPolygonsEqual",
"assertRastersEqual",
"assertRasterSrs",
"banded_strings",
"doRollover",
"doRollOver",
"failureException",
"GetDescription",
"GetDriver",
"GetFileList",
"GetLayer",
"GetMetaData",
"GetMetadata",
"GetSpatialRef",
"longMessage",
"maxDiff",
"ReadAsArray",
"SetMetadata",
"SetMetaData",
"SetMetadataItem",
"SetMetaDataItem",
"setUp",
"setUpClass",
"setUpTestData",
"shortDescription",
"sum_columns",
"tearDown",
"tearDownClass",
"useZ",
]
[tool.ruff.lint.per-file-ignores]
"setup.py" = ["D10"]
"src/init.py" = ["A001", "E402", "F401", "F403"]
"src/**/tests/init.py" = ["D104"]
"test_*.py" = ["ARG002", "D10", "D20", "D21", "D30", "D40", "D41", "PT009", "S101", "TID252"]
[tool.ruff.lint.pydocstyle]
convention = "google"