Add Pylint lints

benchmarks/benchmarks/tools.py

@@ -20,7 +20,7 @@
 
 
 def setup():
-    global adata
+    global adata  # noqa: PLW0603
     adata = pbmc68k_reduced()
     assert "X_pca" in adata.obsm
 

docs/conf.py

@@ -17,7 +17,7 @@
 
 HERE = Path(__file__).parent
 sys.path[:0] = [str(HERE.parent), str(HERE / "extensions")]
-import scanpy  # noqa
+import scanpy
 
 if TYPE_CHECKING:
     from sphinx.application import Sphinx

docs/extensions/debug_docstrings.py

@@ -15,13 +15,13 @@
 _pd_orig = sphinx.ext.napoleon._process_docstring
 
 
-def pd_new(app, what, name, obj, options, lines):  # noqa: PLR0917
+def pd_new(app, what, name, obj, options, lines) -> None:  # noqa: PLR0917
     """Wrap ``sphinx.ext.napoleon._process_docstring``."""
     _pd_orig(app, what, name, obj, options, lines)
     print(*lines, sep="\n")
 
 
-def setup(app: Sphinx):
+def setup(app: Sphinx) -> None:
     """App setup hook."""
     if os.environ.get("DEBUG") is not None:
         sphinx.ext.napoleon._process_docstring = pd_new

docs/extensions/function_images.py

@@ -14,7 +14,7 @@
 
 def insert_function_images(  # noqa: PLR0917
     app: Sphinx, what: str, name: str, obj: Any, options: Options, lines: list[str]
-):
+) -> None:
     """Insert images for plot functions."""
     path = app.config.api_dir / f"{name}.png"
     if what != "function" or not path.is_file():
@@ -27,7 +27,7 @@ def insert_function_images(  # noqa: PLR0917
     ]
 
 
-def setup(app: Sphinx):
+def setup(app: Sphinx) -> None:
     """App setup hook."""
     app.add_config_value("api_dir", Path(), "env")
     app.connect("autodoc-process-docstring", insert_function_images)

pyproject.toml

@@ -226,55 +226,58 @@ docstring-code-format = true
 
 [tool.ruff.lint]
 select = [
-    "B",       # Likely bugs and design issues
-    "D",       # Documentation style
-    "E",       # Error detected by Pycodestyle
-    "EM",      # Traceback-friendly error messages
-    "F",       # Errors detected by Pyflakes
-    "FBT",     # No positional boolean parameters
-    "I",       # Import sorting
-    "ICN",     # Follow import conventions
-    "PIE",     # Syntax simplifications
-    "PLR0917", # Ban APIs with too many positional parameters
-    "PT",      # Pytest style
-    "PTH",     # Pathlib instead of os.path
-    "PYI",     # Typing
-    "SIM",     # Simplify control flow
-    "TC",      # Manage type checking blocks
-    "UP",      # Update legacy syntax
-    "TID251",  # Banned imports
-    "W",       # Warning detected by Pycodestyle
+    "B",      # Likely bugs and design issues
+    "D",      # Documentation style
+    "E",      # Error detected by Pycodestyle
+    "EM",     # Traceback-friendly error messages
+    "F",      # Errors detected by Pyflakes
+    "FBT",    # No positional boolean parameters
+    "I",      # Import sorting
+    "ICN",    # Follow import conventions
+    "PIE",    # Syntax simplifications
+    "PL",     # Pylint
+    "PT",     # Pytest style
+    "PTH",    # Pathlib instead of os.path
+    "PYI",    # Typing
+    "RUF100", # Unused noqa
+    "SIM",    # Simplify control flow
+    "TC",     # Manage type checking blocks
+    "UP",     # Update legacy syntax
+    "TID251", # Banned imports
+    "W",      # Warning detected by Pycodestyle
 ]
+external = [ "PLR0917" ] # preview lint that we use
 ignore = [
-    # line too long -> we accept long comment lines; black gets rid of long code lines
-    "E501",
-    # module level import not at top of file -> required to circumvent circular imports for Scanpys API
-    "E402",
     # E266 too many leading '#' for block comment -> Scanpy allows them for comments into sections
     "E262",
+    # module level import not at top of file -> required to circumvent circular imports for Scanpys API
+    "E402",
+    # line too long -> we accept long comment lines; black gets rid of long code lines
+    "E501",
     # allow I, O, l as variable names -> I is the identity matrix, i, j, k, l is reasonable indexing notation
     "E741",
-    # `Literal["..."] | str` is useful for autocompletion
-    "PYI051",
     # We ban blank lines before docstrings instead of the opposite
     "D203",
     # We want multiline summaries to start on the first line, not the second
     "D213",
     # TODO: replace our current param docs reuse with this and remove it here:
     "D417",
+    # Numbers like “2” aren’t that “magic”.
+    "PLR2004",
+    # `Literal["..."] | str` is useful for autocompletion
+    "PYI051",
 ]
 [tool.ruff.lint.per-file-ignores]
 # Do not assign a lambda expression, use a def
 "src/scanpy/tools/_rank_genes_groups.py" = [ "E731" ]
 # No need for docstrings for all benchmarks
 "benchmarks/**/*.py" = [ "D102", "D103" ]
-# No need for docstrings for all test modules and test functions
-"tests/**/*.py" = [ "D100", "D101", "D103" ]
+# D*: No need for docstrings for all test modules and test functions
+# PLR0913: Test may use many fixtures
+"tests/**/*.py" = [ "D100", "D101", "D103", "PLR0913" ]
 [tool.ruff.lint.isort]
 known-first-party = [ "scanpy", "testing.scanpy" ]
 required-imports = [ "from __future__ import annotations" ]
-[tool.ruff.lint.pydocstyle]
-convention = "numpy"
 [tool.ruff.lint.flake8-tidy-imports.banned-api]
 "pytest.importorskip".msg = "Use the “@needs” decorator/mark instead"
 "pandas.api.types.is_categorical_dtype".msg = "Use isinstance(s.dtype, CategoricalDtype) instead"
@@ -290,6 +293,11 @@ convention = "numpy"
 [tool.ruff.lint.flake8-type-checking]
 exempt-modules = [  ]
 strict = true
+[tool.ruff.lint.pydocstyle]
+convention = "numpy"
+[tool.ruff.lint.pylint]
+max-args = 10
+max-positional-args = 5
 
 [tool.towncrier]
 name = "scanpy"

src/scanpy/_settings.py

@@ -93,7 +93,7 @@ class ScanpyConfig:
     N_PCS: int
     """Default number of principal components to use."""
 
-    def __init__(
+    def __init__(  # noqa: PLR0913
         self,
         *,
         verbosity: Verbosity | int | str = Verbosity.warning,
@@ -406,7 +406,7 @@ def categories_to_ignore(self, categories_to_ignore: Iterable[str]):
         "transparent",
         "ipython_format",
     )
-    def set_figure_params(
+    def set_figure_params(  # noqa: PLR0913
         self,
         *,
         scanpy: bool = True,

src/scanpy/_utils/__init__.py

@@ -188,10 +188,10 @@
     return decorator
 
 
-def _import_name(name: str) -> Any:
+def _import_name(full_name: str) -> Any:
     from importlib import import_module
 
-    parts = name.split(".")
+    parts = full_name.split(".")
     obj = import_module(parts[0])
     for _i, name in enumerate(parts[1:]):
         i = _i
@@ -387,7 +387,7 @@
     asso_matrix: list[list[float]] = []
     for ipred_group, pred_group in enumerate(adata.obs[prediction].cat.categories):
         if "?" in pred_group:
-            pred_group = str(ipred_group)
+            pred_group = str(ipred_group)  # noqa: PLW2901
         # starting from numpy version 1.13, subtractions of boolean arrays are deprecated
         mask_pred = adata.obs[prediction].values == pred_group
         mask_pred_int = mask_pred.astype(np.int8)
@@ -927,7 +927,9 @@
     return groups_order_subset, groups_masks_obs
 
 
-def warn_with_traceback(message, category, filename, lineno, file=None, line=None):  # noqa: PLR0917
+def warn_with_traceback(  # noqa: PLR0917
+    message, category, filename, lineno, file=None, line=None
+) -> None:
     """Get full tracebacks when warning is raised by setting.
 
     warnings.showwarning = warn_with_traceback

src/scanpy/experimental/pp/_highly_variable_genes.py

@@ -128,7 +128,7 @@ def clac_clipped_res_dense(gene: int, cell: int) -> np.float64:
     return residuals
 
 
-def _highly_variable_pearson_residuals(
+def _highly_variable_pearson_residuals(  # noqa: PLR0912, PLR0915
     adata: AnnData,
     *,
     theta: float = 100,
@@ -308,7 +308,7 @@ def _highly_variable_pearson_residuals(
     layer=doc_layer,
     inplace=doc_inplace,
 )
-def highly_variable_genes(
+def highly_variable_genes(  # noqa: PLR0913
     adata: AnnData,
     *,
     theta: float = 100,

src/scanpy/experimental/pp/_recipes.py

@@ -33,7 +33,7 @@
     check_values=doc_check_values,
     inplace=doc_inplace,
 )
-def recipe_pearson_residuals(
+def recipe_pearson_residuals(  # noqa: PLR0913
     adata: AnnData,
     *,
     theta: float = 100,

src/scanpy/external/exporting.py

@@ -33,7 +33,7 @@
     "neighbors_key",
     "overwrite",
 )
-def spring_project(
+def spring_project(  # noqa: PLR0912, PLR0915
     adata: AnnData,
     project_dir: Path | str,
     embedding_method: str,
@@ -90,17 +90,16 @@
     if embedding_method not in adata.obsm_keys():
         if "X_" + embedding_method in adata.obsm_keys():
             embedding_method = "X_" + embedding_method
+        elif embedding_method in adata.uns:
+            embedding_method = (
+                "X_"
+                + embedding_method
+                + "_"
+                + adata.uns[embedding_method]["params"]["layout"]
+            )
         else:
-            if embedding_method in adata.uns:
-                embedding_method = (
-                    "X_"
-                    + embedding_method
-                    + "_"
-                    + adata.uns[embedding_method]["params"]["layout"]
-                )
-            else:
-                msg = f"Run the specified embedding method `{embedding_method}` first."
-                raise ValueError(msg)
+            msg = f"Run the specified embedding method `{embedding_method}` first."
+            raise ValueError(msg)
 
     coords = adata.obsm[embedding_method]
 
@@ -477,8 +476,6 @@
 
     Path(outpath).write_text(json.dumps(PAGA_data, indent=4))
 
-    return None
-
 
 @old_positionals(
     "embedding_keys",
@@ -490,7 +487,7 @@
     "port",
     "do_debug",
 )
-def cellbrowser(
+def cellbrowser(  # noqa: PLR0913
     adata: AnnData,
     data_dir: Path | str,
     data_name: str,

src/scanpy/external/pl.py

@@ -269,7 +269,7 @@ def sam(
     "ax",
 )
 @_doc_params(show_save_ax=doc_show_save_ax)
-def wishbone_marker_trajectory(
+def wishbone_marker_trajectory(  # noqa: PLR0913
     adata: AnnData,
     markers: Collection[str],
     *,

src/scanpy/external/pp/_bbknn.py

@@ -14,7 +14,7 @@
 
 @old_positionals("batch_key", "use_rep", "approx", "use_annoy", "metric", "copy")
 @doctest_needs("bbknn")
-def bbknn(
+def bbknn(  # noqa: PLR0913
     adata: AnnData,
     *,
     batch_key: str = "batch",

src/scanpy/external/pp/_dca.py

@@ -41,7 +41,7 @@
     "return_info",
     "copy",
 )
-def dca(
+def dca(  # noqa: PLR0913
     adata: AnnData,
     mode: Literal["denoise", "latent"] = "denoise",
     *,

src/scanpy/external/pp/_hashsolo.py

@@ -48,7 +48,7 @@
     from numpy.typing import ArrayLike, NDArray
 
 
-def _calculate_log_likelihoods(
+def _calculate_log_likelihoods(  # noqa: PLR0915
     data: np.ndarray, number_of_noise_barcodes: int
 ) -> tuple[NDArray[np.float64], NDArray[np.float64], dict[int, str]]:
     """Calculate log likelihoods for each hypothesis, negative, singlet, doublet.

src/scanpy/external/pp/_magic.py

@@ -23,7 +23,7 @@
 
 
 @doctest_needs("magic")
-def magic(
+def magic(  # noqa: PLR0913
     adata: AnnData,
     name_list: Literal["all_genes", "pca_only"] | Sequence[str] | None = None,
     *,

src/scanpy/external/pp/_mnn_correct.py

@@ -13,7 +13,7 @@
     from anndata import AnnData
 
 
-def mnn_correct(
+def mnn_correct(  # noqa: PLR0913
     *datas: AnnData | np.ndarray,
     var_index: Collection[str] | None = None,
     var_subset: Collection[str] | None = None,

src/scanpy/external/tl/_phate.py

@@ -33,7 +33,7 @@
     "copy",
 )
 @doctest_needs("phate")
-def phate(
+def phate(  # noqa: PLR0913
     adata: AnnData,
     n_components: int = 2,
     *,

src/scanpy/external/tl/_phenograph.py

@@ -41,7 +41,7 @@
     "copy",
 )
 @doctest_needs("phenograph")
-def phenograph(
+def phenograph(  # noqa: PLR0913
     data: AnnData | np.ndarray | SpBase,
     clustering_algo: Literal["louvain", "leiden"] | None = "louvain",
     *,

src/scanpy/external/tl/_sam.py

@@ -30,7 +30,7 @@
     "verbose",
 )
 @doctest_needs("samalg")
-def sam(
+def sam(  # noqa: PLR0913
     adata: AnnData,
     *,
     max_iter: int = 10,

src/scanpy/external/tl/_trimap.py

@@ -27,7 +27,7 @@
     "copy",
 )
 @doctest_needs("trimap")
-def trimap(
+def trimap(  # noqa: PLR0913
     adata: AnnData,
     n_components: int = 2,
     *,

src/scanpy/get/_aggregated.py

@@ -180,7 +180,7 @@ def _power(X: Array, power: float) -> Array:
     return X**power if isinstance(X, np.ndarray) else X.power(power)
 
 
-def aggregate(
+def aggregate(  # noqa: PLR0912
     adata: AnnData,
     by: str | Collection[str],
     func: AggType | Iterable[AggType],

src/scanpy/get/get.py

@@ -28,7 +28,7 @@
 
 
 # TODO: implement diffxpy method, make singledispatch
-def rank_genes_groups_df(
+def rank_genes_groups_df(  # noqa: PLR0912
     adata: AnnData,
     group: str | Iterable[str] | None,
     *,