Add PIE and B lints (#3573)

Author: flying-sheep

benchmarks/benchmarks/_utils.py

@@ -202,7 +202,8 @@ def skip(**skipped: AbstractSet) -> Callable[[C], C]:
         skipped_combs = [
             tuple(record.values())
             for record in (
-                dict(zip(param_names, vals)) for vals in itertools.product(*params)
+                dict(zip(param_names, vals, strict=True))
+                for vals in itertools.product(*params)
             )
             if any(v in skipped.get(n, set()) for n, v in record.items())
         ]

pyproject.toml

@@ -55,7 +55,7 @@ dependencies = [
     "h5py>=3.8",
     "tqdm",
     "scikit-learn>=1.1,<1.6",
-    "statsmodels>=0.13",
+    "statsmodels>=0.14",
     "patsy!=1.0.0",                               # https://github.com/pydata/patsy/issues/215
     "networkx>=2.8",
     "natsort",
@@ -226,22 +226,24 @@ 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
-    "W",       # Warning detected by Pycodestyle
-    "UP",      # pyupgrade
-    "I",       # isort
-    "D",       # pydocstyle
-    "TC",      # manage type checking blocks
-    "TID251",  # Banned imports
+    "FBT",     # No positional boolean parameters
+    "I",       # Import sorting
     "ICN",     # Follow import conventions
-    "PTH",     # Pathlib instead of os.path
-    "PYI",     # Typing
+    "PIE",     # Syntax simplifications
     "PLR0917", # Ban APIs with too many positional parameters
-    "FBT",     # No positional boolean parameters
     "PT",      # Pytest style
+    "PTH",     # Pathlib instead of os.path
+    "PYI",     # Typing
     "SIM",     # Simplify control flow
-    "EM",      # Traceback-friendly error messages
+    "TC",      # Manage type checking blocks
+    "UP",      # Update legacy syntax
+    "TID251",  # Banned imports
+    "W",       # Warning detected by Pycodestyle
 ]
 ignore = [
     # line too long -> we accept long comment lines; black gets rid of long code lines

src/scanpy/_compat.py

@@ -220,15 +220,15 @@ def _delegate(cls) -> None:
             if name.startswith("_") or not callable(meth):
                 continue
 
-            def mk_wrapper(name: str):
+            def mk_wrapper(name: str, meth):
                 # Old pytest versions try to run the doctests
                 @wraps(meth, assigned=set(WRAPPER_ASSIGNMENTS) - {"__doc__"})
                 def wrapper(self: _FakeRandomGen, *args, **kwargs):
                     return getattr(self._state, name)(*args, **kwargs)
 
                 return wrapper
 
-            setattr(cls, name, mk_wrapper(name))
+            setattr(cls, name, mk_wrapper(name, meth))
 
 
 _FakeRandomGen._delegate()

src/scanpy/_utils/__init__.py

@@ -193,7 +193,8 @@ def _import_name(name: str) -> Any:
 
     parts = name.split(".")
     obj = import_module(parts[0])
-    for i, name in enumerate(parts[1:]):
+    for _i, name in enumerate(parts[1:]):
+        i = _i
         try:
             obj = import_module(f"{obj.__name__}.{name}")
         except ModuleNotFoundError:
@@ -203,9 +204,9 @@ def _import_name(name: str) -> Any:
     for name in parts[i + 1 :]:
         try:
             obj = getattr(obj, name)
-        except AttributeError:
+        except AttributeError as e:
             msg = f"{parts[:i]}, {parts[i + 1 :]}, {obj} {name}"
-            raise RuntimeError(msg)
+            raise RuntimeError(msg) from e
     return obj
 
 
@@ -312,7 +313,7 @@ def get_igraph_from_adjacency(adjacency: CSBase, *, directed: bool = False) -> G
         weights = weights.A1
     g = ig.Graph(directed=directed)
     g.add_vertices(adjacency.shape[0])  # this adds adjacency.shape[0] vertices
-    g.add_edges(list(zip(sources, targets)))
+    g.add_edges(list(zip(sources, targets, strict=True)))
     with suppress(KeyError):
         g.es["weight"] = weights
     if g.vcount() != adjacency.shape[0]:
@@ -450,9 +451,9 @@ def identify_groups(ref_labels, pred_labels, *, return_overlaps: bool = False):
 
     """
     ref_unique, ref_counts = np.unique(ref_labels, return_counts=True)
-    ref_dict = dict(zip(ref_unique, ref_counts))
+    ref_dict = dict(zip(ref_unique, ref_counts, strict=True))
     pred_unique, pred_counts = np.unique(pred_labels, return_counts=True)
-    pred_dict = dict(zip(pred_unique, pred_counts))
+    pred_dict = dict(zip(pred_unique, pred_counts, strict=True))
     associated_predictions = {}
     associated_overlaps = {}
     for ref_label in ref_unique:
@@ -736,7 +737,9 @@ def _(
                 )
             )
         ):
-            warnings.warn("Rechunking scaling_array in user operation", UserWarning)
+            warnings.warn(
+                "Rechunking scaling_array in user operation", UserWarning, stacklevel=3
+            )
             scaling_array = scaling_array.rechunk(make_axis_chunks(X, axis))
     else:
         scaling_array = da.from_array(
@@ -889,8 +892,8 @@ def select_groups(
         )
         for iname, name in enumerate(adata.obs[key].cat.categories):
             # if the name is not found, fallback to index retrieval
-            if adata.obs[key].cat.categories[iname] in adata.obs[key].values:
-                mask_obs = adata.obs[key].cat.categories[iname] == adata.obs[key].values
+            if name in adata.obs[key].values:
+                mask_obs = name == adata.obs[key].values
             else:
                 mask_obs = str(iname) == adata.obs[key].values
             groups_masks_obs[iname] = mask_obs

src/scanpy/_version.py

@@ -24,9 +24,9 @@ def _get_version_from_vcs() -> str:  # pragma: no cover
     try:
         # Version can be either statically set in pyproject.toml or computed dynamically:
         return metadata.core.version or metadata.hatch.version.cached
-    except UnknownPluginError:
+    except UnknownPluginError as e:
         msg = "Unable to import hatch plugin."
-        raise ImportError(msg)
+        raise ImportError(msg) from e
 
 
 try:

src/scanpy/cli.py

@@ -6,7 +6,7 @@
 import sys
 from argparse import ArgumentParser, Namespace, _SubParsersAction
 from collections.abc import MutableMapping
-from functools import lru_cache, partial
+from functools import cached_property, partial
 from pathlib import Path
 from shutil import which
 from subprocess import run
@@ -81,8 +81,7 @@ def __eq__(self, other: Mapping[str, ArgumentParser]):
             )
         return self.parser_map == other
 
-    @property
-    @lru_cache
+    @cached_property
     def commands(self) -> frozenset[str]:
         return frozenset(
             binary.name[len(self.command) + 1 :]

src/scanpy/experimental/pp/_highly_variable_genes.py

@@ -150,6 +150,7 @@ def _highly_variable_pearson_residuals(
         warnings.warn(
             "`flavor='pearson_residuals'` expects raw count data, but non-integers were found.",
             UserWarning,
+            stacklevel=3,
         )
     # check theta
     if theta <= 0:

src/scanpy/experimental/pp/_normalization.py

@@ -59,6 +59,7 @@ def _pearson_residuals(
         warn(
             "`normalize_pearson_residuals()` expects raw count data, but non-integers were found.",
             UserWarning,
+            stacklevel=3,
         )
 
     if isinstance(X, CSBase):

src/scanpy/experimental/pp/_recipes.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from types import MappingProxyType
 from typing import TYPE_CHECKING
 
 import numpy as np
@@ -17,6 +18,9 @@
 from scanpy.preprocessing import pca
 
 if TYPE_CHECKING:
+    from collections.abc import Mapping
+    from typing import Any
+
     import pandas as pd
     from anndata import AnnData
 
@@ -39,7 +43,7 @@ def recipe_pearson_residuals(
     chunksize: int = 1000,
     n_comps: int | None = 50,
     random_state: float | None = 0,
-    kwargs_pca: dict = {},
+    kwargs_pca: Mapping[str, Any] = MappingProxyType({}),
     check_values: bool = True,
     inplace: bool = True,
 ) -> tuple[AnnData, pd.DataFrame] | None:

src/scanpy/external/exporting.py

@@ -264,7 +264,7 @@ def _get_edges(adata, neighbors_key=None):
     else:
         matrix = neighbors["connectivities"]
     matrix = matrix.tocoo()
-    edges = [(i, j) for i, j in zip(matrix.row, matrix.col)]
+    edges = [(i, j) for i, j in zip(matrix.row, matrix.col, strict=True)]
 
     return edges
 
@@ -444,7 +444,7 @@ def _export_PAGA_to_SPRING(adata, paga_coords, outpath):
     # make node list
     nodes = []
     for i, name, xy, color, size, cells in zip(
-        range(len(names)), names, coords, colors, sizes, cell_groups
+        range(len(names)), names, coords, colors, sizes, cell_groups, strict=True
     ):
         nodes.append(
             {
@@ -459,7 +459,7 @@ def _export_PAGA_to_SPRING(adata, paga_coords, outpath):
 
     # make link list, avoid redundant encoding (graph is undirected)
     links = []
-    for source, target, weight in zip(sources, targets, weights):
+    for source, target, weight in zip(sources, targets, weights, strict=True):
         if source < target and weight > min_edge_weight_save:
             links.append(
                 {"source": int(source), "target": int(target), "weight": float(weight)}