Use Ruff for code formatting (#851)

Signed-off-by: Lukas Heumos <[email protected]>

Author: Zethson

.pre-commit-config.yaml

@@ -8,18 +8,10 @@ minimum_pre_commit_version: 2.16.0
 ci:
     skip: []
 repos:
-    - repo: https://github.com/psf/black
-      rev: 24.10.0
-      hooks:
-          - id: black
     - repo: https://github.com/rbubley/mirrors-prettier
       rev: v3.4.2
       hooks:
           - id: prettier
-    - repo: https://github.com/asottile/blacken-docs
-      rev: 1.19.1
-      hooks:
-          - id: blacken-docs
     - repo: https://github.com/pre-commit/mirrors-mypy
       rev: v1.14.1
       hooks:
@@ -31,3 +23,4 @@ repos:
       hooks:
           - id: ruff
             args: [--fix, --exit-non-zero-on-fix]
+          - id: ruff-format

benchmarks/spatialdata_benchmark.py

@@ -44,7 +44,6 @@ def time_map_blocks(self, _):
 
 
 class TimeQueries:
-
     params = ([100, 1_000, 10_000], [True, False], [100, 1_000])
     param_names = ["length", "filter_table", "n_transcripts_per_cell"]
 

benchmarks/utils.py

@@ -8,7 +8,7 @@
 - function run_benchmark is used to run the benchmarks.
 
 Performant dataset generation functions so the benchmarks run fast even for large artificial datasets.
-The object is to generate a dataset containing many cells. By copying the same cell values instead of 
+The object is to generate a dataset containing many cells. By copying the same cell values instead of
 doing gaussian blur on the whole image, we can generate the same dataset in a fraction of the time.
 - function labeled_particles is used to generate labeled blobs.
 - function _generate_ball is used to generate a ball of given radius and dimension.

docs/extensions/typed_returns.py

@@ -11,7 +11,7 @@ def _process_return(lines):
         m = re.fullmatch(r"(?P<param>\w+)\s+:\s+(?P<type>[\w.]+)", line)
         if m:
             # Once this is in scanpydoc, we can use the fancy hover stuff
-            yield f'-{m["param"]} (:class:`~{m["type"]}`)'
+            yield f"-{m['param']} (:class:`~{m['type']}`)"
         else:
             yield line
 

pyproject.toml

@@ -103,27 +103,6 @@ filterwarnings = [
     #     "ignore:.*U.*mode is deprecated:DeprecationWarning",
 ]
 
-[tool.black]
-line-length = 120
-target-version = ['py310']
-include = '\.pyi?$'
-exclude = '''
-(
-  /(
-      \.eggs
-    | \.git
-    | \.hg
-    | \.mypy_cache
-    | \.tox
-    | \.venv
-    | _build
-    | buck-out
-    | build
-    | dist
-  )/
-)
-'''
-
 [tool.jupytext]
 formats = "ipynb,md"
 
@@ -148,6 +127,7 @@ exclude = [
     "docs/_build",
     "dist",
     "setup.py",
+    
 ]
 line-length = 120
 target-version = "py310"

src/spatialdata/_core/operations/aggregate.py

@@ -180,9 +180,9 @@ def aggregate(
         ONES_KEY = None
         if value_key is None:
             ONES_KEY = "__ones_column_aggregate"
-            assert (
-                ONES_KEY not in values_.columns
-            ), f"Column {ONES_KEY} is reserved for internal use and cannot be already present in values_"
+            assert ONES_KEY not in values_.columns, (
+                f"Column {ONES_KEY} is reserved for internal use and cannot be already present in values_"
+            )
             values_[ONES_KEY] = 1
             value_key = ONES_KEY
 
@@ -384,10 +384,10 @@ def _aggregate_shapes(
             "agg_func='sum' instead."
         )
         assert not isinstance(values.iloc[0].geometry, Point), (
-            "Fractions cannot be computed when values are points. " "Please use fractions=False."
+            "Fractions cannot be computed when values are points. Please use fractions=False."
         )
     assert not (categorical and agg_func == "mean"), (
-        "Incompatible choice: aggregating a categorical column with " "agg_func='mean'"
+        "Incompatible choice: aggregating a categorical column with agg_func='mean'"
     )
 
     # we need to add a column of ones to the values dataframe to be able to count the number of instances in each zone

src/spatialdata/_core/operations/rasterize.py

@@ -678,9 +678,9 @@ def rasterize_shapes_points(
     elif isinstance(agg_func, str):
         AGGREGATIONS = ["sum", "count", "count_cat", "first"]
 
-        assert np.isin(
-            agg_func, AGGREGATIONS
-        ), f"Aggregation function must be one of {', '.join(AGGREGATIONS)}. Found {agg_func}"
+        assert np.isin(agg_func, AGGREGATIONS), (
+            f"Aggregation function must be one of {', '.join(AGGREGATIONS)}. Found {agg_func}"
+        )
 
         assert agg_func == "count" or value_key is not None, f"value_key cannot be done for agg_func={agg_func}"
 

src/spatialdata/_core/operations/transform.py

@@ -287,9 +287,9 @@ def _(
         if transformation is None and to_coordinate_system is not None:
             return data.transform_to_coordinate_system(target_coordinate_system=to_coordinate_system)
         raise RuntimeError(ERROR_MSG_AFTER_0_0_15)
-    assert bool(transformation is None) != bool(
-        to_coordinate_system is None
-    ), "When maintain_positioning is True, only one of transformation and to_coordinate_system can be None"
+    assert bool(transformation is None) != bool(to_coordinate_system is None), (
+        "When maintain_positioning is True, only one of transformation and to_coordinate_system can be None"
+    )
     new_elements: dict[str, dict[str, Any]] = {}
     for element_type in ["images", "labels", "points", "shapes"]:
         d = getattr(data, element_type)

src/spatialdata/_core/operations/vectorize.py

@@ -113,7 +113,7 @@ def _(element: GeoDataFrame, **kwargs: Any) -> GeoDataFrame:
         return _make_circles(element, obs)
     if isinstance(element.geometry.iloc[0], Point):
         return element
-    raise RuntimeError("Unsupported geometry type: " f"{type(element.geometry.iloc[0])}")
+    raise RuntimeError(f"Unsupported geometry type: {type(element.geometry.iloc[0])}")
 
 
 @to_circles.register(DaskDataFrame)
@@ -281,7 +281,7 @@ def _(gdf: GeoDataFrame, buffer_resolution: int = 16) -> GeoDataFrame:
             return buffered_df
         assert isinstance(gdf.geometry.iloc[0], Polygon | MultiPolygon)
         return gdf
-    raise RuntimeError("Unsupported geometry type: " f"{type(gdf.geometry.iloc[0])}")
+    raise RuntimeError(f"Unsupported geometry type: {type(gdf.geometry.iloc[0])}")
 
 
 @to_polygons.register(DaskDataFrame)

src/spatialdata/_core/query/relational_query.py

@@ -164,9 +164,9 @@ def _filter_table_by_elements(
     """
     assert set(elements_dict.keys()).issubset({"images", "labels", "shapes", "points"})
     assert len(elements_dict) > 0, "elements_dict must not be empty"
-    assert any(
-        len(elements) > 0 for elements in elements_dict.values()
-    ), "elements_dict must contain at least one dict which contains at least one element"
+    assert any(len(elements) > 0 for elements in elements_dict.values()), (
+        "elements_dict must contain at least one dict which contains at least one element"
+    )
     if table is None:
         return None
     to_keep = np.zeros(len(table), dtype=bool)

src/spatialdata/_core/spatialdata.py

@@ -158,7 +158,6 @@ def __init__(
             "For renaming, please see the discussion here https://github.com/scverse/spatialdata/discussions/707 .",
             exc_type=(ValueError, KeyError),
         ) as collect_error:
-
             if images is not None:
                 for k, v in images.items():
                     with collect_error(location=("images", k)):
@@ -1715,7 +1714,6 @@ def get_attrs(
         """
 
         def _flatten_mapping(m: Mapping[str, Any], parent_key: str = "", sep: str = "_") -> dict[str, Any]:
-
             items: list[tuple[str, Any]] = []
             for k, v in m.items():
                 new_key = f"{parent_key}{sep}{k}" if parent_key else k
@@ -2007,7 +2005,7 @@ def h(s: str) -> str:
                 descr += f"{h('empty_line')}"
                 descr_class = v.__class__.__name__
                 if attr == "shapes":
-                    descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class} " f"shape: {v.shape} (2D shapes)"
+                    descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class} shape: {v.shape} (2D shapes)"
                 elif attr == "points":
                     length: int | None = None
                     if len(v.dask) == 1:
@@ -2037,7 +2035,7 @@ def h(s: str) -> str:
                             + ", ".join([str(dim) if not isinstance(dim, Delayed) else "<Delayed>" for dim in v.shape])
                             + ")"
                         )
-                    descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class} " f"with shape: {shape_str} {dim_string}"
+                    descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class} with shape: {shape_str} {dim_string}"
                 elif attr == "tables":
                     descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class} {v.shape}"
                 else:
@@ -2055,7 +2053,7 @@ def h(s: str) -> str:
                             if dims is None:
                                 dims = "".join(vv.dims)
                             shapes.append(shape)
-                        descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class}[{dims}] " f"{', '.join(map(str, shapes))}"
+                        descr += f"{h(attr + 'level1.1')}{k!r}: {descr_class}[{dims}] {', '.join(map(str, shapes))}"
                     else:
                         raise TypeError(f"Unknown type {type(v)}")
             if last_attr is True:

src/spatialdata/_io/_utils.py

@@ -47,7 +47,7 @@ def ome_zarr_logger(level: Any) -> Generator[None, None, None]:
 
 
 def _get_transformations_from_ngff_dict(
-    list_of_encoded_ngff_transformations: list[dict[str, Any]]
+    list_of_encoded_ngff_transformations: list[dict[str, Any]],
 ) -> MappingToCoordinateSystem_t:
     list_of_ngff_transformations = [NgffBaseTransformation.from_dict(d) for d in list_of_encoded_ngff_transformations]
     list_of_transformations = [BaseTransformation.from_ngff(t) for t in list_of_ngff_transformations]
@@ -118,8 +118,7 @@ def overwrite_channel_names(group: zarr.Group, element: DataArray | DataTree) ->
     multiscales_meta = group.attrs["multiscales"]
     if len(multiscales_meta) != 1:
         raise ValueError(
-            f"Multiscale metadata must be of length one but got length {len(multiscales_meta)}. Data might"
-            f"be corrupted."
+            f"Multiscale metadata must be of length one but got length {len(multiscales_meta)}. Data mightbe corrupted."
         )
     multiscales_meta[0]["metadata"]["omero"]["channels"] = channel_metadata
     group.attrs["multiscales"] = multiscales_meta

src/spatialdata/dataloader/datasets.py

@@ -163,9 +163,9 @@ def _validate(
             raise ValueError("`table_name` must be provided if `return_annotations` is not `None`.")
 
         # check that the regions specified in the two dicts are the same
-        assert set(regions_to_images.keys()) == set(
-            regions_to_coordinate_systems.keys()
-        ), "The keys in `regions_to_images` and `regions_to_coordinate_systems` must be the same."
+        assert set(regions_to_images.keys()) == set(regions_to_coordinate_systems.keys()), (
+            "The keys in `regions_to_images` and `regions_to_coordinate_systems` must be the same."
+        )
         self.regions = list(regions_to_coordinate_systems.keys())  # all regions for the dataloader
 
         cs_region_image: list[tuple[str, str, str]] = []  # list of tuples (coordinate_system, region, image)

src/spatialdata/testing.py

@@ -106,9 +106,9 @@ def assert_elements_are_identical(
         if check_transformations:
             assert transformations0.keys() == transformations1.keys()
             for key in transformations0:
-                assert (
-                    transformations0[key] == transformations1[key]
-                ), f"transformations0[{key}] != transformations1[{key}]"
+                assert transformations0[key] == transformations1[key], (
+                    f"transformations0[{key}] != transformations1[{key}]"
+                )
 
     # compare the elements
     if isinstance(element0, AnnData):

src/spatialdata/transformations/operations.py

@@ -63,7 +63,7 @@ def set_transformation(
             _set_transformations(element, transformations)
         else:
             assert isinstance(transformation, dict), (
-                "If set_all=True, transformation must be of type " "dict[str, BaseTransformation]."
+                "If set_all=True, transformation must be of type dict[str, BaseTransformation]."
             )
             assert to_coordinate_system is None, "If set_all=True, to_coordinate_system must be None."
             _set_transformations(element, transformation)

tests/conftest.py

@@ -117,8 +117,7 @@ def full_sdata() -> SpatialData:
 
 @pytest.fixture(
     # params=["labels"]
-    params=["full", "empty"]
-    + ["images", "labels", "points", "table_single_annotation", "table_multiple_annotations"]
+    params=["full", "empty"] + ["images", "labels", "points", "table_single_annotation", "table_multiple_annotations"]
     # + ["empty_" + x for x in ["table"]] # TODO: empty table not supported yet
 )
 def sdata(request) -> SpatialData:

tests/core/operations/test_rasterize_bins.py

@@ -172,7 +172,7 @@ def _get_sdata(n: int):
     table.obs["region"] = regions
     with pytest.raises(
         ValueError,
-        match="Found multiple regions annotated by the table: " "points, shapes.",
+        match="Found multiple regions annotated by the table: points, shapes.",
     ):
         _ = rasterize_bins(
             sdata=sdata,

tests/core/query/test_spatial_query.py

@@ -280,7 +280,11 @@ def test_query_raster(
     model = (
         Labels3DModel
         if is_labels and is_3d
-        else Labels2DModel if is_labels else Image3DModel if is_3d else Image2DModel
+        else Labels2DModel
+        if is_labels
+        else Image3DModel
+        if is_3d
+        else Image2DModel
     )
 
     image_element = model.parse(image)
@@ -420,7 +424,6 @@ def test_query_polygons(is_bb_3d: bool, with_polygon_query: bool, multiple_boxes
         assert isinstance(polygons_result, list)
         assert len(polygons_result) == 2
         if box_outside_polygon:
-
             assert polygons_result[0] is None
             assert polygons_result[1].index[0] == 3
         else:

tests/io/test_metadata.py

@@ -42,8 +42,7 @@ def test_validate_can_write_metadata_on_element(full_sdata, element_name):
         # trying to save metadata before writing the data
         with pytest.warns(
             UserWarning,
-            match="The SpatialData object appears not to be backed by a Zarr storage, so metadata cannot be "
-            "written.",
+            match="The SpatialData object appears not to be backed by a Zarr storage, so metadata cannot be written.",
         ):
             full_sdata._validate_can_write_metadata_on_element(element_name)
 

tests/io/test_multi_table.py

@@ -99,7 +99,7 @@ def test_set_table_annotates_spatialelement(self, full_sdata, tmp_path):
         tmpdir = Path(tmp_path) / "tmp.zarr"
         del full_sdata["table"].uns[TableModel.ATTRS_KEY]
         with pytest.raises(
-            TypeError, match="No current annotation metadata found. " "Please specify both region_key and instance_key."
+            TypeError, match="No current annotation metadata found. Please specify both region_key and instance_key."
         ):
             full_sdata.set_table_annotates_spatialelement("table", "labels2d", region_key="non_existent")
         with pytest.raises(ValueError, match="Instance key column 'non_existent' not found in table.obs."):

tests/models/test_models.py

@@ -213,9 +213,9 @@ def test_labels_model_with_multiscales(self, model):
         assert actual.scale0.image.dtype == image.dtype
         assert actual.scale1.image.dtype == image.dtype
         assert set(np.unique(image)) == set(np.unique(actual.scale0.image)), "Scale0 should be preserved"
-        assert set(np.unique(image)) >= set(
-            np.unique(actual.scale1.image)
-        ), "Subsequent scales should not have interpolation artifacts"
+        assert set(np.unique(image)) >= set(np.unique(actual.scale1.image)), (
+            "Subsequent scales should not have interpolation artifacts"
+        )
 
     @pytest.mark.parametrize("model", [ShapesModel])
     @pytest.mark.parametrize("path", [POLYGON_PATH, MULTIPOLYGON_PATH, POINT_PATH])

tests/transformations/test_transformations.py

@@ -765,7 +765,7 @@ def test_get_affine_for_element(images):
         np.array(
             [
                 # fmt: off
-                #c  y  x       # noqa: E265
+                # c  y  x       # noqa: E265
                 [1, 0, 0, 0],  # c
                 [0, 0, 1, 1],  # x
                 [0, 1, 0, 2],  # y