Add isort rule and lint (#62)

.pre-commit-config.yaml

@@ -1,20 +1,27 @@
+# See https://pre-commit.com for more information
+# See https://pre-commit.com/hooks.html for more hooks
+
 repos:
   - repo: https://github.com/asottile/pyupgrade
     rev: v3.15.2
     hooks:
       - id: pyupgrade
         args:
           - --py38-plus
-  - repo: https://github.com/psf/black
-    rev: 24.4.0
-    hooks:
-      - id: black
-        language_version: python3
-        exclude: versioneer.py
-        args:
-          - --target-version=py38
   - repo: https://github.com/pycqa/flake8
     rev: 7.0.0
     hooks:
       - id: flake8
         language_version: python3
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v2.4.0
+    hooks:
+      - id: trailing-whitespace
+      - id: end-of-file-fixer
+        exclude: tests/.*\.json
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.4.9
+    hooks:
+      - id: ruff
+        args: ["--fix"]
+      - id: ruff-format

pyproject.toml

@@ -84,3 +84,22 @@ ignore_missing_imports = true
 [[tool.mypy.overrides]]
 module = "stac_geoparquet.*"
 disallow_untyped_defs = true
+
+[tool.ruff.lint]
+select = [
+    # Pyflakes
+    "F",
+    # Pycodestyle
+    "E",
+    "W",
+    # isort
+    "I",
+]
+ignore = [
+    "E501", # Line too long
+]
+
+[tool.ruff.lint.extend-per-file-ignores]
+"__init__.py" = [
+    "F401", # Allow unused imports in __init__.py files
+]

stac_geoparquet/_compat.py

@@ -1,5 +1,5 @@
-import pystac
 import packaging.version
+import pystac
 
 PYSTAC_1_7_0 = packaging.version.parse(pystac.__version__) >= packaging.version.Version(
     "1.7.0"

stac_geoparquet/arrow/_util.py

@@ -1,5 +1,6 @@
 import operator
 from functools import reduce
+from itertools import islice
 from typing import (
     Any,
     Dict,
@@ -11,7 +12,6 @@
 )
 
 import pyarrow as pa
-from itertools import islice
 
 
 def update_batch_schema(

stac_geoparquet/cli.py

@@ -2,8 +2,8 @@
 
 import argparse
 import logging
-import sys
 import os
+import sys
 
 from stac_geoparquet import pc_runner
 
@@ -53,6 +53,7 @@ def parse_args(args: list[str] | None = None) -> argparse.Namespace:
 def setup_logging() -> None:
     import logging
     import warnings
+
     import rich.logging
 
     warnings.filterwarnings("ignore", message=".*initial implementation of Parquet.*")

stac_geoparquet/pc_runner.py

@@ -1,11 +1,12 @@
 from __future__ import annotations
 
 import json
+import urllib.parse
 from typing import Any
 
 import azure.data.tables
 import requests
-import urllib.parse
+
 from stac_geoparquet.pgstac_reader import CollectionConfig
 
 PARTITION_FREQUENCIES = {

stac_geoparquet/pgstac_reader.py

@@ -1,25 +1,25 @@
 from __future__ import annotations
-import textwrap
-import hashlib
 
+import collections.abc
+import dataclasses
 import datetime
+import hashlib
+import itertools
 import logging
+import textwrap
 from typing import Any
-import collections.abc
-import itertools
 
+import dateutil.tz
 import fsspec
 import pandas as pd
-import pystac
-import dateutil.tz
-import dataclasses
 import pyarrow.fs
 import pypgstac.db
 import pypgstac.hydration
+import pystac
 import shapely.wkb
 import tqdm.auto
-from stac_geoparquet import to_geodataframe
 
+from stac_geoparquet import to_geodataframe
 
 logger = logging.getLogger(__name__)
 

stac_geoparquet/stac_geoparquet.py

@@ -3,20 +3,19 @@
 """
 
 from __future__ import annotations
-import collections
 
-from typing import Sequence, Any, Literal
+import collections
 import warnings
+from typing import Any, Literal, Sequence
+from urllib.parse import urlparse
 
-import pystac
 import geopandas
+import numpy as np
 import pandas as pd
 import pyarrow as pa
-import numpy as np
+import pystac
 import shapely.geometry
 
-from urllib.parse import urlparse
-
 from stac_geoparquet.utils import fix_empty_multipolygon
 
 STAC_ITEM_TYPES = ["application/json", "application/geo+json"]

stac_geoparquet/utils.py

@@ -3,8 +3,9 @@
 import functools
 from typing import Any
 
-import shapely.geometry
 import pystac
+import shapely.geometry
+import shapely.geometry.base
 
 
 @functools.singledispatch
@@ -36,7 +37,7 @@ def assert_equal_item(
     )
     assert result.bbox == expected.bbox
     assert result.datetime == expected.datetime
-    assert type(result.stac_extensions) == type(expected.stac_extensions)
+    assert isinstance(result.stac_extensions, type(expected.stac_extensions))
     assert sorted(result.stac_extensions) == sorted(expected.stac_extensions)
     assert result.collection_id == expected.collection_id
     assert result.extra_fields == expected.extra_fields
@@ -86,7 +87,7 @@ def assert_link_equal(
 
 
 def fix_empty_multipolygon(
-    item_geometry: dict[str, Any]
+    item_geometry: dict[str, Any],
 ) -> shapely.geometry.base.BaseGeometry:
     # Filter out missing geoms in MultiPolygons
     # https://github.com/shapely/shapely/issues/1407

tests/json_equals.py

@@ -3,7 +3,6 @@
 
 from ciso8601 import parse_rfc3339
 
-
 JsonValue = Union[list, tuple, int, float, dict, str, bool, None]
 
 

tests/test_pgstac_reader.py

@@ -2,15 +2,14 @@
 import json
 import pathlib
 
-import pystac
 import dateutil
 import pandas as pd
+import pystac
 import pytest
 
 import stac_geoparquet.pgstac_reader
-from stac_geoparquet.utils import assert_equal
 from stac_geoparquet._compat import PYSTAC_1_7_0
-
+from stac_geoparquet.utils import assert_equal
 
 HERE = pathlib.Path(__file__).parent
 

tests/test_stac_geoparquet.py

@@ -1,16 +1,16 @@
 import json
 import pathlib
 
-import stac_geoparquet
-import shapely.geometry
+import geopandas
 import pandas as pd
 import pandas.testing
 import pyarrow as pa
 import pystac
-import geopandas
-import requests
 import pytest
+import requests
+import shapely.geometry
 
+import stac_geoparquet
 from stac_geoparquet.stac_geoparquet import to_item_collection
 from stac_geoparquet.utils import assert_equal, fix_empty_multipolygon
 

tests/test_to_dict.py

@@ -1,11 +1,10 @@
 import pathlib
 
-import pytest
 import geopandas
+import pytest
 
 import stac_geoparquet
 
-
 HERE = pathlib.Path(__file__).parent