[CI] Add pre-commit hook pyupgrade to auto upgrade Python syntax

"A tool (and pre-commit hook) to automatically upgrade syntax for newer versions of the language."

https://github.com/asottile/pyupgrade

Author: jbampton

.pre-commit-config.yaml

@@ -10,6 +10,10 @@ repos:
     hooks:
       - id: identity
       - id: check-hooks-apply
+  - repo: https://github.com/asottile/pyupgrade
+    rev: v3.18.0
+    hooks:
+      - id: pyupgrade
   - repo: https://github.com/psf/black-pre-commit-mirror
     rev: 24.10.0
     hooks:

python/sedona/core/jvm/config.py

@@ -28,7 +28,7 @@
 
 from sedona.utils.decorators import classproperty
 
-string_types = (type(b""), type(""))
+string_types = (bytes, str)
 
 
 def is_greater_or_equal_version(version_a: str, version_b: str) -> bool:

python/sedona/maps/SedonaPyDeck.py

@@ -315,7 +315,7 @@ def _create_default_fill_color_(cls, gdf, plot_col):
         :return: fill_color string for pydeck map
         """
         plot_max = gdf[plot_col].max()
-        return "[85, 183, 177, ({0} / {1}) * 255 + 15]".format(plot_col, plot_max)
+        return "[85, 183, 177, ({} / {}) * 255 + 15]".format(plot_col, plot_max)
 
     @classmethod
     def _create_coord_column_(cls, gdf, geometry_col, add_points=False):

python/sedona/utils/decorators.py

@@ -20,7 +20,7 @@
 T = TypeVar("T")
 
 
-class classproperty(object):
+class classproperty:
 
     def __init__(self, f):
         self.f = f

python/setup.py

@@ -21,7 +21,7 @@
 
 from sedona import version
 
-with open("README.md", "r") as fh:
+with open("README.md") as fh:
     long_description = fh.read()
 
 extension_args = {}

python/tests/core/test_avoiding_python_jvm_serde_df.py

@@ -165,24 +165,20 @@ def test_spatial_join_query_flat_to_df(self):
         right_geometries = self.__row_to_list(right_geometries_raw)
 
         # Ignore the ordering of these
-        assert set(geom[0] for geom in left_geometries) == set(
-            [
-                "POLYGON ((0 4, -3 3, -8 6, -6 8, -2 9, 0 4))",
-                "POLYGON ((10 3, 10 6, 14 6, 14 3, 10 3))",
-                "POLYGON ((2 2, 2 4, 3 5, 7 5, 9 3, 8 1, 4 1, 2 2))",
-                "POLYGON ((-1 -1, -1 -3, -2 -5, -6 -8, -5 -2, -3 -2, -1 -1))",
-                "POLYGON ((-1 -1, -1 -3, -2 -5, -6 -8, -5 -2, -3 -2, -1 -1))",
-            ]
-        )
-        assert set(geom[0] for geom in right_geometries) == set(
-            [
-                "POINT (-3 5)",
-                "POINT (11 5)",
-                "POINT (4 3)",
-                "POINT (-1 -1)",
-                "POINT (-4 -5)",
-            ]
-        )
+        assert {geom[0] for geom in left_geometries} == {
+            "POLYGON ((0 4, -3 3, -8 6, -6 8, -2 9, 0 4))",
+            "POLYGON ((10 3, 10 6, 14 6, 14 3, 10 3))",
+            "POLYGON ((2 2, 2 4, 3 5, 7 5, 9 3, 8 1, 4 1, 2 2))",
+            "POLYGON ((-1 -1, -1 -3, -2 -5, -6 -8, -5 -2, -3 -2, -1 -1))",
+            "POLYGON ((-1 -1, -1 -3, -2 -5, -6 -8, -5 -2, -3 -2, -1 -1))",
+        }
+        assert {geom[0] for geom in right_geometries} == {
+            "POINT (-3 5)",
+            "POINT (11 5)",
+            "POINT (4 3)",
+            "POINT (-1 -1)",
+            "POINT (-4 -5)",
+        }
 
     def test_range_query_flat_to_df(self):
         poi_point_rdd = WktReader.readToGeometryRDD(

python/tests/sql/test_function.py

@@ -1911,14 +1911,10 @@ def test_st_collect_on_array_type(self):
         )
 
         # then result should be as expected
-        assert set(
-            [
-                el[0]
-                for el in geometry_df_collected.selectExpr(
-                    "ST_AsText(collected)"
-                ).collect()
-            ]
-        ) == {
+        assert {
+            el[0]
+            for el in geometry_df_collected.selectExpr("ST_AsText(collected)").collect()
+        } == {
             "MULTILINESTRING ((1 2, 3 4), (3 4, 4 5))",
             "MULTIPOINT ((1 2), (-2 3))",
             "MULTIPOLYGON (((1 2, 1 4, 3 4, 3 2, 1 2)), ((0.5 0.5, 5 0, 5 5, 0 5, 0.5 0.5)))",
@@ -1944,14 +1940,10 @@ def test_st_collect_on_multiple_columns(self):
         )
 
         # then result should be calculated
-        assert set(
-            [
-                el[0]
-                for el in geometry_df_collected.selectExpr(
-                    "ST_AsText(collected)"
-                ).collect()
-            ]
-        ) == {
+        assert {
+            el[0]
+            for el in geometry_df_collected.selectExpr("ST_AsText(collected)").collect()
+        } == {
             "MULTILINESTRING ((1 2, 3 4), (3 4, 4 5))",
             "MULTIPOINT ((1 2), (-2 3))",
             "MULTIPOLYGON (((1 2, 1 4, 3 4, 3 2, 1 2)), ((0.5 0.5, 5 0, 5 5, 0 5, 0.5 0.5)))",

python/tests/stats/test_dbscan.py

@@ -92,7 +92,7 @@ def get_actual_results(
 
         clusters = {
             frozenset([y[0] for y in clusters_members if y[1] == x])
-            for x in set([y[1] for y in clusters_members])
+            for x in {y[1] for y in clusters_members}
         }
 
         return clusters