Ruff PERF: fixed all remaining performance linting issues

Author: evertlammerts

duckdb/experimental/spark/exception.py

@@ -1,7 +1,7 @@
 from typing import Optional
 
 
-class ContributionsAcceptedError(NotImplementedError):  # noqa: D100
+class ContributionsAcceptedError(NotImplementedError):
     """This method is not planned to be implemented, if you would like to implement this method
     or show your interest in this method to other members of the community,
     feel free to open up a PR or a Discussion over on https://github.com/duckdb/duckdb.

duckdb/experimental/spark/sql/types.py

@@ -8,7 +8,7 @@
 import re
 import time
 from builtins import tuple
-from collections.abc import Iterator
+from collections.abc import Iterator, Mapping
 from types import MappingProxyType
 from typing import (
     Any,
@@ -18,7 +18,7 @@
     TypeVar,
     Union,
     cast,
-    overload, Mapping,
+    overload,
 )
 
 import duckdb

duckdb/polars_io.py

@@ -235,15 +235,12 @@ def source_generator(
             results = relation_final.fetch_arrow_reader()
         else:
             results = relation_final.fetch_arrow_reader(batch_size)
-        while True:
-            try:
-                record_batch = results.read_next_batch()
-                if predicate is not None and duck_predicate is None:
-                    # We have a predicate, but did not manage to push it down, we fallback here
-                    yield pl.from_arrow(record_batch).filter(predicate)
-                else:
-                    yield pl.from_arrow(record_batch)
-            except StopIteration:
-                break
+
+        for record_batch in iter(results.read_next_batch, None):
+            if predicate is not None and duck_predicate is None:
+                # We have a predicate, but did not manage to push it down, we fallback here
+                yield pl.from_arrow(record_batch).filter(predicate)
+            else:
+                yield pl.from_arrow(record_batch)
 
     return register_io_source(source_generator, schema=schema)

scripts/generate_connection_methods.py

@@ -109,8 +109,7 @@ def create_definition(name, method) -> str:
     body = []
     for method in connection_methods:
         names = method["name"] if isinstance(method["name"], list) else [method["name"]]
-        for name in names:
-            body.append(create_definition(name, method))
+        body.extend(create_definition(name, method) for name in names)
 
     # ---- End of generation code ----
 

scripts/generate_import_cache_cpp.py

@@ -133,10 +133,7 @@ def __init__(self, module: dict) -> None:
         self.classes.reverse()
 
     def get_classes(self):
-        classes = []
-        for item in self.classes:
-            classes.append(item.to_string())
-        return "".join(classes)
+        return "".join(item.to_string() for item in self.classes)
 
     def to_string(self):
         string = f"""
@@ -233,10 +230,8 @@ def get_root_modules(files: list[ModuleFile]):
 
 
 def get_module_file_path_includes(files: list[ModuleFile]):
-    includes = []
-    for file in files:
-        includes.append(f'#include "duckdb_python/import_cache/modules/{file.file_name}"')
-    return "\n".join(includes)
+    template = '#include "duckdb_python/import_cache/modules/{}'
+    return "\n".join(template.format(f.file_name) for f in files)
 
 
 module_includes = get_module_file_path_includes(files)

tests/conftest.py

@@ -207,9 +207,7 @@ def _require(extension_name, db_name="") -> Union[duckdb.DuckDBPyConnection, Non
 
         extension_paths_found = []
         for pattern in extension_search_patterns:
-            extension_pattern_abs = Path(pattern).resolve()
-            for path in extension_pattern_abs.glob("*"):
-                extension_paths_found.append(path)
+            extension_paths_found.extend(list(Path(pattern).resolve().glob("*")))
 
         for path in extension_paths_found:
             print(path)

tests/coverage/test_pandas_categorical_coverage.py

@@ -72,14 +72,10 @@ def check_create_table(category, pandas):
 class TestCategory:
     @pytest.mark.parametrize("pandas", [NumpyPandas()])
     def test_category_string_uint16(self, duckdb_cursor, pandas):
-        category = []
-        for i in range(300):
-            category.append(str(i))
+        category = [str(i) for i in range(300)]
         check_create_table(category, pandas)
 
     @pytest.mark.parametrize("pandas", [NumpyPandas()])
     def test_category_string_uint32(self, duckdb_cursor, pandas):
-        category = []
-        for i in range(70000):
-            category.append(str(i))
+        category = [str(i) for i in range(70000)]
         check_create_table(category, pandas)

tests/fast/api/test_duckdb_connection.py

@@ -87,12 +87,9 @@ def test_cursor_lifetime(self):
         con = duckdb.connect()
 
         def use_cursors() -> None:
-            cursors = []
-            for _ in range(10):
-                cursors.append(con.cursor())
+            cursors = [con.cursor() for _ in range(10)]
 
             for cursor in cursors:
-                print("closing cursor")
                 cursor.close()
 
         use_cursors()

tests/fast/arrow/test_arrow_list.py

@@ -24,11 +24,7 @@ def create_and_register_comparison_result(column_list, duckdb_cursor):
     column_amount = len(column_list)
     assert column_amount
     row_amount = len(column_list[0][2])
-    inserted_values = []
-    for row in range(row_amount):
-        for col in range(column_amount):
-            inserted_values.append(column_list[col][2][row])
-    inserted_values = tuple(inserted_values)
+    inserted_values = tuple(column_list[col][2][row] for row in range(row_amount) for col in range(column_amount))
 
     column_format = ",".join(["?" for _ in range(column_amount)])
     row_format = ",".join([f"({column_format})" for _ in range(row_amount)])

tests/fast/pandas/test_df_object_resolution.py

@@ -372,7 +372,7 @@ def test_structs_in_nested_types(self, pandas, duckdb_cursor):
             "v4": ObjectPair({}, {"key1": 21}),
         }
 
-        for _, pair in pairs.items():
+        for pair in pairs.values():
             check_struct_upgrade("MAP(VARCHAR, INTEGER)[]", construct_list, pair, pandas, duckdb_cursor)
 
         for key, pair in pairs.items():