refactor: Improve code readability with consistent formatting

Author: EnyMan

pyiceberg/table/__init__.py

@@ -1799,16 +1799,20 @@ def projection(self) -> Schema:
         return current_schema.select(*self.selected_fields, case_sensitive=self.case_sensitive)
 
     @abstractmethod
-    def plan_files(self) -> Iterable[ScanTask]: ...
+    def plan_files(self) -> Iterable[ScanTask]:
+        ...
 
     @abstractmethod
-    def to_arrow(self) -> pa.Table: ...
+    def to_arrow(self) -> pa.Table:
+        ...
 
     @abstractmethod
-    def to_pandas(self, **kwargs: Any) -> pd.DataFrame: ...
+    def to_pandas(self, **kwargs: Any) -> pd.DataFrame:
+        ...
 
     @abstractmethod
-    def to_polars(self) -> pl.DataFrame: ...
+    def to_polars(self) -> pl.DataFrame:
+        ...
 
     def update(self: S, **overrides: Any) -> S:
         """Create a copy of this table scan with updated fields."""
@@ -1841,7 +1845,8 @@ def with_case_sensitive(self: S, case_sensitive: bool = True) -> S:
         return self.update(case_sensitive=case_sensitive)
 
     @abstractmethod
-    def count(self) -> int: ...
+    def count(self) -> int:
+        ...
 
 
 class ScanTask:

pyiceberg/table/upsert_util.py

@@ -191,9 +191,7 @@ def has_duplicate_rows(df: pyarrow_table, join_cols: list[str]) -> bool:
     return len(df.select(join_cols).group_by(join_cols).aggregate([([], "count_all")]).filter(pc.field("count_all") > 1)) > 0
 
 
-def _compare_columns_vectorized(
-    source_col: pa.Array | pa.ChunkedArray, target_col: pa.Array | pa.ChunkedArray
-) -> pa.Array:
+def _compare_columns_vectorized(source_col: pa.Array | pa.ChunkedArray, target_col: pa.Array | pa.ChunkedArray) -> pa.Array:
     """
     Vectorized comparison of two columns, returning a boolean array where True means values differ.
 
@@ -223,7 +221,7 @@ def _compare_columns_vectorized(
 
         # PyArrow cannot directly compare struct columns, so we recursively compare each field
         diff_masks = []
-        for i, field in enumerate(col_type):
+        for i, _field in enumerate(col_type):
             src_field = pc.struct_field(source_col, [i])
             tgt_field = pc.struct_field(target_col, [i])
             field_diff = _compare_columns_vectorized(src_field, tgt_field)
@@ -237,7 +235,12 @@ def _compare_columns_vectorized(
         field_diff = functools.reduce(pc.or_, diff_masks)
         return pc.or_(field_diff, struct_null_diff)
 
-    elif pa.types.is_list(col_type) or pa.types.is_large_list(col_type) or pa.types.is_fixed_size_list(col_type) or pa.types.is_map(col_type):
+    elif (
+        pa.types.is_list(col_type)
+        or pa.types.is_large_list(col_type)
+        or pa.types.is_fixed_size_list(col_type)
+        or pa.types.is_map(col_type)
+    ):
         # For list/map types, fall back to Python comparison as PyArrow doesn't support vectorized comparison
         # This is still faster than the original row-by-row approach since we batch the conversion
         source_py = source_col.to_pylist()

tests/table/test_upsert.py

@@ -100,12 +100,14 @@ def gen_target_iceberg_table(
 ) -> Table:
     additional_columns = ", t.order_id + 1000 as order_line_id" if composite_key else ""
 
-    df = ctx.sql(f"""
+    df = ctx.sql(
+        f"""
         with t as (SELECT unnest(range({start_row},{end_row + 1})) as order_id)
         SELECT t.order_id {additional_columns}
             , date '2021-01-01' as order_date, 'A' as order_type
         from t
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     table = catalog.create_table(identifier, df.schema)
 
@@ -166,23 +168,27 @@ def test_merge_scenario_skip_upd_row(catalog: Catalog) -> None:
 
     ctx = SessionContext()
 
-    df = ctx.sql("""
+    df = ctx.sql(
+        """
         select 1 as order_id, date '2021-01-01' as order_date, 'A' as order_type
         union all
         select 2 as order_id, date '2021-01-01' as order_date, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     table = catalog.create_table(identifier, df.schema)
 
     table.append(df)
 
-    source_df = ctx.sql("""
+    source_df = ctx.sql(
+        """
         select 1 as order_id, date '2021-01-01' as order_date, 'A' as order_type
         union all
         select 2 as order_id, date '2021-01-01' as order_date, 'B' as order_type
         union all
         select 3 as order_id, date '2021-01-01' as order_date, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     res = table.upsert(df=source_df, join_cols=["order_id"])
 
@@ -202,23 +208,27 @@ def test_merge_scenario_date_as_key(catalog: Catalog) -> None:
     identifier = "default.test_merge_scenario_date_as_key"
     _drop_table(catalog, identifier)
 
-    df = ctx.sql("""
+    df = ctx.sql(
+        """
         select date '2021-01-01' as order_date, 'A' as order_type
         union all
         select date '2021-01-02' as order_date, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     table = catalog.create_table(identifier, df.schema)
 
     table.append(df)
 
-    source_df = ctx.sql("""
+    source_df = ctx.sql(
+        """
         select date '2021-01-01' as order_date, 'A' as order_type
         union all
         select date '2021-01-02' as order_date, 'B' as order_type
         union all
         select date '2021-01-03' as order_date, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     res = table.upsert(df=source_df, join_cols=["order_date"])
 
@@ -238,23 +248,27 @@ def test_merge_scenario_string_as_key(catalog: Catalog) -> None:
 
     ctx = SessionContext()
 
-    df = ctx.sql("""
+    df = ctx.sql(
+        """
         select 'abc' as order_id, 'A' as order_type
         union all
         select 'def' as order_id, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     table = catalog.create_table(identifier, df.schema)
 
     table.append(df)
 
-    source_df = ctx.sql("""
+    source_df = ctx.sql(
+        """
         select 'abc' as order_id, 'A' as order_type
         union all
         select 'def' as order_id, 'B' as order_type
         union all
         select 'ghi' as order_id, 'A' as order_type
-    """).to_arrow_table()
+    """
+    ).to_arrow_table()
 
     res = table.upsert(df=source_df, join_cols=["order_id"])
 
@@ -1563,5 +1577,3 @@ def test_coarse_match_filter_composite_key_mixed_types() -> None:
     # numeric_id is sparse (density < 10%), so should use In()
     # string_id is non-numeric, so should use In()
     assert isinstance(result, And)
-
-