fix(python): Ensure write_excel recognises the Array dtype and writ…

…es it out as a string (#20994)

py-polars/polars/io/spreadsheet/_write_utils.py

@@ -12,9 +12,7 @@
     Datetime,
     Float64,
     Int64,
-    List,
     Object,
-    Struct,
     Time,
 )
 from polars.datatypes.group import FLOAT_DTYPES, INTEGER_DTYPES
@@ -190,13 +188,13 @@ def _xl_column_range(
     include_header: bool,
     as_range: bool = True,
 ) -> tuple[int, int, int, int] | str:
-    """Return the excel sheet range of a named column, accounting for all offsets."""
+    """Return the Excel sheet range of a named column, accounting for all offsets."""
     col_start = (
         table_start[0] + int(include_header),
         table_start[1] + (df.get_column_index(col) if isinstance(col, str) else col[0]),
     )
     col_finish = (
-        col_start[0] + len(df) - 1,
+        col_start[0] + df.height - 1,
         col_start[1] + (0 if isinstance(col, str) else (col[1] - col[0])),
     )
     if as_range:
@@ -358,7 +356,7 @@ def _map_str(s: Series) -> Series:
     cast_cols = [
         F.col(col).map_batches(_map_str).alias(col)
         for col, tp in df.schema.items()
-        if tp in (List, Struct, Object)
+        if (tp.is_nested() or tp == Object)
     ]
     if cast_cols:
         df = df.with_columns(cast_cols)
@@ -569,7 +567,7 @@ def _xl_setup_workbook(
     workbook: Workbook | BytesIO | Path | str | None,
     worksheet: str | Worksheet | None = None,
 ) -> tuple[Workbook, Worksheet, bool]:
-    """Establish the target excel workbook and worksheet."""
+    """Establish the target Excel workbook and worksheet."""
     from xlsxwriter import Workbook
     from xlsxwriter.worksheet import Worksheet
 

py-polars/polars/io/spreadsheet/functions.py

@@ -948,17 +948,17 @@ def _drop_null_data(
                 col = df[col_name]
                 if (
                     col.dtype == Null
-                    or col.null_count() == len(df)
+                    or col.null_count() == df.height
                     or (
                         col.dtype in NUMERIC_DTYPES
-                        and col.replace(0, None).null_count() == len(df)
+                        and col.replace(0, None).null_count() == df.height
                     )
                 ):
                     null_cols.append(col_name)
         if null_cols:
             df = df.drop(*null_cols)
 
-    if len(df) == 0 and len(df.columns) == 0:
+    if df.height == df.width == 0:
         return _empty_frame(raise_if_empty)
     if drop_empty_rows:
         return df.filter(~F.all_horizontal(F.all().is_null()))

py-polars/tests/unit/io/test_spreadsheet.py

@@ -19,7 +19,12 @@
 if TYPE_CHECKING:
     from collections.abc import Sequence
 
-    from polars._typing import ExcelSpreadsheetEngine, SchemaDict, SelectorType
+    from polars._typing import (
+        ExcelSpreadsheetEngine,
+        PolarsDataType,
+        SchemaDict,
+        SelectorType,
+    )
 
 
 # pytestmark = pytest.mark.slow()
@@ -543,22 +548,6 @@ def test_read_mixed_dtype_columns(
     )
 
 
-@pytest.mark.parametrize("engine", ["calamine", "openpyxl", "xlsx2csv"])
-def test_write_excel_bytes(engine: ExcelSpreadsheetEngine) -> None:
-    df = pl.DataFrame({"colx": [1.5, -2, 0], "coly": ["a", None, "c"]})
-
-    excel_bytes = BytesIO()
-    df.write_excel(excel_bytes)
-
-    df_read = pl.read_excel(excel_bytes, engine=engine)
-    assert_frame_equal(df, df_read)
-
-    # also confirm consistent behaviour when 'infer_schema_length=0'
-    df_read = pl.read_excel(excel_bytes, engine=engine, infer_schema_length=0)
-    expected = pl.DataFrame({"colx": ["1.5", "-2", "0"], "coly": ["a", None, "c"]})
-    assert_frame_equal(expected, df_read)
-
-
 def test_schema_overrides(path_xlsx: Path, path_xlsb: Path, path_ods: Path) -> None:
     df1 = pl.read_excel(
         path_xlsx,
@@ -900,10 +889,21 @@ def test_excel_write_column_and_row_totals(engine: ExcelSpreadsheetEngine) -> No
         assert xldf.row(-1) == (None, 0.0, 0.0, 0, 0, None, 0.0, 0)
 
 
-@pytest.mark.parametrize("engine", ["calamine", "openpyxl", "xlsx2csv"])
-def test_excel_write_compound_types(engine: ExcelSpreadsheetEngine) -> None:
+@pytest.mark.parametrize(
+    ("engine", "list_dtype"),
+    [
+        ("calamine", pl.List(pl.Int8)),
+        ("openpyxl", pl.List(pl.UInt16)),
+        ("xlsx2csv", pl.Array(pl.Int32, 2)),
+    ],
+)
+def test_excel_write_compound_types(
+    engine: ExcelSpreadsheetEngine,
+    list_dtype: PolarsDataType,
+) -> None:
     df = pl.DataFrame(
-        {"x": [[1, 2], [3, 4], [5, 6]], "y": ["a", "b", "c"], "z": [9, 8, 7]}
+        data={"x": [[1, 2], [3, 4], [5, 6]], "y": ["a", "b", "c"], "z": [9, 8, 7]},
+        schema_overrides={"x": pl.Array(pl.Int32, 2)},
     ).select("x", pl.struct(["y", "z"]))
 
     xls = BytesIO()
@@ -966,6 +966,22 @@ def test_excel_read_named_table_with_total_row(tmp_path: Path) -> None:
     assert xldf.row(3) == (None, 0, 0)
 
 
+@pytest.mark.parametrize("engine", ["calamine", "openpyxl", "xlsx2csv"])
+def test_excel_write_to_bytesio(engine: ExcelSpreadsheetEngine) -> None:
+    df = pl.DataFrame({"colx": [1.5, -2, 0], "coly": ["a", None, "c"]})
+
+    excel_bytes = BytesIO()
+    df.write_excel(excel_bytes)
+
+    df_read = pl.read_excel(excel_bytes, engine=engine)
+    assert_frame_equal(df, df_read)
+
+    # also confirm consistent behaviour when 'infer_schema_length=0'
+    df_read = pl.read_excel(excel_bytes, engine=engine, infer_schema_length=0)
+    expected = pl.DataFrame({"colx": ["1.5", "-2", "0"], "coly": ["a", None, "c"]})
+    assert_frame_equal(expected, df_read)
+
+
 @pytest.mark.parametrize("engine", ["xlsx2csv", "openpyxl", "calamine"])
 def test_excel_write_to_file_object(
     engine: ExcelSpreadsheetEngine, tmp_path: Path
@@ -1352,7 +1368,7 @@ def test_drop_empty_rows(
     assert df3.shape == (10, 4)
 
 
-def test_write_excel_select_col_dtype() -> None:
+def test_excel_write_select_col_dtype() -> None:
     from openpyxl import load_workbook
     from xlsxwriter import Workbook