test: add parameterized tests for parquet file sorting functionality

Author: kosiew

python/tests/test_dataframe.py

@@ -34,6 +34,9 @@
     column,
     literal,
 )
+from datafusion import (
+    col as df_col,
+)
 from datafusion import (
     functions as f,
 )
@@ -2753,3 +2756,34 @@ def test_show_from_empty_batch(capsys) -> None:
     ctx.create_dataframe([[batch]]).show()
     out = capsys.readouterr().out
     assert "| a |" in out
+
+
+@pytest.mark.parametrize("file_sort_order", [[["a"]], [[df_col("a")]]])
+def test_register_parquet_file_sort_order(ctx, tmp_path, file_sort_order):
+    table = pa.table({"a": [1, 2]})
+    path = tmp_path / "file.parquet"
+    pa.parquet.write_table(table, path)
+    ctx.register_parquet("t", path, file_sort_order=file_sort_order)
+    assert "t" in ctx.catalog().schema().names()
+
+
+@pytest.mark.parametrize("file_sort_order", [[["a"]], [[df_col("a")]]])
+def test_register_listing_table_file_sort_order(ctx, tmp_path, file_sort_order):
+    table = pa.table({"a": [1, 2]})
+    dir_path = tmp_path / "dir"
+    dir_path.mkdir()
+    pa.parquet.write_table(table, dir_path / "file.parquet")
+    ctx.register_listing_table(
+        "t", dir_path, schema=table.schema, file_sort_order=file_sort_order
+    )
+    assert "t" in ctx.catalog().schema().names()
+
+
+@pytest.mark.parametrize("file_sort_order", [[["a"]], [[df_col("a")]]])
+def test_read_parquet_file_sort_order(tmp_path, file_sort_order):
+    ctx = SessionContext()
+    table = pa.table({"a": [1, 2]})
+    path = tmp_path / "data.parquet"
+    pa.parquet.write_table(table, path)
+    df = ctx.read_parquet(path, file_sort_order=file_sort_order)
+    assert df.collect()[0].column(0).to_pylist() == [1, 2]