fix: Include index coordinates in to_dataframe when name differs from dimension (Fixes #10851)

Author: samay2504

xarray/core/dataset.py

@@ -7202,11 +7202,23 @@ def to_pandas(self) -> pd.Series | pd.DataFrame:
 
     def _to_dataframe(self, ordered_dims: Mapping[Any, int]):
         from xarray.core.extension_array import PandasExtensionArray
-
-        # All and only non-index arrays (whether data or coordinates) should
-        # become columns in the output DataFrame. Excluding indexes rather
-        # than dims handles the case of a MultiIndex along a single dimension.
-        columns_in_order = [k for k in self.variables if k not in self.xindexes]
+        from xarray.core.indexes import PandasIndex, PandasMultiIndex
+
+        # All non-index variables become columns. For indexes, I excluded:
+        # 1. PandasMultiIndex components (A, B from a MultiIndex)
+        # 2. PandasIndex where name matches dim (e.g., 'x' indexing dim 'x')
+        # 3. Any index whose name matches a dimension
+        # This allows PandasIndex coords created via set_xindex with a different
+        # name (e.g., 'pf' indexing dim 'pos') to be included as columns.
+        indexes_to_exclude = set()
+        for name, idx in self.xindexes.items():
+            if (
+                isinstance(idx, PandasMultiIndex)
+                or (isinstance(idx, PandasIndex) and name == idx.dim)
+                or name in self.dims
+            ):
+                indexes_to_exclude.add(name)
+        columns_in_order = [k for k in self.variables if k not in indexes_to_exclude]
         non_extension_array_columns = [
             k
             for k in columns_in_order

xarray/tests/test_issue_10851.py

@@ -0,0 +1,51 @@
+"""Test for issue #10851: Dataset Index not included in to_dataframe when name differs from dimension."""
+import numpy as np
+import pandas as pd
+
+import xarray as xr
+
+
+class TestToDataFrameIndexColumn:
+    """Tests for to_dataframe including index coordinates with different names."""
+
+    def test_to_dataframe_includes_index_with_different_name(self):
+        """Index coordinates with name different from dimension should be in columns."""
+        ds_temp = xr.Dataset(
+            data_vars=dict(temp=(["time", "pos"], np.array([[5, 10, 15, 20, 25]]))),
+            coords=dict(
+                pf=("pos", [1.0, 2.0, 4.2, 8.0, 10.0]),
+                time=("time", [pd.to_datetime("2025-01-01")]),
+            ),
+        ).set_xindex("pf")
+
+        df = ds_temp.to_dataframe()
+
+        assert "pf" in df.columns
+        assert "temp" in df.columns
+        np.testing.assert_array_equal(df["pf"].values, [1.0, 2.0, 4.2, 8.0, 10.0])
+
+    def test_to_dataframe_still_excludes_matching_dim_index(self):
+        """Index coordinates where name matches dimension should not be in columns."""
+        ds = xr.Dataset(
+            data_vars=dict(temp=(["x"], [1, 2, 3])),
+            coords=dict(x=("x", [10, 20, 30])),
+        )
+
+        df = ds.to_dataframe()
+
+        assert "temp" in df.columns
+        assert "x" not in df.columns
+
+    def test_to_dataframe_roundtrip_with_set_xindex(self):
+        """Dataset with set_xindex should roundtrip to DataFrame correctly."""
+        ds = xr.Dataset(
+            data_vars=dict(val=(["dim"], [100, 200, 300])),
+            coords=dict(coord_idx=("dim", ["a", "b", "c"])),
+        ).set_xindex("coord_idx")
+
+        df = ds.to_dataframe()
+
+        assert "coord_idx" in df.columns
+        assert "val" in df.columns
+        assert list(df["coord_idx"]) == ["a", "b", "c"]
+        assert list(df["val"]) == [100, 200, 300]