Fix set_index when an existing dimension becomes a level (pydata#3520)

* Added a test

* Fix set_index

* lint

* black / mypy

* Use _replace method

* whats new

doc/whats-new.rst

@@ -79,6 +79,8 @@ New Features
 
 Bug fixes
 ~~~~~~~~~
+- Fix a bug in `set_index` in case that an existing dimension becomes a level variable of MultiIndex. (:pull:`3520`)
+  By `Keisuke Fujii <https://github.com/fujiisoup>`_.
 - Harmonize `_FillValue`, `missing_value` during encoding and decoding steps. (:pull:`3502`)
   By `Anderson Banihirwe <https://github.com/andersy005>`_. 
 - Fix regression introduced in v0.14.0 that would cause a crash if dask is installed

xarray/core/dataarray.py

@@ -48,7 +48,7 @@
     assert_coordinate_consistent,
     remap_label_indexers,
 )
-from .dataset import Dataset, merge_indexes, split_indexes
+from .dataset import Dataset, split_indexes
 from .formatting import format_item
 from .indexes import Indexes, default_indexes
 from .merge import PANDAS_TYPES
@@ -1601,10 +1601,10 @@ def set_index(
         --------
         DataArray.reset_index
         """
-        _check_inplace(inplace)
-        indexes = either_dict_or_kwargs(indexes, indexes_kwargs, "set_index")
-        coords, _ = merge_indexes(indexes, self._coords, set(), append=append)
-        return self._replace(coords=coords)
+        ds = self._to_temp_dataset().set_index(
+            indexes, append=append, inplace=inplace, **indexes_kwargs
+        )
+        return self._from_temp_dataset(ds)
 
     def reset_index(
         self,

xarray/core/dataset.py

@@ -204,6 +204,7 @@ def merge_indexes(
     """
     vars_to_replace: Dict[Hashable, Variable] = {}
     vars_to_remove: List[Hashable] = []
+    dims_to_replace: Dict[Hashable, Hashable] = {}
     error_msg = "{} is not the name of an existing variable."
 
     for dim, var_names in indexes.items():
@@ -244,7 +245,7 @@ def merge_indexes(
         if not len(names) and len(var_names) == 1:
             idx = pd.Index(variables[var_names[0]].values)
 
-        else:
+        else:  # MultiIndex
             for n in var_names:
                 try:
                     var = variables[n]
@@ -256,15 +257,22 @@ def merge_indexes(
                 levels.append(cat.categories)
 
             idx = pd.MultiIndex(levels, codes, names=names)
+            for n in names:
+                dims_to_replace[n] = dim
 
         vars_to_replace[dim] = IndexVariable(dim, idx)
         vars_to_remove.extend(var_names)
 
     new_variables = {k: v for k, v in variables.items() if k not in vars_to_remove}
     new_variables.update(vars_to_replace)
+
+    # update dimensions if necessary  GH: 3512
+    for k, v in new_variables.items():
+        if any(d in dims_to_replace for d in v.dims):
+            new_dims = [dims_to_replace.get(d, d) for d in v.dims]
+            new_variables[k] = v._replace(dims=new_dims)
     new_coord_names = coord_names | set(vars_to_replace)
     new_coord_names -= set(vars_to_remove)
-
     return new_variables, new_coord_names
 
 

xarray/tests/test_dataarray.py

@@ -1182,6 +1182,16 @@ def test_selection_multiindex_remove_unused(self):
         expected = expected.set_index(xy=["x", "y"]).unstack()
         assert_identical(expected, actual)
 
+    def test_selection_multiindex_from_level(self):
+        # GH: 3512
+        da = DataArray([0, 1], dims=["x"], coords={"x": [0, 1], "y": "a"})
+        db = DataArray([2, 3], dims=["x"], coords={"x": [0, 1], "y": "b"})
+        data = xr.concat([da, db], dim="x").set_index(xy=["x", "y"])
+        assert data.dims == ("xy",)
+        actual = data.sel(y="a")
+        expected = data.isel(xy=[0, 1]).unstack("xy").squeeze("y").drop("y")
+        assert_equal(actual, expected)
+
     def test_virtual_default_coords(self):
         array = DataArray(np.zeros((5,)), dims="x")
         expected = DataArray(range(5), dims="x", name="x")