Avoid auto creation of indexes in concat

xarray/core/concat.py

@@ -8,6 +8,7 @@
 
 from xarray.core import dtypes, utils
 from xarray.core.alignment import align, reindex_variables
+from xarray.core.coordinates import Coordinates
 from xarray.core.duck_array_ops import lazy_array_equiv
 from xarray.core.indexes import Index, PandasIndex
 from xarray.core.merge import (
@@ -646,14 +647,26 @@ def get_indexes(name):
             # preserves original variable order
             result_vars[name] = result_vars.pop(name)
 
-    result = type(datasets[0])(result_vars, attrs=result_attrs)
-
-    absent_coord_names = coord_names - set(result.variables)
+    absent_coord_names = coord_names - set(result_vars)
     if absent_coord_names:
         raise ValueError(
             f"Variables {absent_coord_names!r} are coordinates in some datasets but not others."
         )
-    result = result.set_coords(coord_names)
+    coord_vars = {
+        name: result_var
+        for name, result_var in result_vars.items()
+        if name in coord_names
+    }
+    coords = Coordinates(coord_vars, indexes=result_indexes)
+
+    # TODO: this is just the complement of the set of coord_vars
+    result_data_vars = {
+        name: result_var
+        for name, result_var in result_vars.items()
+        if name not in coord_names
+    }
+
+    result = type(datasets[0])(result_data_vars, coords=coords, attrs=result_attrs)
     result.encoding = result_encoding
 
     result = result.drop_vars(unlabeled_dims, errors="ignore")
@@ -668,6 +681,7 @@ def get_indexes(name):
         result_indexes[dim] = index
 
     # TODO: add indexes at Dataset creation (when it is supported)
+    # TODO: do we actually need this step anymore?
     result = result._overwrite_indexes(result_indexes)
 
     return result

xarray/tests/test_concat.py

@@ -978,6 +978,32 @@ def test_concat_str_dtype(self, dtype, dim) -> None:
 
         assert np.issubdtype(actual.x2.dtype, dtype)
 
+    def test_concat_avoids_index_auto_creation(self) -> None:
+        # TODO once passing indexes={} directly to DataArray constructor is allowed then no need to create coords first
+        coords = Coordinates({"x": np.array([1, 2, 3])}, indexes={})
+        datasets = [
+            Dataset(
+                {"a": (["x", "y"], np.zeros((3, 3)))},
+                coords=coords,
+            )
+            for _ in range(2)
+        ]
+        # should not raise on concat
+        combined = concat(datasets, dim="x")
+        assert combined["a"].shape == (6, 3)
+        assert combined["a"].dims == ("x", "y")
+
+        # nor have auto-created any indexes
+        assert combined.indexes == {}
+
+        # should not raise on stack
+        combined = concat(datasets, dim="z")
+        assert combined["a"].shape == (2, 3, 3)
+        assert combined["a"].dims == ("z", "x", "y")
+
+        # nor have auto-created any indexes
+        assert combined.indexes == {}
+
 
 class TestConcatDataArray:
     def test_concat(self) -> None:
@@ -1051,6 +1077,33 @@ def test_concat_lazy(self) -> None:
         assert combined.shape == (2, 3, 3)
         assert combined.dims == ("z", "x", "y")
 
+    def test_concat_avoids_index_auto_creation(self) -> None:
+        # TODO once passing indexes={} directly to DataArray constructor is allowed then no need to create coords first
+        coords = Coordinates({"x": np.array([1, 2, 3])}, indexes={})
+        arrays = [
+            DataArray(
+                np.zeros((3, 3)),
+                dims=["x", "y"],
+                coords=coords,
+            )
+            for _ in range(2)
+        ]
+        # should not raise on concat
+        combined = concat(arrays, dim="x")
+        assert combined.shape == (6, 3)
+        assert combined.dims == ("x", "y")
+
+        # nor have auto-created any indexes
+        assert combined.indexes == {}
+
+        # should not raise on stack
+        combined = concat(arrays, dim="z")
+        assert combined.shape == (2, 3, 3)
+        assert combined.dims == ("z", "x", "y")
+
+        # nor have auto-created any indexes
+        assert combined.indexes == {}
+
     @pytest.mark.parametrize("fill_value", [dtypes.NA, 2, 2.0])
     def test_concat_fill_value(self, fill_value) -> None:
         foo = DataArray([1, 2], coords=[("x", [1, 2])])