Enable mypy type checking for 5 test files (#10763)

* Enable mypy type checking for 8 test files

Remove 8 test files from mypy exclusion list and fix all type errors:
- test_duck_array_ops.py
- test_indexing.py
- test_sparse.py
- test_units.py
- test_variable.py
- test_merge.py
- test_ufuncs.py
- test_weighted.py

Changes include:
- Added type annotations for function return types and variables
- Fixed variable name conflicts to avoid shadowing
- Added minimal type ignores only where necessary (third-party limitations)
- Changed .get(key) to [key] for better Python idioms
- Fixed scipy_.py import fallback type issue

All tests pass with zero mypy errors.

Co-authored-by: Claude <[email protected]>

* Fix mypy unused-ignore order: assignment before misc

The CI requires 'assignment, misc' order for the type ignore comment.

Co-authored-by: Claude <[email protected]>

* Revert to original type ignore order [misc, assignment]

The CI environment may have different mypy configuration. Using the original order.

Co-authored-by: Claude <[email protected]>

* Remove type ignore from scipy_.py ImportError fallback

The CI has scipy-stubs installed which provides proper typing for scipy.io.netcdf_file,
so the type ignore is unnecessary and causes an 'unused-ignore' error.

Co-authored-by: Claude <[email protected]>

---------

Co-authored-by: Claude <[email protected]>

Author: max-sixty

pyproject.toml

@@ -190,11 +190,6 @@ check_untyped_defs = false
 module = [
   "xarray.tests.test_coding_times",
   "xarray.tests.test_dask",
-  "xarray.tests.test_duck_array_ops",
-  "xarray.tests.test_indexing",
-  "xarray.tests.test_sparse",
-  "xarray.tests.test_units",
-  "xarray.tests.test_variable",
 ]
 
 # Use strict = true whenever namedarray has become standalone. In the meantime

xarray/tests/test_duck_array_ops.py

@@ -4,6 +4,7 @@
 import datetime as dt
 import pickle
 import warnings
+from typing import Any
 
 import numpy as np
 import pandas as pd
@@ -64,13 +65,13 @@ def categorical2():
 
     @pytest.fixture
     def arrow1():
-        return pd.arrays.ArrowExtensionArray(
+        return pd.arrays.ArrowExtensionArray(  # type: ignore[attr-defined]
             pa.array([{"x": 1, "y": True}, {"x": 2, "y": False}])
         )
 
     @pytest.fixture
     def arrow2():
-        return pd.arrays.ArrowExtensionArray(
+        return pd.arrays.ArrowExtensionArray(  # type: ignore[attr-defined]
             pa.array([{"x": 3, "y": False}, {"x": 4, "y": True}])
         )
 
@@ -940,8 +941,8 @@ def test_datetime_to_numeric_cftime(dask):
         result = duck_array_ops.datetime_to_numeric(
             times, datetime_unit="h", dtype=dtype
         )
-    expected = 24 * np.arange(0, 35, 7).astype(dtype)
-    np.testing.assert_array_equal(result, expected)
+    expected2: Any = 24 * np.arange(0, 35, 7).astype(dtype)
+    np.testing.assert_array_equal(result, expected2)
 
     with raise_if_dask_computes():
         if dask:
@@ -951,8 +952,8 @@ def test_datetime_to_numeric_cftime(dask):
         result = duck_array_ops.datetime_to_numeric(
             time, offset=times[0], datetime_unit="h", dtype=int
         )
-    expected = np.array(24 * 7).astype(int)
-    np.testing.assert_array_equal(result, expected)
+    expected3 = np.array(24 * 7).astype(int)
+    np.testing.assert_array_equal(result, expected3)
 
 
 @requires_cftime

xarray/tests/test_indexing.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 import itertools
-from typing import Any
+from typing import Any, Union
 
 import numpy as np
 import pandas as pd
@@ -642,7 +642,7 @@ def setup(self):
 
     def test_arrayize_vectorized_indexer(self) -> None:
         for i, j, k in itertools.product(self.indexers, repeat=3):
-            vindex = indexing.VectorizedIndexer((i, j, k))
+            vindex = indexing.VectorizedIndexer((i, j, k))  # type: ignore[arg-type]
             vindex_array = indexing._arrayize_vectorized_indexer(
                 vindex, self.data.shape
             )
@@ -676,46 +676,58 @@ def test_arrayize_vectorized_indexer(self) -> None:
         np.testing.assert_array_equal(b, np.arange(5)[:, np.newaxis])
 
 
-def get_indexers(shape, mode):
+def get_indexers(
+    shape: tuple[int, ...], mode: str
+) -> Union[indexing.VectorizedIndexer, indexing.OuterIndexer, indexing.BasicIndexer]:
     if mode == "vectorized":
         indexed_shape = (3, 4)
-        indexer = tuple(np.random.randint(0, s, size=indexed_shape) for s in shape)
-        return indexing.VectorizedIndexer(indexer)
+        indexer_v = tuple(np.random.randint(0, s, size=indexed_shape) for s in shape)
+        return indexing.VectorizedIndexer(indexer_v)
 
     elif mode == "outer":
-        indexer = tuple(np.random.randint(0, s, s + 2) for s in shape)
-        return indexing.OuterIndexer(indexer)
+        indexer_o = tuple(np.random.randint(0, s, s + 2) for s in shape)
+        return indexing.OuterIndexer(indexer_o)
 
     elif mode == "outer_scalar":
-        indexer = (np.random.randint(0, 3, 4), 0, slice(None, None, 2))
-        return indexing.OuterIndexer(indexer[: len(shape)])
+        indexer_os: tuple[Any, ...] = (
+            np.random.randint(0, 3, 4),
+            0,
+            slice(None, None, 2),
+        )
+        return indexing.OuterIndexer(indexer_os[: len(shape)])
 
     elif mode == "outer_scalar2":
-        indexer = (np.random.randint(0, 3, 4), -2, slice(None, None, 2))
-        return indexing.OuterIndexer(indexer[: len(shape)])
+        indexer_os2: tuple[Any, ...] = (
+            np.random.randint(0, 3, 4),
+            -2,
+            slice(None, None, 2),
+        )
+        return indexing.OuterIndexer(indexer_os2[: len(shape)])
 
     elif mode == "outer1vec":
-        indexer = [slice(2, -3) for s in shape]
-        indexer[1] = np.random.randint(0, shape[1], shape[1] + 2)
-        return indexing.OuterIndexer(tuple(indexer))
+        indexer_o1v: list[Any] = [slice(2, -3) for s in shape]
+        indexer_o1v[1] = np.random.randint(0, shape[1], shape[1] + 2)
+        return indexing.OuterIndexer(tuple(indexer_o1v))
 
     elif mode == "basic":  # basic indexer
-        indexer = [slice(2, -3) for s in shape]
-        indexer[0] = 3
-        return indexing.BasicIndexer(tuple(indexer))
+        indexer_b: list[Any] = [slice(2, -3) for s in shape]
+        indexer_b[0] = 3
+        return indexing.BasicIndexer(tuple(indexer_b))
 
     elif mode == "basic1":  # basic indexer
         return indexing.BasicIndexer((3,))
 
     elif mode == "basic2":  # basic indexer
-        indexer = [0, 2, 4]
-        return indexing.BasicIndexer(tuple(indexer[: len(shape)]))
+        indexer_b2 = [0, 2, 4]
+        return indexing.BasicIndexer(tuple(indexer_b2[: len(shape)]))
 
     elif mode == "basic3":  # basic indexer
-        indexer = [slice(None) for s in shape]
-        indexer[0] = slice(-2, 2, -2)
-        indexer[1] = slice(1, -1, 2)
-        return indexing.BasicIndexer(tuple(indexer[: len(shape)]))
+        indexer_b3: list[Any] = [slice(None) for s in shape]
+        indexer_b3[0] = slice(-2, 2, -2)
+        indexer_b3[1] = slice(1, -1, 2)
+        return indexing.BasicIndexer(tuple(indexer_b3[: len(shape)]))
+
+    raise ValueError(f"Unknown mode: {mode}")
 
 
 @pytest.mark.parametrize("size", [100, 99])

xarray/tests/test_sparse.py

@@ -661,16 +661,19 @@ def test_concat(self):
             sparse.concatenate([self.sp_ar, self.sp_ar, self.sp_ar], axis=0),
         )
 
-        out = xr.concat([self.sp_xr, self.sp_xr, self.sp_xr], dim="y")
+        out_concat = xr.concat([self.sp_xr, self.sp_xr, self.sp_xr], dim="y")
         assert_sparse_equal(
-            out.data, sparse.concatenate([self.sp_ar, self.sp_ar, self.sp_ar], axis=1)
+            out_concat.data,
+            sparse.concatenate([self.sp_ar, self.sp_ar, self.sp_ar], axis=1),
         )
 
     def test_stack(self):
         arr = make_xrarray({"w": 2, "x": 3, "y": 4})
         stacked = arr.stack(z=("x", "y"))
 
-        z = pd.MultiIndex.from_product([np.arange(3), np.arange(4)], names=["x", "y"])
+        z = pd.MultiIndex.from_product(
+            [list(range(3)), list(range(4))], names=["x", "y"]
+        )
 
         expected = xr.DataArray(
             arr.data.reshape((2, -1)), {"w": [0, 1], "z": z}, dims=["w", "z"]
@@ -753,8 +756,8 @@ def test_dataset_pickle(self):
     def test_coarsen(self):
         a1 = self.ds_xr
         a2 = self.sp_xr
-        m1 = a1.coarsen(x=2, boundary="trim").mean()
-        m2 = a2.coarsen(x=2, boundary="trim").mean()
+        m1 = a1.coarsen(x=2, boundary="trim").mean()  # type: ignore[attr-defined]
+        m2 = a2.coarsen(x=2, boundary="trim").mean()  # type: ignore[attr-defined]
 
         assert isinstance(m2.data, sparse.SparseArray)
         assert np.allclose(m1.data, m2.data.todense())
@@ -781,7 +784,7 @@ def test_rolling_exp(self):
 
     @pytest.mark.xfail(reason="No implementation of np.einsum")
     def test_dot(self):
-        a1 = self.xp_xr.dot(self.xp_xr[0])
+        a1 = self.sp_xr.dot(self.sp_xr[0])
         a2 = self.sp_ar.dot(self.sp_ar[0])
         assert_equal(a1, a2)
 
@@ -835,8 +838,8 @@ def test_reindex(self):
             {"x": [1, 100, 2, 101, 3]},
             {"x": [2.5, 3, 3.5], "y": [2, 2.5, 3]},
         ]:
-            m1 = x1.reindex(**kwargs)
-            m2 = x2.reindex(**kwargs)
+            m1 = x1.reindex(**kwargs)  # type: ignore[arg-type]
+            m2 = x2.reindex(**kwargs)  # type: ignore[arg-type]
             assert np.allclose(m1, m2, equal_nan=True)
 
     @pytest.mark.xfail
@@ -852,12 +855,12 @@ def test_where(self):
         xr.DataArray(a).where(cond)
 
         s = sparse.COO.from_numpy(a)
-        cond = s > 3
-        xr.DataArray(s).where(cond)
+        cond2 = s > 3
+        xr.DataArray(s).where(cond2)
 
         x = xr.DataArray(s)
-        cond = x > 3
-        x.where(cond)
+        cond3: DataArray = x > 3
+        x.where(cond3)
 
 
 class TestSparseCoords: