Add error message when output has no chunks with parallel weights gen…

…eration (#304)

* Add error msg when no chunks in par=T

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix for locstreamout

* add simple test

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: David Huard <[email protected]>

CHANGES.rst

@@ -6,6 +6,7 @@ What's new
 
 Bug fixes
 ~~~~~~~~~
+* Raise a meaningful error messages when the output grid has no chunks with `parallel=True` (:issue:`299`, :pull:`304`). By `Pascal Bourgault <https://github.com/aulemahal>`_.
 * Correct guess of output chunks for the :``SpatialAverager``.
 
 0.8.1 (2023-09-05)

xesmf/frontend.py

@@ -963,6 +963,11 @@ def _init_para_regrid(self, ds_in, ds_out, kwargs):
             ds_out = ds_out.set_coords(['lon_b', 'lat_b'])
         if 'lon_b' in ds_in.data_vars:
             ds_in = ds_in.set_coords(['lon_b', 'lat_b'])
+        if not (set(self.out_horiz_dims) - {'dummy'}).issubset(ds_out.chunksizes.keys()):
+            raise ValueError(
+                'Using `parallel=True` requires the output grid to have chunks along all spatial dimensions. '
+                'If the dataset has no variables, consider adding an all-True spatial mask with appropriate chunks.'
+            )
         # Drop everything in ds_out except mask or create mask if None. This is to prevent map_blocks loading unnecessary large data
         if self.sequence_out:
             ds_out_dims_drop = set(ds_out.variables).difference(ds_out.data_vars)

xesmf/tests/test_frontend.py

@@ -649,6 +649,11 @@ def test_para_weight_gen():
     assert all(regridder_locs.w.data.data == para_regridder_locs.w.data.data)
 
 
+def test_para_weight_gen_errors():
+    with pytest.raises(ValueError, match='requires the output grid to have chunks'):
+        xe.Regridder(ds_in, ds_out, 'conservative', parallel=True)
+
+
 def test_regrid_dataset():
     # xarray.Dataset containing in-memory numpy array
     regridder = xe.Regridder(ds_in, ds_out, 'conservative')