Allow in-memory arrays with open_mfdataset

doc/whats-new.rst

@@ -35,6 +35,8 @@ Deprecations
 Bug fixes
 ~~~~~~~~~
 
+- Allow in-memory arrays with :py:func:`xarray.open_mfdataset` by passing ``chunks=None``. (:pull:`5704`).
+  By `Jimmy Westling <https://github.com/illviljan>`_.
 
 Documentation
 ~~~~~~~~~~~~~

xarray/backends/api.py

@@ -817,7 +817,7 @@ def open_datatree(
 
 def open_mfdataset(
     paths: str | NestedSequence[str | os.PathLike],
-    chunks: T_Chunks | None = None,
+    chunks: T_Chunks | None = {},
     concat_dim: (
         str
         | DataArray
@@ -858,7 +858,7 @@ def open_mfdataset(
         concatenation along more than one dimension is desired, then ``paths`` must be a
         nested list-of-lists (see ``combine_nested`` for details). (A string glob will
         be expanded to a 1-dimensional list.)
-    chunks : int, dict, 'auto' or None, optional
+    chunks : int, dict, 'auto' or None, default: {}
         Dictionary with keys given by dimension names and values given by chunk sizes.
         In general, these should divide the dimensions of each dataset. If int, chunk
         each dimension by ``chunks``. By default, chunks will be chosen to load entire
@@ -1037,7 +1037,7 @@ def open_mfdataset(
             "instead specify combine='nested' along with a value for `concat_dim`.",
         )
 
-    open_kwargs = dict(engine=engine, chunks=chunks or {}, **kwargs)
+    open_kwargs = dict(engine=engine, chunks=chunks, **kwargs)
 
     if parallel:
         import dask

xarray/tests/test_backends.py

@@ -3947,7 +3947,7 @@
     return request.param
 
 
-@pytest.fixture(params=[None, 5])
+@pytest.fixture(params=[None, {}, 5])
 def chunks(request):
     return request.param
 
@@ -3999,16 +3999,21 @@
                 subds.to_zarr(store=tmpfiles[ii])
 
         # check that calculation on opened datasets works properly
+        chunks = chunks if (not chunks and readengine != "zarr") else "auto"
         with open_mfdataset(
             tmpfiles,
             combine="nested",
             concat_dim="x",
             engine=readengine,
             parallel=parallel,
-            chunks=chunks if (not chunks and readengine != "zarr") else "auto",
+            chunks=chunks,
         ) as actual:
             # check that using open_mfdataset returns dask arrays for variables
-            assert isinstance(actual["foo"].data, dask_array_type)
+            # when a chunks parameter has been defined:
+            if chunks is None:
+                assert isinstance(actual["foo"].data, np.ndarray)
+            else:
+                assert isinstance(actual["foo"].data, dask_array_type)
 
             assert_identical(original, actual)
 
@@ -4688,7 +4693,7 @@
             with open_dataset(tmp, chunks=chunks) as dask_ds:
                 assert_identical(data, dask_ds)
                 with create_tmp_file() as tmp2:
                     dask_ds.to_netcdf(tmp2)
                     with open_dataset(tmp2) as on_disk:
                         assert_identical(data, on_disk)