Fixed dask.optimize on datasets

Another attempt to fix pydata#3698. The issue with my fix in is that we hit
`Variable._dask_finalize` in both `dask.optimize` and `dask.persist`. We
want to do the culling of unnecessary tasks (`test_persist_Dataset`) but
only in the persist case, not optimize (`test_optimize`).

xarray/core/dataset.py

@@ -777,10 +777,19 @@ def _dask_postcompute(results, info, *args):
     @staticmethod
     def _dask_postpersist(dsk, info, *args):
         variables = {}
+        # postpersist is called in both dask.optimize and dask.persist
+        # When persisting, we want to filter out unrelated keys for
+        # each Variable's task graph.
+        is_persist = len(dsk) == len(info)
         for is_dask, k, v in info:
             if is_dask:
                 func, args2 = v
-                result = func(dsk, *args2)
+                if is_persist:
+                    name = args2[1][0]
+                    dsk2 = {k: v for k, v in dsk.items() if k[0] == name}
+                else:
+                    dsk2 = dsk
+                result = func(dsk2, *args2)
             else:
                 result = v
             variables[k] = result

xarray/core/variable.py

@@ -501,9 +501,6 @@ def __dask_postpersist__(self):
 
     @staticmethod
     def _dask_finalize(results, array_func, array_args, dims, attrs, encoding):
-        if isinstance(results, dict):  # persist case
-            name = array_args[0]
-            results = {k: v for k, v in results.items() if k[0] == name}
         data = array_func(results, *array_args)
         return Variable(dims, data, attrs=attrs, encoding=encoding)
 

xarray/tests/test_dask.py

@@ -1607,3 +1607,11 @@ def test_more_transforms_pass_lazy_array_equiv(map_da, map_ds):
         assert_equal(map_da._from_temp_dataset(map_da._to_temp_dataset()), map_da)
         assert_equal(map_da.astype(map_da.dtype), map_da)
         assert_equal(map_da.transpose("y", "x", transpose_coords=False).cxy, map_da.cxy)
+
+
+def test_optimize():
+    # https://github.com/pydata/xarray/issues/3698
+    a = dask.array.ones((10, 4), chunks=(5, 2))
+    arr = xr.DataArray(a).chunk(5)
+    (arr2,) = dask.optimize(arr)
+    arr2.compute()