raise TypeError in Variable.quantile if data is a dask array

doc/whats-new.rst

@@ -74,6 +74,10 @@ Bug fixes
   ``rtol`` arguments when called on ``DataArray`` objects.
   By `Stephan Hoyer <https://github.com/shoyer>`_.
 
+- Xarray ``quantile`` methods now properly raise a ``TypeError`` when applied to
+  objects with data stored as ``dask`` arrays (:issue:`1529`).
+  By `Joe Hamman <https://github.com/jhamman>`_.
+
 .. _whats-new.0.9.6:
 
 v0.9.6 (8 June 2017)

xarray/core/variable.py

@@ -1114,9 +1114,9 @@ def quantile(self, q, dim=None, interpolation='linear'):
         """
 
         if isinstance(self.data, dask_array_type):
-            TypeError("quantile does not work for arrays stored as dask "
-                      "arrays. Load the data via .compute() or .load() prior "
-                      "to calling this method.")
+            raise TypeError("quantile does not work for arrays stored as dask "
+                            "arrays. Load the data via .compute() or .load() "
+                            "prior to calling this method.")
         if LooseVersion(np.__version__) < LooseVersion('1.10.0'):
             raise NotImplementedError(
                 'quantile requres numpy version 1.10.0 or later')

xarray/tests/test_variable.py

@@ -1018,6 +1018,14 @@ def test_quantile(self):
                                             axis=axis)
                 np.testing.assert_allclose(actual.values, expected)
 
+    @requires_dask
+    def test_quantile_dask_raises(self):
+        # regression for GH1524
+        v = Variable(['x', 'y'], self.d).chunk(2)
+
+        with self.assertRaisesRegexp(TypeError, 'arrays stored as dask'):
+            v.quantile(0.5, dim='x')
+
     def test_big_endian_reduce(self):
         # regression test for GH489
         data = np.ones(5, dtype='>f4')
@@ -1243,7 +1251,7 @@ def test_full_like(self):
 
     @requires_dask
     def test_full_like_dask(self):
-        orig = Variable(dims=('x', 'y'), data=[[1.5 ,2.0], [3.1, 4.3]],
+        orig = Variable(dims=('x', 'y'), data=[[1.5, 2.0], [3.1, 4.3]],
                         attrs={'foo': 'bar'}).chunk(((1, 1), (2,)))
 
         def check(actual, expect_dtype, expect_values):