Merge pull request #5019 from cpcloud/raise-on-index-hash

ER: give a better error message for hashing indices

Author: cpcloud

pandas/core/index.py

@@ -600,7 +600,7 @@ def __contains__(self, key):
             return False
 
     def __hash__(self):
-        return hash(self.view(np.ndarray))
+        raise TypeError("unhashable type: %r" % type(self).__name__)
 
     def __getitem__(self, key):
         """Override numpy.ndarray's __getitem__ method to work as desired"""
@@ -1852,6 +1852,7 @@ def equals(self, other):
             # e.g. fails in numpy 1.6 with DatetimeIndex #1681
             return False
 
+
 class MultiIndex(Index):
 
     """

pandas/tests/test_index.py

@@ -12,7 +12,8 @@
 import numpy as np
 from numpy.testing import assert_array_equal
 
-from pandas.core.index import Index, Float64Index, Int64Index, MultiIndex, InvalidIndexError
+from pandas.core.index import (Index, Float64Index, Int64Index, MultiIndex,
+                               InvalidIndexError)
 from pandas.core.frame import DataFrame
 from pandas.core.series import Series
 from pandas.util.testing import (assert_almost_equal, assertRaisesRegexp,
@@ -75,7 +76,10 @@ def test_set_name_methods(self):
         self.assertEqual(ind.names, [name])
 
     def test_hash_error(self):
-        self.assertRaises(TypeError, hash, self.strIndex)
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(self.strIndex).__name__):
+            hash(self.strIndex)
 
     def test_new_axis(self):
         new_index = self.dateIndex[None, :]
@@ -661,6 +665,12 @@ def setUp(self):
         self.mixed = Float64Index([1.5, 2, 3, 4, 5])
         self.float = Float64Index(np.arange(5) * 2.5)
 
+    def test_hash_error(self):
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(self.float).__name__):
+            hash(self.float)
+
     def check_is_index(self, i):
         self.assert_(isinstance(i, Index) and not isinstance(i, Float64Index))
 
@@ -736,6 +746,7 @@ def test_astype(self):
         self.assert_(i.equals(result))
         self.check_is_index(result)
 
+
 class TestInt64Index(unittest.TestCase):
     _multiprocess_can_split_ = True
 
@@ -779,6 +790,12 @@ def test_constructor_corner(self):
         arr = np.array([1, '2', 3, '4'], dtype=object)
         self.assertRaises(TypeError, Int64Index, arr)
 
+    def test_hash_error(self):
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(self.index).__name__):
+            hash(self.index)
+
     def test_copy(self):
         i = Int64Index([], name='Foo')
         i_copy = i.copy()
@@ -1155,6 +1172,12 @@ def setUp(self):
                                 labels=[major_labels, minor_labels],
                                 names=self.index_names)
 
+    def test_hash_error(self):
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(self.index).__name__):
+            hash(self.index)
+
     def test_set_names_and_rename(self):
         # so long as these are synonyms, we don't need to test set_names
         self.assert_(self.index.rename == self.index.set_names)
@@ -2231,6 +2254,7 @@ def test_get_combined_index():
     result = _get_combined_index([])
     assert(result.equals(Index([])))
 
+
 if __name__ == '__main__':
     nose.runmodule(argv=[__file__, '-vvs', '-x', '--pdb', '--pdb-failure'],
                    exit=False)

pandas/tseries/tests/test_period.py

@@ -1057,6 +1057,13 @@ class TestPeriodIndex(TestCase):
     def setUp(self):
         pass
 
+    def test_hash_error(self):
+        index = period_range('20010101', periods=10)
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(index).__name__):
+            hash(index)
+
     def test_make_time_series(self):
         index = PeriodIndex(freq='A', start='1/1/2001', end='12/1/2009')
         series = Series(1, index=index)

pandas/tseries/tests/test_timeseries.py

@@ -1718,6 +1718,13 @@ def _simple_ts(start, end, freq='D'):
 class TestDatetimeIndex(unittest.TestCase):
     _multiprocess_can_split_ = True
 
+    def test_hash_error(self):
+        index = date_range('20010101', periods=10)
+        with tm.assertRaisesRegexp(TypeError,
+                                   "unhashable type: %r" %
+                                   type(index).__name__):
+            hash(index)
+
     def test_stringified_slice_with_tz(self):
         #GH2658
         import datetime