Merge pull request #8042 from artemyk/fix-assert_frame_equal-multiindex-dtypes

Fix testing multiindex dtypes in assert_frame_equal and assert_series_equal

Author: jreback

pandas/tests/test_testing.py

@@ -6,9 +6,10 @@
 import nose
 import numpy as np
 import sys
-from pandas import Series
+from pandas import Series, DataFrame
 from pandas.util.testing import (
-    assert_almost_equal, assertRaisesRegexp, raise_with_traceback, assert_series_equal,
+    assert_almost_equal, assertRaisesRegexp, raise_with_traceback, 
+    assert_series_equal, assert_frame_equal,
     RNGContext
 )
 
@@ -174,6 +175,46 @@ def test_less_precise(self):
         self.assertRaises(AssertionError, assert_series_equal, s1, s2)
         self.assertRaises(AssertionError, assert_series_equal, s1, s2, True)
 
+    def test_index_dtype(self):
+        df1 = DataFrame.from_records(
+                {'a':[1,2],'c':['l1','l2']}, index=['a'])
+        df2 = DataFrame.from_records(
+                {'a':[1.0,2.0],'c':['l1','l2']}, index=['a'])
+        self._assert_not_equal(df1.c, df2.c, check_index_type=True)
+
+    def test_multiindex_dtype(self):
+        df1 = DataFrame.from_records(
+                {'a':[1,2],'b':[2.1,1.5],'c':['l1','l2']}, index=['a','b'])
+        df2 = DataFrame.from_records(
+                {'a':[1.0,2.0],'b':[2.1,1.5],'c':['l1','l2']}, index=['a','b'])
+        self._assert_not_equal(df1.c, df2.c, check_index_type=True)
+
+
+class TestAssertFrameEqual(unittest.TestCase):
+    _multiprocess_can_split_ = True
+
+    def _assert_equal(self, x, y, **kwargs):
+        assert_frame_equal(x,y,**kwargs)
+        assert_frame_equal(y,x,**kwargs)
+
+    def _assert_not_equal(self, a, b, **kwargs):
+        self.assertRaises(AssertionError, assert_frame_equal, a, b, **kwargs)
+        self.assertRaises(AssertionError, assert_frame_equal, b, a, **kwargs)
+
+    def test_index_dtype(self):
+        df1 = DataFrame.from_records(
+                {'a':[1,2],'c':['l1','l2']}, index=['a'])
+        df2 = DataFrame.from_records(
+                {'a':[1.0,2.0],'c':['l1','l2']}, index=['a'])
+        self._assert_not_equal(df1, df2, check_index_type=True)
+
+    def test_multiindex_dtype(self):
+        df1 = DataFrame.from_records(
+                {'a':[1,2],'b':[2.1,1.5],'c':['l1','l2']}, index=['a','b'])
+        df2 = DataFrame.from_records(
+                {'a':[1.0,2.0],'b':[2.1,1.5],'c':['l1','l2']}, index=['a','b'])
+        self._assert_not_equal(df1, df2, check_index_type=True)
+
 class TestRNGContext(unittest.TestCase):
 
     def test_RNGContext(self):

pandas/util/testing.py

@@ -608,9 +608,12 @@ def assert_series_equal(left, right, check_dtype=True,
     else:
         assert_index_equal(left.index, right.index)
     if check_index_type:
-        assert_isinstance(left.index, type(right.index))
-        assert_attr_equal('dtype', left.index, right.index)
-        assert_attr_equal('inferred_type', left.index, right.index)
+        for level in range(left.index.nlevels):
+            lindex = left.index.get_level_values(level)
+            rindex = right.index.get_level_values(level)
+            assert_isinstance(lindex, type(rindex))
+            assert_attr_equal('dtype', lindex, rindex)
+            assert_attr_equal('inferred_type', lindex, rindex)
 
 # This could be refactored to use the NDFrame.equals method
 def assert_frame_equal(left, right, check_dtype=True,
@@ -657,9 +660,12 @@ def assert_frame_equal(left, right, check_dtype=True,
                                 check_exact=check_exact)
 
     if check_index_type:
-        assert_isinstance(left.index, type(right.index))
-        assert_attr_equal('dtype', left.index, right.index)
-        assert_attr_equal('inferred_type', left.index, right.index)
+        for level in range(left.index.nlevels):
+            lindex = left.index.get_level_values(level)
+            rindex = right.index.get_level_values(level)
+            assert_isinstance(lindex, type(rindex))
+            assert_attr_equal('dtype', lindex, rindex)
+            assert_attr_equal('inferred_type', lindex, rindex)
     if check_column_type:
         assert_isinstance(left.columns, type(right.columns))
         assert_attr_equal('dtype', left.columns, right.columns)