BUG: Make sure series-series boolean comparions are label based (GH4947)

Author: jreback

doc/source/release.rst

@@ -374,6 +374,8 @@ Bug Fixes
     - appending a 0-len table will work correctly (:issue:`4273`)
     - ``to_hdf`` was raising when passing both arguments ``append`` and ``table`` (:issue:`4584`)
     - reading from a store with duplicate columns across dtypes would raise (:issue:`4767`)
+    - Fixed a bug where ``ValueError`` wasn't correctly raised when column names
+      weren't strings (:issue:`4956`)
   - Fixed bug in tslib.tz_convert(vals, tz1, tz2): it could raise IndexError exception while
     trying to access trans[pos + 1] (:issue:`4496`)
   - The ``by`` argument now works correctly with the ``layout`` argument
@@ -500,8 +502,6 @@ Bug Fixes
   - Fixed a bug with setting invalid or out-of-range values in indexing
     enlargement scenarios (:issue:`4940`)
   - Tests for fillna on empty Series (:issue:`4346`), thanks @immerrr
-  - Fixed a bug where ``ValueError`` wasn't correctly raised when column names
-    weren't strings (:issue:`4956`)
   - Fixed ``copy()`` to shallow copy axes/indices as well and thereby keep
     separate metadata. (:issue:`4202`, :issue:`4830`)
   - Fixed skiprows option in Python parser for read_csv (:issue:`4382`)
@@ -521,6 +521,7 @@ Bug Fixes
   - Fix a bug where reshaping a ``Series`` to its own shape raised ``TypeError`` (:issue:`4554`)
     and other reshaping issues.
   - Bug in setting with ``ix/loc`` and a mixed int/string index (:issue:`4544`)
+  - Make sure series-series boolean comparions are label based (:issue:`4947`)
 
 pandas 0.12.0
 -------------

pandas/tests/test_frame.py

@@ -4523,8 +4523,10 @@ def f():
     def test_logical_with_nas(self):
         d = DataFrame({'a': [np.nan, False], 'b': [True, True]})
 
+        # GH4947
+        # bool comparisons should return bool
         result = d['a'] | d['b']
-        expected = Series([np.nan, True])
+        expected = Series([True, True])
         assert_series_equal(result, expected)
 
         # GH4604, automatic casting here
@@ -4533,10 +4535,6 @@ def test_logical_with_nas(self):
         assert_series_equal(result, expected)
 
         result = d['a'].fillna(False,downcast=False) | d['b']
-        expected = Series([True, True],dtype=object)
-        assert_series_equal(result, expected)
-
-        result = (d['a'].fillna(False,downcast=False) | d['b']).convert_objects()
         expected = Series([True, True])
         assert_series_equal(result, expected)
 

pandas/tests/test_series.py

@@ -2757,6 +2757,64 @@ def test_comparison_different_length(self):
         b = Series([2, 3, 4])
         self.assertRaises(ValueError, a.__eq__, b)
 
+    def test_comparison_label_based(self):
+
+        # GH 4947
+        # comparisons should be label based
+
+        a = Series([True, False, True], list('bca'))
+        b = Series([False, True, False], list('abc'))
+
+        expected = Series([True, False, False], list('bca'))
+        result = a & b
+        assert_series_equal(result,expected)
+
+        expected = Series([True, False, True], list('bca'))
+        result = a | b
+        assert_series_equal(result,expected)
+
+        expected = Series([False, False, True], list('bca'))
+        result = a ^ b
+        assert_series_equal(result,expected)
+
+        # rhs is bigger
+        a = Series([True, False, True], list('bca'))
+        b = Series([False, True, False, True], list('abcd'))
+
+        expected = Series([True, False, False], list('bca'))
+        result = a & b
+        assert_series_equal(result,expected)
+
+        expected = Series([True, False, True], list('bca'))
+        result = a | b
+        assert_series_equal(result,expected)
+
+        # filling
+
+        # vs empty
+        result = a & Series([])
+        expected = Series([False, False, False], list('bca'))
+        assert_series_equal(result,expected)
+
+        result = a | Series([])
+        expected = Series([True, True, True], list('bca'))
+        assert_series_equal(result,expected)
+
+        # vs non-matching
+        result = a & Series([1],['z'])
+        expected = Series([False, False, False], list('bca'))
+        assert_series_equal(result,expected)
+
+        result = a | Series([1],['z'])
+        expected = Series([True, True, True], list('bca'))
+        assert_series_equal(result,expected)
+
+        # identity
+        # we would like s[s|e] == s to hold for any e, whether empty or not
+        for e in [Series([]),Series([1],['z']),Series(['z']),Series(np.nan,b.index),Series(np.nan,a.index)]:
+            result = a[a | e]
+            assert_series_equal(result,a)
+
     def test_between(self):
         s = Series(bdate_range('1/1/2000', periods=20).asobject)
         s[::2] = np.nan