Merge pull request #7015 from cpcloud/bool-raises-on-exprs-err-msg

ERR: more informative error message on bool arith op failures

Author: cpcloud

doc/source/release.rst

@@ -206,6 +206,8 @@ API Changes
 - Added ``factorize`` functions to ``Index`` and ``Series`` to get indexer and unique values (:issue:`7090`)
 - :meth:`DataFrame.describe` on a DataFrame with a mix of Timestamp and string like objects
   returns a different Index (:issue:`7088`). Previously the index was unintentionally sorted.
+- arithmetic operations with **only** ``bool`` dtypes now raise an error
+  (:issue:`7011`, :issue:`6762`, :issue:`7015`)
 
 Deprecations
 ~~~~~~~~~~~~

doc/source/v0.14.0.txt

@@ -228,6 +228,18 @@ Display Changes
   length of the series (:issue:`7101`)
 - Fixed a bug in the HTML repr of a truncated Series or DataFrame not showing the class name with the
   `large_repr` set to 'info' (:issue:`7105`)
+- arithmetic operations with **only** ``bool`` dtypes now raise an error
+  (:issue:`7011`, :issue:`6762`, :issue:`7015`)
+
+  .. code-block:: python
+
+     x = pd.Series(np.random.rand(10) > 0.5)
+     y = True
+     x * y
+
+     # this now raises for arith ops like ``+``, ``*``, etc.
+     NotImplementedError: operator '*' not implemented for bool dtypes
+
 
 .. _whatsnew_0140.groupby:
 

pandas/computation/expressions.py

@@ -158,7 +158,10 @@ def _has_bool_dtype(x):
     try:
         return x.dtype == bool
     except AttributeError:
-        return 'bool' in x.blocks
+        try:
+            return 'bool' in x.blocks
+        except AttributeError:
+            return isinstance(x, (bool, np.bool_))
 
 
 def _bool_arith_check(op_str, a, b, not_allowed=frozenset(('+', '*', '-', '/',

pandas/core/ops.py

@@ -61,25 +61,25 @@ def names(x):
                               default_axis=default_axis, fill_zeros=np.inf),
         # Causes a floating point exception in the tests when numexpr
         # enabled, so for now no speedup
-        mod=arith_method(operator.mod, names('mod'), default_axis=default_axis,
-                         fill_zeros=np.nan),
+        mod=arith_method(operator.mod, names('mod'), None,
+                         default_axis=default_axis, fill_zeros=np.nan),
         pow=arith_method(operator.pow, names('pow'), op('**'),
                          default_axis=default_axis),
         # not entirely sure why this is necessary, but previously was included
         # so it's here to maintain compatibility
-        rmul=arith_method(operator.mul, names('rmul'),
+        rmul=arith_method(operator.mul, names('rmul'), op('*'),
                           default_axis=default_axis),
-        rsub=arith_method(lambda x, y: y - x, names('rsub'),
+        rsub=arith_method(lambda x, y: y - x, names('rsub'), op('-'),
                           default_axis=default_axis),
         rtruediv=arith_method(lambda x, y: operator.truediv(y, x),
-                              names('rtruediv'), truediv=True,
+                              names('rtruediv'), op('/'), truediv=True,
                               fill_zeros=np.inf, default_axis=default_axis),
         rfloordiv=arith_method(lambda x, y: operator.floordiv(y, x),
-                               names('rfloordiv'), default_axis=default_axis,
-                               fill_zeros=np.inf),
-        rpow=arith_method(lambda x, y: y ** x, names('rpow'),
+                               names('rfloordiv'), op('//'),
+                               default_axis=default_axis, fill_zeros=np.inf),
+        rpow=arith_method(lambda x, y: y ** x, names('rpow'), op('**'),
                           default_axis=default_axis),
-        rmod=arith_method(lambda x, y: y % x, names('rmod'),
+        rmod=arith_method(lambda x, y: y % x, names('rmod'), op('%'),
                           default_axis=default_axis),
     )
     new_methods['div'] = new_methods['truediv']
@@ -100,11 +100,11 @@ def names(x):
             and_=bool_method(operator.and_, names('and_'), op('&')),
             or_=bool_method(operator.or_, names('or_'), op('|')),
             # For some reason ``^`` wasn't used in original.
-            xor=bool_method(operator.xor, names('xor')),
+            xor=bool_method(operator.xor, names('xor'), op('^')),
             rand_=bool_method(lambda x, y: operator.and_(y, x),
-                              names('rand_')),
-            ror_=bool_method(lambda x, y: operator.or_(y, x), names('ror_')),
-            rxor=bool_method(lambda x, y: operator.xor(y, x), names('rxor'))
+                              names('rand_'), op('&')),
+            ror_=bool_method(lambda x, y: operator.or_(y, x), names('ror_'), op('|')),
+            rxor=bool_method(lambda x, y: operator.xor(y, x), names('rxor'), op('^'))
         ))
 
     new_methods = dict((names(k), v) for k, v in new_methods.items())
@@ -431,7 +431,7 @@ def maybe_convert_for_time_op(cls, left, right, name):
         return cls(left, right, name)
 
 
-def _arith_method_SERIES(op, name, str_rep=None, fill_zeros=None,
+def _arith_method_SERIES(op, name, str_rep, fill_zeros=None,
                          default_axis=None, **eval_kwargs):
     """
     Wrapper function for Series arithmetic operations, to avoid
@@ -506,7 +506,7 @@ def wrapper(left, right, name=name):
     return wrapper
 
 
-def _comp_method_SERIES(op, name, str_rep=None, masker=False):
+def _comp_method_SERIES(op, name, str_rep, masker=False):
     """
     Wrapper function for Series arithmetic operations, to avoid
     code duplication.
@@ -578,7 +578,7 @@ def wrapper(self, other):
     return wrapper
 
 
-def _bool_method_SERIES(op, name, str_rep=None):
+def _bool_method_SERIES(op, name, str_rep):
     """
     Wrapper function for Series arithmetic operations, to avoid
     code duplication.
@@ -647,7 +647,7 @@ def _radd_compat(left, right):
     return output
 
 
-def _flex_method_SERIES(op, name, str_rep=None, default_axis=None,
+def _flex_method_SERIES(op, name, str_rep, default_axis=None,
                         fill_zeros=None, **eval_kwargs):
     doc = """
     Binary operator %s with support to substitute a fill_value for missing data

pandas/tests/test_expressions.py

@@ -357,6 +357,18 @@ def test_bool_ops_raise_on_arithmetic(self):
                 with tm.assertRaisesRegexp(NotImplementedError, err_msg):
                     f(df.a, df.b)
 
+                with tm.assertRaisesRegexp(NotImplementedError, err_msg):
+                    f(df.a, True)
+
+                with tm.assertRaisesRegexp(NotImplementedError, err_msg):
+                    f(False, df.a)
+
+                with tm.assertRaisesRegexp(TypeError, err_msg):
+                    f(False, df)
+
+                with tm.assertRaisesRegexp(TypeError, err_msg):
+                    f(df, True)
+
 
 if __name__ == '__main__':
     import nose