implement __contains__ for Categorical

Author: fjetter

asv_bench/benchmarks/categoricals.py

@@ -169,3 +169,16 @@ def setup(self, dtype):
 
     def time_isin_categorical(self, dtype):
         self.series.isin(self.sample)
+
+
+class Slicing(object):
+
+    def setup(self):
+        n = 1 * 10**4
+        self.df = pd.DataFrame(
+            dict(A=range(n * 3)),
+            index=pd.CategoricalIndex(list('a' * n + 'b' * n + 'c' * n))
+        )
+
+    def time_loc_categorical(self):
+        self.df.loc['b']

doc/source/whatsnew/v0.23.0.txt

@@ -1080,6 +1080,7 @@ Performance Improvements
 - Improved performance of :func:`Series.isin` in the case of categorical dtypes (:issue:`20003`)
 - Improved performance of ``getattr(Series, attr)`` when the Series has certain index types. This manifiested in slow printing of large Series with a ``DatetimeIndex`` (:issue:`19764`)
 - Fixed a performance regression for :func:`GroupBy.nth` and :func:`GroupBy.last` with some object columns (:issue:`19283`)
+- Improved performance of :meth:`Categorical.loc` (:issue:`20395`)
 
 .. _whatsnew_0230.docs:
 

pandas/core/arrays/categorical.py

@@ -1846,6 +1846,11 @@ def __iter__(self):
         """Returns an Iterator over the values of this Categorical."""
         return iter(self.get_values().tolist())
 
+    def __contains__(self, key):
+        """Returns True if `key` is in this Categorical."""
+        code_values = self.categories.get_indexer([key])
+        return code_values[0] in self.codes
+
     def _tidy_repr(self, max_vals=10, footer=True):
         """ a short repr displaying only max_vals and an optional (but default
         footer)