dropna added for unique method. Performance improvements for nunique method.

ref pandas-dev#7771

Author: lexual

pandas/core/base.py

@@ -288,21 +288,29 @@ def value_counts(self, normalize=False, sort=True, ascending=False,
                                             tz=getattr(self, 'tz', None))
         return result
 
-    def unique(self):
+    def unique(self, dropna=False):
         """
         Return array of unique values in the object. Significantly faster than
         numpy.unique. Includes NA values.
 
+        Parameters
+        ----------
+        dropna : boolean, default False
+            Don't include NaN in the result.
+
         Returns
         -------
         uniques : ndarray
         """
-        from pandas.core.nanops import unique1d
-        values = self.values
-        if hasattr(values,'unique'):
-            return values.unique()
-
-        return unique1d(values)
+        if dropna:
+            return self.dropna().unique()
+        else:
+            if hasattr(self.values, 'unique'):
+                # Categorical Series not supported by unique1d
+                return self.values.unique()
+            else:
+                from pandas.core.nanops import unique1d
+                return unique1d(self.values)
 
     def nunique(self, dropna=True):
         """
@@ -319,7 +327,7 @@ def nunique(self, dropna=True):
         -------
         nunique : int
         """
-        return len(self.value_counts(dropna=dropna))
+        return len(self.unique(dropna=dropna))
 
     def factorize(self, sort=False, na_sentinel=-1):
         """

pandas/core/categorical.py

@@ -869,7 +869,7 @@ def mode(self):
                              fastpath=True)
         return result
 
-    def unique(self):
+    def unique(self, **kwargs):
         """
         Return the unique values.
 

pandas/tseries/index.py

@@ -852,6 +852,23 @@ def take(self, indices, axis=0):
             return self[maybe_slice]
         return super(DatetimeIndex, self).take(indices, axis)
 
+    def unique(self, dropna=False):
+        """
+        Index.unique with handling for DatetimeIndex metadata
+
+        Parameters
+        ----------
+        dropna : boolean, default False
+            Don't include NaN in the result.
+
+        Returns
+        -------
+        result : DatetimeIndex
+        """
+        result = Int64Index.unique(self, dropna=dropna)
+        return DatetimeIndex._simple_new(result, tz=self.tz,
+                                         name=self.name)
+
     def union(self, other):
         """
         Specialized union for DatetimeIndex objects. If combine