SPARK-1168, Added foldByKey to pyspark.

Author: ScrapCodes

python/pyspark/rdd.py

@@ -939,7 +939,21 @@ def _mergeCombiners(iterator):
                     combiners[k] = mergeCombiners(combiners[k], v)
             return combiners.iteritems()
         return shuffled.mapPartitions(_mergeCombiners)
+    
+    def foldByKey(self, zeroValue, func, numPartitions=None):
+        """
+        Merge the values for each key using an associative function "func" and a neutral "zeroValue"
+        which may be added to the result an arbitrary number of times, and must not change 
+        the result (e.g., 0 for addition, or 1 for multiplication.).                
 
+        >>> rdd = sc.parallelize([("a", 1), ("b", 1), ("a", 1)])
+        >>> from operator import add
+        >>> rdd.foldByKey(0, add).collect()
+        [('a', 2), ('b', 1)]
+        """
+        return self.combineByKey(lambda v: func(zeroValue, v), func, func, numPartitions)
+    
+    
     # TODO: support variant with custom partitioner
     def groupByKey(self, numPartitions=None):
         """