Implement support for Keyed producer

Provides support for two partitioners
* Round robin
* Hashed (default as per kafka clients)

Author: mahendra

README.md

@@ -41,6 +41,18 @@ for message in consumer:
 kafka.close()
 ```
 
+## Keyed messages
+from kafka.client import KafkaClient
+from kafka.producer import KeyedProducer
+from kafka.partitioner import HashedPartitioner, RoundRobinPartitioner
+
+kafka = KafkaClient("localhost", 9092)
+
+producer = KeyedProducer(kafka, "my-topic", partitioner=HashedPartitioner)
+producer.send("key1", "some message")
+producer.send("key2", "this methode")
+
+
 ## Low level
 
 ```python

kafka/partitioner.py

@@ -0,0 +1,52 @@
+from itertools import cycle
+
+
+class Partitioner(object):
+    """
+    Base class for a partitioner
+    """
+    def __init__(self, partitions):
+        """
+        Initialize the partitioner
+
+        partitions - A list of available partitions (during startup)
+        """
+        self.partitions = partitions
+
+    def partition(self, key, partitions):
+        """
+        Takes a string key and num_partitions as argument and returns
+        a partition to be used for the message
+
+        partitions - The list of partitions is passed in every call. This
+                     may look like an overhead, but it will be useful
+                     (in future) when we handle cases like rebalancing
+        """
+        raise NotImplemented('partition function has to be implemented')
+
+
+class RoundRobinPartitioner(Partitioner):
+    """
+    Implements a round robin partitioner which sends data to partitions
+    in a round robin fashion
+    """
+    def __init__(self, partitions):
+        self.partitions = cycle(partitions)
+
+    def partition(self, key, partitions):
+        # Refresh the partition list if necessary
+        if self.partitions != partitions:
+            self.partitions = cycle(partitions)
+
+        return self.partitions.next()
+
+
+class HashedPartitioner(Partitioner):
+    """
+    Implements a partitioner which selects the target partition based on
+    the hash of the key
+    """
+    def partition(self, key, partitions):
+        size = len(partitions)
+        idx = hash(key) % size
+        return partitions[idx]

kafka/producer.py

@@ -3,6 +3,7 @@
 
 from kafka.common import ProduceRequest
 from kafka.protocol import create_message
+from kafka.partitioner import HashedPartitioner
 
 log = logging.getLogger("kafka")
 
@@ -23,3 +24,34 @@ def send_messages(self, *msg):
 
         resp = self.client.send_produce_request([req])[0]
         assert resp.error == 0
+
+
+class KeyedProducer(object):
+    """
+    A producer which distributes messages to partitions based on the key
+
+    Args:
+    client - The kafka client instance
+    topic - The kafka topic to send messages to
+    partitioner - A partitioner class that will be used to get the partition
+        to send the message to. Must be derived from Partitioner
+    """
+    def __init__(self, client, topic, partitioner=None):
+        self.client = client
+        self.topic = topic
+        self.client._load_metadata_for_topics(topic)
+
+        if not partitioner:
+            partitioner = HashedPartitioner
+
+        self.partitioner = partitioner(self.client.topic_partitions[topic])
+
+    def send(self, client, key, msg):
+        partitions = self.client.topic_partitions[topic]
+        partition = self.partitioner.partition(key, partitions)
+
+        req = ProduceRequest(self.topic, partition,
+                             messages=[create_message(msg)])
+
+        resp = self.client.send_produce_request([req])[0]
+        assert resp.error == 0