Metadata-Version: 1.1
Name: asynckafka
Version: 0.2.0
Summary: Fast python kafka client for asyncio.
Home-page: http://github.com/jmf-mordis/asynckafka
Author: José Melero Fernández
Author-email: jmelerofernandez@gmail.com
License: MIT
Download-URL: https://github.com/jmf-mordis/asynckafka/archive/0.2.0.tar.gz
Description: **********
        Asynckafka
        **********
        
        .. image:: https://travis-ci.com/jmf-mordis/asynckafka.svg?branch=master
            :target: https://travis-ci.com/jmf-mordis/asynckafka
        
        Fast python Kafka client for asyncio.
        Asynckafka is written in Cython_ on top of Rdkafka_ as Kafka driver.
        
        .. _Cython: cython.org
        .. _Rdkafka: https://github.com/edenhill/librdkafka
        
        The documentation can be found here_.
        
        .. _here: https://jmf-mordis.github.io/asynckafka/
        
        Features
        ========
        
        * Consumer using a balanced group
        * Producer
        
        The library was born as a project to learn Cython, right now it only has the basic
        features implemented, some of the most important missing features are:
        
        * Offset management in the consumer
        * Precise partition management
        
        Examples
        ========
        
        Simple consumer
        ---------------
        
        How to use a consumer::
        
            consumer = Consumer(
                brokers='localhost:9092',
                topics=['my_topic'],
                group_id='my_group_id',
            )
            consumer.start()
        
            async for message in consumer:
                print(f"Received message: {message.payload}")
        
        Simple producer
        ---------------
        
        How to use a producer::
        
            producer = Producer(brokers="localhost:9092")
            producer.start()
            await producer.produce("my_topic", b"my_message")
        
        Benchmark
        #########
        
        The test was performed in June of 2018 using a single Kafka broker without replication.
        The purpose of the benchmark was only to have an idea of the order of magnitude of the
        library's performance under these conditions.
        
        Comparison between asynckafka and aiokafka in production and consumption:
        
        Production
        **********
        
        .. image:: https://github.com/jmf-mordis/asynckafka/raw/master/docs/benchmark/graphs/producer.png
           :width: 800
        
        Consumption
        ***********
        
        .. image:: https://github.com/jmf-mordis/asynckafka/raw/master/docs/benchmark/graphs/consumer.png
           :width: 800
        
        The benchmark used for asynckafka is in benchmark directory while the
        benchmark used for aiokafka is in its own repository, also in the benchmark folder.
Keywords: asyncio,kafka,cython
Platform: *nix
Classifier: Development Status :: 2 - Pre-Alpha
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: License :: OSI Approved :: MIT License
Classifier: Intended Audience :: Developers
Classifier: Framework :: AsyncIO
