Home >Backend Development >Python Tutorial >Detailed explanation of Thrift sample code sharing in Python

Detailed explanation of Thrift sample code sharing in Python

黄舟
黄舟Original
2017-05-07 10:40:593357browse


Preface

Apache Thrift is an efficient framework implemented by Facebook that supports remote service invocation in multiple programming languages. This article will briefly introduce the architecture, development and use of Apache Thrift from the perspective of Python developers.

Thrift introduction

Thrift network stack

Transport

Transport网络读写(socket,http等)抽象,用于和其他thrift组件解耦。
Transport的接口包括:open, close, read, write, flush, isOpen, readAll。
Server端需要ServerTransport(对监听socket的一种抽象),用于接收客户端连接,接口包括:listen, accept, close。
python中Transport的实现包括:TSocket, THttpServer, TSSLSocket, TTwisted, TZlibTransport,都是对某种协议或框架的实现。还有两个装饰器,用于为已有的Transport添加功能,TBufferedTransport(增加缓冲)和TFramedTransport(添加帧)。
在创建server时,传入的时Tranport的工厂,这些Factory包括:TTransportFactoryBase(没有任何修饰,直接返回),TBufferedTransportFactory(返回带缓冲的Transport)和TFramedTransportFactory(返回帧定位的Transport)。

Protocol

Protocol用于对数据格式抽象,在rpc调用时序列化请求和响应。
TProtocol的实现包括:TJSONProtocol,TSimpleJSONProtocol,TBinaryProtocol,TBinaryPotocolAccelerated,TCompactProtocol。

Processor

Processor对stream读写抽象,最终会调用用户编写的handler已响应对应的service。
具体的Processor有compiler生成,用户需要实现service的实现类。

Server

Server创建Transport,输入、输出的Protocol,以及响应service的handler,监听到client的请求然后委托给processor处理。
TServer是基类,构造函数的参数包括:
1) processor, serverTransport
2) processor, serverTransport, transportFactory, protocolFactory
3) processor, serverTransport, inputTransportFactory, outputTransportFactory, inputProtocolFactory, outputProtocolFactory   
TServer内部实际上需要3)所列的参数,1)和2)会导致对应的参数使用默认值。
TServer的子类包括:TSimpleServer, TThreadedServer, TThreadPoolServer, TForkingServer, THttpServer, TNonblockingServer, TProcessPoolServer
TServer的serve方法用于开始服务,接收client的请求。

Code generated

constants.py: 包含声明的所有常量
ttypes.py: 声明的struct,实现了具体的序列化和反序列化
SERVICE_NAME.py: 对应service的描述文件,包含了:
    Iface: service接口定义
    Client: client的rpc调用桩

Usage

Thrift的用法实际上很简单,定义好IDL,然后实现service对应的handler(方法名、参数列表与接口定义一致接口),最后就是选择各个组件。需要选择的包括:Transport(一般都是socket,只是十分需要选择buffed和framed装饰器factory),Protocol,Server。

Example

IDL file

/*
thrift接口定义文件
*/service HelloService {    string say(1:string msg)
}

After editing the definition file, run the following command to generate thrift file. The hello directory can be moved to the current directory to facilitate subsequent calls.

thrift -r -gen py hello.thrift

server

# coding: utf-8"""
thrift_client.py
"""import socketimport sysfrom hello import HelloServicefrom hello.ttypes import *from thrift.transport import TSocketfrom thrift.transport import TTransportfrom thrift.protocol import TBinaryProtocolfrom thrift.server import TServerclass HelloServiceHandler:
    def say(self, msg):
        ret = "Received: " + msg        print ret        return ret


handler = HelloServiceHandler()
processor = HelloService.Processor(handler)
transport = TSocket.TServerSocket("localhost", 9090)
tfactory = TTransport.TBufferedTransportFactory()
pfactory = TBinaryProtocol.TBinaryProtocolFactory()

server = TServer.TSimpleServer(processor, transport, tfactory, pfactory)print "Starting thrift server in python..."server.serve()print "done!"

client

# coding: utf-8"""
thrift_client.py
"""import sysfrom hello import HelloServicefrom thrift import Thriftfrom thrift.transport import TSocketfrom thrift.transport import TTransportfrom thrift.protocol import TBinaryProtocoltry:
    transport = TSocket.TSocket('localhost', 9090)
    transport = TTransport.TBufferedTransport(transport)
    protocol = TBinaryProtocol.TBinaryProtocol(transport)
    client = HelloService.Client(protocol)
    transport.open()    print "client - say"
    msg = client.say("Hello!")    print "server - " + msg

    transport.close()except Thrift.TException, ex:    print "%s" % (ex.message)

Running results

$ ptyhon thrift_client.pyclient - say
server - Received: Hello!

$ python thrift_server.pyStarting thrift server in python...Received: Hello!

Summary

This article is just a simple example, in actual projects , generally based on zookeeper to register and manage the thrift state of the service, and further encapsulate the server and client to facilitate calls in various modules of the project.

The above is the detailed content of Detailed explanation of Thrift sample code sharing in Python. For more information, please follow other related articles on the PHP Chinese website!

Statement:
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn