1.如何安装Thrift编译器，
2.通过一个小例子展示如何使用python通过Thrift调用Hive
必要前提条件：1.已安装配置jdk 2.已安装Hadoop,Hive并已启动、可使用。3.已安装gcc与g++并可使用。
本机环境说明：
jdk1.7
hadoop-1.1.2
hive-0.11.0
操作系统:Ubuntu12.04
一.安装Thrift
1.安装需要的工具和类库
sudo apt-get install libboost-dev libboost-test-dev libboost-program-options-dev libevent-dev automake libtool flex bison pkg-config g++ libssl-dev
2.下载\解压\编译\安装 Thrift
wget https://pypi.python.org/packages/source/t/thrift/thrift-0.9.1.tar.gz tar -zxvf thrift-0.9.1.tar.gz cd thrift-0.9.1 ./configure sudo make && sudo make install
二.将hive_home/lib/py下的文件copy至python可加载的目录
cp -r $HIVE_HOME/lib/py/* /usr/local/lib/python2.7/dist-packages/
至此环境已搭建完毕
三.启动hiveserver
hive --service hiveserver -p 100028 -v
请在已将HIVE_HOME/bin目录配置进系统环境变量的情况下执行该命令，否则请在HIVE_HOME/bin目录下执行
四.测试程序
测试程序如下，在本机环境下测试成功，在生产环境中一定要需要捕获的异常捕获，将所有异常情况控制在自己手里。
#!/usr/bin/env python
#*-*coding:UTF-8 *-*
#Author:JohnWang
#Date:2014-01-15
#Version:1.0
import sys
from hive_service import ThriftHive
from hive_service.ttypes import HiveServerException
from thrift import Thrift
from thrift.transport import TSocket,TTransport
from thrift.transport.TTransport import TTransportException
from thrift.protocol import TBinaryProtocol
class HiveClient(object):
def __init__(self,conInfo):
self.__dict__.update(conInfo)
def connect(self):
transport = None
client = None
try:
transport = TSocket.TSocket(self.ip,self.port)
transport = TTransport.TBufferedTransport(transport)
protocol = TBinaryProtocol.TBinaryProtocol(transport)
client = ThriftHive.Client(protocol)
transport.open()
except TTransportException,error:
print error
return False
self.client = client
self.tranport = transport
if self.db:
return self.execute(‘use %s‘ %self.db)
return True
def execute(self,sql):
try:
self.client.execute(sql)
except HiveServerException,error:
print error
return False
return True
def getOne(self):
pass
def getAll(self):
pass
def close(self):
if self.transport:
self.transport.close()
return True
else:
return False
if __name__ == ‘__main__‘:
conInfo = {‘ip‘:‘127.0.0.1‘,‘port‘:10028,‘user‘:‘‘,‘passwd‘:‘‘,‘db‘:‘test_hive‘}
client = HiveClient(conInfo)
if client.connect():
print ‘connect hive succes‘