python脚本用sqoop把mysql数据导入hive数据仓库中

使用说明：由于项目需要将mysql中的业务数据导入到hive中，这里采用sqoop来做中间桥梁，并且通过supervisor来做时间控制，让python定时启动，导入数据。
安装sqoop1.4.6和supervisor

#! /usr/bin/env python  
# coding:utf-8  
# --------------------------------  
# Created by coco  on 16/2/23  
# ---------------------------------  
# Comment: 主要功能说明 :初始化业务数据库  

import os  
import pyhs2  


conn=pyhs2.connect(host="192.168.8.94",port=10000,authMechanism="PLAIN",user="hdfs")  
mysql_info={"host":"192.168.1.200","port":3306,"user":"root","passwd":"123456"}  
print mysql_info  
def run_hive_query(sql):  
    with conn.cursor()  as cursor:  
        cursor.execute(sql)  
        return cursor.fetchall()  
def mysql_to_hive(host,port,user,passwd,database,table):  
    #os.system("hadoop fs -rm    -r /user/task/%s"%table)  
    if [database] not in run_hive_query("show databases"):  
        with conn.cursor() as cursor:  
            cursor.execute("create database " +database)  
    with conn.cursor() as cursor:  
        cursor.execute("use  "+database)  
    if [table] not in run_hive_query("show tables"):  
        os.system("sqoop   import --connect   jdbc:mysql://%s:%s/%s --username  %s   --password  %s --table %s  --hive-database  %s  -m 10 --create-hive-table --hive-import   --hive-overwrite "%(  
            host,port,database,user,passwd,table,database))  
    else:  
        os.system("sqoop   import --connect   jdbc:mysql://%s:%s/%s --username  %s   --password  %s --table %s  --hive-database  %s  -m 10 --hive-import   --hive-overwrite "%(  
            host,port,database,user,passwd,table,database))  


mysql_to_hive(mysql_info["host"],mysql_info["port"],mysql_info["user"],mysql_info["passwd"].replace("(","\("),"wwn","cm_vip")  

转载自：http://blog.csdn.net/ljphilp/article/details/53884090