pyhs2读取Hive

版本/平台原因，SparkSql无法读取实时ORC分区表，采用Pyhs2读取

import pyhs2
 
#连接hive客户端
def get_hiveserver_connect():
    con=pyhs2.connect(host="localhost",
                      port=8000,
                      authMechanism="KERBEROS",
                      user="xxxx",
                      database="default")
    return con
 
#执行语句，返回 List[Dict]
def get_hive_result(consql):
 
    con = get_hiveserver_connect()
    print con
    cur = con.cursor()
    print cur
    cur.execute(consql)
    columns=[]
    list_data=[]
     
    #获取列字段信息
    for j in cur.getSchema():
        columns.append(j['columnName'])
 
    data_list=cur.fetch()
    #将字段信息和数据映射为字典
    for data in data_list:
        list_data.append(dict(zip(columns,data)))
 
    con.close()
 
    return list_data

consql = "select pv,uv from test1"
print get_hive_result(consql)
# Return: [{'uv': 200, 'pv': 100}, {'uv': 220, 'pv': 110}]