python链接imala工具

# -*- coding:utf-8 -*-
from datetime import datetime, date, timedelta
import time
from impala.dbapi import connect
from krbcontext import krbcontext

log = lambda X: print(f"{datetime.now().strftime('%Y-%m-%d %H:%M:%S')}：{X}")


class ConnectUtils(object):
    def __init__(self, connect_type="impala"):
        '''
        :param connect_type: 默认impala,可以选择 impala 或者 hive
        '''
        self.conn = None
        self.cur = None
        self.connect_type = connect_type
        self.__connect()

    def __connect(self, connect_host='cdh2.slave1.com'):
        '''
        连接大数据平台
        '''
        retry_nums = 10
        port = 21051 if self.connect_type == "impala" else 10000
        while retry_nums:
            try:
                with krbcontext(using_keytab=True,
                                principal='***/****',
                                keytab_file='keytab_file_path'):
                    conn = connect(host=connect_host, port=port, auth_mechanism='GSSAPI', database='spiderdb',
                                   kerberos_service_name=self.connect_type)
                    self.conn = conn
                    self.cur = self.conn.cursor()
                    return
            except Exception as e:
                log(f'error msg: {e}')
                time.sleep(0.5)
                retry_nums -= 1
        exit()

    def implement_sql(self, sql):
        '''
        执行语句，例如：插入语句、更新语句、同步语句
        '''
        try:
            self.cur.execute(sql)
            self.conn.commit()
            log(f"语句执行成功:{sql}")
        except Exception as e:
            log(f"sql语句执行失败，msg:{e}")

    def select(self, sql, ls=None):
        '''
        查询语句
        '''
        try:
            if ls:
                self.cur.executemany(sql, ls)
            else:
                self.cur.execute(sql)
            data = self.cur.fetchall()
            return data
        except Exception as e:
            log(f"数据查询失败，msg:{e}")

    def insert(self, table_name: str, keys: list, data_ls: list, partition=None):
        '''
        数据量较多时使用
        :param table_name: 表名
        :param keys: 插入字段
        :param data_ls: 数据列表
        :param partition: 分区字典
        '''
        cph_ls = "{}" + ','.join(['{}' for i in range(len(keys))])  # 拼接占位符字符串
        # 判断是否加入分区条件
        if partition:
            partition = ','.join([f"{key}='{value}'" for key, value in partition.items()])
            sql = f"INSERT into {table_name}({','.join(keys)}) PARTITION({partition}) values({cph_ls})"
        else:
            sql = f"INSERT into {table_name}({','.join(keys)}) values({cph_ls})"
        try:
            self.cur.executemany(sql, data_ls)
            log(f"数据插入成功，数据量：{len(data_ls)}")
        except Exception as e:
            log(f"数据插入失败，msg:{e}")

    def close(self):
        self.cur.close()
        self.conn.close()
        self.conn, self.cur = None, None

    def __del__(self):
        if self.conn:
            print("关闭连接")
            self.close()