每天迁移MySQL历史数据到历史库Python脚本


Posted in Python onApril 13, 2018

本文实例为大家分享了Python每天迁移MySQL历史数据到历史库的具体代码,供大家参考,具体内容如下

#!/usr/bin/env python 
# coding:utf-8 
__author__ = 'John' 
 

import MySQLdb 
import sys 
import datetime 
import time 
 
class ClassMigrate(object): 
  def _get_argv(self): 
    self.usage = """ 
      usage(): 
      python daily_migration.py --source=192.168.1.4:3306/db_name:tab_name/proxy/password \\ 
                    --dest=192.168.1.150:13301/db_name_archive:tab_name_201601/proxy/password \\ 
                    --delete_strategy=delete --primary_key=auto_id --date_col=ut --time_interval=180 
      """ 
    if len(sys.argv) == 1: 
      print self.usage 
      sys.exit(1) 
    elif sys.argv[1] == '--help' or sys.argv[1] == '-h': 
        print self.usage 
        sys.exit() 
    elif len(sys.argv) > 2: 
      for i in sys.argv[1:]: 
        _argv = i.split('=') 
        if _argv[0] == '--source': 
          _list = _argv[1].split('/') 
          self.source_host = _list[0].split(':')[0] 
          self.source_port = int(_list[0].split(':')[1]) 
          self.source_db = _list[1].split(':')[0] 
          self.source_tab = _list[1].split(':')[1] 
          self.source_user = _list[2] 
          self.source_password = _list[3] 
        elif _argv[0] == '--dest': 
          _list = _argv[1].split('/') 
          self.dest_host = _list[0].split(':')[0] 
          self.dest_port = int(_list[0].split(':')[1]) 
          self.dest_db = _list[1].split(':')[0] 
          self.dest_tab = _list[1].split(':')[1] 
          self.dest_user = _list[2] 
          self.dest_password = _list[3] 
        elif _argv[0] == '--delete_strategy': 
          self.deleteStrategy = _argv[1] 
          if self.deleteStrategy not in ('delete', 'drop'): 
            print (self.usage) 
            sys.exit(1) 
        elif _argv[0] == '--primary_key': 
          self.pk = _argv[1] 
        elif _argv[0] == '--date_col': 
          self.date_col = _argv[1] 
        elif _argv[0] == '--time_interval': 
          self.interval = _argv[1] 
        else: 
          print (self.usage) 
          sys.exit(1) 
 
  def __init__(self): 
    self._get_argv() 
## -------------------------------------------------------------------- 
    self.sourcedb_conn_str = MySQLdb.connect(host=self.source_host, port=self.source_port, user=self.source_user, passwd=self.source_password, db=self.source_db, charset='utf8') 
    self.sourcedb_conn_str.autocommit(True) 
    self.destdb_conn_str = MySQLdb.connect(host=self.dest_host, port=self.dest_port, user=self.dest_user, passwd=self.dest_password, db=self.dest_db, charset='utf8') 
    self.destdb_conn_str.autocommit(True) 
## -------------------------------------------------------------------- 
    self.template_tab = self.source_tab + '_template' 
    self.step_size = 20000 
## -------------------------------------------------------------------- 
    self._migCompleteState = False 
    self._deleteCompleteState = False 
## -------------------------------------------------------------------- 
    self.source_cnt = '' 
    self.source_min_id = '' 
    self.source_max_id = '' 
    self.source_checksum = '' 
    self.dest_cn = '' 
## -------------------------------------------------------------------- 
    self.today = time.strftime("%Y-%m-%d") 
    # self.today = '2016-05-30 09:59:40' 

def sourcedb_query(self, sql, sql_type): 
    try: 
      cr = self.sourcedb_conn_str.cursor() 
      cr.execute(sql) 
      if sql_type == 'select': 
        return cr.fetchall() 
      elif sql_type == 'dml': 
        rows = self.sourcedb_conn_str.affected_rows() 
        return rows 
      else: 
        return True 
    except Exception, e: 
      print (str(e) + "<br>") 
      return False 
    finally: 
      cr.close() 

  def destdb_query(self, sql, sql_type, values=''): 
    try: 
      cr = self.destdb_conn_str.cursor() 
      if sql_type == 'select': 
        cr.execute(sql) 
        return cr.fetchall() 
      elif sql_type == 'insertmany': 
        cr.executemany(sql, values) 
        rows = self.destdb_conn_str.affected_rows() 
        return rows 
      else: 
        cr.execute(sql) 
        return True 
    except Exception, e: 
      print (str(e) + "<br>") 
      return False 
    finally: 
      cr.close() 
 
 def create_table_from_source(self): 
    '''''因为tab_name表的数据需要迁移到archive引擎表,所以不适合使用这种方式。 预留作其他用途。''' 
    try: 
      sql = "show create table %s;" % self.source_tab 
      create_str = self.sourcedb_query(sql, 'select')[0][1] 
      create_str = create_str.replace('CREATE TABLE', 'CREATE TABLE IF NOT EXISTS') 
      self.destdb_query(create_str, 'ddl') 
      return True 
    except Exception, e: 
      print (str(e) + "<br>") 
      return False 

  def create_table_from_template(self): 
    try: 
      sql = 'CREATE TABLE IF NOT EXISTS %s like %s;' % (self.dest_tab, self.template_tab) 
      state = self.destdb_query(sql, 'ddl') 
      if state: 
        return True 
      else: 
        return False 
    except Exception, e: 
      print (str(e + "<br>") + "<br>") 
      return False 

  def get_min_max(self): 
    """ 创建目标表、并获取源表需要迁移的总条数、最小id、最大id """ 
    try: 
      print ("\nStarting Migrate at -- %s <br>") % (datetime.datetime.now().__str__()) 
      sql = """select count(*),IFNULL(min(%s),-1),IFNULL(max(%s),-1) from %s where %s >= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 00:00:00') \ 
           and %s <= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 23:59:59') """ \ 
            % (self.pk, self.pk, self.source_tab, self.date_col, self.today, self.interval, self.date_col, self.today, self.interval) 
      q = self.sourcedb_query(sql, 'select') 
      self.source_cnt = q[0][0] 
      self.source_min_id = q[0][1] 
      self.source_max_id = q[0][2] 
      self.source_checksum = str(self.source_cnt) + '_' + str(self.source_min_id) + '_' + str(self.source_max_id) 
      if self.source_cnt == 0 or self.source_min_id == -1 or self.source_max_id == -1: 
        print ("There is 0 record in source table been matched! <br>") 
        return False 
      else: 
        return True 
    except Exception, e: 
      print (str(e) + "<br>") 
      return False 
 
  def migrate_2_destdb(self): 
    try: 
      get_min_max_id = self.get_min_max() 
      if get_min_max_id: 
        k = self.source_min_id 
        desc_sql = "desc %s;" % self.source_tab 
        # self.filed = [] 
        cols = self.sourcedb_query(desc_sql, 'select') 
        # for j in cols: 
        #   self.filed.append(j[0]) 
        fileds = "%s," * len(cols) # 源表有多少个字段,就拼凑多少个%s,拼接到insert语句 
        fileds = fileds.rstrip(',') 
        while k <= self.source_max_id: 
          sql = """select * from %s where %s >= %d and %s< %d \ 
               and %s >= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 00:00:00') \ 
               and %s <= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 23:59:59') """\ 
             % (self.source_tab, self.pk, k, self.pk, k+self.step_size, self.date_col, self.today, self.interval, self.date_col, self.today, self.interval) 
          print ("\n%s <br>") % sql 
          starttime = datetime.datetime.now() 
          results = self.sourcedb_query(sql, 'select') 
          insert_sql = "insert into " + self.dest_tab + " values (%s)" % fileds 
          rows = self.destdb_query(insert_sql, 'insertmany', results) 
          if rows == False: 
            print ("Insert failed!! <br>") 
          else: 
            print ("Inserted %s rows. <br>") % rows 
          endtime = datetime.datetime.now() 
          timeinterval = endtime - starttime 
          print("Elapsed :" + str(timeinterval.seconds) + '.' + str(timeinterval.microseconds) + " seconds <br>") 
          k += self.step_size 
        print ("\nInsert complete at -- %s <br>") % (datetime.datetime.now().__str__()) 
        return True 
      else: 
        return False 
    except Exception, e: 
      print (str(e) + "<br>") 
      return False 
  
  def verify_total_cnt(self): 
    try: 
      sql = """select count(*),IFNULL(min(%s),-1),IFNULL(max(%s),-1) from %s where %s >= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 00:00:00') \ 
           and %s <= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 23:59:59') """ \ 
            % (self.pk, self.pk, self.dest_tab, self.date_col, self.today, self.interval, self.date_col, self.today, self.interval) 
      dest_result = self.destdb_query(sql, 'select') 
      self.dest_cnt = dest_result[0][0] 
      dest_checksum = str(self.dest_cnt) + '_' + str(dest_result[0][1]) + '_' + str(dest_result[0][2]) 
      print ("source_checksum: %s, dest_checksum: %s <br>") % (self.source_checksum, dest_checksum) 
      if self.source_cnt == dest_result[0][0] and dest_result[0][0] != 0 and self.source_checksum == dest_checksum: 
        self._migCompleteState = True 
        print ("Verify successfully !!<br>") 
      else: 
        print ("Verify failed !!<br>") 
        sys.exit(77) 
    except Exception, e: 
      print (str(e) + "<br>") 
  
  def drop_daily_partition(self): 
    try: 
      if self._migCompleteState: 
        sql = """explain partitions select * from %s where %s >= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 00:00:00') 
               and %s <= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 23:59:59') """\ 
             % (self.source_tab, self.date_col, self.today, self.interval, self.date_col, self.today, self.interval) 
        partition_name = self.sourcedb_query(sql, 'select') 
        partition_name = partition_name[0][3] 
 
 
 
 
        sql = """select count(*),IFNULL(min(%s),-1),IFNULL(max(%s),-1) from %s partition (%s)""" \ 
            % (self.pk, self.pk, self.source_tab, partition_name) 
        q = self.sourcedb_query(sql, 'select') 
        source_cnt = q[0][0] 
        source_min_id = q[0][1] 
        source_max_id = q[0][2] 
        checksum = str(source_cnt) + '_' + str(source_min_id) + '_' + str(source_max_id) 
        if source_cnt == 0 or source_min_id == -1 or source_max_id == -1: 
          print ("There is 0 record in source PARTITION been matched! <br>") 
        else: 
          if checksum == self.source_checksum: 
            drop_par_sql = "alter table %s drop partition %s;" % (self.source_tab, partition_name) 
            droped = self.sourcedb_query(drop_par_sql, 'ddl') 
            if droped: 
              print (drop_par_sql + " <br>") 
              print ("\nDrop partition complete at -- %s <br>") % (datetime.datetime.now().__str__()) 
              self._deleteCompleteState = True 
            else: 
              print (drop_par_sql + " <br>") 
              print ("Drop partition failed.. <br>") 
          else: 
            print ("The partition %s checksum failed !! Drop failed !!") % partition_name 
            sys.exit(77) 
    except Exception, e: 
      print (str(e) + "<br>") 
 
  def delete_data(self): 
    try: 
      if self._migCompleteState: 
        k = self.source_min_id 
        while k <= self.source_max_id: 
          sql = """delete from %s where %s >= %d and %s< %d \ 
               and %s >= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 00:00:00') \ 
               and %s <= CONCAT(DATE_FORMAT(DATE_ADD('%s', INTERVAL -%s day),'%%Y-%%m-%%d'), ' 23:59:59') """ \ 
             % (self.source_tab, self.pk, k, self.pk, k+self.step_size, self.date_col, self.today, self.interval, self.date_col, self.today, self.interval) 
          print ("\n%s <br>") % sql 
          starttime = datetime.datetime.now() 
          rows = self.sourcedb_query(sql, 'dml') 
          if rows == False: 
            print ("Delete failed!! <br>") 
          else: 
            print ("Deleted %s rows. <br>") % rows 
          endtime = datetime.datetime.now() 
          timeinterval = endtime - starttime 
          print("Elapsed :" + str(timeinterval.seconds) + '.' + str(timeinterval.microseconds) + " seconds <br>") 
          time.sleep(1) 
          k += self.step_size 
        print ("\nDelete complete at -- %s <br>") % (datetime.datetime.now().__str__()) 
        self._deleteCompleteState = True 
    except Exception, e: 
      print (str(e) + "<br>") 
       
  def do(self): 
    tab_create = self.create_table_from_template() 
    if tab_create: 
      migration = self.migrate_2_destdb() 
      if migration: 
        self.verify_total_cnt() 
        if self._migCompleteState: 
          if self.deleteStrategy == 'drop': 
            self.drop_daily_partition() 
          else: 
            self.delete_data() 
          print ("\n<br>") 
          print ("====="*5 + '<br>') 
          print ("source_total_cnt: %s <br>") % self.source_cnt 
          print ("dest_total_cnt: %s <br>") % self.dest_cnt 
          print ("====="*5 + '<br>') 
          if self._deleteCompleteState: 
            print ("\nFinal result: Successfully !! <br>") 
            sys.exit(88) 
          else: 
            print ("\nFinal result: Failed !! <br>") 
            sys.exit(254) 
    else: 
      print ("Create table failed ! Exiting. . .") 
      sys.exit(255) 
  
f = ClassMigrate() 
f.do()

以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持三水点靠木。

Python 相关文章推荐
跟老齐学Python之用while来循环
Oct 02 Python
python中map、any、all函数用法分析
Apr 21 Python
用pickle存储Python的原生对象方法
Apr 28 Python
TensorFlow深度学习之卷积神经网络CNN
Mar 09 Python
python实现事件驱动
Nov 21 Python
python创建属于自己的单词词库 便于背单词
Jul 30 Python
python3实现从kafka获取数据,并解析为json格式,写入到mysql中
Dec 23 Python
Pycharm内置终端及远程SSH工具的使用教程图文详解
Mar 19 Python
PyQt5 界面显示无响应的实现
Mar 26 Python
Python requests及aiohttp速度对比代码实例
Jul 16 Python
Pycharm-community-2020.2.3 社区版安装教程图文详解
Dec 08 Python
matplotlib bar()实现百分比堆积柱状图
Feb 24 Python
python实现数据库跨服务器迁移
Apr 12 #Python
解决python3爬虫无法显示中文的问题
Apr 12 #Python
python读取中文txt文本的方法
Apr 12 #Python
基于python 处理中文路径的终极解决方法
Apr 12 #Python
解决Python2.7读写文件中的中文乱码问题
Apr 12 #Python
python 实现对文件夹内的文件排序编号
Apr 12 #Python
pandas数值计算与排序方法
Apr 12 #Python
You might like
社区(php&amp;&amp;mysql)一
2006/10/09 PHP
PHP 中的面向对象编程:通向大型 PHP 工程的办法
2006/12/03 PHP
frename PHP 灵活文件命名函数 frename
2009/09/09 PHP
PHP_NETWORK_GETADDRESSES: GETADDRINFO FAILED问题解决办法
2014/05/04 PHP
PHP中用Trait封装单例模式的实现
2019/12/18 PHP
LazyLoad 延迟加载(按需加载)
2010/05/31 Javascript
基于JQuery的一句代码实现表格的简单筛选
2010/07/26 Javascript
THREE.JS入门教程(1)THREE.JS使用前了解
2013/01/24 Javascript
javascript操纵OGNL标签示例代码
2014/06/16 Javascript
JQuery替换DOM节点的方法
2015/06/11 Javascript
充分发挥Node.js程序性能的一些方法介绍
2015/06/23 Javascript
js实现select下拉框菜单
2015/12/08 Javascript
jQuery插件之Tocify动态节点目录菜单生成器附源码下载
2016/01/08 Javascript
Bootstrap php制作动态分页标签
2016/12/23 Javascript
jquery实现提示语淡入效果
2017/05/05 jQuery
在vue-cli的组件模板里使用font-awesome的两种方法
2018/09/28 Javascript
微信小程序登录态和检验注册过没的app.js写法
2019/05/22 Javascript
vue 解决兄弟组件、跨组件深层次的通信操作
2020/07/27 Javascript
解决ant design vue中树形控件defaultExpandAll设置无效的问题
2020/10/26 Javascript
使用python在校内发人人网状态(人人网看状态)
2014/02/19 Python
小白入门篇使用Python搭建点击率预估模型
2018/10/12 Python
python pygame实现方向键控制小球
2019/05/17 Python
python [:3] 实现提取数组中的数
2019/11/27 Python
使用tqdm显示Python代码执行进度功能
2019/12/08 Python
Python object类中的特殊方法代码讲解
2020/03/06 Python
加拿大百叶窗和窗帘定制网站:Blinds
2017/01/30 全球购物
Under Armour安德玛德国官网:美国高端运动科技品牌
2019/03/09 全球购物
几个人围成一圈的问题
2013/09/26 面试题
工商治理实习生的自我评价
2014/01/15 职场文书
计算机应用应届生求职信
2014/07/12 职场文书
交警失职检讨书
2015/01/26 职场文书
倡议书的格式写法
2015/04/28 职场文书
高效笔记技巧分享:学会这些让你不再困扰
2019/09/04 职场文书
浅谈react useEffect闭包的坑
2021/06/08 Javascript
详解CSS中的特指度和层叠问题
2021/07/15 HTML / CSS
Android Studio 计算器开发
2022/05/20 Java/Android