python实现不同数据库间数据同步功能

python实现不同数据库间数据同步功能

功能描述

数据库间数据同步方式很多,在上篇博文中有总结。本文是用py程序实现数据同步。
A数据库中有几十张表,要汇聚到B数据库中,且表结构一致,需要准实时的进行数据同步,用工具实现时对其控制有限且配置较繁琐,故自写程序,可自由设置同步区间,记录自己想要的日志

代码

本代码实现功能简单,采用面向过程,有需求的同学可以自己优化成面向对象方式,在日志这块缺少数据监控,可根据需求增加。主要注意点:
1、数据抽取时采用区间抽取(按时间区间)、流式游标迭代器+fetchone,避免内存消耗
2、在数据插入时采用executemany(list),加快插入效率

import pymysql
import os
import datetime,time

def update_time(content):
  with open(filepathtime, "w") as f:
    f.writelines(content)

def recode_log(content):
  with open(filepathlog, "a") as f:
    f.writelines(content)

def transferdata():
  #1、获取需要抽取的表,抽取数据的时间点
  with open(filepathtime, "r") as f:
    lines = f.readlines() # 读取所有数据
    print("需要同步的表信息",lines)
    for line in lines:
      startdatetime = time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))
      tablename_list =line.split(",")
      #print(tablename_list)
      #print(tablename_list[-1])
      tablename_list[-1] = tablename_list[-1].replace("
","")
      #print(tablename_list)
      tablename = tablename_list[0]
      updatetime = tablename_list[1]
      #print(tablename,updatetime)

      #2、抽取此表此时间点的数据,同步
      updatetime_s = datetime.datetime.strptime(updatetime, "%Y-%m-%d %H:%M:%S")
      updatetime_e = (updatetime_s + datetime.timedelta(hours=1)).strftime("%Y-%m-%d %H:%M:%S")
      #print(updatetime_s)
      #print(q_sql)
      db = pymysql.connect(host=host_o, port=port_o, user=user_o, passwd=passwd_o, db=db_o)
      cursor = db.cursor()
      q_sql = "select a,b,c from %s where c >= "%s" " % 
          (tablename, updatetime_s)
      #2.1 首先判断下原表中是否有待同步数据,若有则同步且更新同步的时间参考点,若没有则不同步且不更新同步的时间参考点
      try:
        cursor.execute(q_sql)
        results = cursor.fetchone()
        #print(results) #返回是元组
        #print("查询原表数据成功!",tablename)
      except BaseException as e:
        print("查询原表数据失败!",tablename, str(e))
        #记录异常日志
        updatetime_n = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(time.time()))
        eachline_log = updatetime_n + "[erro]:" + tablename + str(e) + "
"
        content_log.append(eachline_log)
        recode_log(content_log)
      db.close()

      if results:
        print("===============================================================================")
        print("有数据可同步",tablename)
        db = pymysql.connect(host=host_o, port=port_o, user=user_o, passwd=passwd_o, db=db_o, charset="utf8", cursorclass=pymysql.cursors.SSDictCursor)
        cursor = db.cursor()
        q_sql1 = "select a,b,c from %s where c >= "%s" and c < "%s" " % 
             (tablename, updatetime_s, updatetime_e)
        #print(q_sql1)
        result_list = []
        try:
          # startdatetime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
          cursor.execute(q_sql1)
          #results = cursor.fetchall()
          # enddatetime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
          # print(results) #返回是元组
          #使用流式游标迭代器+fetchone,减少内存消耗
          while True:
            result = cursor.fetchone()
            if not result:
              print("此区间无数据", q_sql1)
              break
            else:
              one_list = list(result.values())
              # print(result_list)
              result_list.append(one_list)
          print(result_list) #返回是列表
          #print("查询数据成功!", tablename)
        except BaseException as e:
          print("查询数据失败!", tablename, str(e))
          # 记录异常日志
          updatetime_n = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(time.time()))
          eachline_log = updatetime_n + "[erro]:" + tablename + str(e) + "
"
          content_log.append(eachline_log)
          recode_log(content_log)
        db.close()

        results_len = (len(result_list))
        if results_len>0:
          #3、将数据插入到目标表中,利用list提高插入效率
          i_sql = "insert into table_t(a,b,c) values (%s,%s,%s)"
          #print(i_sql)
          db = pymysql.connect(host=host_d, port=port_d, user=user_d, passwd=passwd_d, db=db_d)
          cursor = db.cursor()
          try:
            #startdatetime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
            cursor.executemany(i_sql, result_list)
            db.commit()
            #enddatetime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
            print("插入成功!",tablename)
          except BaseException as e:
            db.rollback()
            print("插入失败!", tablename,str(e))
            #记录异常日志
            updatetime_n = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(time.time()))
            eachline_log = updatetime_n + "[erro]:" + tablename + str(e) + "
"
            content_log.append(eachline_log)
            recode_log(content_log)
          db.close()
        enddatetime = time.strftime("%Y-%m-%d %H:%M:%S",time.localtime(time.time()))

        #4、如果有数据同步,则更新参考点时间为下一个节点时间
        eachline_time = tablename+","+updatetime_e+"
" #此时间点是下一个时间点updatetime_e
        content_time.append(eachline_time)
        print("更新表时间点",content_time)

        # 5、记录成功日志
        eachline_log = enddatetime + "[success]:" + tablename + "开始时间" + startdatetime + 
          "结束时间" + enddatetime + ",同步数据量"+str(results_len)+",当前参考点" + updatetime_e + "
"
        content_log.append(eachline_log)
        print("日志信息",content_log)
        #print("===============================================================================")
      else:
        print("===============================================================================")
        print("无数据可同步",tablename)
        #db.close()
        enddatetime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(time.time()))
        # 4、如果无数据同步,则参考点时间不更新
        eachline_time = tablename + "," + updatetime + "
" #此时间点还是原时间updatetime
        content_time.append(eachline_time)
        print("不更新表时间点",content_time)

        # 5、成功日志信息
        eachline_log = enddatetime + "[success]:" + tablename + "开始时间" + startdatetime + 
          "结束时间" + enddatetime + ",同步数据量0"+ ",当前参考点" + updatetime_e + "
"
        content_log.append(eachline_log)
        print("日志信息",content_log)
        #print("===============================================================================")

    #更新配置文件,记录日志
    update_time(content_time)
    recode_log(content_log)

if __name__ == "__main__":
  filepathtime = "D:/test/table-time.txt"
  filepathlog = "D:/test/table-log.txt"
  host_o = "localhost"
  port_o = 3306
  user_o = "root"
  passwd_o = "root@123"
  db_o = "csdn"
  host_d = "localhost"
  port_d = 3306
  user_d = "root"
  passwd_d = "root@123"
  db_d = "csdn"
  content_time = []
  content_log = []
  transferdata()

  #每5分钟执行一次同步
  # while True:
  #   transferdata()
  #   time.sleep(300)

table-time.txt配置文件,格式说明:
每行包括源库表名、此表的最小时间time,以逗号分隔
若多个表,可配置多个时间
每次脚本执行后,同步更新时间time。时间间隔设置为1小时,可根据情况在updatetime_e中对增量进行修改

table-log.txt
记录每次同步任务执行的结果,或执行中发生异常的日志
此文件需要定期进行清理

到此这篇关于python实现不同数据库间数据同步功能的文章就介绍到这了,更多相关python实现数据同步内容请搜索云海天教程以前的文章或继续浏览下面的相关文章希望大家以后多多支持云海天教程!