數(shù)據(jù)庫(kù)間數(shù)據(jù)同步方式很多,在上篇博文中有總結(jié)。本文是用py程序?qū)崿F(xiàn)數(shù)據(jù)同步。
A數(shù)據(jù)庫(kù)中有幾十張表,要匯聚到B數(shù)據(jù)庫(kù)中,且表結(jié)構(gòu)一致,需要準(zhǔn)實(shí)時(shí)的進(jìn)行數(shù)據(jù)同步,用工具實(shí)現(xiàn)時(shí)對(duì)其控制有限且配置較繁瑣,故自寫(xiě)程序,可自由設(shè)置同步區(qū)間,記錄自己想要的日志
本代碼實(shí)現(xiàn)功能簡(jiǎn)單,采用面向過(guò)程,有需求的同學(xué)可以自己優(yōu)化成面向?qū)ο蠓绞剑谌罩具@塊缺少數(shù)據(jù)監(jiān)控,可根據(jù)需求增加。主要注意點(diǎn):
1、數(shù)據(jù)抽取時(shí)采用區(qū)間抽取(按時(shí)間區(qū)間)、流式游標(biāo)迭代器+fetchone,避免內(nèi)存消耗
2、在數(shù)據(jù)插入時(shí)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、獲取需要抽取的表,抽取數(shù)據(jù)的時(shí)間點(diǎn)
with open(filepathtime, 'r') as f:
lines = f.readlines() # 讀取所有數(shù)據(jù)
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('\n','')
#print(tablename_list)
tablename = tablename_list[0]
updatetime = tablename_list[1]
#print(tablename,updatetime)
#2、抽取此表此時(shí)間點(diǎn)的數(shù)據(jù),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 首先判斷下原表中是否有待同步數(shù)據(jù),若有則同步且更新同步的時(shí)間參考點(diǎn),若沒(méi)有則不同步且不更新同步的時(shí)間參考點(diǎn)
try:
cursor.execute(q_sql)
results = cursor.fetchone()
#print(results) #返回是元組
#print("查詢?cè)頂?shù)據(jù)成功!",tablename)
except BaseException as e:
print("查詢?cè)頂?shù)據(jù)失敗!",tablename, str(e))
#記錄異常日志
updatetime_n = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
eachline_log = updatetime_n + '[erro]:' + tablename + str(e) + '\n'
content_log.append(eachline_log)
recode_log(content_log)
db.close()
if results:
print("===============================================================================")
print("有數(shù)據(jù)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) #返回是元組
#使用流式游標(biāo)迭代器+fetchone,減少內(nèi)存消耗
while True:
result = cursor.fetchone()
if not result:
print("此區(qū)間無(wú)數(shù)據(jù)", 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("查詢數(shù)據(jù)成功!", tablename)
except BaseException as e:
print("查詢數(shù)據(jù)失敗!", tablename, str(e))
# 記錄異常日志
updatetime_n = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
eachline_log = updatetime_n + '[erro]:' + tablename + str(e) + '\n'
content_log.append(eachline_log)
recode_log(content_log)
db.close()
results_len = (len(result_list))
if results_len>0:
#3、將數(shù)據(jù)插入到目標(biāo)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("插入失敗!", tablename,str(e))
#記錄異常日志
updatetime_n = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
eachline_log = updatetime_n + '[erro]:' + tablename + str(e) + '\n'
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、如果有數(shù)據(jù)同步,則更新參考點(diǎn)時(shí)間為下一個(gè)節(jié)點(diǎn)時(shí)間
eachline_time = tablename+','+updatetime_e+'\n' #此時(shí)間點(diǎn)是下一個(gè)時(shí)間點(diǎn)updatetime_e
content_time.append(eachline_time)
print("更新表時(shí)間點(diǎn)",content_time)
# 5、記錄成功日志
eachline_log = enddatetime + '[success]:' + tablename + '開(kāi)始時(shí)間' + startdatetime + \
'結(jié)束時(shí)間' + enddatetime + ',同步數(shù)據(jù)量'+str(results_len)+',當(dāng)前參考點(diǎn)' + updatetime_e + '\n'
content_log.append(eachline_log)
print("日志信息",content_log)
#print("===============================================================================")
else:
print("===============================================================================")
print("無(wú)數(shù)據(jù)可同步",tablename)
#db.close()
enddatetime = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
# 4、如果無(wú)數(shù)據(jù)同步,則參考點(diǎn)時(shí)間不更新
eachline_time = tablename + ',' + updatetime + '\n' #此時(shí)間點(diǎn)還是原時(shí)間updatetime
content_time.append(eachline_time)
print("不更新表時(shí)間點(diǎn)",content_time)
# 5、成功日志信息
eachline_log = enddatetime + '[success]:' + tablename + '開(kāi)始時(shí)間' + startdatetime + \
'結(jié)束時(shí)間' + enddatetime + ',同步數(shù)據(jù)量0'+ ',當(dāng)前參考點(diǎn)' + updatetime_e + '\n'
content_log.append(eachline_log)
print("日志信息",content_log)
#print("===============================================================================")
#更新配置文件,記錄日志
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分鐘執(zhí)行一次同步
# while True:
# transferdata()
# time.sleep(300)
到此這篇關(guān)于python實(shí)現(xiàn)不同數(shù)據(jù)庫(kù)間數(shù)據(jù)同步功能的文章就介紹到這了,更多相關(guān)python實(shí)現(xiàn)數(shù)據(jù)同步內(nèi)容請(qǐng)搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家!