一、背景

        最近项目搞重构，将原有的系统拆分成了多个子系统。但是有数据表需要在不同系统中数据，同时为了解决项目性能最了一个很简单的方案，就是公共数据存在每个系统之中。

二、分析

        分析这些表，这些表相比源数据表，表结构完全相同，除了名称加了MD前缀。但数据却要相同。这里我们可以从几个维度去进行比对：

        a、表结构

        b、表数据量

        c、表主键最大id

        d、表数据

三、编码

因最开始是以脚本的形式来写，所有没做配置文件。

1、数据库相关

数据库连接信息

# coding=utf8

"""
数据库配置信息
"""

mysql_data = {
    "CRM": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "PLM": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306

    },
    "MES": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": ********",
        "db": "db_name",
        "port": 3306
    },
    "SCM": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "UC": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "GDM": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "ESM": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "LOCAL": {
        "host": "127.0.0.1",
        "user": "user_name",
        "password": "********",
        "db": "db_name",
        "port": 3306
    },
    "78": {
        'host': '127.0.0.1',
        "user": 'user_name',
        'password': '********',
        "db": "db_name",
        "port": 3306
    },
    "PLM-PRE": {
        'host': '127.0.0.1',
        "user": 'user_name',
        'password': '********',
        "db": "db_name",
        "port": 3306
    }
}

数据库连接池

# coding=utf8
import pymysql
from pymysql import err

"""
获取crm的数据到plm
1、项目相关数据
2、合同相关数据
"""


class MSP_Connect:
    def __init__(self, mysql_info: dict):
        try:
            self.host = mysql_info['host']
            self.user = mysql_info['user']
            self.password = mysql_info['password']
            self.db = mysql_info['db']
            self.con = pymysql.connect(
                host=self.host,
                user=self.user,
                password=self.password,
                db=self.db,
                charset="utf8"
            )
            self.cur = self.con.cursor()
        except KeyError:
            print("mysql_info键名错误，key值分别【host, user, password, db】")
        except err.OperationalError as e:
            if e.args[0] == 1045:
                print("MySql密码错误")
            else:
                print("发生了其它连接错误")

    # 执行sql
    def execute(self, sql, values=None):
        if values == None:
            self.cur.execute(sql)
            return self.cur.fetchall()
        return self.cur.execute(sql, values)

    # 执行多条sql
    def execute_many(self, sql, values):
        return self.cur.executemany(sql, values)

    # 提交事务
    def commit(self):
        self.con.commit()

    # 释放资源
    def close(self):
        self.cur.close()
        self.con.close()

2、检查处理

# coding=utf8
import pymysql
import os
from datetime import datetime, timedelta
from msp_script.database.connect_config import MSP_Connect
from msp_script.database.db_info import mysql_data
from openpyxl import Workbook, load_workbook


def check_msp_table_data(target_db_name, file_name=None):
    """
    :description: 检查数据同步
    :param target_db_name: 目标数据
    :param file_name: 文件名称
    :return:
    """
    print('*******************************************     [' + target_db_name + ']     *******************************************')
    delete_excel_file()
    target_db = MSP_Connect(mysql_data[target_db_name])
    write_result = []
    # 1、获取目标系统中的主数据表
    print('1、获取目标系统中的主数据表')
    md_tables = get_table_index_by_md(target_db)
    print('\t主数据表为' + str(list(table[1] for table in md_tables)))
    info = []
    # 2、对比源数据、目标数据库表结构
    print('2、对比源数据、目标数据库表结构')
    for table in md_tables:

        target_table_name = table[1]  # 获取当前检查的数据表名词
        source_table_info = get_source_table_name(target_table_name)  # 获取原数据表名称以及在哪个系统
        source_table_name = source_table_info[0]
        source_db_name = source_table_info[1]
        if source_db_name in ('CRM', 'PLM', 'MES', 'SCM', 'GDM', 'ESM'):
            source_db = MSP_Connect(mysql_data[source_db_name])
            # 检查表及表数量
            check_rusult = check_table_info(source_db, source_table_name, target_db, target_table_name)
            check_rusult.append(target_db_name)
            check_rusult.append(source_db_name)
            write_result.append(check_rusult)
            if check_rusult[8] != "" and check_rusult[8] != None:
                info.append(check_rusult[8])
    source_db.close()
    target_db.close()
    # 写入报告
    print('3、写入比对结果到Excel')
    write_excel(write_result, file_name)
    return info


def get_source_table_name(target_table_name):
    """
    通过目标表名称获取源数据表名称
    :param target_table_name: 目标数据库名称
    :return: 源数据表名词，系统的简称
    """
    try:
        source_table_name = target_table_name.split('_', 2)[2]
        sys_name = target_table_name.split('_', 2)[1].upper()
        return source_table_name, sys_name
    except Exception as e:
        print("[" + target_table_name + "] 可能是非主数据表")


def get_table_index_by_md(target_db: MSP_Connect):
    """
    :description: 获取目标数据库中的主数据表
    :param target_db: 目标数据库连接对象
    :return:
    """
    get_md_table_sql = """
    SELECT
        TABLE_SCHEMA AS '数据库名',
        TABLE_NAME AS '表名',
        TABLE_ROWS AS '行数',
        DATA_LENGTH AS '数据大小',
        INDEX_LENGTH AS '索引大小',
        CREATE_TIME AS '创建时间',
        table_comment AS '表描述' 
    FROM
        information_schema.TABLES WHERE TABLE_NAME LIKE 'md_%'
    """
    result = target_db.execute(get_md_table_sql, None)
    return result


# 检验msp数据同步
def check_table_info(source_db: MSP_Connect, source_table_name, target_db: MSP_Connect, target_table_name):
    """
    :param source_db: 源数据库连接对象
    :param source_table_name: 源数据表
    :param target_db: 目标数据库连接对象
    :param target_table_name: 目标数据表
    :return:
    """
    # 获取表字段
    source_sql = "SHOW COLUMNS FROM `" + source_table_name + "`"
    target_sql = "SHOW COLUMNS FROM `" + target_table_name + "`"
    source_result = source_db.execute(source_sql, None)
    target_result = target_db.execute(target_sql, None)

    # 获取表数量
    source_count_sql = "select count(*) from `" + source_table_name + "`"
    target_count_sql = "select count(*) from `" + target_table_name + "`"
    source_count = source_db.execute(source_count_sql, None)
    target_count = target_db.execute(target_count_sql, None)

    # 获取最大id
    source_max_id_sql = "select max(id) from `" + source_table_name + "`"
    target_max_id_sql = "select max(id) from `" + target_table_name + "`"
    source_max_id = source_db.execute(source_max_id_sql, None)
    target_max_id = target_db.execute(target_max_id_sql, None)

    if source_result == target_result:
        flag = '相同'
    else:
        flag = '不同'

    msg = "目标系统表【" + target_table_name + "】与源系统表【" + source_table_name + "】表结构【 " + flag + " 】最大值分别为【" + str(target_max_id[0][0]) + ", " + str(source_max_id[0][0]) + "】, 数量分别为【" + str(target_count[0][0]) +", "+ str(source_count[0][0]) +"】"
    # print(msg)
    result = False
    remark = ''
    check_data_result = check_data(source_db, source_table_name, target_db, target_table_name)
    if result == False:
        if source_result != target_result:
            remark = '目标表【' + target_table_name + '】与源系统表【' + source_table_name + '】表结构不一致'
            print('\t' + remark)
        elif source_count != target_count:
            remark = '目标表【' + target_table_name + '】与源系统表【' + source_table_name + '】数据量不一致，数据量差额为：' + str(abs(source_count[0][0] - target_count[0][0])) + '条'
            print('\t' + remark)
        elif source_max_id != target_max_id:
            remark = '目标表【' + target_table_name + '】与源系统表【' + source_table_name + '】表最大值不一致'
            print('\t' + remark)
        elif check_data_result[0] == False and len(check_data_result) == 2:
            remark = '目标表【' + target_table_name + '】与源系统表【' + source_table_name + '】最近的100条数据不相同，ID为：[' + check_data_result[1] + ']'
            print('\t' + remark)
        elif check_data_result[0] == False and len(check_data_result) == 3:
            remark = check_data_result[1]
            print('\t' + remark)
        else:
            result = True

    return [
        flag,
        target_table_name,
        source_table_name,
        target_count[0][0],
        source_count[0][0],
        target_max_id[0][0],
        source_max_id[0][0],
        result,
        remark
    ]


# 将比对结果写入到Excel
def write_excel(write_result: list, file_name):
    """
    :param write_result: 写入的数据
    :param file_name: 文件名称无后缀
    :return:
    """
    parent_dir = os.path.abspath(os.path.join(os.getcwd(), os.pardir))
    file_dir = os.path.join(parent_dir, 'file')
    file_path = os.path.join(file_dir, file_name) + '.xlsx'
    if os.path.exists(file_path):
        work_book = load_workbook(file_path)
        work_sheet = work_book.active
        max_row = work_sheet.max_row
    else:
        work_book = Workbook()
        work_sheet = work_book.active
        headers = ['序号', '目标系统', '目标系统表', '源系统', '源系统表', '表结构是否相同', '目标系统表数量', '原系统表数量', '目标系统最大值', '原系统最大值', '比对结果', '备注']
        for col, headers in enumerate(headers, start=1):
            work_sheet.cell(row=1, column=col, value=headers)
        max_row = 2

    for num, result in enumerate(write_result, start=max_row):
        raw = []
        raw.append(num - 1)     # 序号
        raw.append(result[9])   # 目标系统
        raw.append(result[1])   # 目标系统表
        raw.append(result[10])  # 源系统
        raw.append(result[2])   # 源系统表
        raw.append(result[0])   # 表结构是否相同
        raw.append(result[3])   # 目标系统表数量
        raw.append(result[4])   # 原系统表数量
        raw.append(result[5])   # 目标系统最大值
        raw.append(result[6])   # 原系统最大值
        raw.append(result[7])   # 比对结果
        raw.append(result[8])   # 备注
        work_sheet.append(raw)

    work_book.save(parent_dir + "/file/" + file_name + '.xlsx')


# 删除前一天产生的Excel文件
def delete_excel_file():
    format_time = (datetime.now() - timedelta(days=1)).strftime('%Y%m%d')
    parent_path = os.path.abspath(os.path.join(os.getcwd(), os.pardir)) + '/file'
    for file in os.listdir(parent_path):
        if file.endswith('.xlsx') and "数据比对结果_" + format_time in file:
            os.remove(os.path.join(parent_path, file))


def check_data(source_db: MSP_Connect, source_table_name, target_db: MSP_Connect, target_table_name):
    """
    :param source_db: 源数据库连接
    :param source_table_name: 源表
    :param target_db: 目标数据库连接
    :param target_table_name: 目标表
    :return: 错误信息
    """
    res = []
    try:
        source_sql = "select * from `" + source_table_name + "` order by update_at desc, id desc limit 0, 100"
        target_sql = "select * from `" + target_table_name + "` order by update_at desc, id desc limit 0, 100"
        source_result = source_db.execute(source_sql)
        target_result = target_db.execute(target_sql)
        # return True if source_result == target_result else False
        if target_result == source_result:
            res.append(True)
            return res
        else:
            ids = ''
            for result in target_result:
                if result not in source_result:
                    ids += (str(result[0]) + ',')
            res.append(False)
            res.append(ids[:-1])
            return res
    except pymysql.err.OperationalError as e:
        if e.args[0] == 1054:
            res.append(False)
            res.append("源系统表【" + source_table_name + "】表缺少update_at字段")
            res.append('error')
            print("\t源系统表【" + source_table_name + '】表缺少update_at字段')
            return res


def check_data_custom(source_db, target_db: list, tables: list):
    """
    :param source_db: 原数据库别名
    :param target_db: 目标数据库别名
    :param tables: 需要检查的的表名
    :return:
    """
    result = []
    for db in target_db:
        tar_db = MSP_Connect(mysql_data[db.upper()])
        dict = {}
        list = []
        for table in tables:
            res = check_table_info(MSP_Connect(mysql_data[source_db]), table, tar_db, 'md_' + source_db.lower() + '_' + table)
            if res[7] == False:
                list.append({
                    # 'res': res[7],
                    'mark': res[8],
                    'table': table
                })
                dict[db.upper()] = list
        result.append(dict) if dict else None
        tar_db.close()
    return result


if __name__ == '__main__':
    # db_name = ['PLM', 'CRM', 'GDM', 'MES', 'SCM', 'ESM']
    db_name = ['PLM', 'CRM']
    # format_time = time.strftime('%Y%m%d%H%M%S', time.localtime(time.time()))
    format_time = datetime.now().strftime('%Y%m%d%H%M%S')
    file_name = '数据比对结果_' + format_time
    # print(file_name)
    for name in db_name:
        print()
        print('****************************************************************************************')
        print(name + "系统数据开始检验...")
        check_msp_table_data(name, file_name)

3、api

# coding=utf8
import logging

from flask import Flask, jsonify, request, send_file
from msp_script.public.check_msp_syn import check_msp_table_data, check_data_custom
from datetime import datetime, timedelta
import os

app = Flask(__name__)
app.debug = True


@app.route('/msp/api/check_syn_data', methods=['GET'])
def check_syn_data():
    """
    :param target_db_name 目标数据库名称，必要参数
    :param file_name 报告文件名称，必要参数
    """
    target_db_name = request.args.get('target_db_name')
    file_name = request.args.get('file_name')
    info = {}

    if file_name == None or file_name == '':
        info['code'] = 500
        info['message'] = '[file_name]文件名称不能为空！'
        return jsonify(info)
    else:
        file_name = file_name + '_' + datetime.now().strftime('%Y%m%d%H%M%S')

    if target_db_name == None or target_db_name == '':
        info['code'] = 500
        info['message'] = '[target_db_name]数据库别名不能为空！'
        return jsonify(info)

    if target_db_name.__contains__(','):
        db_name = target_db_name.split(',')
        for db in db_name:
            if db.upper() not in ('CRM', 'PLM', 'MES', 'SCM', 'ESM', 'GDM'):
                info['code'] = 500
                info['message'] = '数据库别名只能是[CRM, PLM, MES, SCM, ESM, GDM]！'
                return jsonify(info)

        for name in db_name:
            data = check_msp_table_data(name.upper(), file_name)
            info[name] = data
        info['code'] = 200
        info['file'] = '/msp/api/download/' + file_name + '.xlsx'
        return jsonify(info)
    elif target_db_name.upper() not in ('CRM', 'PLM', 'MES', 'SCM', 'ESM', 'GDM'):
        info['code'] = 500
        info['message'] = '数据库别名只能是[CRM, PLM, MES, SCM, ESM, GDM]！'
        return jsonify(info)
    else:
        data = check_msp_table_data(target_db_name.upper(), file_name)
        info['code'] = 200
        info['file'] = '/msp/api/download/' + file_name + '.xlsx'
        info[target_db_name] = data
        return jsonify(info)


@app.route('/msp/api/download/<string:file_name>', methods=['GET'])
def download_file(file_name):
    """
    :param file_name: 文件名称
    :param suffix: 文件后缀
    :return:
    """
    info = {}
    if file_name == "" or file_name == None:
        info['code'] = 500
        info['message'] = '文件名称不能为空！'
        return jsonify(info)
    else:
        parent_dir = os.path.abspath(os.path.join(os.getcwd(), os.pardir))
        file_dir = os.path.join(parent_dir, 'file')
        file_path = os.path.join(file_dir, file_name)

        if os.path.exists(file_path) and os.path.isfile(file_path):
            return send_file(file_path, as_attachment=True)
        else:
            info['code'] = 500
            info['message'] = '文件不存在！'
            return jsonify(info)


@app.route('/msp/api/check_table_data', methods=['POST'])
def check_tables():
    """
    :param source_db 源数据库名称，必要参数
    :param target_db 目标数据库名称，必要参数
    :param tables 需要检查的源数据表名称，必要参数
    """
    source_db = request.json.get('source_db').upper()
    target_db = [db.upper() for db in request.json.get('target_db')]
    tables = request.json.get('tables')
    result = {}

    if source_db == '' or source_db == None:
        result['code'] = 500
        result['message'] = 'source_db原数据库名称不能为空'
        return jsonify(result)
    elif source_db not in ['CRM', 'PLM', 'MES', 'SCM', 'GDM', 'ESM']:
            result['code'] = 500
            result['message'] = '原数据库名称[' + source_db + ']不存在'
            return jsonify(result)

    if target_db == '' or target_db == None or len(target_db) == 0:
        result['code'] = 500
        result['message'] = 'target_db目标数据库名称不能为空'
        return jsonify(result)
    else:
        for t_db in target_db:
            if t_db not in ['CRM', 'PLM', 'MES', 'SCM', 'GDM', 'ESM']:
                result['code'] = 500
                result['message'] = '目标数据库名称[' + t_db + ']不存在'
                return jsonify(result)

    if source_db in target_db:
        result['code'] = 500
        result['message'] = 'source_db源数据库' + source_db + '不允许同时存在目标数据库列表' + str(target_db) + '中'
        return jsonify(result)

    if tables == '' or tables == None or len(tables) == 0:
        result['code'] = 500
        result['message'] = 'tables需要验证的表不能为空'
        return jsonify(result)
    else:
        res = check_data_custom(source_db, target_db, tables)
        result['code'] = 200
        result['message'] = '成功'
        result['source_db'] = source_db
        result['target_db'] = target_db
        result['tables'] = tables
        result['no_match_data'] = res
        return jsonify(result)


@app.route("/demo", methods=['GET'])
def demo():
    dict = {
        "a": 1,
        "v": 2
    }
    return jsonify(dict)


if __name__ == '__main__':
    app.run()

四、测试

这里一共提供三个接口

1、全量检查数据

/msp/api/check_syn_data   GET

参数：target_db_name   需要检查的数据库名称

           file_name    执行完输出的报告名称

2、自定义检查系统表数据

/msp/api/check_table_data   POST

参数：source_db  源数据库名称

           target_db   目标数据库名称

           tables   需要检查的表

3、下载检查结果

/msp/api/download/<文件名称>    GET