Python实现SQL Server数据库实现对象同步轻量级

缘由

日常工作中经常遇到类似的问题:把某个服务器上的某些指定的表同步到另外一台服务器。

类似需求用SSIS或者其他ETL工作很容易实现,比如用SSIS的话,就会会存在相当一部分反复的手工操作。

建源的数据库信息,目标的数据库信息,如果是多个表,需要一个一个地拉source和target,然后一个一个地mapping,然后运行实现数据同步。

然后很可能,这个workflow使用也就这么一次,就寿终正寝了,却一样要浪费时间去做这个ETL。

快速数据同步实现

于是在想,可不可能快速实现类似需求,尽最大程度减少重复的手工操作?类似基于命令行的方式,简单快捷,不需要太多的手动操作。

于是就有了本文,基于Python(目的是顺便熟悉一下Python的语法),快速实现SQL Server的数据库之间的数据同步操作,后面又稍微扩展了一下,可以实现不同服务器的数据库之间的表结构,表对应的数据,存储过程,函数,用户自定义类型表(user define table type)的同步

目前支持在两个SQL Server数据源之间:每次同步一张或者多张表/存储过程,也可以同步整个数据库的所有表/存储过程(以及表/存储过程依赖的其他数据库对象)。

支持sqlserver2012以上版本

需要考虑到一些基本的校验问题:在源服务器上,需要同步的对象是否存在,或者输入的对象是否存在于源服务器的数据库里。

在目标服务器上,对于表的同步:

1,表的存在依赖于schema,需要考虑到表的schema是否存在,如果不存在先在target库上创建表对应的schema

2,target表中是否有数据?如果有数据,是否以覆盖的方式执行

对于存储过程的同步:

1,类似于表,需要考虑存储过程的schema是否存在,如果不存在先在target库上创建表对应的schema
2,类似于表,arget数据库中是否已经存在对应的存储过程,是否以覆盖的方式执行
3,存储过程可能依赖于b表,某些函数,用户自定义表变量等等,同步存储过程的时候需要先同步依赖的对象,这一点比较复杂,实现过程中遇到在很多很多的坑

  可能存在对象A依赖于对象B,对象B依赖于对象C……,这里有点递归的意思

  这一点导致了重构大量的代码,一开始都是直来直去的同步,无法实现这个逻辑,切实体会到代码的“单一职责”原则

参数说明

参数说明如下,大的包括四类:

1,源服务器信息 (服务器地址,实例名,数据库名称,用户名,密码),没有用户名密码的情况下,使用windows身份认证模式

2,目标服务器信息(服务器地址,实例名,数据库名称,用户名,密码),没有用户名密码的情况下,使用windows身份认证模式

3,同步的对象类型以及对象

4,同步的对象在目标服务器上存在的情况下,是否强制覆盖

其实在同步数据的时候,也可以把需要同步的行数提取出来做参数,比较简单,这里暂时没有做。

比如需要快速搭建一个测试环境,需要同步所有的表结构和每个表的一部分数据即可。

表以及数据同步

表同步的原理是,创建目标表,遍历源数据的表,生成insert into values(***),(***),(***)格式的sql,然后插入目标数据库,这里大概步骤如下:

1,表依赖于schema,所以同步表之前先同步schema

2,强制覆盖的情况下,会drop掉目标表(如果存在的话),防止目标表与源表结构不一致,非强制覆盖的情况下,如果字段不一致,则抛出异常

3,同步表结构,包括字段,索引,约束等等,但是无法支持外键,刻意去掉了外键,想想为什么?因吹斯汀。

4,需要筛选出来非计算列字段,insert语句只能是非计算列字段(又导致重构了部分代码)

5,转义处理,在拼凑SQL的时候,需要进行转义处理,否则会导致SQL语句错误,目前处理了字符串中的’字符,二进制字段,时间字段的转义处理(最容易发生问题的地方)

6,鉴于insert into values(***),(***),(***)语法上允许的最大值是1000,因此每生成1000条数据,就同步一次

7,自增列的identity_insert 标识打开与关闭处理

使用如下参数,同步源数据库的三张表到目标数据库,因为这里是在本机命名实例下测试,因此实例名和端口号输入

执行同步的效果

说明:

1,如果输入obj_type=”tab” 且-obj=为None的情况下,会同步源数据库中的所有表。
2,这个效率取决于机器性能和网络传输,本机测试的话,每秒中可以提交3到4次,也就是每秒钟可以提交3000~4000行左右的数据。

已知的问题:

1,当表的索引为filter index的时候,无法生成包含where条件的索引创建语句,那个看起来蛋疼的表结构导出语句,暂时没时间改它。
2,暂时不支持其他少用的类型字段,比如地理空间字段什么的。

存储过程对象的同步

存储过程同步的原理是,在源数据库上生成创建存储过程的语句,然后写入目标库,这里大概步骤如下:
1,存储过程依赖于schema,所以同步存储过程之前先同步schema(同表)
2,同步的过程会检查依赖对象,如果依赖其他对象,暂停当前对象同步,先同步依赖对象
3,重复第二步骤,直至完成
4,对于存储过程的同步,如果是强制覆盖的话,强制覆盖仅仅对存储过程自己生效(删除&重建),对依赖对象并不生效,如果依赖对象不存在,就创建,否则不做任何事情

使用如下参数,同步源数据库的两个存储过程到目标数据库,因为这里是在本机命名实例下测试,因此实例名和端口号输入

说明:测试要同步的存储过程之一为[dbo].[sp_test01],它依赖于其他两个对象:dbo.table01和dbo.fn_test01()

create proc [dbo].[sp_test01]
as
begin
 set no count on;
 delete from dbo.table01 where id = 1000
 select dbo.fn_test01()
end

而dbo.fn_test01()的如下,依赖于另外一个对象:dbo.table02

create function [dbo].[fn_test01]
(
)
RETURNS int
AS
BEGIN 
 declare @count int = 0
 select @count = count(1) from dbo.table02
 return @count
END

因此,这个测试的[dbo].[sp_test01]就依赖于其他对象,如果其依赖的对象不存在,同步的时候,仅仅同步这个存储过程本身,是没有意义的

同步某一个对象的依赖对象,使用如下SQL查出来对象依赖信息,因此这里就层层深入,同步依赖对象。
这里就类似于同步A的时候,A依赖于B和C,然后停止同步A,先同步B和C,同步B或者C的时候,可能又依赖于其他对象,然后继续先同步其依赖对象。

效果如下

如果输入obj_type=”sp” 且-obj=为None的情况下,会同步源数据库中的所有存储过程以及其依赖对象

已知的问题:

1,加密的存储过程或者函数是无法实现同步的,因为无法生成创建对象的脚本
1,table type的同步也是一个蛋疼的过程,目前支持,但是支持的并不好,原因是创建table type之前,先删除依赖于table type的对象,否则无法删除与创建。

特别说明

依赖对象的解决,还是比较蛋疼的

如果在默认schema为dbo的对象,在存储过程或者函数中没有写schema(参考如下修改后的sp,不写相关表的schema dbo,dbo.test01==>test01),
使用 sys.dm_sql_referenced_entities这个系统函数是无法找到其依赖的对象的,奇葩的是可以找到schema的类型,却没有返回对象本身。

这一点导致在代码中层层深入,进行了长时间的debug,完全没有想到这个函数是这个鸟样子,因为这里找到依赖对象的类型,却找不到对象本身,次奥!!!
另外一种情况就是动态SQL了,无法使用 sys.dm_sql_referenced_entities这个系统函数找到其依赖的对象。

其他对象的同步

  支持其他数据库对象的同步,比如function,table type等,因为可以在同步其他存储过程对象的时候附带的同步function,table type,这个与表或者存储过程类似,不做过多说明。 

已知问题:

1,201906122030:经测试,目前暂时不支持Sequence对象的同步。

需要改进的地方

1,代码结构优化,更加清晰和条例的结构(一开始用最直接简单粗暴的方式快速实现,后面重构了很多代码,现在自己看起来还有很多不舒服的痕迹)
2,数据同步的效率问题,对于多表的导入导出操作,依赖于单线程,多个大表导出串行的话,可能存在效率上的瓶颈,如何根据表的数据量,尽可能平均地分配多多个线程中,提升效率
3,更加友好清晰的异常提示以及日志记录,生成导出日志信息。
4,异构数据同步,MySQL《==》SQL Server《==》Oracle《==》PGSQL

代码端午节写好了,这几天抽空进行了一些测试以及bug fix,应该还潜在不少未知的bug,工作量比想象中的大的多了去了。

# -*- coding: utf-8 -*-
# !/usr/bin/env python3

__author__ = 'MSSQL123'
__date__ = '2019-06-07 09:36'

import os
import sys
import time
import datetime
import pymssql
from decimal import Decimal

usage = '''
     -----parameter explain-----

     source database parameter
     -s_h      : soure database host                     ----- must require parameter
     -s_i      : soure database instace name                 ----- default instance name MSSQL
     -s_d      : soure database name                     ----- must require parameter
     -s_u      : soure database login                     ----- default windows identifier
     -s_p      : soure database login password                ----- must require when s_u is not null
     -s_P      : soure database instance port                 ----- default port 1433

     target database parameter
     -t_h      : target database host                     ----- must require parameter
     -t_i      : target database instace name                 ----- default instance name MSSQL
     -t_d      : target database name                     ----- must require parameter
     -t_u      : target database login                    ----- default windows identifier
     -t_p      : target database login password                ----- must require when s_u is not null
     -t_P      : target database instance port                ----- default port 1433

     sync object parameter
     -obj_type   : table or sp or function or other databse object       ----- tab or sp or fn or tp
     -obj      : table|sp|function|type name                 ----- whick table or sp sync

     overwirte parameter
     -f       : force overwirte target database object              ----- F or N 


     --help: help document
     Example:
     python DataTransfer.py -s_h=127.0.0.1 -s_P=1433 -s_i="MSSQL" -s_d="DB01" -obj_type="tab" -obj="dbo.t1,dbo.t2" -t_h=127.0.0.1 -t_P=1433 -t_i="MSSQL" -t_d="DB02" -f="Y"          
     python DataTransfer.py -s_h=127.0.0.1 -s_P=1433 -s_i="MSSQL" -s_d="DB01" -obj_type="sp" -obj="dbo.sp1,dbo.sp2" -t_h=127.0.0.1 -t_P=1433 -t_i="MSSQL" -t_d="DB02" -f="Y"
     '''


class SyncDatabaseObject(object):
  # source databse
  s_h = None
  s_i = None
  s_P = None
  s_u = None
  s_p = None
  s_d = None

  # obj type
  s_obj_type = None
  # sync objects
  s_obj = None

  # target database
  t_h = None
  t_i = None
  t_P = None
  t_u = None
  t_p = None
  t_d = None

  f = None

  file_path = None

  def __init__(self, *args, **kwargs):
    for k, v in kwargs.items():
      setattr(self, k, v)

  # connect to sqlserver
  def get_connect(self, _h, _i, _P, _u, _p, _d):
    cursor = False
    try:
      if (_u) and (_p):
        conn = pymssql.connect(host=_h,
                    server=_i,
                    port=_P,
                    user=_u,
                    password=_p,
                    database=_d)
      else:
        conn = pymssql.connect(host=_h,
                    server=_i,
                    port=_P,
                    database=_d)
      if (conn):
        return conn
    except:
      raise
    return conn

  # check connection
  def validated_connect(self, _h, _i, _P, _u, _p, _d):
    if not (self.get_connect(_h, _i, _P, _u, _p, _d)):
      print("connect to " + str(_h) + " failed,please check you parameter")
      exit(0)

  '''
  this is supposed to be a valid object name just like _name,or dbo._name,or [schema]._name or schema.[_name]
  then transfer this kind of valid object name to format object name like [dbo].[_name](give a default dbo schema name when no schema name)
  other format object name consider as unvalid,will be rasie error in process
  format object name
    1,_name       ======> [dbo].[_name]
    2,dbo._name     ======> [dbo].[_name]
    3,[schema]._name   ======> [dbo].[_name]
    3,schema._name    ======> [schema].[_name]
    4,[schema].[_name]  ======> [schema].[_name]
    5,[schema].[_name  ======> rasie error format message
  '''

  @staticmethod
  def format_object_name(name):
    format_name = ""
    if ("." in name):
      schema_name = name[0:name.find(".")]
      object_name = name[name.find(".") + 1:]
      if not ("[" in schema_name):
        schema_name = "[" + schema_name + "]"
      if not ("[" in object_name):
        object_name = "[" + object_name + "]"
      format_name = schema_name + "." + object_name
    else:
      if ("[" in name):
        format_name = "[dbo]." + name
      else:
        format_name = "[dbo]." + "[" + name + "]"
    return format_name

  '''
  check user input object is a valid object
  '''

  def exits_object(self, conn, name):
    conn = conn
    cursor_source = conn.cursor()
    # get object by name from source db
    sql_script = r'''select top 1 1 from
                (
                  select concat(QUOTENAME(schema_name(schema_id)),'.',QUOTENAME(name)) as obj_name from sys.objects 
                  union all
                  select concat(QUOTENAME(schema_name(schema_id)),'.',QUOTENAME(name)) as obj_name from sys.types 
                )t where obj_name = '{0}'
                 '''.format(self.format_object_name(name))
    cursor_source.execute(sql_script)
    result = cursor_source.fetchall()
    if not result:
      return 0
    else:
      return 1
    conn.cursor.close()
    conn.close()

  # table variable sync
  def sync_table_variable(self, tab_name, is_reference):
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()

    if (self.exits_object(conn_source, self.format_object_name(tab_name))) > 0:
      pass
    else:
      print("----------------------- warning message -----------------------")
      print("--------warning: object " + tab_name + " not existing in source database ------------")
      print("----------------------- warning message -----------------------")
      print()
      return

    exists_in_target = 0
    sql_script = r'''select top 1 1
               from sys.table_types tp
               where is_user_defined = 1 
               and concat(QUOTENAME(schema_name(tp.schema_id)),'.',QUOTENAME(tp.name)) = '{0}' ''' \
      .format((self.format_object_name(tab_name)))
    # if the table schema exists in target server,skip
    cursor_target.execute(sql_script)
    exists_in_target = cursor_target.fetchone()

    # weather exists in target server database
    if (self.f == "Y"):
      if (is_reference != "Y"):
        # skiped,table type can not drop when used by sp
        sql_script = r'''
                     if OBJECT_ID('{0}') is not null
                       drop type {0}
                   '''.format(self.format_object_name(tab_name))
        cursor_target.execute(sql_script)
        conn_target.commit()
    else:
      if exists_in_target:
        print("----------------------- warning message -----------------------")
        print("the target table type " + tab_name + " exists ,skiped sync table type from source")
        print("----------------------- warning message -----------------------")
        print()
        return

    sql_script = r'''   
                  DECLARE @SQL NVARCHAR(MAX) = ''

                  SELECT @SQL =
                  'CREATE TYPE ' + '{0}' + 'AS TABLE' + CHAR(13) + '(' + CHAR(13) +
                  STUFF((
                          SELECT CHAR(13) + '  , [' + c.name + '] ' +
                            CASE WHEN c.is_computed = 1
                              THEN 'AS ' + OBJECT_DEFINITION(c.[object_id], c.column_id)
                              ELSE
                                CASE WHEN c.system_type_id != c.user_type_id
                                  THEN '[' + SCHEMA_NAME(tp.[schema_id]) + '].[' + tp.name + ']'
                                  ELSE '[' + UPPER(y.name) + ']'
                                END +
                                CASE
                                  WHEN y.name IN ('varchar', 'char', 'varbinary', 'binary')
                                    THEN '(' + CASE WHEN c.max_length = -1


                                            THEN 'MAX'
                                            ELSE CAST(c.max_length AS VARCHAR(5))
                                          END + ')'
                                  WHEN y.name IN ('nvarchar', 'nchar')
                                    THEN '(' + CASE WHEN c.max_length = -1
                                            THEN 'MAX'
                                            ELSE CAST(c.max_length / 2 AS VARCHAR(5))
                                          END + ')'
                                  WHEN y.name IN ('datetime2', 'time2', 'datetimeoffset')
                                    THEN '(' + CAST(c.scale AS VARCHAR(5)) + ')'
                                  WHEN y.name = 'decimal'
                                    THEN '(' + CAST(c.[precision] AS VARCHAR(5)) + ',' + CAST(c.scale AS VARCHAR(5)) + ')'
                                  ELSE ''
                                END +
                                CASE WHEN c.collation_name IS NOT NULL AND c.system_type_id = c.user_type_id
                                  THEN ' COLLATE ' + c.collation_name
                                  ELSE ''
                                END +
                                CASE WHEN c.is_nullable = 1
                                  THEN ' NULL'
                                  ELSE ' NOT NULL'
                                END +
                                CASE WHEN c.default_object_id != 0
                                  THEN ' CONSTRAINT [' + OBJECT_NAME(c.default_object_id) + ']' +
                                      ' DEFAULT ' + OBJECT_DEFINITION(c.default_object_id)
                                  ELSE ''
                                END
                            END
                          From sys.table_types tp
                          Inner join sys.columns c on c.object_id = tp.type_table_object_id
                          Inner join sys.types y ON y.system_type_id = c.system_type_id 
                          WHERE tp.is_user_defined = 1 and y.name<>'sysname'
                          and concat(QUOTENAME(schema_name(tp.schema_id)),'.',QUOTENAME(tp.name)) = '{0}'
                          ORDER BY c.column_id

                  FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 7, '   ')
                    + ');'

                  select @SQL as script
                '''.format(self.format_object_name(self.format_object_name((tab_name))))
    cursor_target = conn_target.cursor()
    cursor_source.execute(sql_script)
    row = cursor_source.fetchone()
    try:
      if not exists_in_target:
        # execute the script on target server
        cursor_target.execute(str(row[0])) # drop current stored_procudre if exists
        conn_target.commit()
        print("*************table type " + self.format_object_name(tab_name) + " synced *********************")
        print() # give a blank row when finish
    except:
      print("----------------------- error message -----------------------")
      print("-----------table type " + self.format_object_name(tab_name) + " synced error ---------------")
      print("----------------------- error message -----------------------")
      print()
      # raise

    cursor_source.close()
    conn_source.close()
    cursor_target.close()
    conn_target.close()

  # schema sync
  def sync_schema(self):
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()
    arr_schema = []
    # get all table in database when not define table name
    schema_result = cursor_source.execute(r'''
                         select name from sys.schemas where schema_id>4 and schema_id<16384
                        ''')
    for row in cursor_source.fetchall():
      cursor_target.execute(r''' if not exists(select * from sys.schemas where name = '{0}')
                         begin
                           exec('create schema [{0}]')
                         end
                       '''.format(str(row[0])))
      conn_target.commit()

    cursor_source.close()
    conn_source.close()
    cursor_target.close()
    conn_target.close()

  def sync_table_schema_byname(self, tab_name, is_reference):
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()

    if (self.exits_object(conn_source, self.format_object_name(tab_name)) == 0):
      print("----------------------- warning message -----------------------")
      print("---------------warning: object " + tab_name + " not existing in source database ----------------")
      print("----------------------- warning message -----------------------")
      print()
      return

    # if exists a reference table for sp,not sync the table agagin
    if (self.exits_object(conn_target, self.format_object_name(tab_name)) > 0):
      if (self.f != "Y"):
        print("----------------------- warning message -----------------------")
        print("---------------warning: object " + tab_name + " existing in target database ----------------")
        print("----------------------- warning message -----------------------")
        print()
        return

    sql_script = r''' select top 1 1 from sys.tables 
                   where type_desc = 'USER_TABLE' 
                   and concat(QUOTENAME(schema_name(schema_id)),'.',QUOTENAME(name)) = '{0}' 
                 '''.format((self.format_object_name(tab_name)))
    # if the table schema exists in target server,skip
    cursor_target.execute(sql_script)
    exists_in_target = cursor_target.fetchone()


    if exists_in_target:
      if (self.f == "Y"):
        if (is_reference != "Y"):
          cursor_target.execute("drop table {0}".format(tab_name))
      else:
        print("----------------------- warning message -----------------------")
        print("the target table " + tab_name + " exists ,skiped sync table schema from source")
        print("----------------------- warning message -----------------------")
        print()
        return

    sql_script = r'''   DECLARE 
                   @object_name SYSNAME
                  , @object_id INT

                SELECT 
                   @object_name = '[' + s.name + '].[' + o.name + ']'
                  , @object_id = o.[object_id]
                FROM sys.objects o WITH (NOWAIT)
                JOIN sys.schemas s WITH (NOWAIT) ON o.[schema_id] = s.[schema_id]
                WHERE QUOTENAME(s.name) + '.' + QUOTENAME(o.name) = '{0}'
                  AND o.[type] = 'U'
                  AND o.is_ms_shipped = 0

                DECLARE @SQL NVARCHAR(MAX) = ''

                ;WITH index_column AS 
                (
                  SELECT 
                     ic.[object_id]
                    , ic.index_id
                    , ic.is_descending_key
                    , ic.is_included_column
                    , c.name
                  FROM sys.index_columns ic WITH (NOWAIT)
                  JOIN sys.columns c WITH (NOWAIT) ON ic.[object_id] = c.[object_id] AND ic.column_id = c.column_id
                  WHERE ic.[object_id] = @object_id
                ),
                fk_columns AS 
                (
                   SELECT 
                     k.constraint_object_id
                    , cname = c.name
                    , rcname = rc.name
                  FROM sys.foreign_key_columns k WITH (NOWAIT)
                  JOIN sys.columns rc WITH (NOWAIT) ON rc.[object_id] = k.referenced_object_id AND rc.column_id = k.referenced_column_id 
                  JOIN sys.columns c WITH (NOWAIT) ON c.[object_id] = k.parent_object_id AND c.column_id = k.parent_column_id
                  WHERE k.parent_object_id = @object_id
                )
                SELECT @SQL = 'CREATE TABLE ' + @object_name + '' + '(' + '' + STUFF((
                  SELECT '' + ', [' + c.name + '] ' + 
                    CASE WHEN c.is_computed = 1
                      THEN 'AS ' + cc.[definition] 
                      ELSE UPPER(tp.name) + 
                        CASE WHEN tp.name IN ('varchar', 'char', 'varbinary', 'binary', 'text')
                            THEN '(' + CASE WHEN c.max_length = -1 THEN 'MAX' ELSE CAST(c.max_length AS VARCHAR(5)) END + ')'
                           WHEN tp.name IN ('nvarchar', 'nchar')
                            THEN '(' + CASE WHEN c.max_length = -1 THEN 'MAX' ELSE CAST(c.max_length / 2 AS VARCHAR(5)) END + ')'
                           WHEN tp.name IN ('datetime2', 'time2', 'datetimeoffset') 
                            THEN '(' + CAST(c.scale AS VARCHAR(5)) + ')'
                           WHEN tp.name = 'decimal' 
                            THEN '(' + CAST(c.[precision] AS VARCHAR(5)) + ',' + CAST(c.scale AS VARCHAR(5)) + ')'
                          ELSE ''
                        END +
                        CASE WHEN c.collation_name IS NOT NULL THEN ' COLLATE ' + c.collation_name ELSE '' END +
                        CASE WHEN c.is_nullable = 1 THEN ' NULL' ELSE ' NOT NULL' END +
                        CASE WHEN dc.[definition] IS NOT NULL THEN ' DEFAULT' + dc.[definition] ELSE '' END + 
                        CASE WHEN ic.is_identity = 1 THEN ' IDENTITY(' + CAST(ISNULL( /*ic.seed_value*/ 1, '0') AS CHAR(1)) + ',' + CAST(ISNULL(ic.increment_value, '1') AS CHAR(1)) + ')' ELSE '' END 
                    END + ''
                  FROM sys.columns c WITH (NOWAIT)
                  JOIN sys.types tp WITH (NOWAIT) ON c.user_type_id = tp.user_type_id
                  LEFT JOIN sys.computed_columns cc WITH (NOWAIT) ON c.[object_id] = cc.[object_id] AND c.column_id = cc.column_id
                  LEFT JOIN sys.default_constraints dc WITH (NOWAIT) ON c.default_object_id != 0 AND c.[object_id] = dc.parent_object_id AND c.column_id = dc.parent_column_id
                  LEFT JOIN sys.identity_columns ic WITH (NOWAIT) ON c.is_identity = 1 AND c.[object_id] = ic.[object_id] AND c.column_id = ic.column_id
                  WHERE c.[object_id] = @object_id
                  ORDER BY c.column_id
                  FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, '' + ' ')
                  + ISNULL((SELECT '' + ', CONSTRAINT [' + k.name + '] PRIMARY KEY (' + 
                          (SELECT STUFF((
                             SELECT ', [' + c.name + '] ' + CASE WHEN ic.is_descending_key = 1 THEN 'DESC' ELSE 'ASC' END
                             FROM sys.index_columns ic WITH (NOWAIT)
                             JOIN sys.columns c WITH (NOWAIT) ON c.[object_id] = ic.[object_id] AND c.column_id = ic.column_id
                             WHERE ic.is_included_column = 0
                               AND ic.[object_id] = k.parent_object_id 
                               AND ic.index_id = k.unique_index_id   
                             FOR XML PATH(N''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, ''))
                      + ')' + ''
                      FROM sys.key_constraints k WITH (NOWAIT)
                      WHERE k.parent_object_id = @object_id 
                        AND k.[type] = 'PK'), '') + ')' + ''
                  + ISNULL((SELECT (
                    SELECT '' +
                       'ALTER TABLE ' + @object_name + ' WITH' 
                      + CASE WHEN fk.is_not_trusted = 1 
                        THEN ' NOCHECK' 
                        ELSE ' CHECK' 
                       END + 
                       ' ADD CONSTRAINT [' + fk.name + '] FOREIGN KEY(' 
                       + STUFF((
                        SELECT ', [' + k.cname + ']'
                        FROM fk_columns k
                        WHERE k.constraint_object_id = fk.[object_id] and 1=2
                        FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, '')
                        + ')' +
                       ' REFERENCES [' + SCHEMA_NAME(ro.[schema_id]) + '].[' + ro.name + '] ('
                       + STUFF((
                        SELECT ', [' + k.rcname + ']'
                        FROM fk_columns k
                        WHERE k.constraint_object_id = fk.[object_id]
                        FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, '')
                        + ')'
                      + CASE 
                        WHEN fk.delete_referential_action = 1 THEN ' ON DELETE CASCADE' 
                        WHEN fk.delete_referential_action = 2 THEN ' ON DELETE SET NULL'
                        WHEN fk.delete_referential_action = 3 THEN ' ON DELETE SET DEFAULT' 
                        ELSE '' 
                       END
                      + CASE 
                        WHEN fk.update_referential_action = 1 THEN ' ON UPDATE CASCADE'
                        WHEN fk.update_referential_action = 2 THEN ' ON UPDATE SET NULL'
                        WHEN fk.update_referential_action = 3 THEN ' ON UPDATE SET DEFAULT' 
                        ELSE '' 
                       END 
                      + '' + 'ALTER TABLE ' + @object_name + ' CHECK CONSTRAINT [' + fk.name + ']' + ''
                    FROM sys.foreign_keys fk WITH (NOWAIT)
                    JOIN sys.objects ro WITH (NOWAIT) ON ro.[object_id] = fk.referenced_object_id
                    WHERE fk.parent_object_id = @object_id
                    FOR XML PATH(N''), TYPE).value('.', 'NVARCHAR(MAX)')), '')
                  + ISNULL(((SELECT
                     '' + 'CREATE' + CASE WHEN i.is_unique = 1 THEN ' UNIQUE' ELSE '' END 
                        + ' NONCLUSTERED INDEX [' + i.name + '] ON ' + @object_name + ' (' +
                        STUFF((
                        SELECT ', [' + c.name + ']' + CASE WHEN c.is_descending_key = 1 THEN ' DESC' ELSE ' ASC' END
                        FROM index_column c
                        WHERE c.is_included_column = 0
                          AND c.index_id = i.index_id
                        FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, '') + ')' 
                        + ISNULL('' + 'INCLUDE (' + 
                          STUFF((
                          SELECT ', [' + c.name + ']'
                          FROM index_column c
                          WHERE c.is_included_column = 1
                            AND c.index_id = i.index_id
                          FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)'), 1, 2, '') + ')', '') + ''
                    FROM sys.indexes i WITH (NOWAIT)
                    WHERE i.[object_id] = @object_id
                      AND i.is_primary_key = 0
                      AND i.[type] = 2
                    FOR XML PATH(''), TYPE).value('.', 'NVARCHAR(MAX)')
                  ), '')

                select @SQL as script '''.format(self.format_object_name(tab_name))
    cursor_target = conn_target.cursor()
    cursor_source.execute(sql_script)
    row = cursor_source.fetchone()

    if not row[0]:
      return

    try:
      cursor_target.execute(row[0]) # drop current table schema if exists
      conn_target.commit()
      print("*************schema " + self.format_object_name(tab_name) + " synced *************")
      print() # give a blank row when finish
    except:
      print("----------------------- warning message -----------------------")
      print("-----------schema " + self.format_object_name(tab_name) + " synced failed---------------")
      print("----------------------- warning message -----------------------")
      print()

    cursor_source.close()
    conn_source.close()
    cursor_target.close()
    conn_target.close()

  def get_table_column(self, conn, tab_name):
    column_names = ""
    conn = conn
    cursor_source = conn.cursor()
    # get object by name from source db
    sql_script = r'''select name from sys.columns 
             where object_id = object_id('{0}') and is_computed=0 order by object_id
                 '''.format(self.format_object_name(tab_name))
    cursor_source.execute(sql_script)
    result = cursor_source.fetchall()
    for row in result:
      column_names = column_names + row[0] + ","

    return column_names[0:len(column_names) - 1]
    conn.cursor.close()
    conn.close()

  def sync_table_schema(self):
    #default not sync by referenced other object
    is_reference = "N"
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()
    arr_table = []
    if (self.s_obj):
      for tab_name in self.s_obj.split(","):
        if (tab_name) and (self.exits_object(conn_source, tab_name)>0):
          self.sync_table_schema_byname(tab_name, is_reference)
        else:
          print("----------------------- warning message -----------------------")
          print("-----------schema " + self.format_object_name(tab_name) + " not existing in source database---------------")
          print("----------------------- warning message -----------------------")
          print()
    else:
      # sync all tables
      # get all table in database when not define table name
      sql_script = ''' SELECT QUOTENAME(s.name)+'.'+ QUOTENAME(o.name)
               FROM sys.objects o WITH (NOWAIT)
               JOIN sys.schemas s WITH (NOWAIT) ON o.[schema_id] = s.[schema_id]
               WHERE o.[type] = 'U' AND o.is_ms_shipped = 0
             '''
      cursor_source.execute(sql_script)
      for row in cursor_source.fetchall():
        self.sync_table_schema_byname(str(row[0]), is_reference)

  # sync data from soure table to target table
  def sync_table_data(self):
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()
    arr_table = []
    if (self.s_obj):
      arr_table = self.s_obj.split(',')
      for tab_name in arr_table:
        if (self.exits_object(conn_target, self.format_object_name(tab_name)) == 0):
          arr_table.remove(tab_name)
          print("----------------- warning message -----------------------")
          print("----------------- warning: table " + tab_name + " not existing in target database ---------------------")
          print("----------------- warning message -----------------------")
    else:
      # get all table in database when not define table name
      tab_result = cursor_source.execute(r''' SELECT QUOTENAME(s.name)+'.'+ QUOTENAME(o.name)
                          FROM sys.objects o WITH (NOWAIT)
                          JOIN sys.schemas s WITH (NOWAIT) ON o.[schema_id] = s.[schema_id]
                          WHERE o.[type] = 'U'
                            AND o.is_ms_shipped = 0
                      ''')
      for row in cursor_source.fetchall():
        arr_table.append(str(row[0]))

    insert_columns = ""
    insert_columns = self.get_table_column(conn_source, tab_name)

    for tab_name in arr_table:
      if (self.f != "Y"):
        sql_script = "select top 1 {0} from {1} ".format(insert_columns, tab_name)
        # if exists data in target table,break
        cursor_target.execute(sql_script)
        exists = cursor_target.fetchone()
        if exists:
          print("----------------------- warning message -----------------------")
          print("the target table " + tab_name + " exists data,skiped sync table type from source")
          print("----------------------- warning message -----------------------")
          print()
          continue
      else:
        sql_script = "truncate table {0} ".format(tab_name)
        # if exists data in target table,break
        cursor_target.execute(sql_script)
        conn_target.commit()

      insert_columns = ""
      insert_columns = self.get_table_column(conn_source, tab_name)

      insert_prefix = ""

      # weather has identity column
      cursor_source.execute(r'''select 1 from sys.columns 
                  where object_id = OBJECT_ID('{0}') and is_identity =1
                  '''.format(tab_name))
      exists_identity = None
      exists_identity = cursor_source.fetchone()
      if (exists_identity):
        insert_prefix = "set identity_insert {0} on; ".format(tab_name)

      # data source
      insert_sql = ""
      values_sql = ""
      current_row = ""
      counter = 0
      sql_script = r''' select {0} from {1}  '''.format(insert_columns, tab_name)
      cursor_source.execute(sql_script)

      # create insert columns
      '''
      for field in cursor_source.description:
        insert_columns = insert_columns + str(field[0]) + ","
      insert_columns = insert_columns[0:len(insert_columns) - 1]
      '''
      insert_prefix = insert_prefix + "insert into {0} ({1}) values ".format(tab_name, insert_columns)

      for row in cursor_source.fetchall():
        counter = counter + 1
        for key in row:
          if (str(key) == "None"):
            current_row = current_row + r''' null, '''
          else:
            if (type(key) is datetime.datetime):
              current_row = current_row + r''' '{0}', '''.format(str(key)[0:23])
            elif (type(key) is str):
              # 我槽!!!,这里又有一个坑:https://blog.csdn.net/dadaowuque/article/details/81016127
              current_row = current_row + r''' '{0}', '''.format(
                key.replace("'", "''").replace('\u0000', '').replace('\x00', ''))
            elif (type(key) is Decimal):
              d = Decimal(key)
              s = '{0:f}'.format(d)
              current_row = current_row + r''' '{0}', '''.format(s)
            elif (type(key) is bytes):
              # print(hex(int.from_bytes(key, 'big', signed=True) ))
              current_row = current_row + r''' {0}, '''.format(
                hex(int.from_bytes(key, 'big', signed=False)))
            else:
              current_row = current_row + r''' '{0}', '''.format(key)

        current_row = current_row[0:len(current_row) - 2] # remove the the last one char ","
        values_sql = values_sql + "(" + current_row + "),"
        current_row = ""

        # execute the one batch when
        if (counter == 1000):
          insert_sql = insert_prefix + values_sql
          insert_sql = insert_sql[0:len(insert_sql) - 1] # remove the the last one char ","
          if (exists_identity):
            insert_sql = insert_sql + " ;set identity_insert {0} off;".format(tab_name)
          try:
            cursor_target.execute(insert_sql)
          except:
            print(
              "----------------------error " + tab_name + " data synced failed-------------------------")
            raise

          conn_target.commit()
          insert_sql = ""
          values_sql = ""
          current_row = ""
          counter = 0
          print(time.strftime("%Y-%m-%d %H:%M:%S",
                    time.localtime()) + "*************** " + self.format_object_name(
            tab_name) + " " + str(1000) + " rows synced *************")

      if (values_sql):
        insert_sql = insert_prefix + values_sql
        insert_sql = insert_sql[0:len(insert_sql) - 1] # remove the the last one char ","
        if (exists_identity):
          insert_sql = insert_sql + " ; set identity_insert {0} off;".format(tab_name)
        # execute the last batch
        try:
          cursor_target.execute(insert_sql)
        except:
          print("------------------error " + tab_name + " data synced failed------------------------")
          raise

        conn_target.commit()
        insert_sql = ""
        values_sql = ""
        current_row = ""
        print(time.strftime("%Y-%m-%d %H:%M:%S",
                  time.localtime()) + "*************** " + self.format_object_name(
          tab_name) + " " + str(
          counter) + " rows synced *************")
        print(time.strftime("%Y-%m-%d %H:%M:%S",
                  time.localtime()) + "----------------synced " + self.format_object_name(
          tab_name) + " data finished---------------")
        print()
    cursor_source.close()
    conn_source.close()
    cursor_target.close()
    conn_target.close()

  def sync_dependent_object(self, obj_name):
    # 强制覆盖,不需要对依赖对象生效,如果是因为属于依赖对象而被同步的,先检查target中是否存在,如果存在就不继续同步,这里打一个标记来实现
    is_refernece = "Y"
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()
    '''
    find dependent objects
    if exists dependent objects,sync Dependent objects objects in advance
    '''
    sql_check_dependent = r'''       
               SELECT * FROM
              (
                SELECT
                distinct rtrim(lower(s.type)) COLLATE Chinese_PRC_CI_AS as obj_type, 
                QUOTENAME(d.referenced_schema_name)+'.'+QUOTENAME(d.referenced_entity_name) COLLATE Chinese_PRC_CI_AS as obj
                FROM sys.dm_sql_referenced_entities('{0}','OBJECT') as d
                  inner join sys.sysobjects s on s.id = d.referenced_id
                union all
                SELECT
                distinct rtrim(lower(d.referenced_class_desc)) COLLATE Chinese_PRC_CI_AS as obj_type,
                QUOTENAME(d.referenced_schema_name)+'.'+QUOTENAME(d.referenced_entity_name) COLLATE Chinese_PRC_CI_AS as obj
                FROM sys.dm_sql_referenced_entities('{0}','OBJECT') as d
                  inner join sys.types s on s.user_type_id = d.referenced_id
              )t
               '''.format(self.format_object_name(obj_name))

    cursor_source.execute(sql_check_dependent)
    result = cursor_source.fetchall()
    for row in result:
      if row[1]:
        if (row[0] == "u"):
          if (row[1]):
            self.sync_table_schema_byname(row[1], is_refernece)
        elif (row[0] == "fn" or row[0] == "if"):
          if (row[1]):
            self.sync_procudre_by_name("f", row[1], is_refernece)
        elif (row[0] == "type"):
          if (row[1]):
            self.sync_table_variable(row[1], is_refernece)

  def sync_procudre_by_name(self, type, obj_name, is_reference):
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()

    if (self.exits_object(conn_source, self.format_object_name(obj_name)) == 0):
      print("---------------warning message----------------")
      print("---------------warning: object " + obj_name + " not existing in source database ----------------")
      print("---------------warning message----------------")
      print()
      return

    if (self.exits_object(conn_target, self.format_object_name(obj_name)) > 0):
      if (self.f != "Y"):
        print("---------------warning message----------------")
        print("---------------warning: object " + obj_name + " existing in target database ----------------")
        print("---------------warning message----------------")
        print()
        return

    ''' 
    本来想直接生成删除语句的:  
    这里有一个该死的转义,怎么都弄不好,中午先去吃饭吧,
    下午回来想了一下,换一种方式,不要死磕转义问题了
    sql_script = 
          select
          'if object_id('+''''+QUOTENAME(schema_name(uid))+ '' + QUOTENAME(name)+''''+') is not null '
          +' drop proc '+QUOTENAME(schema_name(uid))+ '.' + QUOTENAME(name) ,
          OBJECT_DEFINITION(id)
          from sys.sysobjects where xtype = 'P' and uid not in (16,19)

    '''
    sql_script = r'''
          select
          QUOTENAME(schema_name(uid))+'.'+QUOTENAME(name),
          OBJECT_DEFINITION(id)
          from sys.sysobjects where xtype in ('P','IF','FN') and uid not in (16,19)
         '''
    if (obj_name):
      sql_script = sql_script + " and QUOTENAME(schema_name(uid))+ '.' + QUOTENAME(name) ='{0}' ".format(
        self.format_object_name(obj_name))

    cursor_source.execute(sql_script)
    row = cursor_source.fetchone()
    try:
      if type == "f":
        sql_script = r'''
                if object_id('{0}') is not null 
                  drop function {0}
              '''.format(self.format_object_name(row[0]))
      elif type == "p":
        sql_script = r'''
                if object_id('{0}') is not null 
                  drop proc {0}
              '''.format(self.format_object_name(row[0]))

      cursor_target.execute(sql_script) # drop current stored_procudre if exists
      conn_target.commit()

      # sync dependent object
      if (is_reference != "N"):
        self.sync_dependent_object(self.format_object_name(row[0]))

      # sync object it self
      cursor_target.execute(str(row[1])) # execute create stored_procudre script
      conn_target.commit()
      print("*************sync sp: " + self.format_object_name(row[0]) + " finished *****************")
      print()
    except:
      print("---------------error message----------------")
      print("------------------ sync " + row[0] + "sp error --------------------------")
      print("---------------error message----------------")
      print()

    cursor_source.close()
    conn_source.close()
    cursor_target.close()
    conn_target.close()

  def sync_procudre(self, type):
    is_reference = "N"
    conn_source = self.get_connect(self.s_h, self.s_i, self.s_P, self.s_u, self.s_p, self.s_d)
    conn_target = self.get_connect(self.t_h, self.t_i, self.t_P, self.t_u, self.t_p, self.t_d)
    cursor_source = conn_source.cursor()
    cursor_target = conn_target.cursor()
    if (self.s_obj):
      for proc_name in self.s_obj.split(","):
        self.sync_dependent_object(proc_name)
        self.sync_procudre_by_name(type, proc_name, is_reference)
    # sync all sp and function
    else:
      sql_script = r'''
            select
            QUOTENAME(schema_name(uid))+'.'+QUOTENAME(name),
            OBJECT_DEFINITION(id)
            from sys.sysobjects where xtype = upper('{0}') and uid not in (16,19)
           '''.format(type)
      cursor_source.execute(sql_script)
      for row in cursor_source.fetchall():
        self.sync_dependent_object(row[0])
        self.sync_procudre_by_name(type, row[0], is_reference)


if __name__ == "__main__":

  '''
  sync = SyncDatabaseObject(s_h="127.0.0.1",
               s_i = "sql2017",
               s_P = 49744,
               s_d="DB01",

               t_h="127.0.0.1",
               t_i="sql2017",
               t_P=49744,
               t_d="DB02",

               s_obj_type = "sp",
               s_obj = "dbo.sp_test01",

               f="Y")
  sync.sync_procudre("p")
  '''

  p_s_h = ""
  p_s_i = "MSSQL"
  p_s_P = 1433
  p_s_d = ""

  p_s_u = None
  p_s_p = None

  p_s_obj = ""
  p_type = ""

  p_t_s = ""
  p_t_i = "MSSQL"
  p_t_P = "1433"
  p_t_d = ""

  p_t_u = None
  p_t_p = None

  # force conver target database object,default not force cover target database object
  p_f = "N"

  # sync obj type table|sp
  p_obj_type = None
  # sync whick database object
  p_obj = None

  if len(sys.argv) == 1:
    print(usage)
    sys.exit(1)
  elif sys.argv[1] == '--help':
    print(usage)
    sys.exit()
  elif len(sys.argv) >= 2:
    for i in sys.argv[1:]:
      _argv = i.split('=')

      # source server name
      if _argv[0] == '-s_h':
        p_s_h = _argv[1]
      # source server instance name
      if _argv[0] == '-s_i':
        if (_argv[1]):
          p_s_i = _argv[1]
      # source server instance PORT
      if _argv[0] == '-s_P':
        if (_argv[1]):
          p_s_P = _argv[1]
      # source database name
      if _argv[0] == '-s_d':
        p_s_d = _argv[1]

      if _argv[0] == '-s_u':
        p_s_u = _argv[1]

      if _argv[0] == '-s_p':
        p_s_p = _argv[1]

      if _argv[0] == '-t_h':
        p_t_h = _argv[1]

      if _argv[0] == '-t_i':
        if (_argv[1]):
          p_t_i = _argv[1]

      if _argv[0] == '-t_P':
        if (_argv[1]):
          p_t_P = _argv[1]

      if _argv[0] == '-t_d':
        p_t_d = _argv[1]

      if _argv[0] == '-t_u':
        p_t_u = _argv[1]

      if _argv[0] == '-t_p':
        p_t_p = _argv[1]

      if _argv[0] == '-f':
        if (_argv[1]):
          p_f = _argv[1]

      # object type
      if _argv[0] == '-obj_type':
        if not (_argv[1]):
          print("-obj_type can not be null (-obj=tab|-obj=sp|-obj=fn|-obj=type)")
          exit(0)
        else:
          p_obj_type = _argv[1]

      # object name
      if _argv[0] == '-obj':
        if (_argv[1]):
          p_obj = _argv[1]

    # require para
    if p_s_h.strip() == "":
      print("source server host cannot be null")
      exit(0)
    if p_s_d.strip() == "":
      print("source server host database name cannot be null")
      exit(0)

    if p_t_h.strip() == "":
      print("target server host cannot be null")
      exit(0)

    if p_t_d.strip() == "":
      print("target server host database name cannot be null")
      exit(0)

    sync = SyncDatabaseObject(s_h=p_s_h,
                 s_i=p_s_i,
                 s_P=p_s_P,
                 s_d=p_s_d,
                 s_u=p_s_u,
                 s_p=p_s_p,

                 s_obj=p_obj,

                 t_h=p_t_h,
                 t_i=p_t_i,
                 t_P=p_t_P,
                 t_d=p_t_d,
                 t_u=p_t_u,
                 t_p=p_t_p,

                 f=p_f)

    sync.validated_connect(p_s_h, p_s_i, p_s_P, p_s_d, p_s_u, p_s_p)
    sync.validated_connect(p_t_h, p_t_i, p_t_P, p_t_d, p_t_u, p_t_p)

    if (p_f.upper() == "Y"):
      confirm = input("confirm you want to overwrite the target object? ")
      if confirm.upper() != "Y":
        exit(0)

    print("-------------------------- sync begin ----------------------------------")
    print()

    if (p_obj_type == "tab"):
      # sync schema
      sync.sync_schema()
      # sync table schema
      sync.sync_table_schema()
      # sync data
      sync.sync_table_data()
    elif (p_obj_type == "sp"):
      # sync schema
      sync.sync_schema()
      # sync sp
      sync.sync_procudre("p")
    elif (p_obj_type == "fn"):
      # sync schema
      sync.sync_schema()
      # sync sp
      sync.sync_procudre("fn")
    elif (p_obj_type == "tp"):
      # sync schema
      sync.sync_schema()
      # sync sp
      sync.sync_table_variable()
    else:
      print("-obj_type is not validated")

    print()
    print("-------------------------- sync finish ----------------------------------")

总结

以上所述是小编给大家介绍的Python实现SQL Server数据库实现对象同步轻量级,希望对大家有所帮助,如果大家有任何疑问欢迎给我留言,小编会及时回复大家的!


数据运维技术 » Python实现SQL Server数据库实现对象同步轻量级