如何在使用flywaydb时对版本控制存储过程进行版本控制？

本文介绍了如何在使用flywaydb时对版本控制存储过程进行版本控制？的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！ 问题描述

我目前有一个使用Microsoft SQL Server后端的Java Web应用程序。数据库迁移目前由sqlcmd手动完成，它使用：r命令引用我们的存储过程和视图的脚本文件。

I currently have a Java web app which uses a Microsoft SQL Server back end. Database migrations are currently being done manually by sqlcmd which makes use of the :r command to reference script files of our stored procedures and views.

这样每个存储过程都有他的自己的文件proc_someprocedure.sql迁移是由升级脚本6.1 upgrade.sql完成的，该脚本将proc文件引用为drop并在其当前版本中重新创建。

This way each stored procedure has his own file "proc_someprocedure.sql" The migration is done by an upgrade script "6.1 upgrade.sql" which refrences the proc file to drop and recreate it in it's current version.

我们希望迁移到flyway并让应用程序在部署时迁移它自己的数据库。但是，我找不到任何简单的方法来为每个存储过程维护单个源代码文件。如果存储过程发生更改，我需要在版本控制中对V_6_1__change.sql文件和proc_someprocedure.sql文件进行更改。

We're looking to migrate to flyway and have the application migrate it's own database upon deployment. However, I can't find any easy way to maintain a single source code file for each stored procedure. In the event that a stored procedure is changed, I need to make the change in both a V_6_1__change.sql file and the proc_someprocedure.sql file in version control.

我'我试图避免复制/粘贴的sql代码，因为它在过去困扰我们。其他人如何处理这种情况？

I'm trying to avoid copy/pasted sql code as it's haunted us in the past.. How are others handling this situation?

推荐答案

从3.0开始，没有一流的支持。但是有三种方法可以达到这个目的：

As of 3.0 there is no first class support for this. There are however 3 ways to achieve this:

• 使用自定义MigrationResolver挂钩到你使用的任何源控件并检测那些方式的改变
• 使用通过afterMigrate重新创建所有存储过程的FlywayCallback
• 在单独的模式中使用单独的Flyway实例并将cleanOnValidationError设置为true。每次校验和更改时，都会清除架构并重新运行该Flyway实例管理的所有迁移