如何在团队中同步存储过程文件和代码

Question

I have been studying on this matter for few hours and still studying on it. I'm actually getting lots of info and sources that are not relevant to what I need as I'm not too sure what to search for in google.

In my company, we use SVN to update and commit our source code and allow each developer in the team who work at the same project to get latest code from each other. This practice works fine without any issue so far.

The only problem that concerns me is how to synchronize stored procedures files in the similar method?

In the past, we face issue like forgetting to get the latest stored procedures from others and we don't even know who change what stored procedure, and deployed the files that without other people latest changes to the client. So our only workable manual method is to make sure we send our latest stored procedure physical files to everyone to update and make sure they remember to update, which is not so practical and unsafe, because people do forget sometimes.

We thought about SVN, but not really work for us because we don't stored our stored procedures in the IDE, and it's not a good concept to store in our IDE as stored procedure is not really code file.

Highly appreciate if someone could suggest some good practice to share these type of files across people who work in the same project.

Just for information, the IDE we working at is visual studio and we code in c# .Net.

Answer 1

My team is similar to yours. Actually we manage all stored procedures and table schema SQL files in a folder which is also saved in TFS/SVN. Every time when a developer want to change a store procedure, he should get the latest source stored procedure file and edit it in SQL Server Management Studio to edit it. Then he need to commit the stored procedure files with his code changes. It is also the same when he wants to delete or add stored procedure files.

Then after each deployment, when a developer finds there is db error blocking his ongoing works, he will find the stored procedure files in the source and run the most recently modified SQL files. The blocking issue should be solved.

Answer 2

We're storing all procedures as .sql files in the same folder with the sources that uses it. The procedures are installed automatically to correct database with each build and a version / build number if added to the end of the procedure so that different versions of the same application can be used with the same database (for testing environments).

Having your procedures in version control helps a lot when you have to track down what has been changed or who changed it and of course to get the correct versions installed at the correct time.

In addition to procedures we also store other objects like views, functions, triggers, constraints etc. into version control. You could store tables too, but for that we have a special handling because tables can't be re-created, it needs alter table clauses to be executed.

We're not using SVN, but I would assume the same basic idea would work with it too.