Published 20 August 2019
Since projects only contain SQL scripts and Visual Studio configuration files, they can be committed to version control without any extra configuration.
When committing to version control you should ignore the following:
[project name].dbmdl file
[project name].jfm file
[project name].sqlproj.user file
It is especially important to ignore the user file as this contains user-specific settings which can cause unexpected behavior for other users.
Working with Git
Sample .gitignore file for SQL Change Automation
A .gitignore file containing rules to ignore the files listed above can be downloaded here.
Place your .gitignore file in the same folder as your SQL Change Automation project file (.sqlproj file).
Git Branching Strategies for SQL Change Automation
Git encourages workflows that branch and merge often. We encourage Git users to embrace branching and merging for their database code for the same reasons many people do this for application code. Read more about this in "The Managers Guide to Git Training for Database Administrators."
SQL Change Automation works well with many popular branching strategies, including Git Flow, Release Flow, and other patterns. If you are not sure which pattern to begin using, we often advise:
- If a Git branching strategy is already working well for a team you work closely with, consider adopting this strategy and asking a member of that team to serve as a coach for your group
- If this is not the case, the Azure DevOps Release Flow model works well for many teams who are first starting out