I’m using worktree to generate two different versions of a complex feature and then compare them to see which implement the most robust functionalities. Both plans generated by the two different models involve making changes to a database. I wonder if I actually implement the two versions of code changes, will there be code or database related conflicts? Is the worktree mode suitable or dangerous for changes involving database adjustments? If I continue to use worktree, how can I prevent database related conflicts?
If I understood you correctly, you need separate test databases.
1 Like