Git squashing and duplicate commits
Hi guys, I'm a bit confused on how a proper workflow should look for squashing. I have the following issue:
1. I'll go and make commits on a feature / fix branch and squash & merge onto the main branch
2. I'll pull my changes from main into my local feature / fix branch to keep it up to date
3. I'll make some more commits on the local feature / fix branch and then do the same steps in 1 & 2
This duplicates the commits since the commits in my local are not squashed. What should I be doing to keep it in sync with main and not duplicate commits?
3 Replies
generally you only squash and merge at the end of a feature branch, and pull from main as you're working on the branch and your colleagues solve issues on the main branch.
each new feature gets a new branch that's fresh, and features that are completed have their changes merged and their branches discarded
Ohh got ya, I've been creating these long-life feature branches where I'll have multiple squash & merge lifecycles... so I guess it's generally best practice like you were saying to delete the branch after a squash and then create a new one
Thank you!
Also for anyone in the future who happens to stumble upon this, I deleted my duplicate commits by using
git rebase -i HEAD~10 where 10 is the number of commits to look back on in the commit history. Then I deleted the duplicate commits in the terminal by backspacing / deleting each commit message I wanted to delete and then force pushing to main (risky, I know)yeah, generally you work on a feature in its branch until its done and then squash/merge. THe mess doesn't matter if it's just you looking at it anyway.
And feature / bugfix branches are generally just named for the ticket in the bugtracker they're fixing, so that's why you discard the old ones