Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I tried rerere once and it felt too much like magic to me, i.e. too complicated in a way that I didn't trust. Experience with conflicts has led me to eschew magic merge tools and rely on the simplest strategies: 1. minimize conflicts; 2. bite the bullet and deal with them manually. (Edit: my question about rerere is: how identical is "identical"? How can I be sure that it will redo what I did before in exactly the way I would do it now? Doesn't it have to understand my intent to achieve that?)

The diligent-rebasing-along-the-way workflow I proposed is all about #1. You still have to deal with intermediate conflicts this way too, but at least they're minimized. If something you commit to master conflicts with my B49, I have to fix B1..B49 but at least I can write B50..B100 in a way that takes your work into account.



There is nothing really magic about rerere. During a merge, it records each conflict. When you commit the merge, it records your resolution of the conflict. If that _identical_ conflict occurs again, it re-applies the same resolution. You can choose whether it marks the the file as resolved or not, which allows you to easily review what was done before committing the merge.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: