Ah yes, the vegetables and sports git merging strategies. I personally prefer rebase myself, but that’s okay.
I never thought about it and instantly wanted to reply “wait why can’t you do that‽” but now that I thought about it, what would you want the history to look like in that case? A slightly weird rebase? A single commit which seemingly copy pasted the entire other branch with no relation to it left behind?
Sometimes you really don’t want to look over the commit history of your colleagues. As long as it’s a small feature, a single commit is a pretty good option.
Rather than:
- implemented X
- forgot this
- oh, this was not needed
- now tests actually pass
- oops
- fixed this
- should be ready
That’s basically my commit history for every repo where I need the pipeline to run to see if everything works.
When I do that I always have a Dev branch that I use as the production branch to run the actual calculations.
When I get something working I merge it off, clean up the history a little bit, rebase main onto it and then rebase de onto main.
It’s fine if the changes belong in a single commit. Otherwise, an interactive rebase to craft a clean, quality history before merging is much, much better.
I’m not a fan of changing history in general. Rebase can also he dangerous.
I think ultimately it’s a matter of scale. Sometimes it can be useful to look into the details of the development of a single feature, but in a large project, that rarely happens. I’m not a fan of squashing, but for large projects, it helps to keep your history manageable.
Rebasing is not dangerous. You can always go back if something is not to your liking.
You don’t rebase shared history, you use rebases to craft a clean, quality commit history for your own branches before merging. If everyone does this, then squashing is unnecessary, because garbage commits don’t exist. It is the far superior way of doing things if you actually care about having good commits.
Keeping a quality history rather than squashing also makes many other git tools much better, such as
git blame,git revert,git bisect, and so on.Rebasing is dangerous if you rebase shared history. If you rebase a local branch, you have to be aware of how much of that local branch you may already have shared.
On top of that, if you’ve got a lot of commits you’re rebasing in a merge conflict that can become extremely repetitive.
So ideally, you only rebase single commits that you haven’t pushed yet. As long as you do that: always pull main and rebase on top of that before you push single commits, rebasing is fine. But the more you deviate from that, the riskier it becomes.
You don’t share feature branches. So you always know precisely what is shared history: the commit you branched from.
The workflow is branch from shared history, rebase your branch as many times as necessary during development to craft a quality history, then merge back.
I rebase dozens of times a day and have never had a single issue with it.
If you’re bothered by repetitive merge conflicts (which, in my experience, are quite rare if you’re doing things correctly), that’s what git rerere is for.
Rebasing is for crafting a quality history of your own commits (or getting your branch up to date with the trunk). Merging is for integrating your commits with the shared history.
This is what I’m saying: it depends on your workflow.
And I"m saying that doing with merges (and squashing) what should be done with rebases is bad. You can do it that way, but you shouldn’t, because it makes for worse history and less usable git tools.
I constantly rebase my feature branches regardless of how many commits there are and whether I’ve pushed any of them. So long as no one else has checked out my branch it’s perfectly safe. Personally I find rebase merge conflicts far easier to work with. Traditional merge conflicts are “Here’s someone else’s changes, figure out how to merge them into your feature branch.” Rebase merge conflicts are “The main branch has changed since you made your changes. Re-apply your changes to the new base.” For me/my brain, the latter is so much easier. The only time I ever run into problems is when there are merges in the history I’m rebasing. Which I avoid by never merging into my feature branches, only rebasing.
And if it goes wrong, just
git rebase —abort. Or if you already completed the rebase,git reset —hard origin/YOUR-BRANCH. Or if you majorly fucked up, usegit reflogto find a good commit and reset to that. Zero risk if you know what you’re doing.
Yeah, I’m with you. I mean, git isn’t magic. You “can” squash anything, including a merge commit, by just being at the end result, running
git reset <commit you want to be squashed off of>and then running a manual git add and commit there. That’s basically all a squash is.But what you’ll be left with us a single commit that contains all of the code from the branch you’re squashing and also all the code pulled in from every branch you merged, all written as though it all came from this one commit. And maybe that’s what you want? But it feels like also maybe it’s not?
I think it doesn’t really make sense. Because you can’t “squash” one commit. squash is taking multiple commits and making them one.
When you do a “squash merge” you are really saying “squash all the commits that are on this branch and not the target” then merge.
So you can’t “squash a merge commit” you need at least one additional commit to squash in.
Can you get squash (juice concentrate) in the US? I took a shine to it in the UK, but all I can think that we have here is the frozen concentrate and maybe like some lemonade concentrate in a carton.
You forgot read-only filesystems
If you were gonna replace something you should’ve replaced “vegetables” because squash is a fruit.
The term “vegetable” is a culinary term, and squash is prepared like a vegetable. For another example, tomatoes are fruits but are prepared like vegetables. Squash and tomatoes can be both fruits AND vegetables. This is my position on the “is X a fruit or vegetable?” issue.
I mean, the idea of a “vegetable” isn’t a well defined group of plant parts like fruits are. Vegetables are a mix of seeds, roots, leaves, stems, etc. all of which are quite different. It’s just “parts of a plant that can be cooked as part of a meal”:
“a usually herbaceous plant (such as the cabbage, bean, or potato) grown for an edible part that is usually eaten as part of a meal also : such an edible part” according to the Merriam-Webster dictionary https://www.merriam-webster.com/dictionary/vegetable (similar definitions exist for other dictionaries, some highlight that vegetables are usually used to make non-sweet dishes)
The TLDR is that vegetables are loosely defined as “plant parts that are used to prepare meals, usually non-sweet dishes” and is a culinary term rather than a botanical one like fruits can be. So an item (like tomatoes or squash) can be both a vegetable and a fruit, the former culinary and the latter botanically. They aren’t mutually exclusive.
Depends on which classification system you use. Botaically it is a fruit. But culinarily it is a vegetable.
As long as it merges it’s fiiiine!
But all intersections are squash.
Pickleball?
And I guess there’s cherry-pick, but that’s considered a fruit.
I dunno, I love eating a nice fresh-grown water polo
I think the other intersections should exclude the non-overlapping circle.
+Veggie +Git -Sport: Prune
+Veggie -Git +Sport: Pickle
-Veggie +Git +Sport: I dunno… Does fetch count?
Replicating git history for a file takes 1 merge commit and 3 commits, and this is propably one of the most complex workflows I have encountered:
(might not be correct...) git checkout -b work git mv file file.tmp git commit git checkout -b copy HEAD^ git mv file file2 git commit git checkout work # can be skipped if you merge "work" instead. git merge copy # "work" and "copy" must conflict, stage file.tmp and file2 and commit the result. git mv file.tmp file git commit <git blame is identical for file and file2>I would love to squash this into a single commit, but git doesn’t have a copy operation or detection. :(
Huh. I have never in my 19 year career using git, ever wanted to copy a file and pretend all of the history of that file is also the history of the new file. I mean, I don’t think I’ve ever even wanted to copy a file? Why are you copying a file?
Like, maybe I’m just too familiar with git to see the forest for the trees, but what the heck are you doing over there? 😅
And just in case it’s useful, a tip is that you can use
git blame -Cto have the blame algorithm use a heuristic to try and find a “source” line if it was moved, including from another file, during a commit, and then continue following the history of that line, to try and get the real commit where this was written, not just the last time it was moved around.I can come up with some contrived examples. Maybe someone screwed up the history and they’re trying to repair it such that no one needs to worry about a rebase on their next pull? “compliance”/legal/cya reasons? I also wish to know!
Why are you copying a file?
I’m splitting a several thousand LOC file, which I don’t have previous history in.
Like, maybe I’m just too familiar with git to see the forest for the trees, but what the heck are you doing over there?
Normally copying a file and committing transfers the authorship to you, because the copy just appears from nothing as a brand new file, never known to git. This would prevent browsing the per-line “who changed this last” history past the copy and obfuscate who wrote what and when.
(why the downvote?)
Interesting. Yeah, sounds like what
git blame -Cis for, so I’ve never made copies when splitting files, I’ve just moved lines between files naively. But I guess if one’s tools are limited and doesn’t have the ability to-C, then I guess I could respect the hack that is that solution?I mean, I’m 99% sure git doesn’t store blame or authorship info in the pack files, even as a convenience cache, and just guesses by traversing the patch log with heuristics live when you run blame anyway, so the history mostly doesn’t matter there, but the way you’ve done it does seem to have tricked the heuristics into doing what you want without relying on an option, so that’s neat! It’s an interesting hack, and I like interesting hacks 😛
By the way, if there are down votes, they’re not from me!
I’m not a copyright-lawyer, but I think there are implications on who has authored the code, so preserving this detail can be important. The fancy copy reduced my blame by +90% on the final result.
git blame output can be affected by e.g. ignoring white-space changes.
Right, but what I’m saying is that git doesn’t store authorship information or line-by-line history, no matter how it’s done. Figuring out which line came from where is an algorithm the git blame command does every time you request it, and that algorithm can give different results depending on which options you give the blame command. And so what you’ve found here is a collection of commits that produces a situation the default blame algorithm can follow, without any optional flags, which is neat! Maybe not great for git history, but neat!
You seem to be making this very complex. But it really isn’t. Yes, git doesn’t track renames. So you are working around it by splitting your operation into 2 commits.
- A pure rename.
- A file change.
This way 1 is always considered a rename and 2 is just a regular file change with the same path. You may also consider tweaking the default rename detection threshold with flags like
--find-renamesor options likediff.renameLimit.Would it be nice if Git tracked renames? Probably. But that isn’t how the data model works so it is unlikely to happen soon. But maybe they could add some metadata.
