A much simpler solution: don’t use the git CLI. And in my professional life I don’t know a single person who does. The shortcomings of git have long been abstracted away and as problematic as the CLI is, it’s now just an internal library of the tools we actually use.
Also the git pull criticism is weird. Yeah it exists on paper, and year every so often once in a blue moon there’s a conflict after a pull with rebase, but… this doesn’t even begin to dent the oodles of time saved from just doing Ctrl+T in IntelliJ and be up-to-date with no further input. Why waste 20 minutes 40x-100x a day instead of 45 minutes once every 3-6 months? Especially this case:
My favorite version of this is when the novice has followed someone’s dodgy advice to set pull.rebase = true, then they pull a shared branch that they’re collaborating on, into which their coworker has just merged origin/main. Instant Sorcerer’s Apprentice-scale chaos!
I’m sorry, but are you collaborating or competing on a shared branch? If it is a collaborative effort, maybe just talk about it? And in fact, unless the other person is an utter asshole, they’ll have done so before merging in the new changes from main. That’s not even to mention that in 99,95% of cases or so, that exact scenario is perfectly fine and gets resolved without any issues whats-o-ever and no user input necessary. Bringing us once again to the situation where you save a moderate amount of time multiple times a day by always just pulling.
(edit)
Don’t get me wrong, all of this criticism is of course valid. But it feels like a very arcane case, as no project should be able to produce the issues frequently unless there’s some underlying problem in either the mode of collaboration or the structure of the project in the first place, and the usage of git is long abstracted away and the tools handle virtually any and all edge case, including making merging far smarter than if you were to use the CLI.
I’ve used the git cli exclusively for more than a decade, professionally. I guess it varies wildly by team, but CLIs are the only unambiguous way to communicate instructions, both for humans and computers. That being said, I still don’t mess around with rebase for anything, and I do use a gui diff tool for merge conflict resolution. Practically everything you need to do with git can be done with like 10 commands (I’m actually being generous here, including reset, stash, and tag).
I had to check and make sure I didn’t type the comment above because it sounds exactly like me.
All UIs do things slightly differently, the CLI is always exactly the same… Everywhere. UI for non trivial conflict resolution? Definitely. For everything else, CLI.
And, I’m also reticent to use rebase unless I have to. Gimme that good ole FF :)
UI for non trivial conflict resolution? Definitely.
I dont know about that… Never found they help that much in conflict resolution. They give you some nice buttons for accept their or accept our changes but really I find more often than not those are what breaks code as you often want a mash-up of both sides - which needs to be manually done even in UIs.
Otherwise it is just find the marked sections in the file, and make it look like what you want it to after the merge/rebase. And that is the hardest part - figuring out what it should look like. Which is made easier if you only ever have small commits and merge back to master frequently minimizing the amount your branches drift from each other.
That being said, I still don’t mess around with rebase for anything
Rebasing has a worse reputation than it deserves. It’s something you just get used to - just like how git use is, when you started using it. There are a couple of strategies to make it easier and less anxiety inducing:
- Before starting a rebase of a long branch, create a new branch. That way in case you seriously mess up, you can just delete the rebasing branch and rename the old branch to restore everything (you can usually get away with rebase abort. This is just added safety). Even in case of a successful rebasing, you can just keep the backup branch around, as a faithful record of actual development history.
- Do only one (or max 2) operations in a single rebases. Do this over multiple rebases to get what you want.
After a while, rebasing becomes as simple as commit or merging.
Rebasing and merge conflicts are the top ways that git can turn into a mess. I know that rebasing could (in some circumstances) make merge conflicts less of an issue, but I just mostly think the value of “commit grooming” is overrated. I don’t want to argue about this, if you like doing it, go ahead.
I usually use a gui but I know plenty of colleagues who exclusively use cli. I’ve never understood if it’s an ego thing or what but it’s an incredibly popular way to use git
Which to me is just wild unless you’re doing something you wouldn’t want to use an IDE for - and that’s not actually that many professional things, if I’m being honest. But if you use an IDE, then it’s far easier, faster and importantly doesn’t take you out of your mental flow to just use the built-in git abstraction of that IDE.
I exclusively use CLI, it’s not ego at all, I simply find typing what I want to be quicker than clicking buttons. I’ve written a bunch of aliases to automate my common workflows.
When I need to help a colleague who’s made a mess of something, I can easily give them the command to fix it rather than finding the right options in their GUI of choice and it’s often because of some broken abstraction in the GUI they got into the mess in the first place.
Yeah there are totally commands I use daily, but the visualization involved in looking at the log and available branches (which is a constant use case) is much easier in a gui for me. In fact I’d go as far as saying logs, diffs, and branching in cli are neigh unusable. The buttons I click while in the gui (like fetch/pull/commit) are largely used because at that point (after finding and checking out the right branch, etc) it would be slower to switch back to cli.
I only mentioned ego because I’ve seen multiple junior devs struggling with the command line resist using a gui even when it solves a specific problem they are having quite easily. To each their own though.
I do have a huge ego, but I’ll claim it’s a total coincidence that I use CLI git.
My main reason to use mostly CLI is the better error messages I get when something goes wrong.
My secondary reason is that my preferred GUI tools for git didn’t used to have support for operations I do often such as ‘cherry-pick’ and ‘rebase’. I think that is mostly solved now, but my habits change slow and I’m used to the CLI.
Haha fair. My experience though, is that cherry picking is easily done in gui and I’ve honestly never attempted on cli because it only takes me three clicks in Fork
I disagree, hard.
I disagree with the general conclusion - I think it’s very easy to understand*: each repo has a graph of commits. Each commit includes the diff and metadata (like parent commits). There is a difference between you repo seeing the state of another repo (fetch) and copying commits from another repo into your repo (merge; pull is just a combination of fetch and pull). Tags are pointers to specific commits, branches are pointers to specific commits that get updated when you add a child commit to this commit. That’s a rather small set of very clear concepts for such a complex problem.
I also disagree with a lot of the reasoning. Like “If a commit has the same content but a different parent, it’s NOT the same commit” is not an “alien concept”. When I apply the same change to different parents, I end up with different versions. Which would be kinda bad for a Version Control System.
“This in turn means that you need to be comfortable and fluent in a branching many-worlds cosmology” - yes, if you need to handle different versions, you need to switch between them. That’s the complexity of what you’re doing, not the tool. And I like that Git is not trying to hide things that I need to know to understand what’s happening.
“distinguish between changes and snapshots that have the same intent and content but which are completely non-interchangeable and imply entirely different flows of historical events” How do you even end up in a situation like that? Anyway, sounds like you should be able to merge them without conflicts, if they are in fact completely interchangeable?
“The natural mental model is that names denote global identity.” Why should another repo care, which names I use? How would you even synchronize naming across different repos without adding complexity, e.g. if two devs created a branch “experimental” or “playground”. Why on earth should they be treated as the same branch?
“Git uses the cached remote content, but that’s likely out of date” I actually agree that this can lead to some errors and confusion. But automation exists - you can just fetch every x minutes.
“Branches aren’t quite branches, they’re more like little bookmark go-karts.” A dev describing what basically is just a pointer in this way leads to the suspicion that it might not be Git’s mental model that is alien.
“My favorite version of this is when the novice has followed someone’s dodgy advice to set pull.rebase = true” Maybe don’t do stupid stuff you don’t understand? We know what fetch is, we know what merge is. Pull is basically fetch & merge.
““Pull” presents the illusion that you can just ask Git to make everything okay for you” Just… what? The rest of the sentence doesn’t really fix this error in expectations.
- except the CLI of course, but I can use GUI-tools for most tasks
I also disagree with a lot of the reasoning. Like “If a commit has the same content but a different parent, it’s NOT the same commit” is not an “alien concept”. When I apply the same change to different parents, I end up with different versions. Which would be kinda bad for a Version Control System.
It’s also intuitive, it’s how frames in video compression work, too. And in fact if you have two kids that look virtually identical but are from different families, they are very clearly not the same person. Context matters, most people more-than-intuitively understand that.
“Git uses the cached remote content, but that’s likely out of date” I actually agree that this can lead to some errors and confusion. But automation exists - you can just fetch every x minutes.
Yeah and nevermind that virtually any tool does that for you. So this is a long-solved problem.
Hot take: Git is hard for people who do not know how to read a documentation.
The Git book is very easy to read and only takes a couple of hours to read the most significant chapters. That’s how I learnt it myself.
Git is meant for developers, i.e. people who are supposed to be good at looking up online how stuff works.
Each commit includes the diff and metadata (like parent commits).
Commits don’t store diffs, so you’re wrong from the start here.
Hence why people say “git is hard”
Yeah, you’re right, technically it’s not a “diff”, it’s the changed files.
I don’t think this technical detail has any consequences for the general mental model of Git though - as evidenced by the fact that I have been using Git for years without knowing this detail, and without any problems.
One problem, I think, is that git names are kinda bad. A git branch is just a pointer to a commit, it really doesn’t correspond to what we’d naturally think of as a branch in the context of a physical tree or even in a graph.
That’s a bit problematic for explaining git to programming newbies, because grokking pointers is famously one of the stumbling blocks people have, along with recursion. Front-end web developers who never learned C might not really grok pointers due to never really having to deal with them much.
Some other version control systems like mercurial have both a branch in a more intuitive sense (commits have a branch as a bit of metadata), as well as pointers to commits (mercurial, for example, calls them bookmarks).
As an aside, there’s a few version control systems like darcs where instead of the first-class concept being snapshots, it’s diffs. There’s no separate cherrypick command in darcs, it’s just one way you can use the regular commands.
A git branch is just a pointer to a commit, it really doesn’t correspond to what we’d naturally think of as a branch in the context of a physical tree or even in a graph.
But as the article points out, a commit includes all of its ancestors. Therefore pointing to a commit effectively is equivalent to a branch in the context of a tree.
Some other version control systems like mercurial have both a branch in a more intuitive sense (commits have a branch as a bit of metadata), as well as pointers to commits (mercurial, for example, calls them bookmarks).
I mean, git has bookmarks too, they’re called tags.
What happens after you merge a feature branch into main and delete it? What happens to the branch?
Afterwords, what git commands can you run to see what commits were made as part of the feature branch and which were previously on main?
Mercurial bookmarks correspond to git branches, while mercurial tags correspond to git tags.
My favorite version of this is when the novice has followed someone’s dodgy advice to set pull.rebase = true, then they pull a shared branch that they’re collaborating on, into which their coworker has just merged origin/main. Instant Sorcerer’s Apprentice-scale chaos!
Why are you doing that? Don’t do that.
In my, rather short (5ish years profesionally), career I needed to rebase once. And it was due to some releasing fuck up, a branch had to be released earlier and hence needed to be rebased on another feature branch scheduled for release.
Otherwise, fetch » pull » merge, all day, every day.
I rebase almost daily. I (almost) never merge the main branch into a feature branch, always rebase. I don’t see the point of polluting the history with this commit (assuming I’m the only dev on this branch). I also almost always do an interactive rebase before actually pushing a branch for the first time, in order to clean up commits. I mostly recreate my commits from scratch before pushing, but even then I sometimes forget to include a change in a commit I’ve just made so I then do an interactive rebase to fold fixup commits into the commits they should’ve been in.
I like merging for actually adding commits from a feature branch to main (or release or …)
And anyway… it’s trivial to fix. If you still have the commit ID of the tip of the branch before the pull, go back to that. If not, look it up in the reflog. If that’s too much of a hassle, list the commits you only have locally, stash any changes, reset to the origin/the_branch and cherry-pick your commits again and/or apply the stash.
I really embraced git once I understood that whatever I did locally, it’s most of the time relatively easy to recover from cock-ups. And it’s really difficult to lose work from the moment you’ve added it to a (local) commit or stashed it.
I do understand that git is daunting however, and there is plenty where I think the defaults are bad. Too often I’ve seen merge commits where someone merged a the remote of a branch into the local copy of the same branch, or even this on main. And once this stuff gets pushed it’s neigh impossible to go back.
