Commit a file to a Different Branch Without Checkout

Question

Is it possible to commit a file in a git branch without checking out that branch? If so how?

Essentially I want to be able to save a file in my github pages branch without switching branches all the time. Any thoughts?

I needed to do this atomically without changing the current directory, so multiple commands won't work, even if they can be used as a one liner. Otherwise, you can end up with race conditions.

Update: It's impossible to do what I want (see comments below for use case). I ended up programmatically cloning my current directory to a tmp directory, then checking out my branch in that tmp directory (doesn't affect my working directory) and committing my files to the tmp directory clone. When I'm done, I push back to my working directory and delete the tmp directory. Sucks, but it's the only way to commit files to another branch without changing the current working branch of the working directory. If anyone has a better solution, please feel free to add it below. I'll accept yours if it's better than 'it cannot be done'.

Update update: For full details of why I asked to be able to do this 11 years ago and why I wanted it to be atomic see this repo https://github.com/schneems/git_test. Back in the day we didn't have CI services. Developers at the company I worked for were expected to run tests locally, but not everyone did it. The idea behind that project was to track test run history. This would allow you to see which commit tests (or if they weren't run at all).

CI now handles this functionality but since I asked an "impossible question", I'm always curious if there is an answer.

Answer 1

It's not possible.

The changes you commit are related to the current working copy. If you want to commit to another branch it means that you could commit changes from your working copy, but base them from another copy state.

This is not a natural way of versioning your work, and this is why you need to make different steps (stash changes, checkout the branch, pop stash and commit) to accomplish it.

As for your specific use case, a simple way is to keep two copies of your work, one checked out at master branch, and the other at pages branch.

In the pages working copy, add the master copy as a remote repo.

• You commit pages on master
• Pull from master on the pages copy
• push to GitHub
• reset the master branch at its previous state.

Answer 2

As several others have said, it is literally possible, but impractical.

However, as of Git 2.5 (with some important fixes in 2.6 and minor ones since then), there is a practical method for doing this using git worktree add.

Let's say, for instance, that you want to work on branches main and doc "at the same time", or branches develop and test "at the same time", but the two branches in question deliberately contain different things. (For instance, the doc branch has documentation that exists outside or alongside the code, or the test branch has tests that will be run against the code, but not distributed, or which are expected to have failures for which tests are deliberately skipped on the develop side, or whatever.)

Instead of just:

git clone -b develop <source> theclone

followed by working in theclone with constant switching back and forth between the two branches, you would:

git clone -b develop <source> theclone

but then:

cd theclone
git worktree add ../ct test  # check out branch test in ../ct

or just:

git worktree add ../test     # check out branch test in ../test

Now you can run your tests in ../test while developing in theclone. You can merge and/or rebase changes from one branch to the other in the usual way: the underlying repository is already shared, so no git push or git fetch is required. You simply have both branches checked out, into two separate work-trees, named theclone and test from the top level.

Answer 3

It can be done by reimplementing git commit.

This can be done with various call to git hash-object

But this is hard to achieve.

Please read progit chapter 9 for more details and a full example of how to simulate a commit.

Answer 4

So long as you don't have anything in your current index that differs from your HEAD that you want to keep you can so something like this. (If you do want to keep your index you could temporarily export the GIT_INDEX_FILE environment variable to point at a temporary file for the duration of these commands.)

# Reset index and HEAD to otherbranch
git reset otherbranch

# make commit for otherbranch
git add file-to-commit
git commit "edited file"

# force recreate otherbranch to here
git branch -f otherbranch

# Go back to where we were before
# (two commits ago, the reset and the commit)
git reset HEAD@{2}

We've never actually checked out otherbranch and our working tree files haven't been touched.

Answer 5

Most answers here say that it can't be done, or use stashing. It can be done without stashing, but it's not easy and it's not something you want to incorporate in your daily routine, I think.

For this example, I started with a main branch and a foo branch.

o--o--o (main)
    \
     o (foo)

I want to commit a file to main while foo is checked out. The commands I write here are all done without checking out main.

git add examplefile
git write-tree  
    # this returns a hash: 2ceb75e10358ba27aaf2291e4e302d759fbedd55
git commit-tree 2ceb75e10358ba27aaf2291e4e302d759fbedd55 -p main -m "commit message"  
    # this makes a commit with commit the main branch points to as parent
    # it too returns a hash: 56823299cdc8b6c1ab30095f6b2c2143f3bb2122
git branch -f main 56823299cdc8b6c1ab30095f6b2c2143f3bb2122
    # this points the main branch to the new commit

So what did we do? Essentially, we did all the steps a normal add-commit routine does, with some tweaks.

First, we add a file to the staging area just as normal. Next, we run git write-tree. That writes the staging area to a tree and returns the hash of that tree. A commit is not much more than a pointer to a tree, a pointer to its parent(s) and a message.

Then, we actually make the commit by giving it the three things it needs: the tree, the parent (main) and the commit message. Note on my computer this command doesn't open an editor for the commit message, so I need to give it in the command line. Your experience may vary.

The situation is then as follows:

        o (new commit)
       / 
o--o--o (main)
    \
     o (foo)

We made the new commit at the right spot. However, it's dangling, meaning there is no branch (or tag) pointing to it. We now have to point main to the new commit.

The last command does exactly that. It takes the hash of the newly created commit and points main towards it.

Done!

As I said at the start: this is not something you want to do often. The normal add-commit strategy is well thought out. This strategy might mess up something you're not aware of. It doesn't check for conflicts, it just overwrites files without mentioning it.

Finally, a one-liner to do everything in one fell swoop:

git branch -f main $(git commit-tree $(git write-tree) -p main -m "commit message")

The hash output from write-tree is used immediately as an input for commit-tree, and its output is used immediately in moving the branch.

Answer 6

I made a little tool that does exactly this: https://github.com/qwertzguy/git-quick

It let's you edit specific files from another branch without checking out the other branch completely (just the files you want to edit) and commit them. All this without ever affecting your working copy or staging area.

Behind the scenes it uses a combination of git worktree and sparse checkout. The source is fairly small, so you can read through.

Answer 7

I cannot agree it is not possible. By mixing git stash push, git stash pop, git checkout, git checkout, git add and git commit this is possible.

How I understand problem:

You are on branch master and you made some modifications to file patched.txt and you would like to commit this file to other branch.

What you would like to do is:

• save all changes in this repo by doing git stash
• checkout file.txt from stashed stack
• add file patched (and only this file) to new branch
• get back to state of repo before modyfing file.txt

This can be achieved by executing following commands:

destBranch=patch
thisBranch=master
FileToPutToOtherBranch="file1.txt file2.txt 'file with space in name.txt'"
message="patched files $FileToPutToOtherBranch"
                                                                                  #assumption: we are on master to which modifications to file.txt should not belong
git stash &&\                                                                     #at this point we have clean repository to $thisBranch
git checkout -b $destBranch &&\           
git checkout stash@{0} -- $FileToPutToOtherBranch &&                              #if there are many files, repeat this step                                         #create branch if does not exist (param -b)
git add $FileToPutToOtherBranch &&\                                               # at this point this is equal to git add . --update
git commit -m "$message" &&\
git checkout $thisBranch &&\
git stash apply &&\                                                               # or pop if want to loose backup
git checkout $thisBranch -- $FileToPutToOtherBranch                               # get unpatched files from previous branch

The reason why I am using "&&" and the end is if somebody will copy&paste this snippet into terminal, even if one error occurs, next commands will be executed, which is not good. \ is for informing shell that command is continued in next line.

To proove this works I provide testing environment for this snippet

mkdir -p /tmp/gitcommitToAnyBranch && cd /tmp/gitcommitToAnyBranch &&\
git init 
echo 'this is master file' > file1.txt
echo 'this is file we do not want to have modified in patch branch because it does not     patches any feature' > docs.txt
git add file1.txt && git commit -m "initial commit" 
echo 'now this file gets patched' > file1.txt
git status

Now, if you run my script with parameters

destBranch=patch
thisBranch=`git rev-parse --abbrev-ref HEAD`
FileToPutToOtherBranch="file1.txt"
message="patched file $FileToPutToOtherBranch"

You will have file1.txt modified only in patch branch, for more see gitk --all

Answer 8

While there is currently no single command to do this, there are at least two other options.

1. You could use the github api to create the commit. This post details creating a commit in a github repo.

2. Create github pages as a submodule.

3. Use a series of plumbing commands to create the commit.
   The git book has a description of plumbing commands used to create a commit

note: the command is now mktree not mk-tree

Answer 9

If you accidently modified things in the wrong branch, here's some simple steps :

1. Commit these changes ;
2. Merge them into the right branch ;
3. Checkout the branch your were on at first and reset it to the commit before ;
4. Cleanup your modifications with "git checkout -- .".

Everything should be fine after this. You can also merge, reset and cleanup your modifications selectively.

Answer 10

No you cannot, however what you are doing by hard copying files is implemented by git stash using commits. The stash works as a stack of changes that are saved "somewhere outside the branches spaces", hence can be used to move modifications between branches.

• Make your changes on <file>
• git add <file>
• git stash
• git checkout <branch you want to commit>
• git stash pop
• git commit -m "<commit message>"

you can probably scrip these simple commands down if you do this constantly.

Answer 11

This is how I do it:

if I've acidentally committed, roll back one commit:

git reset --soft 'HEAD^'

add the files you want to add

git add .

create a new temporary branch:

git checkout -b oops-temp

commit your changes:

git commit -m "message about this commit"

checkout the branch you meant to check out:

git checkout realbranch

merge the old branch:

git merge oops-temp

delete the old branch:

git branch -D oops-temp