JavaEar 专注于收集分享传播有价值的技术资料

How to revert Git repository to a previous commit?

How do I revert from my current state to a snapshot made on a certain commit?

If I do git log, then I get the following output:

$ git log
commit a867b4af366350be2e7c21b8de9cc6504678a61b`
Author: Me <me@me.com>
Date:   Thu Nov 4 18:59:41 2010 -0400

blah blah blah...

commit 25eee4caef46ae64aa08e8ab3f988bc917ee1ce4
Author: Me <me@me.com>
Date:   Thu Nov 4 05:13:39 2010 -0400

more blah blah blah...

commit 0766c053c0ea2035e90f504928f8df3c9363b8bd
Author: Me <me@me.com>
Date:   Thu Nov 4 00:55:06 2010 -0400

And yet more blah blah...

commit 0d1d7fc32e5a947fbd92ee598033d85bfc445a50
Author: Me <me@me.com>
Date:   Wed Nov 3 23:56:08 2010 -0400

Yep, more blah blah.

How do revert to the commit from November 3, i.e. commit 0d1d7fc?

30个回答

    最佳答案
  1. This depends a lot on what you mean by "revert".

    Temporarily switch to a different commit

    If you want to temporarily go back to it, fool around, then come back to where you are, all you have to do is check out the desired commit:

    # This will detach your HEAD, that is, leave you with no branch checked out:
    git checkout 0d1d7fc32
    

    Or if you want to make commits while you're there, go ahead and make a new branch while you're at it:

    git checkout -b old-state 0d1d7fc32
    

    To go back to where you were, just check out the branch you were on again. (If you've made changes, as always when switching branches, you'll have to deal with them as appropriate. You could reset to throw them away; you could stash, checkout, stash pop to take them with you; you could commit them to a branch there if you want a branch there.)

    Hard delete unpublished commits

    If, on the other hand, you want to really get rid of everything you've done since then, there are two possibilities. One, if you haven't published any of these commits, simply reset:

    # This will destroy any local modifications.
    # Don't do it if you have uncommitted work you want to keep.
    git reset --hard 0d1d7fc32
    
    # Alternatively, if there's work to keep:
    git stash
    git reset --hard 0d1d7fc32
    git stash pop
    # This saves the modifications, then reapplies that patch after resetting.
    # You could get merge conflicts, if you've modified things which were
    # changed since the commit you reset to.
    

    If you mess up, you've already thrown away your local changes, but you can at least get back to where you were before by resetting again.

    Undo published commits with new commits

    On the other hand, if you've published the work, you probably don't want to reset the branch, since that's effectively rewriting history. In that case, you could indeed revert the commits. With Git, revert has a very specific meaning: create a commit with the reverse patch to cancel it out. This way you don't rewrite any history.

    # This will create three separate revert commits:
    git revert a867b4af 25eee4ca 0766c053
    
    # It also takes ranges. This will revert the last two commits:
    git revert HEAD~2..HEAD
    
    #Similarly, you can revert a range of commits using commit hashes:
    git revert a867b4af..0766c053 
    
    # Reverting a merge commit
    git revert -m 1 <merge_commit_sha>
    
    # To get just one, you could use `rebase -i` to squash them afterwards
    # Or, you could do it manually (be sure to do this at top level of the repo)
    # get your index and work tree into the desired state, without changing HEAD:
    git checkout 0d1d7fc32 .
    
    # Then commit. Be sure and write a good message describing what you just did
    git commit
    

    The git-revert manpage actually covers a lot of this in its description. Another useful link is this git-scm.com section discussing git-revert.

    If you decide you didn't want to revert after all, you can revert the revert (as described here) or reset back to before the revert (see the previous section).

    You may also find this answer helpful in this case:
    How to move HEAD back to a previous location? (Detached head)

  2. 参考答案2
  3. Reverting Working Copy to Most Recent Commit

    To revert to a previous commit, ignoring any changes:

    git reset --hard HEAD
    

    where HEAD is the last commit in your current branch

    Reverting The Working Copy to an Older Commit

    To revert to a commit that's older than the most recent commit:

    # Resets index to former commit; replace '56e05fced' with your commit code
    git reset 56e05fced 
    
    # Moves pointer back to previous HEAD
    git reset --soft HEAD@{1}
    
    git commit -m "Revert to 56e05fced"
    
    # Updates working copy to reflect the new commit
    git reset --hard
    

    Credits go to a similar Stack Overflow question, Revert to a commit by a SHA hash in Git?.

  4. 参考答案3
  5. Lots of complicated and dangerous answers here, but it's actually easy:

    git revert --no-commit 0766c053..HEAD
    git commit
    

    This will revert everything from the HEAD back to the commit hash, meaning it will recreate that commit state in the working tree as if every commit since had been walked back. You can then commit the current tree, and it will create a brand new commit essentially equivalent to the commit you "reverted" to.

    (The --no-commit flag lets git revert all the commits at once- otherwise you'll be prompted for a message for each commit in the range, littering your history with unnecessary new commits.)

    This is a safe and easy way to rollback to a previous state. No history is destroyed, so it can be used for commits that have already been made public.

  6. 参考答案4
  7. The best option for me and probably others is the Git reset option:

    git reset --hard <commidId> && git clean -f
    

    This has been the best option for me! It is simple, fast and effective!


    Note : As mentioned in comments don't do this if you're sharing your branch with other people who have copies of the old commits

    Also from the comments, if you wanted a less 'ballzy' method you could use

    git clean -i

  8. 参考答案5
  9. I have tried a lot of ways to revert local changes in Git, and it seems that this works the best if you just want to revert to the latest commit state.

    git add . && git checkout master -f
    

    Short description:

    • It will NOT create any commits as git revert does.
    • It will NOT detach your HEAD like git checkout <commithashcode> does.
    • It WILL override all your local changes and DELETE all added files since the last commit in the branch.
    • It works only with branches names, so you can revert only to latest commit in the branch this way.

    I found a much more convenient and simple way to achieve the results above:

    git add . && git reset --hard HEAD
    

    where HEAD points to the latest commit at you current branch.

    It is the same code code as boulder_ruby suggested, but I have added git add . before git reset --hard HEAD to erase all new files created since the last commit since this is what most people expect I believe when reverting to the latest commit.

  10. 参考答案6
  11. If you want to "uncommit", erase the last commit message, and put the modified files back in staging, you would use the command:

    git reset --soft HEAD~1
    
    • --soft indicates that the uncommitted files should be retained as working files opposed to --hard which would discard them.
    • HEAD~1 is the last commit. If you want to rollback 3 commits you could use HEAD~3. If you want to rollback to a specific revision number, you could also do that using its SHA hash.

    This is an extremely useful command in situations where you committed the wrong thing and you want to undo that last commit.

    Source: http://nakkaya.com/2009/09/24/git-delete-last-commit/

  12. 参考答案7
  13. Before answering lets add some background, explaining what is this HEAD.

    First of all what is HEAD?

    HEAD is simply a reference to the current commit (latest) on the current branch.
    There can only be a single HEAD at any given time. (excluding git worktree)

    The content of HEAD is stored inside .git/HEAD and it contains the 40 bytes SHA-1 of the current commit.


    detached HEAD

    If you are not on the latest commit - meaning that HEAD is pointing to a prior commit in history its called detached HEAD.

    enter image description here

    On the command line it will look like this- SHA-1 instead of the branch name since the HEAD is not pointing to the the tip of the current branch

    enter image description here


    A few options on how to recover from a detached HEAD:


    git checkout

    git checkout <commit_id>
    git checkout -b <new branch> <commit_id>
    git checkout HEAD~X // x is the number of commits t go back
    

    This will checkout new branch pointing to the desired commit.
    This command will checkout to a given commit.
    At this point you can create a branch and start to work from this point on.

    # Checkout a given commit. 
    # Doing so will result in a `detached HEAD` which mean that the `HEAD`
    # is not pointing to the latest so you will need to checkout branch
    # in order to be able to update the code.
    git checkout <commit-id>
    
    # create a new branch forked to the given commit
    git checkout -b <branch name>
    

    git reflog

    You can always use the reflog as well.
    git reflog will display any change which updated the HEAD and checking out the desired reflog entry will set the HEAD back to this commit.

    Every time the HEAD is modified there will be a new entry in the reflog

    git reflog
    git checkout HEAD@{...}
    

    This will get you back to your desired commit

    enter image description here


    git reset HEAD --hard <commit_id>

    "Move" your head back to the desired commit.

    # This will destroy any local modifications.
    # Don't do it if you have uncommitted work you want to keep.
    git reset --hard 0d1d7fc32
    
    # Alternatively, if there's work to keep:
    git stash
    git reset --hard 0d1d7fc32
    git stash pop
    # This saves the modifications, then reapplies that patch after resetting.
    # You could get merge conflicts, if you've modified things which were
    # changed since the commit you reset to.
    
    • Note: (Since Git 2.7)
      you can also use the git rebase --no-autostash as well.

    This schema illustrate which command does what.
    As you can see there reset && checkout modify the HEAD.

    enter image description here

  14. 参考答案8
  15. You can do this by the following two commands:

    git reset --hard [previous Commit SHA id here]
    git push origin [branch Name] -f
    

    It will remove your previous Git commit.

    If you want to keep your changes, you can also use:

    git reset --soft [previous Commit SHA id here]
    

    Then it will save your changes.

  16. 参考答案9
  17. Say you have the following commits in a text file named ~/commits-to-revert.txt (I used git log --pretty=oneline to get them)

    fe60adeba6436ed8f4cc5f5c0b20df7ac9d93219
    0c27ecfdab3cbb08a448659aa61764ad80533a1b
    f85007f35a23a7f29fa14b3b47c8b2ef3803d542
    e9ec660ba9c06317888f901e3a5ad833d4963283
    6a80768d44ccc2107ce410c4e28c7147b382cd8f
    9cf6c21f5adfac3732c76c1194bbe6a330fb83e3
    fff2336bf8690fbfb2b4890a96549dc58bf548a5
    1f7082f3f52880cb49bc37c40531fc478823b4f5
    e9b317d36a9d1db88bd34831a32de327244df36a
    f6ea0e7208cf22fba17952fb162a01afb26de806
    137a681351037a2204f088a8d8f0db6e1f9179ca
    

    Create a Bash shell script to revert each of them:

    #!/bin/bash
    cd /path/to/working/copy
    for i in `cat ~/commits-to-revert.txt`
    do
        git revert $i --no-commit
    done
    

    This reverts everything back to the previous state, including file and directory creations, and deletions, commit it to your branch and you retain the history, but you have it reverted back to the same file structure. Why Git doesn't have a git revert --to <hash> is beyond me.

  18. 参考答案10
  19. Extra Alternatives to Jefromi's Solutions

    Jefromi's solutions are definitely the best ones, and you should definitely use them. However, for the sake of completeness, I also wanted to show these other alternative solutions that can also be used to revert a commit (in the sense that you create a new commit that undoes changes in previous commit, just like what git revert does).

    To be clear, these alternatives are not the best way to revert commits, Jefromi's solutions are, but I just want to point out that you can also use these other methods to achieve the same thing as git revert.

    Alternative 1: Hard and Soft Resets

    This is a very slightly modified version of Charles Bailey's solution to Revert to a commit by a SHA hash in Git?:

    # Reset the index to the desired commit
    git reset --hard <commit>
    
    # Move the branch pointer back to the previous HEAD
    git reset --soft HEAD@{1}
    
    # Commit the changes
    git commit -m "Revert to <commit>"
    

    This basically works by using the fact that soft resets will leave the state of the previous commit staged in the index/staging-area, which you can then commit.

    Alternative 2: Delete the Current Tree and Replace with the New One

    This solution comes from svick's solution to Checkout old commit and make it a new commit:

    git rm -r .
    git checkout <commit> .
    git commit
    

    Similarly to alternative #1, this reproduces the state of <commit> in the current working copy. It is necessary to do git rm first because git checkout won't remove files that have been added since <commit>.

  20. 参考答案11
  21. Here is a much simpler way to go back to a previous commit (and have it in an uncommited state, to do with it whatever you like):

    git reset HEAD~1
    

    So, no need for commit ids and so on :)

  22. 参考答案12
  23. Assuming you're talking about master and on that respective branch (that said, this could be any working branch you're concerned with):

    # Revert local master branch to November 3rd commit ID
    git reset --hard 0d1d7fc32e5a947fbd92ee598033d85bfc445a50
    
    # Revert remote master branch to November 3rd commit ID
    git push -f origin 0d1d7fc32e5a947fbd92ee598033d85bfc445a50:master
    

    I found the answer from in blog post Delete remote Git repo to specific commit.

  24. 参考答案13
  25. OK, going back to previous commit in git is quite easy...

    Revert back without keeping the changes:

    git reset --hard <commit>
    

    Revert back with keeping the changes:

    git reset --soft <commit>
    

    Explain: using git reset, you can reset to a specific state, it's common using it with a commit hash as you see above.

    But as you see the difference is using the two flags --soft and --hard, by default git reset using --soft flag, but it's a good practice always using the flag, I explain each flags:


    --soft

    The default flag as explained, not need to provide it, does not change the working tree, but add all changes files ready to commit, so you go back to the commit status which changes to files get unstaged.


    --hard

    Be careful with this flag, it resets the working tree and all changes to tracked files and all will be gone!


    I also created the image below that may happens in a real life working with git:

    git reset to a commit

  26. 参考答案14
  27. There is a command (not a part of core Git, but it is in the git-extras package) specifically for reverting and staging old commits:

    git back
    

    Per the man page, it can also be used as such:

    # Remove the latest three commits
    git back 3
    
  28. 参考答案15
  29. After all the changes, when you push all these commands, you might have to use:

    git push -f ...
    

    And not only git push.

  30. 参考答案16
  31. Nothing here worked for me apart from this exact combination:

    git reset --hard <commit_hash>
    git push origin <branch_name> --force
    

    Key here is forcing the push, no extra commits/commit messages etc.

  32. 参考答案17
  33. Select your required commit, and check it by

    git show HEAD
    git show HEAD~1
    git show HEAD~2 
    

    till you get the required commit. To make the HEAD point to that, do

    git reset --hard HEAD~1
    

    or git reset --hard HEAD~2 or whatever.

  34. 参考答案18
  35. To keep the changes from the previous commit to HEAD and move to the previous commit, do:

    git reset <SHA>
    

    If changes are not required from the previous commit to HEAD and just discard all changes, do:

    git reset --hard <SHA>
    
  36. 参考答案19
  37. To completely clean a coder's directory up from some accidental changes, we used:

    git add -A .
    git reset --hard HEAD
    

    Just git reset --hard HEAD will get rid of modifications, but it won't get rid of "new" files. In their case they'd accidentally dragged an important folder somewhere random, and all those files were being treated as new by Git, so a reset --hard didn't fix it. By running the git add -A . beforehand, it explicitly tracked them all with git, to be wiped out by the reset.

  38. 参考答案20
  39. This is one more way to directly reset to a recent commit

    git stash
    git stash clear
    

    It directly clears all the changes that you have been making since the last commit.

    PS: It has a little problem; it also deletes all you recently stored stash changes. Which I guess in most cases should not matter.

  40. 参考答案21
  41. 参考答案22
  42. Revert to Most Recent commit and ignoring all local changes

    git reset --hard HEAD
    
  43. 参考答案23
  44. Revert is the command to rollback the commits.

    git revert <commit1> <commit2> 
    

    Sample:

    git revert 2h3h23233

    It is capable of taking range from the HEAD like below. Here 1 says "revert last commit."

    git revert HEAD~1..HEAD

    and then do git push

  45. 参考答案24
  46. You can complete all these initial steps yourself and push back to git repo.

    1. Pull the latest version of your repository from Bitbucket using the git pull --all command.

    2. Run the git log command with -n 4 from your terminal. The number after the -n determines the number of commits in the log starting from the most recent commit in your local history.

      $ git log -n 4

    3. Reset the head of your repository's history using the git reset --hard HEAD~N where N is the number of commits you want to take the head back. In the following example the head would be set back one commit, to the last commit in the repository history:

    4. Push the change to git repo using git push --force to force push the change.

    If you want git repository to a previous commit

    git pull --all
    git reset --hard HEAD~1
    git push --force
    
  47. 参考答案25
  48. If the situation is an urgent one, and you just want to do what the questioner asked in a quick and dirty way, assuming your project is under directory "my project":

    1. Copy the whole directory and call it something else, like "my project - copy"

    2. Do:

      git reset --hard [first-4-letters&numbers-of-commit's-SHA]
      

    You then have two versions on your system... you can examine or copy or modify files of interest, or whatever, from the previous commit. You can completely discard the files under "my project - copy", if you have decided the new work was going nowhere...

    The obvious thing if you want to carry on with the state of the project without actually discarding the work since this retrieved commit is to rename your directory again: Delete the project containing the retrieved commit (or give it a temporary name) and rename your "my project - copy" directory back to "my project". Then probably do another commit fairly soon.

    Git is a brilliant creation but you can't just "pick it up on the fly": also people who try to explain it far too often assume prior knowledge of other VCS [Version Control Systems] and delve far too deep far too soon, and commit other crimes, like using interchangeable terms for "checking out" - in ways which sometimes appear almost calculated to confuse a beginner.

    To save yourself much stress you have to pretty much have to read a book on Git - I'd recommend "Version Control with Git". And if you can trust me (or rather my scars) when I say "have to", it follows that you might as well do it NOW. Much of the complexity of Git comes from branching and then remerging. But from your question there's no reason why people should be blinding you with science.

    Especially if, for example, this is a desperate situation and you're a newbie with Git!

    PS: One other thought: It is (now) actually quite simple to keep the Git repository ("repo") in a directory other than the one with the working files. This would mean you would not have to copy the entire Git repository using the above quick & dirty solution. See the answer by Fryer using --separate-git-dir here. Be warned, though: If you have a "separate-directory" repository which you don't copy, and you do a hard reset, all versions subsequent to the reset commit will be lost forever, unless you have, as you absolutely should, regularly backed up your repository, preferably to the cloud (e.g. Google Drive) among other places.

  49. 参考答案26
  50. As your commits are pushed remotely, you need to remove them. Let me assume your branch is develop and it is pushed over origin.

    You first need to remove develop from origin:

    git push origin :develop (note the colon)
    

    Then you need to get develop to the status you want, let me assume the commit hash is EFGHIJK:

    git reset --hard EFGHIJK
    

    Lastly, push develop again:

    git push origin develop
    
  51. 参考答案27
  52. If you want to correct some error in the last commit a good alternative would be using git commit --amend command. If the last commit is not pointed by any reference, this will do the trick, as it create a commit with the same parent as the last commit. If there is no reference to the last commit, it will simply be discarded and this commit will be the last commit. This is a good way of correcting commits without reverting commits. However it has its own limitations.

  53. 参考答案28
  54. Try resetting to the desired commit -

    git reset <COMMIT_ID>

    (to check COMMIT_ID use git log)

    This will reset all changed files to un-added state.

    Now you can checkout all un-added files by

    git checkout .

    Check git log to verify your changes.

    UPDATE

    If you have one and only commit in your repo, try

    git update-ref -d HEAD

  55. 参考答案29
  56. Caution! This command can cause losing commit history, if user put the wrong commit mistakenly. Always have en extra backup of your git some where else just in case if you do mistakes, than you are a bit safer. :)

    I have had similar issue and wanted to revert back to earlier Commit. In my case I was not intetessered to keep newer commit hence I used Hard.

    This is how I did it:

    git reset --hard CommitId && git clean -f
    

    This will revert on local repository, here after using git push -f will update remote repository.

    git push -f
    
  57. 参考答案30
  58. Yet another simplest solution; you have to change branch to do this, but afterwards you can just run:

    git branch -f <<branchname>> 0d1d7fc32e5a947fbd92ee598033d85bfc445a50