revert

I have a commit in a remote+local branch and I want to throw that commit out of the history and put some of them into an own branch. Basically, right now I have: D---E---F---G master And I want: E---G topic / D master That should be both in my local and in the (there is only one, called origin) remote repository. Which is the cleanest way to get that? Also, there are also other people who have cloned that repo and who have checked out the master branch. If I would do such a change in the remote repo, would 'git pull' work for them to get also to the same state? If you've published then you are

when I try to execute command like this (from a command-line or Perl script - it doesn't matter): svn revert "build\myfile@test.meta" SVN skips this file and outputs: Skipped 'build\myfile' I tried doing: svn revert "build\*.meta" But it gives the same result. I can revert these files from the GUI. And I can revert these files by doing (but it reverts more than I want): svn revert --recursive "build" Is there a workaround for this? The @ sign in filenames in Subversion actually has a special meaning - a pegged revision number. To quote the Subversion book : The perceptive reader is probably

Github for Windows features these two commands described as: revert this commit - create a new commit that reverts the changes of this commit rollback this commit - roll back this commit leaving all changes made in this and later commits in your working directory Could you explain the exact meaning of these two commands and how they can be used. Specifically I fail to understand what is the purpose of the second one. It makes no sense to me. Is it possible to revert to a previous commit check it out and if I don't like it, go back to where it was initially? This gui seems to feature a very

If I make changes to the working tree and have not yet committed, and I would like to revert the changes I have made, is there a difference between git reset --hard HEAD and git checkout . ? git checkout -- . will obviously only work on the current directory (and subdirectories thereof), git reset --hard will operate on the complete working tree. git checkout -- . will only update the working tree and leave already staged files as is, whereas git reset --hard will match index and working tree with the HEAD commit. when used with a refspec: reset will set the current branch head to the given

I'm experiencing painfully slow operations with one of our SVN repositories/projects. For example, it's taking 5-10 minutes to revert the changes in one small file (10 KB). Or about 40-60 minutes to check out the project of 100 MB. There are about 30 other projects on the same server, some vastly bigger than this one, and none of them preform like this. One thing to note is that this project is a Magento project. It's not very large in terms of disk space, but I have 23k Files and 11k folders, and I have read SVN preforms badly when there are lots of little files; is this true? And is there

I just committed some changes into one of my feature branches ("feedback_tab") then, checked out "master" and merged them in there. I actually meant to merge them into my "development" branch. Now, master is ahead of 'origin/master' (its remote) by 17 commits - I haven't pushed the merge up (and don't want to, obviously). How can I revert master back to the same state as before the accidental merge? I'm confused between git revert and git reset with this stuff. I looked in my git log and there's no entry for merging feedback_tab into master. I'd have thought it would be the top entry? Bit