pull

I removed a file in my local Git repo. I want the file back by updating my Git repo. I run the following unsuccessfully git pull It says upToDate, but I did not get the newest file. The public repo at github does not use SVN, so the problem cannot be SVN. How can you update your local Git repo , such that you get your removed File back to your computer? August has the probably best way for your problem. You could also revert to the last version you pulled: git reset --hard If it hasn't been committed yet: git checkout path/to/file 来源： https://stackoverflow.com/questions/842915/unable-to-git

I created local hg repository for new project: hg init ; hg addremove , then I created empty upstream repo, and pushed first version there: hg push https://remoterepo . Now, I want to set that https://remoterepo as default upstream, so I can just do hg push/pull without giving it. I was surprised Google did not give me a straight answer with "set mercurial upstream", and I didn't find a direct answer here at SO either. So, for the benefit of all the people using SO as a howto, what's the right way to do this? You can do that by adding the upstream URL to /project/.hg/hgrc like this: [paths]

In git, how do you exempt certain files when pulling from an upstream origin (i.e. the original project)? I have a project that I'm working on that was originally forked from a repository that is very active. I've added the original as a remote named "upstream" so that it's possible to run: git pull upstream and update the project to the most recent commit. The problem: there are some files (e.g. my Gruntfile.js ) that I do not want to update alongside the original project. Whenever I pulled these commits, I would get merge conflicts because I've changed the files in my own version. I do want

有时候，需要把git仓库（包括当前状态和历史记录）共享给车间工友，而你和工友之间又不能通过git服务器进行交互时，网上邻居共享便成为了最原始也最有效的手段。那具体怎么操作呢？ 打包整个目录 即把工作区和.git一起打成一个压缩包共享。解压之后即形成完全一样的工作环境。 毋庸置疑，这是最差的方式，不解释。 打包.git目录 即只把.git目录打成压缩包共享。这样解压之后，去.git的同级目录执行git checkout，即可生成整个工作区。 这里利用了git最重要的特性：.git目录包含了所有信息。 已经很好了，但并不完美。 直接共享.git目录 这是正解。它与打包.git目录的区别在于：被共享的.git目录可以直接被用作远程仓库。 也就是说，如果你的机器名是Machine，被共享的.git目录的共享名为Rep，则 //Machine/Rep 可以直接作为仓库URL来使用。 (1) 克隆 git clone //Machine/Rep <dir> (2) 在现有仓库中添加为远程仓库 git add <name> //Machine/Rep git fetch <name> git checkout master git merge <name>/master git push <name> master:mymaster 可以被添加为远程仓库，是直接共享.git目录的最大优势

I am having great success with git push --mirror to make backup copies to a bare repo. But after search on SO and elsewhere, I cannot find a way to clone the thing locally with all branches. I do not want to use git clone since I don't want my local repo to know about the bare repo. If I use git pull it only brings down the HEAD branch. Guessing: git pull /data/Dropbox/backup/that_stuff.git * gets me nowhere, of course. How do I get the entire repo with all branches back again? I realize I could probably just copy the bare repo to my .git directory, but that seems like a bad idea. VonC Try git