This is a post that tells you what the git is and gives explanations of some basic git commands. I also suggests some good materials that you can used for learning git better.
A short history of Git
Linus Torvalds is the creator and, for a long time, principal developer, of the Linux kernel. The management of different versions of Linux, however, was becoming more and more complicated due to the increasing contributors from all over the world. In 2002, the Linux kernel project began using a proprietary distributed version control system called BitKeeper. Unfortunately, In 2005, the relationship between the community that developed the Linux kernel and commercial company (the BitKeeper) broke down. Linus Torvalds had no choice but created a version control system by himself. He gave this system a name of Git. Until now, it has become the most famous version control system.
Git is not the first version control system. The major difference between Git and the other similar systems (CVS, Subversion, Perforce, and so on) is the way Git thinks about data. Most other systems store information as a list of file-based changes. Git thinks about its data more like a stream of snapshots.
Distributed system Most operations in Git need local files and resources to operate. No information is needed from another computer on your network. It makes checking or updating your repository really simple and instant.
Integrity Everything in Git is check-summed. It is impossible to change the contents of your repository without Git knowing it. It is the philosophy of Git system.
Git only adds data Git adds data to the repository when you take any action with Git. It means everything in your repository is undoable.
Basics of Commands
$ git config –global user.name “your name”
$ git config –global user.email firstname.lastname@example.org
git init: initials your directory by git, it will automatically create an invisible directory called
git status: checks the status of your git directory
git add filename: tells git start to tracking this file it also can with the wildcard like
git add *.txt, it will start to tracking all the .txt files in the main directory.
git commit -m "message": let git take a snapshot of your directory, and leave a message for it
git log: to see the messages in this directory.
git remote add origin https://github.com/try-git/try_git.git: add local git directory into the github server. The
originis only the name of your remote, but typically we call it
git remote set-url origin email@example.com:USERNAME/REPOSITORYNAME: you can create a new repository in the server from your side.
git push -u origin master: push our directory to GitHub, where the
originis the name of the remote repository with URL
masteris the name of our local branch. the argument
-uis telling git to remember this combination, so next time you only need to type
git pull origin master: after sometimes your remote repository might be modified by others, you can use this to get the updates
git stash apply: sometimes you might not want to commit the modification on your own directory, you can use
git stashto hide them and after pull, you can re-apply them by
git stash apply
git diff HEAD: let you diff the newest version with the last commit. The
HEADhere is a pointer which points to the last commit, otherwise, you need the
git diff --stage: to see the changes you just staged.
git reset filename: to unstage files
git checkout -- finename: after reset the file, it will be unstaged, but if you want a file back to the status when last time you committed, you can you this commands. the
--means no more options after the
checkout, otherwise, if you happen to have a branch called the same name by filename, it won’t check out the branch.
git branch branch_name: when you want to debug, you can create a copy of you code with this command. after debugging, you can merge the branch back to the main branch and push to the server.
git branch: you can see the info about branches you have
git checkout branchname: you can switch to the branch called
git checkout -b branchname: you can create a new branch and switch to it in one command instead of the combination of steps 14 and 16
git rm filename: git will remove the
filename, after that, you need to
commityour change. if you want to remove a file, you need the argument
-r, but after that you still need to
git rmall the files inside. but a smart way is by adding an argument
-awhen you commit.
git checkout master: after you fixed something in the branch, you need to back to the main branch called master.
git merge branchname: merge the fixed things in
branchnameto the main branch.
git branch -d branchname: after merge you might need to delete the branch. however if you haven’t merged, you really want to delete, you need argument
git push: to update your modification to the remote server.