Git is a distributed, fast and secure source code manager (SCM). The official Git website is gitscm.org. The Git software has initially been created by Linus Torvalds for the Linux project. The current maintainer is now Junio Hamano. The development of Git itself is performed on the kernel.org plateform.
Contrary to CVS and SVN managers, and similarly to Mercurial, Git works in a distributed way. This means that developpers do not depend on a central repository to commit their changes. Each Git repository contains the full history of the project. Each Git object (changeset, file, directory) is a leaf of a tree with multiple branches. Developping a project with Git is based on a model in which one branch corresponds to one feature. Many revisions of the feature may be archived before the branch gets merged with the main trunk. For a detailed explanation of branch development, we recommend reading this page.
One especially interesting technique is cherry-picking, which is essentially taking part of a branch and merging it to another one. This page describes how to use this technique with Git.
Generally, a project developper must be able to:
- clone or create the repository;
- make changes;
- commit changes;
- push changes toward the original repository.
Since Git is distributed, there may not be an authoritative repository.
Summary of commands
|git config||Configures git|
|git init||Creates a new repository|
|git clone||Clones an existing repository|
|git add||Adds a file or directory to a repository|
|git rm||Deletes a file or directory from the repository|
|git commit||Commits changes to the repository|
|git push||Pushes commited changes to a different repository|
|git pull||Pulls changes from a different repository and merges them with your own repository|
|git fetch||Fetches changes from a different repository without merging them to yours|
|git merge||Merges changes to the repository|
|git blame||Gives the origin of each change|
|git log||Displays changes history|
|git diff||Compares two versions|
|git status||Displays status of the current files|
|git show||Displays various git objects|
|git cat-file||Displays the content, type or size of objects|
|git branch||Manages development branches|
|git tag||Manages version tags|
|git remote||Manages remote repositories|
|git checkout||Checks out a branch or a path|
|git reset||Changes the head of a branch|
|git format-patch||Creates a patch|
|git am||Applies a patch|
|git send-email||Sends a patch by email|
|git bisect||Used to diagnose problems|
|git gc||Collects garbage objects|
|git rebase||Rebases history of the repository|
|git grep||Searches for content|
Creating or cloning a repository
The first step usually is to create your own repository, or to clone an existing one.
To create a repository:
[name@server $] git init my-project
To clone a repository:
[name@server $] git clone git://github.com/git/git.git
Commiting a change
When the repository is ready, you change directory and edit the file
[name@server $] cd my-project [name@server $] nano file.txt
When work is done, you add the file:
[name@server $] git add file.txt
and commit the change
[name@server $] git commit
It is then possible to push changes to the origin repository with:
[name@server $] git push origin master
In the above command, origin is the remote repository and master is the current branch that will be pushed.