- Git is a distributed version-control system for tracking changes in source code during software development.
- Git Book -- IMPORTANT
- Learn Git from Atlassian
- Github is a US-based global company that provides hosting for software development version control using Git.
Installation file: 01-installation.md
A Git repository is a directory containing all of a project's files, and it is managed by Git, where you can see a .git directory in all Git repositories, this directory allows Git to keep track of each file's revision history, and other Git data like branches and merges.
To initialize Git in an existing project, go to that project's directory and run the command
git initThis will create the .git directory allowing Git to keep track of your files' changes.
To a Git managed project, run the following command
git init <directory-name>This will create a new directory with the specified name with .git directory inside of it allowing Git to keep track of your files' changes.
Since Git is a distributed version controlling system, repositories may exist on some remote host like GitHub, Gitlab, or Bitbucket.
For example cloning this repository
git clone https://github.com/jordanopensource/git-trainingThis will create a new directory named git-training with the content of the clone repository's files.
Forking; and its name gives it away, it's making a fork, like road forks, where a main road is forked side road(s), and eventually it looks like an eating fork, the same concept applied to Git repositories, where there's a main branch (road) on a repository, and you create a fork branch (side road) of it.
All Git hosts support forking a public repository, where you copy the repository into your account or organization.
Each file in a Git repository goes through the four Git file stages:
- Untracked: The file is newly created or moved from another directory to a Git directory, and Git doesn't keep any track of it.
- Unmodified: The file is tracked by Git, i.e has been committed before, and has no new changes.
- Modified: An uncommitted file that has been modified but not committed yet.
- Staged: A modified file that has been added to be committed.
Here's a figure showing the relations between the four stages.
After creating a file (untracked) or editing an existing one (unmodified), you can add the file to the staging area, by running this command.
git add <file-name>Or using this command to add all of the files in the current directory.
git add .Now to move the file into the unmodified stage, commit the file with a commit message
Commit messages should describe what happened to a file, in an imperative way e.g "Create "
git commit -m "Create <file-name>"Now the file that you have just committed is in the unmodified stage, where if you made a new modification to it it will get moved to the modified stage, and the cycle continues...
If you modified a file, after a commit, and then realized that the changes weren't needed, you can rollback to the latest tracked version by Git, by running the following command.
git checkout -- <file-name>When you make changes on a local repository, you need to synchronize those changes with the remote repository, and the way to do that is pushing your changes
git pushAnd vice versa, when you want to update your local repository from the remote repository, you pull the changes to your local repository
git pullYou can view the full commit history of a repository's branch by running the command
git logAnd you can check the repository's current status (untracked, modified and staged files), by running the command
git statusTo remove a file you accidentally staged use the following command:
git restore --staged <file_name>This will move the file or directory from staged to untracked or modified state.
To undo a commit
git reset --soft HEAD~This will undo a commit by 1 you can specify how much commits you want to remove by adding a number after the ~ like so HEAD~2.
To modify a commit message
git commit --amend