- Elementary commands

- Elementary commands#

clone view

Running git clone <url> performs the following operations:

Copy the remote repository
Create tracking branches for all remote branches.
Check out the active branch from the remote repository.

Note

git clone is run typically only once

fetch view

git fetch without further arguments runs:

Updates all remote tracking branches (“fetches the changes”).

Note

Running git fetch is always a save operation

merge view

In short, git merge is used to combine the changes from one branch into another branch.

The git merge procedure:

Starting from the following setup:

       C'-----D' main
      /
 A---B---C---D---E origin/main

A git merge will perform:

Create a new commit on top of the current branch that contains all the changes from the commits from origin/main

              main 
              ∨
       C'-----D'---F
      /           /
 A---B---C---D---E origin/main

Set the head of your local branch to this new commit

       C'-----D'---F main
      /           /
 A---B---C---D---E origin/main

Note

If your current branch is main, a git merge will be equivalent to git merge origin/main.

rebase view

In short, git rebase is used to apply the changes from one branch onto another branch. It differs from git merge in that it rewrites the commit history of the current branch. Meaning that the commits from the current branch are applied on top of the other branch.

The git rebase procedure:

Starting from the following setup:

       C'-----D' main
      /
 A---B---C---D---E origin/main

A git rebase will perform:

Apply each commit from your local branch onto the remote branch:

                  C''---D'' main
                 /         
A---B---C---D---E origin/main

Note

You might have to resovle conflicts for each commit from the local branch (i.e. C' and D').
With git rebase -i you can also rewrite the history of a single branch, see the Rewriting History article on the official website.

add view

Running git add . will:

Register all changes in the current content of your workspace (mind the .!) with respect to the current branch. This is also called “staging”.

Note

git add <path-to-file> is used to stage changes in a specific file.
git rm is not the counterpart of git add as it removes a file from the tracking and even from the workspace (unless you use --cached).
“unstageing” changes in a file can be done with git restore --stage <path-to-file> (see restore for details)

git add . -p

The -p option allows to partially add files. Running this command will allow to you to individual (de-)select

clone view

git commit -m 'my message' will:

Record all staged changes to the repository
Update the reference of the current branch to this last change

Note

Commit messages contribute drastically to the accessibility of a repository
If you struggle to create meaningful commit messages, consider reviewing what changes you pack into a single commit

push view

Running a git push will:

Update the corresponding branch on the remote repository with the changes from the current local branch

¹If a new commit can be reached by following the history from another commit, then git can ‘fast-forward’ any reference to the new commit.

Note

You might encounter most hickups when doing a git push. However, this is only because the remote refuses any updates that cannot be fast-forwarded ¹ ¹If a new commit can be reached by following the history from another commit, then git can ‘fast-forward’ any reference to the new commit. This operation highlights that your local branch is not up-to-date.