Blog of Funky

Appearance

git not GitHub ​

git 101 ​

Initial Account Configuration ​

bash
git config --global --add user.name "MyName"
git config --global --add user.email "abc@gg.cc"

Initialising a Git repository ​

bash
git init

Adding files to staging area ​

bash
git add filename
git add --all
git add -A # Same as --all
git add .

Restore the changes to status quo ​

bash
git restore .
git restore --staged . # To restore staged changes
git rm --cached filename # To remove from staging area altogether

To remove untracked files / directories ​

bash
git clean -fd

-f - force -d - directories too

See current status, staged/unstaged files etc ​

bash
git status
git status --short

Commit staged changes ​

bash
git commit -m "Better write good messages"
git commit -a -m "Merge without adding/staging"
git commit # Open the default editor set by $EDITOR

View commit history ​

bash
git log
git log --oneline
git log --format=full --decorate --graph

Last Commit Info ​

bash
git show

Differences between commits/branch ​

bash
git diff first-branch second-branch
git diff # Show un-staged changes

Getting help ​

bash
git commit --help
git help --all

Removing a git repository ​

bash
cd intoTheDirectoryFolder
rm .git

Branch ​

Creating new branch ​

bash
git branch new-branch

Creating a orphan(empty) branch ​

bash
git checkout --orphan new-branch

Moving to new branch/commit(time-travelling) ​

bash
git checkout new-branch
git checkout -b new-branch # New branch is created and choosed if -b is used
git checkout commithash # time travelling

Changing branches, changes files accordingly. So, ls will have different result in different branches.

Merging branches ​

bash
git checkout master
git merge new # merge new into master

Deleting branches ​

bash
git branch -d new

List all branches ​

bash
git branch # List local branches
git branch -a # List all branches
git branch -r # For remote branches

Switch to other branch ​

bash
git branch new 
git switch new # Alternative

Create and switch to other branch ​

bash
git switch -c new

git with GitHub ​

Connecting git to github ​

bash
git remote -v # List all remote refs
git remote add origin https://github.com/username/reponame.git # Add new remote ref
git remote set-url origin https://github.com/username/reponame.git # For editing
git remote get-url origin # For getting
git remote rename origin upstream # Renaming
git remote remove origin # Deleting

Setting master to default branch and pushing changes to it ​

bash
git push --set-upstream origin master # Sending from master branch to origin
git push # For subsequent runs
git push origin branch-name

Track changes from upstream ​

bash
git fetch origin # Changes need to merged, also

Git pull is a combination of fetch and merge

Use pull instead of fetch and merge together

bash
git pull origin

See remote log ​

bash
git log origin/master

Remote track a branch ​

To use a different remote branch, rather than the default one.

bash
git switch -c branchname origin/branchname

If used once locally, from next time, one can checkout or switch to that branch only.

GitHub Flow ​

  • Create a new Branch
  • Make changes and add Commits
  • Open a Pull Request
  • Review
  • Deploy
  • Merge

Working with credentials ​

Cache credentials saved ​

bash
git config credential.helper cache
git push http://example.com/repo.git

GCM for credentials ​

bash
git config --global credential.credentialStore secretservice

Remove that GCM configuration ​

bash
git config --unset credential.helper

GitHub : Forking and cloning ​

Cloning a repo ​

bash
git clone url.git # use ssh rather than https one, if ssh has been setup
git clone url.git --depth=1 # Faster , latest commit only

INFO

Note: According to Git naming conventions, it is recommended to name your own repository origin, and the one you forked for upstream origin : fork (R/W) upstream : original (Read only)

Excluding files from tracking (.gitignore) ​

Git will not track files and folders specified in .gitignore. However, the .gitignore file itself is tracked by Git. It is also possible to have additional .gitignore files in subdirectories. These only apply to files or folders within that directory.

It is also possible to ignore files or folders but not show it in the distributed .gitignore file.

These kinds of ignores are specified in the .git/info/exclude file. It works the same way as .gitignore but are not shown to anyone else.

Here are some basic gitignore syntax, one might find helpful

** : All directories check
* : All file check
? : Any single character
[a-z] : regex for single character
[!abc] : not a character from abc
name : Everything with name

Red Zone ​

DANGER

Proceed with caution

Reverting an error ​

bash
git revert HEAD --no-edit # revert last change and commit --no-edit is to use last message only
git revert HEAD~x # where x is x+1th commit one wanna rollback to

Squashing ​

bash
git rebase --interactive HEAD~n # n is the number of commit you want to squash
# calling above command will open a terminal
# replace `pick` with `s` to squash a commit i.e., to remove.
git push --force

Amend a change to previous commit ​

Amend combines changes in the staging environment with the latest commit, and creates a new commit.

bash
git add filename # Add to staging area
git commit --amend # Used to modify the most recent commit.

Amend and replace last commit ​

bash
git commit --amend -m "Added lines to README.md"

Moving back n commits ​

bash
git reset --soft HEAD~n # Soft reset, n is the number of commit you want to reset 
git reset --hard HEAD~n # Hard reset, n is the number of commit you want to reset 
git push -f

Reset ​

WARNING

Not recommended to used with remote repos

bash
git reset commithash
  • commithash being the first 7 characters of the commit hash we found in the log
  • one can rollback the reset, if he/she knows the final hash

See also ​