This is a collection of tools I use with Git that make life easier.

For a basic installation, run the installation script, like this:

sudo ./install-ak-git-tools

Using the above command as-is would install these tools into /usr/bin. The installation prefix (by default /usr) can be overridden by passing something as the first argument.

The installation also adds an uninstaller, called __uninstall-ak-git-tools. Upon installation, the uninstaller is automatically ran first. To uninstall manually, simply use the generated uninstaller:

__uninstall-ak-git-tools /usr

Additionally, some Git configurations can be set in the global domain. To set them, run:

./set-ak-global-config

There is no uninstaller for the configurations set by set-ak-global-config.

Note: Some of these commands are aliases. Running install-ak-git-tools does not include the aliases; run set-ak-global-config to set the aliases.

• git graph ...: (alias) A primitive version of gitk for use in a terminal.

• git scrub ...: (alias) The same as git clean, except instead of running in the current working directory, this starts in the top-level of the superproject, and is recursive across all submodules. To blow away all untracked files everywhere in the entire superproject, run:

    git scrub -xfd
  

• git su ...: (alias) Update all submodules to the commit ID at which they are intended to be checked out according to HEAD in the current repository. To initialize all submodules recursively starting with the current repository, run:

    git su --init
  

• git env ...: Run the given command at the top of the current repository. Add --recursive to also run in all submodules.

• git tenv ...: Run the given command at the top of the superproject. Add --recursive to also run in all submodules.

• git disconnect-alternates: If the current repository is using alternates, run git repack -a and delete the alternates file.

A collection of scripts for cloning faster. See git-repocache (below).

The "repocache" is a single bare repository that contains many remotes, and is used to speed up cloning by reducing the number of objects that must travel over the network. If you're interested in the technicalities, read up on the --reference parameter to git-clone(1).

To use git-repocache, you must set the property repocache.path in Git's global or system configuration.

The following is a list of all the commands supported by git-repocache:

• git repocache: Invoke the given Git command in the repocache. Example:

    git repocache remote add git git://github.com/git/git.git
  

• git repocache-clone: Performs a normal clone, but uses the repocache. Example:

    git repocache-clone git://github.com/git/git.git
  

  Upon successful exit, you should then run git disconnect-alternates inside the newly cloned repository to prevent the repository from becoming corrupt.

• git repocache-clone-by-name: Performs a normal clone, but instead of providing the URI, you provide the name of the remote in the repocache. This command then looks up the URI used in the repocache, and clones using that URI. Example:

    git repocache-clone-by-name git
  

  Upon successful exit, you should then run git disconnect-alternates inside the newly cloned repository to prevent the repository from becoming corrupt.

• git repocache-env: Runs the given command inside the repocache. Example:

    git repocache-env pwd
  

• git repocache-init: Runs git init inside the repocache. Creates the directory first if necessary.

• git repocache-path: Retrieves the path to the repocache from Git and prints it to stdout.

• git repocache-submodule: Invokes the given git-submodule(1) command in the current working directory. Example:

    git repocache-submodule update --init --recursive
  

• git repocache-update: Performs a fetch in the repocache.