Open source developers often have to jump between projects, either to read code, or to craft patches. My Repo Pins reduces the friction so that it becomes trivial to do so.
The idea of the plugin is based on this idea; if the repository URLs can be translated to a filesystem location, the local disk can be used like a cache. My Repo Pins lazily clones the repo to the filesystem location if needed, and then jumps into the project in one single command. You don’t have to remember where you put the project on the local filesystem because it’s always using the same location. Something like this:
~/code-root ├── codeberg.org │ └── Freeyourgadget │ └── Gadgetbridge └── github.com ├── BaseAdresseNationale │ └── fantoir ├── mpv-player │ └── mpv └── NinjaTrappeur ├── cinny └── my-repo-pins
This Emacs plugin aims to help you navigate this repository tree and clone new repositories at the right place in the tree.
Show me What You’ve Got
As always, a small demo is worth a thousand words!
Jump to a local repository you already cloned:
Find a repository in a remote forge, clone it, and jump to it:
Alternatively, you can also specify a absolute git URL you want to clone:
Via MELPA (recommended)
M-x package-install my-repo-pins
Clone this repository, open the
my-repo-pins.el file, then, install the package using
The minimal configuration consists in setting the directory in which you want to clone all your git repositories via the
Let’s say you’d like to store all your git repositories in the
~/code-root directory. You’ll want to add the following snippet in your Emacs configuration file:
(require 'my-repo-pins) (setq my-repo-pins-code-root "~/code-root")
You can then call the
M-x my-repo-pins command to open a project living in your
~/code-root directory or clone a new project in your code root.
Binding this command to a global key binding might make things a bit more convenient. I personally like to bind it to
M-h. You can add the following snippet to your Emacs configuration to set up this key binding:
(global-set-key (kbd "M-h") 'my-repo-pins)
my-repo-pins-code-root - REQUIRED
Path to the directory containing all your projects.
my-repo-pins organize the git repos you’ll clone in a tree fashion.
All the code fetched using
my-repo-pins will end up in this root directory. A tree of subdirectories will be created mirroring the remote clone URL.
For instance, after checking out https://git.savannah.gnu.org/git/emacs/org-mode.git, the source code will live in the my-repo-pins-code-root/git.savannah.gnu.org/git/emacs/org-mode/ local directory
Maximum search depth starting from the
Set this variable to nil if you don’t want any limit.
This is a performance stop gap. It’ll prevent my repo pins from accidentally walking too deep if it fails to detect a project boundary.
By default, this limit is set to 2 to materialize the
<forge>/<username> directories that are supposed to contain the projects.
We won’t search further once we reach this limit. A warning message is issued to the
*Messages* buffer to warn the user the limit has been reached.
Path pointing to the git binary. By default, it’ll look for git in the current
Alist in the form of
("FORGE NAME" . FETCH-FUNCTION) where
FETCH-FUNCTION is a function in charge of retrieving a potential remote clone URL. More about this function in the Fetchers section.
When a repository cannot be found in the code root directory,
my-repo-pins will try to download it from different forges. By default, it’ll try to find it on github.com, gitlab.com, git.sr.ht, and codeberg.org.
Re-Using the Default Fetchers for your own Forge Instance
My-repo-pins provides some generic fetchers for Gitlab, Sourcehut, and Gitea.
You can re-use these generic fetchers for your own forge instance using the following functions:
These functions share the same 4 input arguments:
instance-url: your instance FQDN. For instance:
user-name: the user name for which we want to clone the repository.
repo-name: name of the repository we want to clone.
my-repo-pinswill use to clone the repository once we retrieved the various clone URLs. The callback takes an alist as parameter. The alist being of the form of :
((ssh . SSH-CHECKOUT-URL) (https . HTTPS-CHECKOUT-URL)).
You can re-use these functions by instantiating them for a specific forge, then by appending this instantiation to the
my-repo-pins-forge-fetchers variable in your Emacs configuration.
Let’s say you want to retrieve repositories from the Gnome Gitlab instance living at
gitlab.gnome.org. You’ll have to add the following snippet to your Emacs configuration:
(setq my-repo-pins-forge-fetchers `(("gitlab.gnome.org" (lambda (owner repo cb)(my-repo-pins--query-gitlab-owner-repo "gitlab.gnome.org" owner repo cb))) ,my-repo-pins-forge-fetchers))
Writing your Forge Fetcher from Scratch
You may also want to support a forge for which
my-repo-pins currently does not provide any generic fetcher. In that case, you’ll have to write a function in the form of:
(defun your-custom-fetcher (owner repo) ; (..) the actual implementation )
The function needs to accept two input parameters:
owner: string containing the name of the owner of the query repository. IE.
repository: string containing the name of the query repository. IE.
This function will return either
nil in case the query couldn’t be found on the remote forge. An alist containing the SSH and HTTPS clone URLs in the form of:
'((ssh . SSH-CHECKOUT-URL) (https . HTTPS-CHECKOUT-URL))
- Add the my-repo-pins-max-depth customization variable. This variable prevents my repo pins from accidentally recursing too far in the code root directory. Solves a performance issue with git subtrees.
- When cloning a project from a fullly qualified URL, we make sure the project does not exists locally first. If it does, we jump to the local clone without trying to clone it again.
- This plugin is heavily inspired by Zimbatm’s h. Thanks a lot for this amazing tool, it made my life easier for years!
- Thanks a lot to Wiwi who found the terrible pun we used to name this project.