Sub is a model for setting up shell programs that use subcommands, like
rbenv using bash. Making a sub does not require you to write shell scripts in bash, you can write subcommands in any scripting language you prefer.
A sub program is run at the command line using this style:
$ [name of program] [subcommand] [(args)]
Here's some quick examples:
$ rbenv # prints out usage and subcommands $ rbenv versions # runs the "versions" subcommand $ rbenv shell 1.9.3-p194 # runs the "shell" subcommand, passing "1.9.3-p194" as an argument
Each subcommand maps to a separate, standalone executable program. Sub programs are laid out like so:
. ├── bin # contains the main executable for your program ├── completions # (optional) bash/zsh completions ├── libexec # where the subcommand executables are └── share # static data storage
Each subcommand executable does not necessarily need to be in bash. It can be any program, shell script, or even a symlink. It just needs to run.
Here's an example of adding a new subcommand. Let's say your sub is named
touch libexec/rush-who chmod a+x libexec/rush-who
Now open up your editor, and dump in:
#!/usr/bin/env bash set -e who
Of course, this is a simple example...but now
rush who should work!
$ rush who qrush console Sep 14 17:15
You can run any executable in the
libexec directly, as long as it follows the
NAME-SUBCOMMAND convention. Try out a Ruby script or your favorite language!
You get a few commands that come with your sub:
commands: Prints out every subcommand available
completions: Helps kick off subcommand autocompletion.
help: Document how to use each subcommand
init: Shows how to load your sub with autocompletions, based on your shell.
shell: Helps with calling subcommands that might be named the same as builtin/executables.
If you ever need to reference files inside of your sub's installation, say to access a file in the
share directory, your sub exposes the directory path in the environment, based on your sub name. For a sub named
rush, the variable name will be
Here's an example subcommand you could drop into your
libexec directory to show this in action: (make sure to correct the name!)
#!/usr/bin/env bash set -e echo $_RUSH_ROOT
You can also use this environment variable to call other commands inside of your
libexec directly. Composition of this type very much encourages reuse of small scripts, and keeps scripts doing one thing simply.
Each subcommand can opt into self-documentation, which allows the subcommand to provide information when
sub help [SUBCOMMAND] is run.
This is all done by adding a few magic comments. Here's an example from
rush who (also see
sub commands for another example):
#!/usr/bin/env bash # Usage: sub who # Summary: Check who's logged in # Help: This will print out when you run `sub help who`. # You can have multiple lines even! # # Show off an example indented # # And maybe start off another one? set -e who
Now, when you run
sub, the "Summary" magic comment will now show up:
usage: sub <command> [<args>] Some useful sub commands are: commands List all sub commands who Check who's logged in
sub help who will show the "Usage" magic comment, and then the "Help" comment block:
Usage: sub who This will print out when you run `sub help who`. You can have multiple lines even! Show off an example indented And maybe start off another one?
That's not all you get by convention with sub...
Your sub loves autocompletion. It's the mustard, mayo, or whatever topping you'd like that day for your commands. Just like real toppings, you have to opt into them! Sub provides two kinds of autocompletion:
- Automatic autocompletion to find subcommands (What can this sub do?)
- Opt-in autocompletion of potential arguments for your subcommands (What can this subcommand do?)
Opting into autocompletion of subcommands requires that you add a magic comment of (make sure to replace with your sub's name!):
# Provide YOUR_SUB_NAME completions
and then your script must support parsing of a flag:
--complete. Here's an example from rbenv, namely
#!/usr/bin/env bash set -e [ -n "$RBENV_DEBUG" ] && set -x # Provide rbenv completions if [ "$1" = "--complete" ]; then echo --path exec rbenv shims --short fi # lots more bash...
--complete flag to this subcommand short circuits the real command, and then runs another subcommand instead. The output from your subcommand's
--complete run is sent to your shell's autocompletion handler for you, and you don't ever have to once worry about how any of that works!
init subcommand after you've prepared your sub to get your sub loading automatically in your shell.
Creating shortcuts for commands is easy, just symlink the shorter version you'd like to run inside of your
Let's say we want to shorten up our
rush who to
rush w. Just make a symlink!
cd libexec ln -s rush-who rush-w
rush w should run
libexec/rush-who, and save you mere milliseconds of typing every day!
Clone this repo:
git clone git://github.com/37signals/sub.git [name of your sub] cd [name of your sub] ./prepare.sh [name of your sub]
The prepare script will run you through the steps for making your own sub. Also, don't call it
sub, by the way! Give it a better name.
So you've prepared your own sub, now how do you use it? Here's one way you could install your sub in your
cd git clone [YOUR GIT HOST URL]/sub.git .sub
For bash users:
echo 'eval "$($HOME/.sub/bin/sub init -)"' >> ~/.bash_profile exec bash
For zsh users:
echo 'eval "$($HOME/.sub/bin/sub init -)"' >> ~/.zshenv source ~/.zshenv
You could also install your sub in a different directory, say
/usr/local. This is just one way you could provide a way to install your sub.