Like virtualenv + pip for Ruby.

Installs and manages RubyGems, git repositories, and more.

We're currently in a developer-mode rewrite: rip2.

The best way to install it right now is to clone with git then add to your $PATH.

Got hub?

git clone defunkt/rip

Old school style:

git clone git://github.com/defunkt/rip.git

Now set it up in your ~./bashrc (or whatever):

export RUBYLIB="$RUBYLIB:$HOME/Projects/rip/lib"
export PATH="$PATH:$HOME/Projects/rip/bin"
eval `rip-config`

Replace $HOME/Projects/rip with the path to your clone of rip. To see what the eval statement is executing each time your shell opens, run rip-config by hand:

$ rip-config
function rip-push() {
  eval `rip-sh-push $1`;
};
function rip-pop() {
  eval `rip-sh-pop $1`;
};
function rip-use {
  export RIPENV=$1;
  eval `rip-config`;
};
RIPVERBOSE=1
RIPDIR=${RIPDIR:-"$HOME/.rip"}
RUBYLIB="$RIPDIR/active/lib:$RUBYLIB"
PATH="$RIPDIR/active/bin:$PATH"
MANPATH="$RIPDIR/active/man:$MANPATH"
export RIPVERBOSE RIPDIR RUBYLIB PATH MANPATH

That's what I get. So, let's test this thing out by seeing what envs we have.

Let's try it:

$ rip-envs
$RIPDIR not set. Please eval `rip-shell`

Whoops. We need to restart our shell so the ~/.bashrc changes take hold. We can also run that eval command by hand, but it's easier just to open a new shell.

Now, in our new shell:

$ rip-envs
/Users/chris/.rip not found. Please run `rip-setup`

Okay.

$ rip-setup
$ rip-envs
* base

We're ready to roll.

Many people installed Rip, played with it, realized it wasn't done, then forgot about it - only to become frustrated when some package they installed with Rip screwed up their development environment. Those were the kinds of problems Rip was supposed to solve! What does rip2 have to say about this?

Two words: rip-off. rip2 can be enabled or disabled at any time using the rip-off and rip-on commands:

$ rip-env
choices
$ ruby -r choice -e ''
$ rip-off
$ rip-env
rip is currently disabled. `rip-on` to re-enable.
$ ruby -r choice -e ''
ruby: no such file to load -- choice (LoadError)
$ rip-on
$ ruby -r choice -e ''
$ rip-env
choices

$ rip-env -c emailing
$ rip-install mail
fetching mail 2.2.0
installed activesupport (2.3.5)
installed mime-types (1.16)
installed json_pure (1.4.2)
installed rubyforge (2.0.4)
installed minitest (1.6.0)
installed rake (0.8.7)
installed hoe (2.6.0)
installed polyglot (0.3.1)
installed treetop (1.4.5)
installed mail (2.2.0)

It only fetched the packages I hadn't already installed in another environment. Thanks, rip.

Now we can use our new library:

$ irb -r mail
>> Mail
=> Mail

When we move to another ripenv, we're in a new world:

$ rip-env base
base
$ irb -r mail
`require': no such file to load -- mail (LoadError)
  from /ruby/ree-1.8.7/lib/ruby/1.8/irb/init.rb:254:in `load_modules'

When installing a RubyGem, rip respects dependencies. Installing from a git repository? rip will check for a deps.rip and use that.

By default, rip assumes you don't want to overwrite installed packages:

$ rip-install rack 1.0.0
installed rack (1.0.0)
$ rip-install rack 1.1.0
Conflicts
  rack
    1.1.0 requested, 1.0.0 already installed

Use -f to force the install:

$ rip-install -f rack 1.1.0
installed rack (1.1.0)

(This operation may need some polish, but it works.)

Installation:

$ rip-install RUBYGEM
$ rip-install RUBYGEM VERSION
$ rip-install GIT_REPO
$ rip-install GIT_REPO TAG_OR_VERSION
$ rip-remove PACKAGE_NAME

Introspection:

$ rip-list
$ rip-config
$ rip-installed
$ rip-installed package

Environments:

$ rip-envs
$ rip-env -c NEW_RIPENV
$ rip-env -d RIPENV_TO_DELETE
$ rip-env RIPENV_TO_SWITCH_TO

Spring cleaning:

$ rip-gc
$ rip-fsck

I'm now using rip for all my development. This includes GitHub. Here are fun things to do with rip today;

• rip-readme
• Gemfile => deps.rip
• josh/rip-bundle
• josh/rip-externals
• stacking ripenvs

rip strives to be compatible with:

• POSIX
• bash 3
• Ruby 1.8.7
• Ruby 1.9.1
• RubyGems 1.3.7
• rpg 0.2.0
• git 1.7.0
• the Ruby Packaging Standard: http://chneukirchen.github.com/rps/

To run the test suite:

$ rake

Once you've made your great commits:

1. Fork rip
2. Create a topic branch - git checkout -b my_branch
3. Push to your branch - git push origin my_branch
4. Create an Issue with a link to your branch
5. That's it!

To join the list simply send an email to [email protected]. This will subscribe you and send you information about your subscription, including unsubscribe information.

The archive can be found at http://librelist.com/browser/.