Is it impossible to use Guard with RubyMine?

When you run guard with RubyMine for tests, it is extremely useful to configure a separate database environment for guard spec, or else you'll experience strange issues (one process or the other freezes or gives inconsistent results.

Name your guard spec environment "ci" and create an additional entry in database.yml. I use "ci" for Continuous Automation.

Then put this in your Guardfile. The key thing is

'RAILS_ENV' => 'ci'

Here's how I have mine setup:

group :spec do
  guard :spork, :rspec_port => 1234, :cucumber_env => { 'RAILS_ENV' => 'ci' }, :rspec_env => { 'RAILS_ENV' => 'ci' } do
    watch('config/application.rb')
    watch('config/environment.rb')
    watch(%r{^config/environments/.+.rb$})
    watch(%r{^config/initializers/.+.rb$})
    watch('spec/spec_helper.rb')
    watch(%r{app/models/.+\.rb})
    watch(%r{app/views/.+\.haml})
    watch('Gemfile')
    watch('Gemfile.lock')
    watch('test/test_helper.rb')
  end

  # environment is 'ci'
  guard :rspec, :cli => '--drb --drb-port 1234', :version => 2, :all_after_pass => false, :notification => true, :environment => 'ci' do
    watch(%r{^spec/.+_spec.rb$})
    watch(%r{^lib/(.+).rb$})     { |m| "spec/lib/#{m[1]}_spec.rb" }
    watch('spec/spec_helper.rb')  { "spec" }
  # Rails example
    watch(%r{^spec/.+_spec.rb$})
    watch(%r{^app/(.+).rb$})                           { |m| "spec/#{m[1]}_spec.rb" }
    watch(%r{^lib/(.+).rb$})                           { |m| "spec/lib/#{m[1]}_spec.rb" }
    watch(%r{^app/controllers/(.+)_(controller).rb$})  do |m|
      ["spec/routing/#{m[1]}_routing_spec.rb",
       "spec/#{m[2]}s/#{m[1]}_#{m[2]}_spec.rb",
       "spec/acceptance/#{m[1]}_spec.rb",
       "spec/requests/#{m[1]}_spec.rb"]
    end

    watch(%r{^spec/support/(.+).rb$})                  { "spec" }
    watch('config/routes.rb')                          { "spec/routing" }
    watch('app/controllers/application_controller.rb')  { "spec/controllers" }
    # Capybara request specs
    watch(%r{^app/views/(.+)/.*.(erb|haml)$})          { |m| "spec/requests/#{m[1]}_spec.rb" }
  end
end

I then run

bundle exec guard -g spec

I also don't mind having RubyMine save files automatically every 60 seconds even if that kicks off Guard, as my a new MBP Retina does not noticeably slow down when running Guard.

BTW, Guard running specs is really great as you will find failing tests much faster than trying to run the tests yourself in RubyMine. I.e., the test basically is about done by the time my finger releases from cmd-s to save.

I run this from the terminal. I haven't tried running with RubyMine. Anybody want to comment on the advantages of doing that? I guess having the stack dump clickable would be nice.

This is a really frustrating problem, and makes it very hard to use RubyMine with guard or grunt or other task-runners.

I created a bit of a crazy workaround, but it's working well enough for me:

1. Add an External Tool to run the venerable Unix "touch" command, which updates the modified timestamp of a file, thereby generating a filesystem event. It should look like this:
2. Assign the External Tool a keyboard shortcut.
3. Assign a keyboard shortcut to the Save Document command.
4. Record a macro that runs Save Document, then the "touch" External Tool. You need to use the keyboard shortcuts you assigned because Save Document does not appear in the application's menus.
5. Assign Cmd-S as a keyboard shortcut for your new macro.

My full blog post about this is at http://singlebrook.com/blog/saving-files-in-rubymine

Or you can add a run_guard.rb file at the root of your project with the following code:

exec 'guard'

Watch out! Rubymine has a problem with your PATH. For example, /usr/local/bin is filtered ...

I run it as a Gem command which alto allows me to tap into Bundler at the same time so Guard's version is correct for the project.