Merged pull request #338 from mcmire/revert_readme_changes.

Clear things up in README again re: spec/acceptance vs. spec/requests
This commit is contained in:
Jo Liss 2011-04-28 17:08:08 -07:00
commit 7b69aa669e
1 changed files with 20 additions and 16 deletions

View File

@ -88,7 +88,7 @@ by adding the following line (typically to your <tt>spec_helper.rb</tt> file):
require 'capybara/rspec'
You can now use it in your examples:
You can now write your specs like so:
describe "the signup process", :type => :request do
before :each do
@ -104,16 +104,21 @@ You can now use it in your examples:
end
end
Capybara is only included for examples with <tt>:type => :request</tt> (or
<tt>:acceptance</tt> for compatibility).
Capybara is only included in example groups tagged with
<tt>:type => :request</tt> (or <tt>:acceptance</tt> for compatibility with Steak).
If you use the <tt>rspec-rails</tt> gem, <tt>:type => :request</tt> is
automatically set on all files under <tt>spec/requests</tt>. Essentially, these
are Capybara-enhanced Rails request specs, so it's a good idea to place your
Capybara specs here because within request specs you gain a few additional
features, such as the ability to refer to named route helpers. If you do not
need these, then you may simply use <tt>spec/acceptance</tt> and you will still
get access to Capybara methods.
If you are testing a Rails app and using the <tt>rspec-rails</tt> gem, these
<tt>:request</tt> example groups may look familiar to you. That's because they
are RSpec versions of Rails integration tests. So, in this case essentially what you are getting are Capybara-enhanced request specs. This means that you can
use the Capybara helpers <i>and</i> you have access to things like named route
helpers in your tests (so you are able to say, for instance, <tt>visit
edit_user_path(user)</tt>, instead of <tt>visit "/users/#{user.id}/edit"</tt>,
if you prefer that sort of thing). A good place to put these specs is
<tt>spec/requests</tt>, as <tt>rspec-rails</tt> will automatically tag them with
<tt>:type => :request</tt>. (In fact, <tt>spec/integration</tt> and
<tt>spec/acceptance</tt> will work just as well.)
<tt>rspec-rails</tt> will also automatically include Capybara in <tt>:controller</tt> and <tt>:mailer</tt> example groups.
RSpec's metadata feature can be used to switch to a different driver. Use
<tt>:js => true</tt> to switch to the javascript driver, or provide a
@ -124,7 +129,7 @@ RSpec's metadata feature can be used to switch to a different driver. Use
it 'will switch to one specific driver', :driver => :celerity
end
Capybara also comes with a built in DSL for creating descriptive acceptance tests:
Finally, Capybara also comes with a built in DSL for creating descriptive acceptance tests:
feature "Signing up" do
background do
@ -140,11 +145,10 @@ Capybara also comes with a built in DSL for creating descriptive acceptance test
end
end
Essentially, this is just a shortcut for making a request spec, where
<tt>feature</tt> is a shortcut for <tt>describe ..., :type => :request</tt>,
<tt>background</tt> is an alias for <tt>before :each</tt>, and <tt>scenario</tt>
is an alias for <tt>it</tt>/<tt>example</tt>. Again, you are encouraged to place
these within <tt>spec/requests</tt> rather than <tt>spec/acceptance</tt>.
This is, in fact, just a shortcut for making a request spec, where
<tt>feature</tt> is an alias for <tt>describe ..., :type => :request</tt>,
<tt>background</tt> is an alias for <tt>before</tt>, and <tt>scenario</tt>
is an alias for <tt>it</tt>/<tt>specify</tt>.
Note that Capybara's built in RSpec support only works with RSpec 2.0 or later.
You'll need to roll your own for earlier versions of RSpec.