Andrew Premdas wrote:
> The key phrase is semantic meaning. Its good to use CSS to give semantic
> meaning to things that appear on web pages. Not only can these id's or
> classes be useful for using with screen readers, they should be stable
> things that don't change even if the design of the page does. It should
This is an example of my current practice, consequential on the
discussions I have had on this list:
<h2 id="authentication_request">To Proceed Please Authenticate
Yourself</h2>
<div id="authentication_fields">
<form action="/user_session"
class="new_user_session"
id="new_user_session"
method="post"
>
<label for="user_session_username">User Name</label> <br />
<input id="user_session_username"
name="user_session[username]"
size="30" type="text" />
...
<input id="user_session_submit"
name="commit"
type="submit"
value="Authenticate" />
</form>
</div>
My intent is that id values are tied to the presentation elements they
deal with, not to the layout they find themselves in. So I test the UI
by calling the ids in webrat, rather than the labels. This will ease
any future transition to I18n.t calls in the templates.
As a side issue, in getting the input box ids to work with label ids as
I specified in the templates, I discovered that I had misread the
FormHelper api. I uncovered my mistake by writing a few tests to
exercise this feature as Rails own test/form_helper_test.rb did not.
The point is that I have submitted these tests as a patch to Rails and
they are awaiting verification for inclusion. Patches require three
reviewers for consideration by the core team. My submission has had one
positive review already. I would appreciate a couple of more. The
lighthouse ticket is 2096.
http://rails.lighthouseapp.com/projects/8994/tickets/2096-added-select-method-tests-to-form_helper_testrb
--
Posted via http://www.ruby-forum.com/.
_______________________________________________
rspec-users mailing list
rspec-users@rubyforge.org
http://rubyforge.org/mailman/listinfo/rspec-users
