[ANN] assert2 presents assert_no_rjs_, and .should send_js_to

Rails
1

assert_no_rjs_ now negates all the JavaScript assertions listed here:

http://groups.google.com/group/rubyonrails-talk/browse_thread/thread/49fd7c9b9107d577

And assert_rjs_ now works in RSpec:

 it 'should use Ajax to replace person 45' do
   xhr :get, :replace_person, :id => 45
   @response.body.should send_js_to(:replace_html, 'person_45') do
     strong 'person_45'
   end
 end

Once again, the do-end block calls Nokogiri::HTML::Builder to build a
sample of
HTML. The assertion then matches this to the HTML delivered in the
payload of
the JS Element.update() call.

The committee is unaware of any other RJS solution for RSpec…

(Sorry - no .should not_send_js_to yet…:wink:

2

assert_no_rjs_ now negates all the JavaScript assertions listed here:

http://groups.google.com/group/rubyonrails-talk/browse_thread/thread/49fd7c9b9107d577

(Sorry - no .should not_send_js_to yet…:wink:

Version 0.5.1 fixes that: .should_not send_js_to

Next, assert_xhtml (and .should be_html_with) add the any! keyword:

 assert_xhtml do
   ul.kalika do
     any! 'Billings report'
   end
 end

It evaluates to the XPath * operator, so the above passes if any element
below
the

    contains any “Billings report”.

    This example fails if “Billings report” appears anywhere in the document
    (assuming we reeally don’t like it!;):

 assert_xhtml do
   without!{ any! /Billings report/ }
 end

If anyone can think of new features to add to (or take out of) these
DSLs,
I’m all ears!

–
Quark

3

Phlip,

Is there any plan to allow for negative assertions (e.g., should_not
be_html_with)? I use this in a custom matcher, and can’t think of a way
to
support negative assertions unless be_html_with does.

I’m of course going by the output of negative_failure_message saying not
to
use it that way.

Brandon

4

Brandon O. wrote:

Is there any plan to allow for negative assertions (e.g., should_not
be_html_with)? I use this in a custom matcher, and can’t think of a way to
support negative assertions unless be_html_with does.

I can’t think of a reason not to use .should be_html_with{ without!{
…} }

And it seems that .should_not be_html_with() would, in theory, pass if a
page
weren’t HTML…

5

…} }
It could, if I could determine whether the matcher was being called with
.should or .should_not. I don’t know of a way to do this?

Anyway I’m excited about this new release. Thanks.

Brandon