Just remember that the assert* functions are ONLY available two ways ...
 
either do a
 
require 'test/unit/assertions'
include Test::Unit::Assertions
 
# now you have access to the assert* functions.
 
or
 
put your code inside of an object derived from Test::Unit::TestCase
 
...
 
Hope that helps.
 
j.
 

 
On 10/26/05, Pierre Garigue <[EMAIL PROTECTED]> wrote:
you may want to look at the log4r http://log4r.sourceforge.net/ to have an oject that take care of loging your reports. It may be overkill.

Most of my test use the 'rescue => e' way

example

           expected = "hello" #some text on the homepage
               @@log.info "Expected: Home page Page"
               begin
                      @@ie.goto(homepageurl)

                       assert(@@ie.contains_text(expected))
                       @@log.info "PASS"
               rescue => e
                       @@log.info "FAIL"
               end



I'm not sure the way below is better, here is a fragment

                               begin
                                       assert_nothing_raised(){
                                               (@@ ie.link(:text, link_name).click)
                                       }
                                       @@log.info "PASS: got down to #{link_name}"
                               rescue => e
                                       @@log.info "FAIL: did not get to #{link_name}"
                               end

cheers
Pierre garigue



-----Original Message-----
From: Sy Ali [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, October 26, 2005 1:30 PM
To: [email protected]
Subject: Re: [Wtr-general] getting the HTTP status


On 10/21/05, Pierre Garigue <[EMAIL PROTECTED]> wrote:
> I because I 'rescue' everything, getting a 500 or 404 would just give me a fail. I would like to add info to the log that is failed and got this http status.

I too am using the basic "rescue" technique, which doesn't really give
me a whole lot to work with.  I read through the technical docs[1] and
couldn't find anything more specific.  Perhaps this feature does not
exist?

I wonder what would happen if I turned on friendly error messages ..
then if rescue catches the error, could I scan the page or title to
determine the nature of the error?

Here is my "go to" method, used with
goto("http://example.org/foo/bar.html") or goto(variable).  It's
different than yours because it notices succesfully-loaded, but
redirected pages:

def goto(url)
       begin
               puts "* going to " + url
               $ie.goto(url)
       rescue
               puts ">>>ERROR"
       else
               puts "   ok, destination loads"
       end
       if $ie.url() == url
               puts "   ok, I stayed at the destination url"
       else
               puts "   warning: rewriting or redirection took effect"
       end
end


[1] http://wtr.rubyforge.org/rdoc/index.html

_______________________________________________
Wtr-general mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-general


_______________________________________________
Wtr-general mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-general



--
"http://ruby-lang.org -- do you ruby?"

Jeff Wood
_______________________________________________
Wtr-general mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-general

Reply via email to