[Webtest] Re: Iteration through an XPath results set

Dierk Koenig webtest@lists.canoo.com
Tue, 6 Apr 2004 14:17:04 +0200

I would like to underline Christopher's points and add one:
- in case you are trying to assert special properties of
  all nodes in the result set, you can use the
  doubled negation trick.
  Try to select one node that _breaks_ the property.
  Put that into a "not" step.


> -----Original Message-----
> From: webtest-admin@lists.canoo.com
> [mailto:webtest-admin@lists.canoo.com]On Behalf Of Christopher
> Painter-Wakefield
> Sent: Montag, 5. April 2004 21:12
> To: webtest@gate2.canoo.com
> Subject: [Webtest] Re: Iteration through an XPath results set
> No useful iteration of the type you are asking about is possible, to my
> knowledge.  There is some kind of repeat step, but that again depends on
> you knowing how many elements you need to test.
> In general, this kind of non-deterministic testing is of limited
> usefulness.  You can't tell at a glance that you have tested anything
> worthwhile; for instance, if your search returns no results, you
> won't have
> any bad links :)
> To avoid this problem, control everything - your test data, your test
> environment, etc., as much as possible.  Since we can use all of Ant, we
> use Ant SQL tasks to set up data for each set of tests on our
> database-driven application.  The tests then can be very focused and
> specific for each case, by setting up each case in the database.  For the
> other kind of testing, we use eyeballs.
> That being said, some kind of looping constructs would come in handy in
> certain cases.  Also, I have considered extending verifylinks to allow
> verification of all links matching a particular pattern (regex or xpath);
> my webapp is part of a larger suite of webapps, so our pages have a layout
> that includes links to the other systems, and of course the obligatory
> links to our institution and so forth.  I would like to be able to verify
> all links on the page that are just internal to my application.  This
> should be a fairly easy extension, but my need has not been strong enough
> to find the time to do it.
> Good luck!
> -Christopher
> Message: 2
> From: "Oster Nathan" <Oster_Nathan@bah.com>
> To: <webtest@gate2.canoo.com>
> Date: Sat, 3 Apr 2004 00:13:23 -0500
> Subject: [Webtest] Iteration through an XPath results set
> Reply-To: webtest@lists.canoo.com
> I'm a newbie to both XPath and WebTest, so I apologize if this is too
> basic.
> Is is possible to loop/iterate through the results of an XPath
> expression that returns a set of elements?
> For example, if the XPath expression returns a set of product href
> links, can I iterate through all of them and call "verifylinks"?
> If iteration through a results set is not possible, can someone please
> provide me with an alternative test design for cases where you need to
> validate a set of randomly generated page elements.  This includes
> verifying search results, product lists, a list of navbar buttons, etc.
> Thanks!
> Nate
> _______________________________________________
> WebTest mailing list
> WebTest@lists.canoo.com
> http://lists.canoo.com/mailman/listinfo/webtest