At 08:07 PM 10/23/2001 -0700, you wrote:
>Michael-
>
>If you check the logs, I had this problem about a week
>ago. See:
>http://www.mail-archive.com/[email protected]/msg00624.html
>
>The gist of it is that the old version actually
>performed redirects where the new version does not.
>(On purpose) This allows you to now test that a 302 is
>returned to the browser, but may complicate your
>redirection testing slightly.
>
>Vincent had asked if anyone was interested in a flag
>allowing you to "turn on" redirects on demand. Sounds
>like you may be a +1 for that feature?
Yes I'm definitely +1 for that :-) The system I was using was also doing
functional testing in a slightly different way [esp before HTTPUnit was
integrated]
1 - Get the struts action output
2 - Forward it on to a servlet that would render that output into an XML
format.
3 - Send THAT back to the the client side.
4 - The endTest() then parses the XML and allows you to use XPath, and
other utilities to examine the output for correctness.
So it is a wee bit of a difference. I'll probably refactor so I just do
all my testing "white-box" and peek into the session variables that Struts
generates.
Thank you very much for the reply, and I'm sorry to pipe up with the same
question that had just been asked, forgot to check the archives :-), and I
hadn't by and large been following the list mails.
-Mike
