Hi Deryl,
We are using 1.8 but the same issue is arising as we are not at all using the
form-beans. Is this a know bug or there is some way out of it?
Hoping for some faster reply,
regards,
Maniranjan Kumar
Hi Deryl,
It's not causing me any great issue at the moment (on 1.5) so no panic for the patch.
Alan
Deryl Seale <[EMAIL PROTECTED]> wrote: You should not have to change the code to test
it,
it's true. And this should be fixed in the 1.6
version of StrutsTestCase -- specifically, addressing
the perfectly valid case when validate() returns
a null value.
If it is not fixed, or if you guys are using v1.5,
let me know, and I'll post a patch. I'm sorry to
be so absent, but my job is kicking my ass right
now.
-d.
On Tue, 19 Mar 2002 11:45:00 -0800 (PST), BALA KRISHNA wrote:
>
>Hi Alan,
>
>Thanks for the solution. My problem too got solved
>when I override the validate function in the action
>form. But I have a question:
>
>1. Why should one change the actual code for the sole
>reason of being able to test it? The application works
>fine even if "validate" method is not there.
>
>Thanks,
>balki
>
>
>--- Alan Halley wrote:
>>
>>Hi,
>>In my case the problem was caused by the default
>>validation routine in ActionForm. If this method is
>>not overidden in your code then the default method
>>returns a null and this seems to be th root of the
>>null pointer problem.
>>I overrode this method as follows:-
>>public ActionErrors validate(ActionMapping
>>mapping,HttpServletRequest request)
>>{
>>return new ActionErrors();
>>}
>>This appeared to cure the problem.
>>Alan
>>
