Re: shell32/tests: Fix a test on Win2k platform and above.

[Paul Vriens]

Nicolas Le Cam wrote:
> On Win2k (and according to test.winehq.org on above platforms) an
> ABN_POSCHANGED message is sent to the last window that is removed.
>
> This patch change the ok to a win_skip in this case.
>
> Tell me if you have a better solution.
>
> Nicolas Le Cam

I can see several extra notifications on W2K and below.

You are now doing things different. Previously the test was to check if info is 
not NULL. Now you are only doing the test if info is NULL.

I think a broken() would be more appropriate here. So something like:

ok(info != NULL ||
    broken(!info && !registered_count), /* W2K */
    "got unexpected ABN_POSCHANGED notification\n");

Could you check the behavior on Win9x/NT4?

I guess Vincent Povirk can comment better on the patch as he wrote the initial 
tests.

shell32/tests: Fix a test on Win2k platform and above.

[Nicolas Le Cam]

2009/3/18 Paul Vriens :
> Nicolas Le Cam wrote:
> > On Win2k (and according to test.winehq.org on above platforms) an
> > ABN_POSCHANGED message is sent to the last window that is removed.
> >
> > This patch change the ok to a win_skip in this case.
> >
> > Tell me if you have a better solution.
> >
> > Nicolas Le Cam
>
> I can see several extra notifications on W2K and below.
>
> You are now doing things different. Previously the test was to check if info
> is not NULL. Now you are only doing the test if info is NULL.
>
> I think a broken() would be more appropriate here. So something like:
>
> ok(info != NULL ||
>   broken(!info && !registered_count), /* W2K */
>   "got unexpected ABN_POSCHANGED notification\n");
>
> Could you check the behavior on Win9x/NT4?
>
> I guess Vincent Povirk can comment better on the patch as he wrote the
> initial tests.
>
> --
> Cheers,
>
> Paul.

You're right broken is a better solution.
I will redo my patch this evening and try to set up a Win98SE VM as I
don't own NT4.

shell32/tests: Fix a test on Win2k platform and above.

[Vincent Povirk]

On Wed, Mar 18, 2009 at 3:45 AM, Paul Vriens  wrote:
> I guess Vincent Povirk can comment better on the patch as he wrote the
> initial tests.

I don't think I can because I don't understand what this patch is doing.