Re: [NSApp stopModalWithCode:0];

[Marc Bruenink]

Fred Kiefer wrote:
> Marc Brünink schrieb:
>
> > if I send a [NSApp stopModalWithCode:0]; while I'm not in a modal
> > session I get a NSInvalidArgumentException. This is not engraved in the
> > Openstep Spec and behaves different in Cocoa. So I think this should be
> > changed. Anyone disagree? The bad thing about this one is that I do not
> > get the exception but a panel. So I am not able to catch it. Or......
> > I'm too sleepy. :-)
>
>
> I think you are just wrong here. stopModelWithCode: should only be
> called from within a modal session. It would not do much harm to not
> raise an exception here, but some errors may go undetected then. You
> should be glad that GNUstep tells you that your code is wrong, where
> Cocoa seems to leave you in the dark.

Its just not stated anywhere. Neither the spec nor the documentation 
mentioned an exception. At least as far as I had read them :-)

> I don't have an idea what you are trying to do, but the documentation
> states clearly that you should not call this method from outside the
> event processing of a modal window. If you really need to call this
> method you could work around by checking [NSApp modalWindow]first. But I
> think this is wrong, but I cannot tell you a correct solution without
> knowing the actual problem you are going to solve.
No. Its not about working around at all. This would not be a problem. 
Its about understanding the behavior of GNUstep :-)
So I investigated a bit further. And now it starts to become really 
weird. Ive the following code:

   NS_DURING
   [NSApp stopModalWithCode:0];
   NS_HANDLER
   NSLog(@"localException> %@", localException);
   NS_ENDHANDLER

Which works as expected.
However as soon as I put this code into windowWillClose: it does not 
work at all. The exception is caught, but the alert panel is raised 
nevertheless.


Ciao
Marc