[haiku-appserver] Re: partly paiting

On 2005-10-23 at 14:12:03 [+0200], Axel Dörfler wrote:
> Adi Oanca <adioanca@xxxxxxxxx> wrote:
> > On 10/17/05, Axel Dörfler <axeld@xxxxxxxxxxxxxxxx> wrote:
> > > >       I agree sending one message only. Let's stay with this
> > > > solution and
> > > > see if all works OK. I just am afraid of ::DetachCurrentMessage()
> > > > because, if this method is called from ::FrameMoved/Resized() the
> > > > user will be dumped into debugger with a seg fault; more checking
> > > > is
> > > > required for "case B_VIEW_RESIZED/MOVED:"
> > > I don't think you have to, as I don't think there is any point in
> > > using
> > > this method from these methods - and if you really think otherwise,
> > > we
> > > could still emulate the single message approach, as Stephan
> > > proposed.
> > > I don't think it's used that often anyway in any function that
> > > deals
> > > with BView messages.
> >    You may never know what the user wants to do. What it matters is
> > that we have a bug and it should be fixed.
> 
> It's not really a bug - it's a deviation from standard behaviour that
> should just be documented. I cannot imagine anyone wanting to use
> DetachCurrentMessage() in that case, anyway.
> 
> But even if, we could still have some code like:
> current = CurrentMessage();
> FrameResized();
> if (current != CurrentMessage())
>     /* do whatever is appropriate */
> 
> >    How are you going to emulate a single message? ::CurrentMessage()
> > would return the original message anyway.
> 
> And we can't change this, having the source and all?

Should be fairly easy. BLooper::CurrentMessage() returns fLastMessage, 
which one can simply set to the "fake delivered" message in question. If 
that message is created on the heap, even DetachCurrentMessage() should 
work just fine.

CU, Ingo

Other related posts: