Re: [MiNT] WM_REDRAW

[OL <o.l@lutece.net>]

Hello,
> Am Dienstag, den 27.11.2012, 09:12 +0100 schrieb "J. F. Lemaire" 
> <jflemaire@skynet.be>:
>
> > I'm not sure I understood what you wanted. You're talking about
> > auto-sent WM_REDRAW messages, right?
>
> Nope, probaly the opposite. Of course I also care about aut-sent 
> messages, but
> I was mostly thinking about sending WM_REDRAWS to my own application 
> because maybe the AES
> can handle the list of redraw areas better than I do with my internal 
> list
> (read the previous answer to OL).
>
> Netsurf requests redraws via callback... currently I add them to an 
> internal list and process
> the redraw when the event loop has no more events to process.
> But maybe it would be better to send an WM_REDRAW... at least that 
> would make the event loop
> more "standard". But I only would do that when every AES is ready to 
> handle a bunch of WM_REDRAW'S
> in an optimized fashion. I don't want to get 2Messages for 2 
> overlapping regions.... because each redraw
> results in an wind_get(WF_FIRSRXYWH...) while{ wind_get(WF_NEXTXYWH) 
> ... }
Really I not understand your problem, if you need update an area, of 
course you not need send WM_REDRAW to yourself this a lost of time, AES 
not do anything of it it will just transfer  the message to yourself, 
now in past I use it because sometime the source code start to be a 
little bit complicated and send a message is a good way to simplify it.
You need on only use wind_get(WF_FIRSTXYWH or WF_NEXTXYWH) or better 
wind_get(WF_FIRSTAREAXYWH or WF_NEXTXYWH)



Olivier
> > You're trying to figure out the
> > best way to force a redraw of your application's work area in the most
> > efficient way?
>
> right ;)
>
> Greets,
> Ole