[sldev] [VIEWER] OpenGL clipping in the GUI

Lawson English lenglish5 at cox.net
Sun Sep 9 10:26:34 PDT 2007


Callum Lerwick wrote:
> On Sun, 2007-09-09 at 06:37 -0700, Lawson English wrote: 
>   
>> In the long run, if possible, I think that the GUI needs to be revised 
>> from top to bottom in order to enforce a traditional GUI view hierarchy, 
>> including inherited clipping rectangles and even regions. I can see no 
>> reason why we can't have round windows if we want them for some purpose. 
>> Certainly, if the client-world interface converges the way I think it 
>> could (and probably should), I can foresee plenty of situations where 
>> the interface should be more organic in appearance which would require 
>> "windows" that have no set-in-stone shape, and a clip-view-hierarchy 
>> would be far more important in such a situation.
>>     
>
> I'd like to see slviewer use Cairo for... well every last thing it
> possibly can. Rendering GUI widgets and, hey... maybe Cairo can solve
> all our text rendering problems!
>
> http://cairographics.org/
> http://cairographics.org/samples/
>
> And the whiz-bang OpenGL example page:
>
> http://cairographics.org/OpenGL/
>
> Cairo's text rendering is based on Pango:
>
> http://www.pango.org/
>
> Which nicely wraps the use of native fonts on Window and OSX, and
> fontconfig on X11.
>
> I'd put code where my mouth is on this but I'm really doing plenty of
> that as it is. :) (OpenJPEG, OpenAL...)
>   
Cairo certainly seems useful, and it is crossplatform, open source, and 
so on. As a MacOS X user, I'm kinda uncomfortable with the lack of true 
ATSUI support, but Apple chose to orphan its superior font technology 
during the NeXT takeover (probably in return for support from Adobe in 
the backroom political battles) and if Apple won't support a technology, 
why should open source people who have probably never even  heard of it? 
The fact that MacOS X support  itself is an afterthought merely reflects 
market forces. AFterall, everyone knows that GTK is used by more 
consumers... ;-)


ALl that aside, Cairo, DOES seem like a good thing to use. What 
specifically have been reasons why the LIndens aren't interested? Lack 
of resources? Lack of experience? Time and energy already spent on a 
different system?

What does using Cairo give them that using OpenGL doesn't? Pango 
text-handlign  is a separate issue, according to the website hype...







More information about the SLDev mailing list