Re: We need OpenGL bindings.

["Bruce Detter" <bruce.detter@lmco.com>]

Mr. Richards,
Thanks for the response.  I will check out SGI further, but Mr. Dewars
indicates that the SGI bindings may be for their development environment and
compiler.  A quick search of the SGI site implies the same.  But they may be
open for suggestions.  Here is what I changed (or will change) in the GLX
bindings (the GL and GLU bindings have been fine as is).

in the GLX spec and body files I deleted the following section as Mesa
specific
****************************************************************************
*************
--
-- Mesa GLX Extensions
--

function GLX_Create_GLX_Pixmap_Mesa (Dpy    : in Display_Pointer;
                                     Visual : in X_Visual_Info;
                                     Pixmap : Pixmap_ID;
                                     CMap   : Colormap_ID)
         return GLX_Pixmap;

function GLX_Release_Buffers_Mesa (Dpy : in Display_Pointer;
                                   W   : in Window_ID)
         return boolean;
****************************************************************************
************

the following functions/procedures were just stubbed out and I created the
body's for them.
GLX_Choose_Visual
GLX_Make_Current
GLX_Create_GLX_Pixmap
GLX_Query_Version
GLX_Is_Direct
GLX_Get_Config
GLX_Query_Extensions_String
GLX_Query_Server_String
GLX_Get_Client_String

I had a problem with the pragma inlines in the body files, my compiler (OC
Systems Powerada) just rejected them so I deleted them.  I am also having
trouble with the GLX_Config type getting the compiler to accept the EDT
(GLX_Doublebuffer etc.) as entries in a GLX_GL_Attribute_Array (I just hard
coded the values for now).

I am by no means an expert and have probably over looked a few obvious
solutions, but everything is working at this early stage.  Any help you can
provide will be much appreciated (and an honor).  If you have any questions
or comments you can post or e-mail me.

Thanks again...
BCD3

Chip Richards <chipr@niestu.com> wrote in message
news:7frd8c$juv@tomquartz.niestu.com...
> In article <7fla6s$b64@svlss.lmms.lmco.com>,
> "Bruce Detter" <bruce.detter@lmco.com> writes:
>
> > Yes, they gave me the URL (http://www.niestu.com/languages/oglada) but
we
> > ran into the problems I noted in the last message.  OC Systems is still
> > trying to find some, but I think they are searching the net just like I
am.
>
> As the nominal author and maintainer of those bindings, let me express my
> regret that they aren't meeting your needs.
>
> > They need to be for OpenGL, not Mesa.
>
> I am nearly certain that SGI has such bindings for sale.  There are, so I
am
> told, also some fairly complete(?) and presumably tested(?) OpenGL (not
Mesa)
> bindings as part of the larger Win32 API bindings, if Microsoft is your
> persuasion.
>
> I would, however, be curious to hear what Mesa-specific code exists in the
> NiEstu bindings that make them unsuitable for your needs.  I'd like to
limit
> or eradicate that sort of thing.
>
> > We would prefer to use complete and tested bindings if they can be
found.
>
> Man, me too!  But there weren't any that I could find, so I took those
> bindings as far as I needed them to go at the time.  Thus parts of them
(like
> glX) are still incomplete.  Like most open-source software, they scratch a
> particular set of itches and may leave others unabraded.  I'd gladly
> incorporate any suitable changes or enhancements that you or your team
needed,
> but it sounds like you are looking for a more, hm, commercial solution.
>
> But if you can't find what you are looking for, building on our foundation
> might be better than starting totally from scratch.  If you should find
> yourself considering such a course of action, I would very much appreciate
> keeping a single code base if that's at all possible.  So keep in touch.
>
> > Any help will be greatly appreciated...
>
> Ditto.  You are welcome to join the OGLADA-L mailing list (details on the
> above-mentioned somewhat outdated web page) long enough to ask your
question
> there (or forever, if you wish), or else I can post it there for you and
we
> can see if any of the subscribers happen to have any relevant information.
>
> --
> Chip