On Thu, 16 Mar 2000, Rodolphe Ortalo wrote:
>
> On Wed, 15 Mar 2000, Jon M. Taylor wrote:
> > http://tanuki.dhs.org/libxmi.tar.gz
>
> I have a compilation problem with the version above:
I have reimplemented LibXMI on top of the LibEDemo framework, and
that solved the problem. Now I can hook the API functions properly.
I'll clean up the new tree and release a new tarball tonight.
> > No pixels on the screen yet, but we are getting closer.
>
> I see. Btw, why did you define a display-x target immediately ?
Placeholder, and LibEDemo uses X as one of its sample targets.
> Isn't it
> more a 'LibGGI' display target ?
Sure, a LibGGI target will probably be the second accelerated
target which is written (and the most useful...). However, since the
LibXMI API is so close to the X API, I think it will be a _LOT_ easier and
quicker to prototype using a X target first. Also, there will be many
other internal APIs besides the API which corresponds to the LibXMI API,
which will be exported to the targets (span rendering, buffer typing and
composition, ROPs, etc), and those APIs as well can be better prototyped
on an X target.
> (Unless you intend to use X-specific
> functions later on to access directly the X windows buffer ? Wouldn't it
> be as simple to use a GGI DirectBuffer ?)
All of that is target-specific.
> > stubs/*.c. I wasted hours on this last night and finally gave up for the
> > night. A fresh pair of eyeballs would help out a lot here.
>
> I'll try to understand the problem. But I have to figure out the overall
> thing first (and you may find the problem as soon as you wake up...:-))
Wait for the new tarball.
> I'm very pleased to see you starting this one... Thank you very much.
> I'm willing to help if possible as soon as you stabilize the framework...
Thanks. Hopefully stabilization will happen soon |->.
Jon
---
'Cloning and the reprogramming of DNA is the first serious step in
becoming one with God.'
- Scientist G. Richard Seed
