[MLton-user] _ffi and Windows stdcall symbols

Anoq of the Sun anoq@HardcoreProcessing.com
Tue, 22 Jul 2003 05:20:51 +0300


Stephen Weeks wrote:
> 
> For other readers of MLton-user, Mike is attempting to port mosmlgl,
> http://www.home.gil.com.au/~mthomas/, a Moscow ML binding to OpenGL,
> so that it works with MLton.

Interesting :)


\begin{off-topic-remark-that-I-cannot-keep-to-myself}

The "real" API to use for 3D graphics is still RenderMan though ;)

It's like C vs. ML or something - ML is really nice but a well-kept
secret, so people just use C. RenderMan is like ML if OpenGL is like C ;)

And of course there aren't really any available real-time RenderMan
implementations ATM. :(

And my ML implementation (which is likely to become a
MLton-binary-library only) will not be released in the
very near future either :(

\end{off-topic-remark-that-I-cannot-keep-to-myself}



> With 20030720, the following program now compiles on Cygwin.
> 
>         val glClear = _import "glClear" stdcall: word -> unit;
>         val _ = glClear 0w0

Speaking of Windows and MLton... I'm going for vacation tomorrow
until Sunday and after that I will REALLY port MLton to MinGW.
I have survived 2 harddisk crashes and a lot of exams, and I
now have the time required to do this port (finally!) :)


Cheers
-- 
http://www.HardcoreProcessing.com