[Axiom-developer] RE: Running Axiom in Texmacs for Windows (was: [Texmacs-dev] Plug ins)

[Page, Bill]

Dan, Mike,

On Thursday, December 02, 2004 7:02 AM Mike wrote:
> 
> I've attached some code which shows how to launch a
> subprocess with redirected stdin/out on Windows, which
> you might like to incorporate into tm_axiom.c.  I haven't
> run it for a couple of years but I recall that it
> worked when last I tried!
> ...

On Thursday, December 02, 2004 9:57 AM you Dan wrote:
> 
> After looking at tm_axiom.c, it looks like it would be very 
> easy to re-write this for windows.  You can use Mike's provided
> sample code, or you may even use my pipe.c/h which provides a
> very simple API to do this type of thing (since texmacs does
> it already).  I have attached them.
> 

Thank you both very much for your help. These examples
were very instructional and inspiring. Really. But as
Tim Daly always says: "There ain't no such thing as
a simple job!" ... :) After about a days work, however
I do have a radically revised version of the tm_axiom
program that seems to work very well under Windows. I
have attached the zip file containing the source program
and compiled code (compiled on MinGW/MSYS).

In the end I basically opted for the style and example
provided by Mike (though I am Dan's approach would have
probably worked nearly as well). The thing that I enjoyed
about Mike's version was the use of threads. It seems to
me that this "light weight" kind of process handling is
one of the things that Windows actually does very well
and so for a Windows version, I decided I would like to
learn more about this. Too my great pleasure some simple
experiments seemed to show that it was indeed faster than
the Cygwin version that uses the old Unix fork and pipe
approach. Cygwin TeXmacs users should be able to benefit
directly from this new version. But it might also be
worthwhile investigating backporting this new version of
tm_axiom (I called it tmwin_axiom) back to Unix.

The program is written in old fashioned vanilla C.

Perhaps this experience will be relevant when it comes
to porting some of the other components (graphics
hyperdoc, etc.) to Windows.

Anyway, I would be very pleased if you could give this
version a try. I would like to include it in the pre-
release Windows distribution of Axiom that we now have
on the MathAction web site.

Regards,
Bill Page.

tmwin_axiom.zip
Description: Binary data