Hi James,
I've taken a stab at this occasionally. I want to try building GCL on NT
soon with Cygwin, and if this works I shall try Maxima. Other commitments
have prevented me from doing this lately, but time is freeing up now...
BTW, I was once able to get Maxima 5.6, etc., to compile with the old
Cygwin B21. However, more recent changes by Dr. Schelter seemed to prevent
this. Just a historical note!
Dave Holmgren
----- Original Message -----
From: "James Amundson" <amundson at fnal>
To: "Mike Thomas" <miketh@brisbane.paradigmgeo.com>
Cc: "Maxima List" <maxima@www.ma.utexas.edu>; <gcl-devel@gnu.org>; "Keuchel,
Rainer" <rkeuchel@allgeier.com>
Sent: Thursday, March 21, 2002 8:16 PM
Subject: Re: [Maxima] Fw: [Gcl-devel] Re: gcl on ms windows
> Dear Mike,
>
> I am glad to hear you are working on getting Maxima to work on Windows.
> Other people on the Maxima list have discussed various Windows build
> issues. If they would care to step forward we could try to organize the
> effort.
>
> Did you use the old build system, the new build system, or something
> else? Please be aware that not everything is working in the Unix system,
> particularly the various internal path settings. I don't think xmaxima
> is working right now, either. All these things will be fixed by the time
> we release 5.9.0. In the meantime, it's worth asking on the list if the
> problems you find exist outside of Windows.
>
> --Jim
>
>
> On Tue, 2002-03-19 at 14:48, Mike Thomas wrote:
> > Hi all.
> >
> > I'm pleased to announce that I now have a native Win32 (text mode only)
> > build of Maxima 5.9 as well as GCL 2.5, both based on the latest CVS
> > sources. The build was hosted under Cygwin using the standalone Mingw32
1.1
> > gcc compiler. I have done some simple polynomial factorisation, linear
and
> > non-linear equation manipulation, linear algebra and integration to do a
> > first, successful, check of the viability of this setup.
> >
> > Thanks to Camm (various bits of help), Rainer Keuchel (for the memory
> > checking code from his WinCE/32 port of GCL and inspiration) and Dan
Stanger
> > (pointer about fasd format switch and other knowledge bootstrapping).
> >
> > In compiling Maxima, I found that there is a bug in GCL's MAKE-PATHNAME
> > function which I have not had time to fix - it doesn't handle executable
> > names with .exe extensions properly. I worked around this by using
> > STRING-RIGHT-TRIM on the result of SI::ARGV in "init_max1.lisp".
> >
> > I will be unable to do anything further on this for a couple of days,
but I
> > hope to check everything into CVS as soon as possible.
> >
> > I guess after that the next steps will be:
> >
> > - making the help system work (more path trouble I think),
> > - making binaries available for testing,
> > - reviving XMaxima,
> > - tuning the configuration process, and
> > - using MSYS rather than Cygwin to host the build.
> >
> > I make no promises on time frames for this list.
> >
> > Cheers
> >
> > Mike Thomas.
> >
> > _______________________________________________
> > Maxima mailing list
> > Maxima@www.math.utexas.edu
> > http://www.math.utexas.edu/mailman/listinfo/maxima
>
> _______________________________________________
> Maxima mailing list
> Maxima@www.math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima