Bug#710547: wxmaxima, maxima, gcl error(s) in relation to bug #593548

Patrick Meehan pm.vmail at virginmedia.com
Fri May 31 18:39:53 UTC 2013


Package: wxmaxima
Version: 12.04.0-1
Severity: important

Package: maxima
Version: 5.27.0-3
Severity: important

Package: gcl
Version: 2.6.7+dfsga-1
Severity: important

Hello,
        I have experienced several problems with the wxMaxima package. The
first problem is connected with an already reported bug; #593548.
[W]xmaxima produces a 'the gui has lost connection to the maxima process'
error when running the following code:-

load(fractals);
n:1000000;
plot2d([discrete, fernfale(n)], [style, dots]);

If I attempt to run this example code from the command line, an error
similar to the following is produced:-

'gazonk_12794.lsp has failed to compile in /tmp'

Also, when running the following maxima code:-

load(fractals);
plot3d(mandelbrot_set, [x, -2.5, 1], [y, -1.5, 1.5], [gnuplot_preamble,
"set view map"], [gnuplot_pm3d, true], [grid, 250, 250]);

compilation succeeds, but maxima gets stuck in an inifinite loop or
iteration. I have left this code running for 30+ minutes with no sign of a
mandelbrot plot.


The next problem I ran into concerns the use of parametric equations. When
trying to produce a 3D plot of the 'Kuen' geometric shape, using the
following code:-

expr_1:(2*(cos(s)+s*sin(s))*sin(t))/(1+s^2*sin(t)^2);
expr_2:(2*(sin(s)-s*cos(s))*sin(t))/(1+s^2*sin(t)^2);
expr_3:log(tan(t/2.0))+((2*cos(t))/(1+s^2*sin(t)^2));
plot3d([expr_1, expr_2, expr_3], [s, -4.5, 4.5], [t, 0, %pi], [grid, 50,
50]);

the x & y axes appear to be correct, but the z axis values are totally
incorrect.

I have run all of the above example code using wXmaxima on Fedora 18, which
uses SBCL (Steel Bank Common Lisp), with no problems.
Also, I have sucessfully produced a 'Kuen' plot using MaximaOnAndroid,
which uses ECL (Embedded Common Lisp).

The conclusion that I have drawn from my testing is that the lisp backend,
in particular 'gcl', is the culprit. I therefore downloaded and installed
the upstream maxima source code tarball with SBCL as the lisp backend. I
used imaxima within Emacs and all of the above code sucessfully executes.




                                              Best Regards,
                                                                   Patrick
Meehan.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/debian-science-maintainers/attachments/20130531/9b343b52/attachment.html>


More information about the debian-science-maintainers mailing list