octave bug with replot?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

octave bug with replot?

mhagger-2
Hi,

I'm having some problems with octave, basically I have a program called
"print.m" which does the following:

gset terminal postscript color
gset output "file.ps"
replot
gset terminal x11

Now, with octave-2.0.9 all was fine, however with octave-2.0.11 the
resulting postscript file is a bit strange and comprises a blank page
followed by the required plot.  It seems that "replot" is the cause of
this problem, since if the replot is replaced by the original plot command
then all is fine.

Any thoughts?

Mark



Reply | Threaded
Open this post in threaded view
|

octave bug with replot?

John W. Eaton-6
On 12-Mar-1998, [hidden email] <[hidden email]> wrote:

| I'm having some problems with octave, basically I have a program called
| "print.m" which does the following:
|
| gset terminal postscript color
| gset output "file.ps"
| replot
| gset terminal x11
|
| Now, with octave-2.0.9 all was fine, however with octave-2.0.11 the
| resulting postscript file is a bit strange and comprises a blank page
| followed by the required plot.  It seems that "replot" is the cause of
| this problem, since if the replot is replaced by the original plot command
| then all is fine.
|
| Any thoughts?

In trying to avoid one gnuplot bug, I ran into another.

The quick fix is to change the definition of GPLOT_CMD_REPLOT in
src/pt-plot.cc back to "rep" instead of "cle;rep" and rebuild Octave.
But that will bring back the problem with replot in multiplot mode
that I was trying to avoid.  Perhaps a better fix is the patch I just
posted to http://www.che.wisc.edu/octave/2.0.11-replot-patch (there is
also a link on the News page).

Thanks,

jwe