Mon 19 Oct 2015 10:26:48 PM UTC, comment #6:
Thanks Avinoam for confirming it works.
Closing report.
|
Mon 19 Oct 2015 09:01:25 PM UTC, comment #5:
Attached foo.svg, result of
from comment #1, run with MXE-Octave, compiled with gl2ps ver. 1.3.9.
(with ver 1.3.8 it was empty figure file...)
(file #35232)
|
Sat 17 Oct 2015 05:14:32 PM UTC, comment #4:
gl2ps has been patched and the current development version produces properly scaled images. AFAIU there will be a a release of gl2ps soon [1].
I think we can close this report as it has been fixed upstream
[1] http://www.geuz.org/pipermail/gl2ps/2015/000421.html
|
Mon 28 Apr 2014 02:53:22 AM UTC, comment #3:
I've done some more testing and I'm pretty certain this is a problem with gl2ps rather than Octave. Until they do something about this images just aren't going to output well to svg.
|
Thu 28 Feb 2013 01:48:46 AM UTC, comment #2:
Alas, I'm not certain this isn't an Octave bug with the printing of 'image' data to svg. I used rectangle() to create 4 differently-colored boxes on an ordinary plot background. When I print this the svg is correct. I'm attaching that file as svg_rect_tst.m
Secondly, the white polygons are present at the beginning of every svg file. They represent the axis background. You can verify this with
Now I get the following in the svg file.
(file #27540)
|
Sun 18 Nov 2012 02:22:12 AM UTC, comment #1:
I looked at the content of the svg-file, and found that the four rectangles are being drawn as white.
This appears to be a bug or missing feature in gl2ps. The files where the bug is likely to be found are libinterp/interp-core/gl2ps.c and/or libinterp/interp-core/gl2ps.h. When I have time, I hope to take a look. In the event someeone gets to this before me, the bug is found the change should be made upstream at the link below.
http://geuz.org/gl2ps/
|
Thu 15 Nov 2012 06:58:57 PM UTC, original submission:
produces incorrect output (basically an empty figure).
|