On Sun, 4 Oct 2009 18:28:44 -0500 Steve M. Robbins wrote: > Hello Francesco,
Hi!
[...]
> On Sun, Oct 04, 2009 at 10:42:37PM +0200, Francesco Poli wrote:
> > On Sun, 20 Sep 2009 13:50:54 +0200 Francesco Poli (t1000) wrote:
> >
> > [...]
> > > I noticed a problem with the PostScript snapshot saving option.
> > >
> > > The option seems to work in some cases, but causes geomview to
> > > segfault in other cases.
> >
> > Hello, is there any progress on this bug?
> > Did you succeed in reproducing it?
>
> I am able to reproduce it.
Good, that's something.
>
> > How can it be fixed?
>
> I'm going to forward this to the geomview user list to see whether
> it is a known problem.
By looking at geomview-users list web archives, I see that your message
got one reply by Frank Peters. He suggests:
| A good workaround for this problem would be to save the snapshot
| as a PPM file and then convert to Postscript using pnmtops (from
| netpbm) or ImageMagick/GraphicsMagick.
Well, since I want a vector EPS file, and not a raster image converted
into EPS format, I would not define this as a *good* workaround.
Actually, I would say it's a *last resort* workaround...
Anyway, I've just tried with the three different PPM choices.
a) PPM software snapshot fails just like PostScript snaphot
(segfaults and produces an empty PPM file)
b) PPM Screen snapshot seems to work (but it obviously produces a
raster image, which looks terrible, when converted to EPS format)
c) PPM GLX off-screen snapshot generates strange errors, but
otherwise seems to work (it produces a raster image *identical* to
the previous one)
The errors generated by PPM GLX off-screen snapshot are the following:
X Error of failed request: BadDrawable (invalid Pixmap or Window parameter)
Major opcode of failed request: 136 (DRI2)
Minor opcode of failed request: 4 (DRI2DestroyDrawable)
Resource id in failed request: 0x30001b4
Serial number of failed request: 5507
Current serial number in output stream: 5510
The three files are:
$ ls -1 --size --si test*ppm
615k test_offscreen.ppm
615k test_screen.ppm
0 test_software.ppm
$ diff test_offscreen.ppm test_screen.ppm -sq
Files test_offscreen.ppm and test_screen.ppm are identical
[...]
> A patch is most welcome.
Frankly speaking, I feel a bit discouraged from spending time in trying
to understand the geomview code base, find the bug and prepare a patch
to fix it.
I fear that it will end up like in bug #357445: that bug was filed back
in 2006; after some discussions you proposed a patch to the
geomview-users list, your patch was criticized because it lacked
back-compatibility, I proposed a back-compatible amendment, but nobody
replied since 2007. I raised the issue again on the geomview-users
list on last August: I was suggested (by Frank Peters) to redirect my
question to the geomview-devel list, which I did on August, 19th.
Well, so far mine is the *only* message on that list, since July 2007!
I am beginning to suspect that I am the only subscriber on
geomview-devel, which is kinda awkward, as I am *not* a geomview
developer! :-(
It really seems that patches are *not* welcome... :-(
At least, not to upstream authors.
I really think that the geomview Debian package should begin
integrating patches, without waiting for upstream to accept them.
Maybe geomview upstream authors will wake up in the future, but, in the
meanwhile, Debian users should not wait for ages to see bugs fixed!
Steve, please, do not take this badly.
I am not ranting at you: I am just frustrated, as I see a good and
useful piece of Free Software almost left abandoned: I would rather
*avoid* seeing this package removed from Debian.
So please, try hard to find some time (and love) to dedicate to this
package, as it really deserves it!
Could you please start by fixing #357445, which should not be too
time-consuming?
--
New location for my website! Update your bookmarks!
http://www.inventati.org/frx
..................................................... Francesco Poli .
GnuPG key fpr == C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4
pgpGMvcjk1xJt.pgp
Description: PGP signature
