Red Hat Bugzilla – Bug 62389
performance analysis for printing a lot of glyphs
Last modified: 2015-01-07 18:55:45 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.9) Gecko/20020314
Description of problem:
I tried printing of the following file:
that file was needed about 3 hours for outputting PS on my machine. it's
unbelievable! and PS file which is actually output can't be shown with gs
because gnome-print has a bug.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
2.run gedit with LANG=zh_CN.gb18030
3.read it and printing
Actual Results: for printing, it needs a lot times.
Expected Results: gb18030's font is too big. so I think that about 2 or 3
minutes would be better.
Created attachment 51484 [details]
fix patch is here
the above patch saved a lot of times. actually it finished printing that file in
about 2 or 3 minutes on my machine. it's reasonable, I think. it also contains
gnome-print's bug fix.
if (dx == 0 && dy == 0) return NULL;
if (dx == 0 && dy == 0) retval = g_strdup ("");
this code should not return NULL. (null) was output by this into PS file. I will
file this bug and patch to bugzilla.gnome.org too.
Tagoh you're our gnome-print expert so I'll trust your patch is correct.
Can you rebuild gnome-print? I'm going to be gone this week for guadec.
ok, I have built gnome-print with this patch. this problem will be fixed in
the improvement of printing performance patch was missed in 0.37 - that patch
has been merged in 1.x series though. so we have this problem again in recent beta.
I submitted a patch again to the upstream bugzilla. see
the upstream marked it as WONTFIX, because they don't plan to release the 0.x
series anymore. but it's apparently affected our products.
This patch was applied to our old release, and it has been merged to the GNOME
2.x releases. so can you apply it to our gnome-print?
I'm going to be away for the next few days, but if you (or someone
else) want to apply the patch to our gnome-print RPM, I'm OK with
[ Irrelevant comment -- I think the code would be a whole lot clearer
and also more efficient if was using a GString as a buffer.
I'm *not* suggesting you change this for our RPM. ]
I don't have enough time to improve it until last public beta.
original patch was applied. this problem should be fixed in 0.37-4.
The original patch is fine, thanks.