Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!

Bug 462224

Summary: app-text/cuneiform-1.1.0-r1: fails to run: PUMA_XFinalrecognition failed.
Product: Gentoo Linux Reporter: Juergen Rose <rose>
Component: Current packagesAssignee: No maintainer - Look at https://wiki.gentoo.org/wiki/Project:Proxy_Maintainers if you want to take care of it <maintainer-needed>
Status: RESOLVED TEST-REQUEST    
Severity: normal CC: jstein, nikoli, proxy-maint, rich0, yaleks
Priority: Normal    
Version: unspecified   
Hardware: All   
OS: Linux   
URL: https://bugs.launchpad.net/cuneiform-linux/+bug/1156727
Whiteboard:
Package list:
Runtime testing required: ---
Bug Depends on: 462366    
Bug Blocks:    

Description Juergen Rose 2013-03-18 15:51:27 UTC
Cuneiform fails with:

rose@leopard:/home_leopard/rose/Txt/projects/IVD_soeren(76)$ identify page-000.jpg 
page-000.jpg JPEG 2321x3399 2321x3399+0+0 8-bit sRGB 2.926MB 0.160u 0:00.169
rose@leopard:/home_leopard/rose/Txt/projects/IVD_soeren(77)$ cuneiform -l ger -f hocr -o page-000.hocr page-000.jpg 
Cuneiform for Linux 1.1.0
PUMA_XFinalrecognition failed.

If I convert the jpeg file to a bmp file, I get the same result.
I found an old discussion with the same subject at 
https://bugs.launchpad.net/cuneiform-linux/+bug/349110 .
Comment 1 Juergen Rose 2013-03-18 15:56:40 UTC
A rather current description of the bug with the same version of cuneiform is at http://debian.2.n7.nabble.com/Bug-687089-cuneiform-Always-terminating-with-PUMA-XFinalrecognition-failed-td1102565.html.
Comment 2 Juergen Rose 2013-03-18 16:03:50 UTC
The same issue happens if I convert the image to tiff and if I try other outputformats, as smarttext or text.
Comment 3 Juergen Rose 2013-03-18 17:45:33 UTC
I sent this bug also upstream at https://bugs.launchpad.net/cuneiform-linux/+bug/1156727 .
Comment 4 Juergen Rose 2013-03-19 16:38:29 UTC
If I handle the same input files with cuneiform-1.1.0 from Linux Mint, it works like a charm.
Comment 5 Aleksandr Yakimov 2013-03-25 17:30:53 UTC
(In reply to comment #4)
> If I handle the same input files with cuneiform-1.1.0 from Linux Mint, it
> works like a charm.

Please attach this file. This may be a bug in imagemagick.
Comment 6 Juergen Rose 2013-03-26 00:33:36 UTC
The file is at http://141.89.194.24/~rose/page-000.jpg.
Comment 7 Aleksandr Yakimov 2013-03-26 16:25:39 UTC
(In reply to comment #6)
> The file is at http://141.89.194.24/~rose/page-000.jpg.

This one works fine on my hosts. Have you tried to rebuild imagemagick&cuneiform? Please try with graphicsmagic (as in ubuntu builds).
Comment 8 Juergen Rose 2013-03-26 17:01:32 UTC
Even after a fresh rebuild of imagemagick and cuneiform (still with imagemagick USE flag) I get the same issue:

rose@impala:/home_impala/rose(4)$ ll page-000.*
-rw-r----- 1 rose rose   59716 26. Mär 01:23 page-000.hocr
-rw-r----- 1 rose rose 2925608 26. Mär 01:06 page-000.jpg
rose@impala:/home_impala/rose(5)$ identify page-000.jpg
page-000.jpg JPEG 2321x3399 2321x3399+0+0 8-bit sRGB 2.926MB 0.210u 0:00.209
rose@impala:/home_impala/rose(6)$ emerge -pvD cuneiform

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild   R   #] app-text/cuneiform-1.1.0-r1::sabayon  USE="imagemagick -debug -graphicsmagick" 0 kB

Total: 1 package (1 reinstall), Size of downloads: 0 kB
rose@impala:/home_impala/rose(7)$ cuneiform -l ger -f hocr -o page-000.hocr page-000.jpg 
Cuneiform for Linux 1.1.0
PUMA_XFinalrecognition failed.
rose@impala:/home_impala/rose(8)$ genlop -t imagemagick | tail

     Tue Mar 12 09:36:56 2013 >>> media-gfx/imagemagick-6.7.9.9
       merge time: 3 minutes and 4 seconds.

     Tue Mar 12 15:23:17 2013 >>> media-gfx/imagemagick-6.7.9.9
       merge time: 3 minutes and 5 seconds.

     Tue Mar 26 17:43:12 2013 >>> media-gfx/imagemagick-6.7.9.9
       merge time: 3 minutes and 3 seconds.

rose@impala:/home_impala/rose(9)$ genlop -t cuneiform | tail

     Tue Jul 31 11:01:45 2012 >>> app-text/cuneiform-1.1.0-r1
       merge time: 3 minutes and 25 seconds.

     Tue Mar 26 01:01:37 2013 >>> app-text/cuneiform-1.1.0-r1
       merge time: 3 minutes and 38 seconds.

     Tue Mar 26 17:45:07 2013 >>> app-text/cuneiform-1.1.0-r1
       merge time: 1 minute and 6 seconds.

rose@impala:/home_impala/rose(10)$ ll page-000.*
-rw-r----- 1 rose rose   59716 26. Mär 01:23 page-000.hocr
-rw-r----- 1 rose rose 2925608 26. Mär 01:06 page-000.jpg



Then I rebuilt graphicsmagick and built cuneiform with graohicsmagick USE flag:

rose@impala:/home_impala/rose(14)$ genlop -t graphicsmagick | tail

     Sat Mar  2 11:45:44 2013 >>> media-gfx/graphicsmagick-1.3.17
       merge time: 2 minutes and 8 seconds.

     Tue Mar 12 14:45:28 2013 >>> media-gfx/graphicsmagick-1.3.17
       merge time: 2 minutes and 8 seconds.

     Tue Mar 26 17:57:14 2013 >>> media-gfx/graphicsmagick-1.3.17
       merge time: 2 minutes and 2 seconds.

rose@impala:/home_impala/rose(15)$ genlop -t cuneiform | tail

     Tue Mar 26 17:45:07 2013 >>> app-text/cuneiform-1.1.0-r1
       merge time: 1 minute and 6 seconds.

     Tue Mar 26 17:54:00 2013 >>> app-text/cuneiform-1.1.0-r1
       merge time: 1 minute and 5 seconds.

     Tue Mar 26 17:58:45 2013 >>> app-text/cuneiform-1.1.0-r1
       merge time: 1 minute.

rose@impala:/home_impala/rose(16)$ emerge -pvD cuneiform

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild   R   #] app-text/cuneiform-1.1.0-r1::sabayon  USE="graphicsmagick -debug -imagemagick" 0 kB

Total: 1 package (1 reinstall), Size of downloads: 0 kB



And now cuneiform does not complain:

rose@impala:/home_impala/rose(17)$ cuneiform -l ger -f hocr -o page-000.hocr page-000.jpg 
Cuneiform for Linux 1.1.0
rose@impala:/home_impala/rose(18)$ echo $?
0

rose@impala:/home_impala/rose(20)$ ll page-000.*
-rw-r----- 1 rose rose   59699 26. Mär 18:00 page-000.hocr
-rw-r----- 1 rose rose 2925608 26. Mär 01:06 page-000.jpg
Comment 9 Aleksandr Yakimov 2013-03-26 18:42:13 UTC
OK, are you agree to assign this issue to imagemagick-related?

We should mask imagemagick useflag on amd64 hosts.
Comment 10 Richard Freeman gentoo-dev 2013-03-26 18:56:11 UTC
(In reply to comment #9)
> OK, are you agree to assign this issue to imagemagick-related?
> 
> We should mask imagemagick useflag on amd64 hosts.

Seems to be the best solution to me.  

We probably would also want to force the graphicsmagick flag as well, unless we want to make it the default on x86 as well.
Comment 11 Jonas Stein gentoo-dev 2017-07-09 09:02:30 UTC
Is this fixed in the meantime, what is the status now?