3dpcp/.svn/pristine/5a/5aa77389e036f11e2cf192f2084f6636ae805b64.svn-base
2012-09-16 14:33:11 +02:00

55 lines
2.2 KiB
Text

BUGS
Reporting
---------
Please do not hesitate to report bugs, and if possible their fixes! If you
send an example file, please make sure it's small. To report bugs, do one
of the following:
* go to http://sourceforge.net/bugs/?func=addbug&group_id=7147.
This is the preferred way to report bugs.
* send it to one of the authors. Note that sometimes we may be busy, and
we won't reply it for days. If you post using the previous method, surely
one of the authors will read it.
* use
diff -ru gocr_origin/ gocr_changed/ >patch
to create patches
* if you have compiling problems, do not forget to send your configure-output
and the config.log file
Known bugs (see jocr.SF.net page too)
----------
v0.48 cutting of double melted chars will fail (example: serif MN)
v0.43 on dithered images gocr runs extremely long (seems to hang)
v0.41 linker error using g++ and netpbm under SuSE-9.3
v0.3.5
- segfault on some systems which do not support ifalpha(256+x)
- hexcode not read from database
v0.2.5 german umlauts and i-dots are not handled correctly
problems high resolution fonts
v0.2.4 I guess, there are still bugs.
Some systems do not handle stack in good manner (AmigaOS?).
gocr does extensively consume stack for recursive functions.
Therefore you can get memory protection failures or strange results.
The worst case is a huge black area. If that is a problem for you
request for changing it.
--- --- --- --- only for linux freaks --- --- ---
By mistake I programmed an endless rekursiv function and ...
SuSE6.4+linux2.2.12/13 got several "out of mem" and system CRASHED!!!
ulimit: stack=unlimited
- if text is framed, frame should be ignored, but it is not
v0.2.3 still problems with segmentation
- gcc 2.95.2 (SuSE6.4) error in load_db(), => fixed (thx to jasper)
v0.2.1
- some people have problems running gocr on DOS/Win95
I guess: stack overflow. Is someone able to analyze or fix this?
- large black areas on pbm-files cause a segfault on
Ultra/Sparc (64bit) machines running Linux (2.1.126).
There is a recursive function in the program which causes a
stack overflow, which is not detected by the linux-kernel (BUG?).
I look for a better solution.