The input file is temporarily available at
http://fleksem.klf.uw.edu.pl/~jsbien/tmp/in.djvu.

Best regards

JSB

ocrodjvu --engine cuneiform --language pol --clear-text -o out.djvu in.djvu
Processing 'in.djvu':
- Page #1
PUMA_XFinalrecognition failed.
Exception in thread Thread-2:
Traceback (most recent call last):
  File "/usr/lib/python2.5/threading.py", line 486, in __bootstrap_inner
    self.run()
  File "/usr/lib/python2.5/threading.py", line 446, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/share/ocrodjvu/lib/_ocrodjvu.py", line 424, in page_thread
    result = self.process_page(page)
  File "/usr/share/ocrodjvu/lib/_ocrodjvu.py", line 391, in process_page
    with self._engine.recognize(pfile, language=self._options.language, details=self._options.details) as result:
  File "/usr/share/ocrodjvu/lib/_ocrodjvu.py", line 210, in recognize
    return cuneiform.recognize(pbm_file, language)
  File "/usr/share/ocrodjvu/lib/cuneiform.py", line 72, in recognize
    worker.wait()
  File "/usr/share/ocrodjvu/lib/ipc.py", line 58, in wait
    raise CalledProcessError(return_code, self.__command)
CalledProcessError: Command 'cuneiform' returned non-zero exit status 1

clone 572081 -1
retitle 572081 ocrodjvu: assumes that foreground/background separation is correct
reassign -1 cuneiform 0.7.0+dfsg-5
retitle -1 cuneiform: fails with an unhelpful error message on empty pages
thanks

* jsbien@mimuw.edu.pl <jsbien@mimuw.edu.pl>, 2010-03-01, 14:46:

There are two problems here:

1. ocrodjvu assumes that foreground/background separation has been done
correctly and feeds an OCR engine with page masks only. However, in case
of this document all page masks are solid white.

2. Cuneiform fails (with an unhelpful error message) on a completely
blank page.