tools/poclean.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3 #
   4 # Copyright 2007 Zuza Software Foundation
   5 #
   6 # This file is part of translate.
   7 #
   8 # translate is free software; you can redistribute it and/or modify
   9 # it under the terms of the GNU General Public License as published by
  10 # the Free Software Foundation; either version 2 of the License, or
  11 # (at your option) any later version.
  12 #
  13 # translate is distributed in the hope that it will be useful,
  14 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16 # GNU General Public License for more details.
  17 #
  18 # You should have received a copy of the GNU General Public License
  19 # along with translate; if not, write to the Free Software
  20 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  21
  22 """Produces a clean file from an unclean file (Trados/Wordfast) by stripping
  23 out the tw4win indicators.
  24
  25 This does not convert an RTF file to PO/XLIFF, but produces the target file
  26 with only the target text in from a text version of the RTF.
  27 """
  28
  29 from translate.storage import factory
  30 from translate.misc.multistring import multistring
  31 import re
  32
  33 tw4winre = re.compile(r"\{0>.*?<\}\d{1,3}\{>(.*?)<0\}", re.M | re.S)
  34
  35 def cleanunit(unit):
  36     """cleans the targets in the given unit"""
  37     if isinstance(unit.target, multistring):
  38         strings = unit.target.strings
  39     else:
  40         strings = [unit.target]
  41     for index, string in enumerate(strings):
  42         string = string.replace("\par", "")
  43         strings[index] = tw4winre.sub(r"\1", string)
  44     if len(strings) == 1:
  45         unit.target = strings[0]
  46     else:
  47         unit.target = strings
  48
  49 def cleanfile(thefile):
  50     """cleans the given file"""
  51     for unit in thefile.units:
  52         cleanunit(unit)
  53     return thefile
  54
  55 def runclean(inputfile, outputfile, templatefile):
  56     """reads in inputfile, cleans, writes to outputfile"""
  57     fromfile = factory.getobject(inputfile)
  58
  59     cleanfile(fromfile)
  60 #    if fromfile.isempty():
  61 #        return False
  62     outputfile.write(str(fromfile))
  63     return True
  64
  65 def main():
  66     from translate.convert import convert
  67     formats = {"po":("po", runclean), "xlf":("xlf", runclean), None:("po", runclean)}
  68     parser = convert.ConvertOptionParser(formats, usetemplates=False, description=__doc__)
  69     parser.run()
  70
  71 if __name__ == '__main__':
  72     main()