convert/txt2po.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3 #
   4 # Copyright 2004-2006 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
  23 """convert plain text (.txt) files to Gettext PO localization files
  24
  25 See: http://translate.sourceforge.net/wiki/toolkit/txt2po for examples and
  26 usage instructions
  27 """
  28
  29 from translate.storage import txt
  30 from translate.storage import po
  31
  32 class txt2po:
  33     def __init__(self, duplicatestyle="msgctxt"):
  34         self.duplicatestyle = duplicatestyle
  35
  36     def convertstore(self, thetxtfile):
  37         """converts a file to .po format"""
  38         thetargetfile = po.pofile()
  39         targetheader = thetargetfile.makeheader(charset="UTF-8", encoding="8bit")
  40         targetheader.addnote("extracted from %s" % thetxtfile.filename, "developer")
  41         thetargetfile.addunit(targetheader)
  42         for txtunit in thetxtfile.units:
  43             newunit = thetargetfile.addsourceunit(txtunit.source)
  44             newunit.addlocations(txtunit.getlocations())
  45         thetargetfile.removeduplicates(self.duplicatestyle)
  46         return thetargetfile
  47
  48 def converttxt(inputfile, outputfile, templates, duplicatestyle="msgctxt", encoding="utf-8", flavour=None):
  49     """reads in stdin using fromfileclass, converts using convertorclass, writes to stdout"""
  50     inputstore = txt.TxtFile(inputfile, encoding=encoding, flavour=flavour)
  51     convertor = txt2po(duplicatestyle=duplicatestyle)
  52     outputstore = convertor.convertstore(inputstore)
  53     if outputstore.isempty():
  54         return 0
  55     outputfile.write(str(outputstore))
  56     return 1
  57
  58 def main(argv=None):
  59     from translate.convert import convert
  60     from translate.misc import stdiotell
  61     import sys
  62     sys.stdout = stdiotell.StdIOWrapper(sys.stdout)
  63     formats = {"txt":("po", converttxt), "*":("po", converttxt)}
  64     parser = convert.ConvertOptionParser(formats, usepots=True, description=__doc__)
  65     parser.add_option("", "--encoding", dest="encoding", default='utf-8', type="string",
  66             help="The encoding of the input file (default: UTF-8)")
  67     parser.passthrough.append("encoding")
  68     parser.add_option("", "--flavour", dest="flavour", default="plain",
  69             type="choice", choices=["plain", "dokuwiki", "mediawiki"],
  70             help="The flavour of text file: plain (default), dokuwiki, mediawiki",
  71             metavar="FLAVOUR")
  72     parser.passthrough.append("flavour")
  73     parser.add_duplicates_option()
  74     parser.run(argv)
  75