convert/po2html.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 Gettext PO localization files to HTML files
  24
  25 see: http://translate.sourceforge.net/wiki/toolkit/po2html for examples and
  26 usage instructions
  27 """
  28
  29 from translate.storage import po
  30 try:
  31     import textwrap
  32 except:
  33     textwrap = None
  34
  35 try:
  36     import tidy
  37 except:
  38     tidy = None
  39
  40 class po2html:
  41     """po2html can take a po file and generate html. best to give it a template file otherwise will just concat msgstrs"""
  42     def __init__(self, wrap=None):
  43         self.wrap = wrap
  44
  45     def wrapmessage(self, message):
  46         """rewraps text as required"""
  47         if self.wrap is None:
  48             return message
  49         return "\n".join([textwrap.fill(line, self.wrap, replace_whitespace=False) for line in message.split("\n")])
  50
  51     def convertstore(self, inputstore, includefuzzy):
  52         """converts a file to .po format"""
  53         htmlresult = ""
  54         for inputunit in inputstore.units:
  55             if inputunit.isheader():
  56                 continue
  57             if includefuzzy or not inputunit.isfuzzy():
  58                 htmlresult += self.wrapmessage(inputunit.target) + "\n" + "\n"
  59             else:
  60                 htmlresult += self.wrapmessage(inputunit.source) + "\n" + "\n"
  61         return htmlresult.encode('utf-8')
  62
  63     def mergestore(self, inputstore, templatetext, includefuzzy):
  64         """converts a file to .po format"""
  65         htmlresult = templatetext.replace("\n", " ")
  66         if isinstance(htmlresult, str):
  67             #TODO: get the correct encoding
  68             htmlresult = htmlresult.decode('utf-8')
  69         # TODO: use the algorithm from html2po to get blocks and translate them individually
  70         # rather than using replace
  71         for inputunit in inputstore.units:
  72             if inputunit.isheader():
  73                 continue
  74             msgid = inputunit.source
  75             msgstr = None
  76             if includefuzzy or not inputunit.isfuzzy():
  77                 msgstr = self.wrapmessage(inputunit.target)
  78             else:
  79                 msgstr = self.wrapmessage(inputunit.source)
  80             if msgstr.strip():
  81                 # TODO: "msgid" is already html-encoded ("&" -> "&amp;"), while
  82                 #   "msgstr" is not encoded -> thus the replace fails
  83                 #   see test_po2html.py in line 67
  84                 htmlresult = htmlresult.replace(msgid, msgstr, 1)
  85         htmlresult = htmlresult.encode('utf-8')
  86         if tidy:
  87             htmlresult = str(tidy.parseString(htmlresult))
  88         return htmlresult
  89
  90 def converthtml(inputfile, outputfile, templatefile, wrap=None, includefuzzy=False):
  91     """reads in stdin using fromfileclass, converts using convertorclass, writes to stdout"""
  92     inputstore = po.pofile(inputfile)
  93     convertor = po2html(wrap=wrap)
  94     if templatefile is None:
  95         outputstring = convertor.convertstore(inputstore, includefuzzy)
  96     else:
  97         templatestring = templatefile.read()
  98         outputstring = convertor.mergestore(inputstore, templatestring, includefuzzy)
  99     outputfilepos = outputfile.tell()
 100     outputfile.write(outputstring)
 101     return 1
 102
 103 def main(argv=None):
 104     from translate.convert import convert
 105     from translate.misc import stdiotell
 106     import sys
 107     sys.stdout = stdiotell.StdIOWrapper(sys.stdout)
 108     formats = {("po", "htm"):("htm", converthtml), ("po", "html"):("html", converthtml), ("po", "xhtml"):("xhtml", converthtml), ("po"):("html", converthtml)}
 109     parser = convert.ConvertOptionParser(formats, usetemplates=True, description=__doc__)
 110     if textwrap is not None:
 111         parser.add_option("-w", "--wrap", dest="wrap", default=None, type="int",
 112                 help="set number of columns to wrap html at", metavar="WRAP")
 113         parser.passthrough.append("wrap")
 114     parser.add_fuzzy_option()
 115     parser.run(argv)
 116
 117
 118 if __name__ == '__main__':
 119     main()