Package translate :: Package storage :: Module qm
[hide private]
[frames] | no frames]

Source Code for Module translate.storage.qm

  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   
 23  """Module for parsing Qt .qm files 
 24   
 25  @note: based on documentation from Gettext's .qm implementation 
 26  (see write-qt.c) and on observation of the output of lrelease. 
 27  @note: Certain deprecated section tags are not implemented.  These will break 
 28  and print out the missing tag.  They are easy to implement and should follow 
 29  the structure in 03 (Translation).  We could find no examples that use these 
 30  so we'd rather leave it unimplemented until we actually have test data. 
 31  @note: Many .qm files are unable to be parsed as they do not have the source 
 32  text.  We assume that since they use a hash table to lookup the data there is 
 33  actually no need for the source text.  It seems however that in Qt4's lrelease 
 34  all data is included in the resultant .qm file. 
 35  @todo: We can only parse, not create, a .qm file.  The main issue is that we 
 36  need to implement the hashing algorithm (which seems to be identical to the 
 37  Gettext hash algorithm).  Unlike Gettext it seems that the hash is required, 
 38  but that has not been validated. 
 39  @todo: The code can parse files correctly.  But it could be cleaned up to be 
 40  more readable, especially the part that breaks the file into sections. 
 41   
 42  U{http://qt.gitorious.org/+kde-developers/qt/kde-qt/blobs/master/tools/linguist/shared/qm.cpp} 
 43  U{Plural information<http://qt.gitorious.org/+kde-developers/qt/kde-qt/blobs/master/tools/linguist/shared/numerus.cpp>} 
 44  U{QLocale languages<http://docs.huihoo.com/qt/4.5/qlocale.html#Language-enum>} 
 45  """ 
 46   
 47  import codecs 
 48  import struct 
 49  import sys 
 50   
 51  from translate.storage import base 
 52  from translate.misc.multistring import multistring 
 53   
 54  QM_MAGIC_NUMBER = (0x3CB86418L, 0xCAEF9C95L, 0xCD211CBFL, 0x60A1BDDDL) 
 55   
 56   
57 -def qmunpack(file_='messages.qm'):
58 """Helper to unpack Qt .qm files into a Python string""" 59 f = open(file_) 60 s = f.read() 61 print "\\x%02x" * len(s) % tuple(map(ord, s)) 62 f.close()
63 64
65 -class qmunit(base.TranslationUnit):
66 """A class representing a .qm translation message.""" 67
68 - def __init__(self, source=None):
69 super(qmunit, self).__init__(source)
70 71
72 -class qmfile(base.TranslationStore):
73 """A class representing a .qm file.""" 74 UnitClass = qmunit 75 Name = _("Qt .qm file") 76 Mimetypes = ["application/x-qm"] 77 Extensions = ["qm"] 78 _binary = True 79
80 - def __init__(self, inputfile=None, unitclass=qmunit):
81 self.UnitClass = unitclass 82 base.TranslationStore.__init__(self, unitclass=unitclass) 83 self.units = [] 84 self.filename = '' 85 if inputfile is not None: 86 self.parsestring(inputfile)
87
88 - def __str__(self):
89 """Output a string representation of the .qm data file""" 90 return ""
91
92 - def parse(self, input):
93 """parses the given file or file source string""" 94 if hasattr(input, 'name'): 95 self.filename = input.name 96 elif not getattr(self, 'filename', ''): 97 self.filename = '' 98 if hasattr(input, "read"): 99 qmsrc = input.read() 100 input.close() 101 input = qmsrc 102 if len(input) < 16: 103 raise ValueError("This is not a .qm file: file empty or too small") 104 magic = struct.unpack(">4L", input[:16]) 105 if magic != QM_MAGIC_NUMBER: 106 raise ValueError("This is not a .qm file: invalid magic number") 107 startsection = 16 108 sectionheader = 5 109 while startsection < len(input): 110 section_type, length = struct.unpack(">bL", input[startsection:startsection + sectionheader]) 111 if section_type == 0x42: 112 #print "Section: hash" 113 hashash = True 114 hash_start = startsection + sectionheader 115 hash_data = struct.unpack(">%db" % length, input[startsection + sectionheader:startsection + sectionheader + length]) 116 elif section_type == 0x69: 117 #print "Section: messages" 118 hasmessages = True 119 messages_start = startsection + sectionheader 120 messages_data = struct.unpack(">%db" % length, input[startsection + sectionheader:startsection + sectionheader + length]) 121 elif section_type == 0x2f: 122 #print "Section: contexts" 123 hascontexts = True 124 contexts_start = startsection + sectionheader 125 contexts_data = struct.unpack(">%db" % length, input[startsection + sectionheader:startsection + sectionheader + length]) 126 startsection = startsection + sectionheader + length 127 pos = messages_start 128 source = target = None 129 while pos < messages_start + len(messages_data): 130 subsection, = struct.unpack(">b", input[pos:pos + 1]) 131 if subsection == 0x01: # End 132 #print "End" 133 pos = pos + 1 134 if not source is None and not target is None: 135 newunit = self.addsourceunit(source) 136 newunit.target = target 137 source = target = None 138 else: 139 raise ValueError("Old .qm format with no source defined") 140 continue 141 #print pos, subsection 142 pos = pos + 1 143 length, = struct.unpack(">l", input[pos:pos + 4]) 144 if subsection == 0x03: # Translation 145 if length != -1: 146 raw, = struct.unpack(">%ds" % length, 147 input[pos + 4:pos + 4 + length]) 148 string, templen = codecs.utf_16_be_decode(raw) 149 if target: 150 target.strings.append(string) 151 else: 152 target = multistring(string) 153 pos = pos + 4 + length 154 else: 155 target = "" 156 pos = pos + 4 157 #print "Translation: %s" % target.encode('utf-8') 158 elif subsection == 0x06: # SourceText 159 source = input[pos + 4:pos + 4 + length].decode('iso-8859-1') 160 #print "SourceText: %s" % source 161 pos = pos + 4 + length 162 elif subsection == 0x07: # Context 163 context = input[pos + 4:pos + 4 + length].decode('iso-8859-1') 164 #print "Context: %s" % context 165 pos = pos + 4 + length 166 elif subsection == 0x08: # Disambiguating-comment 167 comment = input[pos + 4:pos + 4 + length] 168 #print "Disambiguating-comment: %s" % comment 169 pos = pos + 4 + length 170 elif subsection == 0x05: # hash 171 hash = input[pos:pos + 4] 172 #print "Hash: %s" % hash 173 pos = pos + 4 174 else: 175 if subsection == 0x02: # SourceText16 176 subsection_name = "SourceText16" 177 elif subsection == 0x04: # Context16 178 subsection_name = "Context16" 179 else: 180 subsection_name = "Unkown" 181 print >> sys.stderr, "Unimplemented: %s %s" % \ 182 (subsection, subsection_name) 183 return
184
185 - def savefile(self, storefile):
186 raise Exception("Writing of .qm files is not supported yet")
187