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

Source Code for Module translate.storage.tmx

  1  #!/usr/bin/env python 
  2  # -*- coding: utf-8 -*- 
  3  # 
  4  # Copyright 2005-2009 Zuza Software Foundation 
  5  # 
  6  # This file is part of the Translate Toolkit. 
  7  # 
  8  # This program 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  # This program 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 this program; if not, see <http://www.gnu.org/licenses/>. 
 20   
 21  """module for parsing TMX translation memeory files""" 
 22   
 23  from lxml import etree 
 24   
 25  from translate import __version__ 
 26  from translate.storage import lisa 
 27   
 28   
29 -class tmxunit(lisa.LISAunit):
30 """A single unit in the TMX file.""" 31 rootNode = "tu" 32 languageNode = "tuv" 33 textNode = "seg" 34
35 - def createlanguageNode(self, lang, text, purpose):
36 """returns a langset xml Element setup with given parameters""" 37 if isinstance(text, str): 38 text = text.decode("utf-8") 39 langset = etree.Element(self.languageNode) 40 lisa.setXMLlang(langset, lang) 41 seg = etree.SubElement(langset, self.textNode) 42 # implied by the standard: 43 # lisa.setXMLspace(seg, "preserve") 44 seg.text = text 45 return langset
46
47 - def getid(self):
48 """Returns the identifier for this unit. The optional tuid property is 49 used if available, otherwise we inherit .getid(). Note that the tuid 50 property is only mandated to be unique from TMX 2.0.""" 51 id = self.xmlelement.get("tuid", "") 52 return id or super(tmxunit, self).getid()
53
54 - def istranslatable(self):
55 return bool(self.source)
56
57 - def addnote(self, text, origin=None, position="append"):
58 """Add a note specifically in a "note" tag. 59 60 The origin parameter is ignored""" 61 if isinstance(text, str): 62 text = text.decode("utf-8") 63 note = etree.SubElement(self.xmlelement, self.namespaced("note")) 64 note.text = text.strip()
65
66 - def getnotelist(self, origin=None):
67 """Private method that returns the text from notes. 68 69 The origin parameter is ignored..""" 70 note_nodes = self.xmlelement.iterdescendants(self.namespaced("note")) 71 note_list = [lisa.getText(note) for note in note_nodes] 72 73 return note_list
74
75 - def getnotes(self, origin=None):
76 return '\n'.join(self.getnotelist(origin=origin))
77
78 - def removenotes(self):
79 """Remove all the translator notes.""" 80 notes = self.xmlelement.iterdescendants(self.namespaced("note")) 81 for note in notes: 82 self.xmlelement.remove(note)
83
84 - def adderror(self, errorname, errortext):
85 """Adds an error message to this unit.""" 86 #TODO: consider factoring out: some duplication between XLIFF and TMX 87 text = errorname 88 if errortext: 89 text += ': ' + errortext 90 self.addnote(text, origin="pofilter")
91
92 - def geterrors(self):
93 """Get all error messages.""" 94 #TODO: consider factoring out: some duplication between XLIFF and TMX 95 notelist = self.getnotelist(origin="pofilter") 96 errordict = {} 97 for note in notelist: 98 errorname, errortext = note.split(': ') 99 errordict[errorname] = errortext 100 return errordict
101
102 - def copy(self):
103 """Make a copy of the translation unit. 104 105 We don't want to make a deep copy - this could duplicate the whole XML 106 tree. For now we just serialise and reparse the unit's XML.""" 107 #TODO: check performance 108 new_unit = self.__class__(None, empty=True) 109 new_unit.xmlelement = etree.fromstring(etree.tostring(self.xmlelement)) 110 return new_unit
111 112
113 -class tmxfile(lisa.LISAfile):
114 """Class representing a TMX file store.""" 115 UnitClass = tmxunit 116 Name = _("TMX Translation Memory") 117 Mimetypes = ["application/x-tmx"] 118 Extensions = ["tmx"] 119 rootNode = "tmx" 120 bodyNode = "body" 121 XMLskeleton = '''<?xml version="1.0" encoding="utf-8"?> 122 <!DOCTYPE tmx SYSTEM "tmx14.dtd"> 123 <tmx version="1.4"> 124 <header></header> 125 <body></body> 126 </tmx>''' 127
128 - def addheader(self):
129 headernode = self.document.getroot().iterchildren(self.namespaced("header")).next() 130 headernode.set("creationtool", "Translate Toolkit - po2tmx") 131 headernode.set("creationtoolversion", __version__.sver) 132 headernode.set("segtype", "sentence") 133 headernode.set("o-tmf", "UTF-8") 134 headernode.set("adminlang", "en") 135 #TODO: consider adminlang. Used for notes, etc. Possibly same as targetlanguage 136 headernode.set("srclang", self.sourcelanguage) 137 headernode.set("datatype", "PlainText")
138 #headernode.set("creationdate", "YYYYMMDDTHHMMSSZ" 139 #headernode.set("creationid", "CodeSyntax" 140
141 - def addtranslation(self, source, srclang, translation, translang):
142 """addtranslation method for testing old unit tests""" 143 unit = self.addsourceunit(source) 144 unit.target = translation 145 tuvs = unit.xmlelement.iterdescendants(self.namespaced('tuv')) 146 lisa.setXMLlang(tuvs.next(), srclang) 147 lisa.setXMLlang(tuvs.next(), translang)
148
149 - def translate(self, sourcetext, sourcelang=None, targetlang=None):
150 """method to test old unit tests""" 151 return getattr(self.findunit(sourcetext), "target", None)
152