Allow higher unicode characters in XMB files.

grit/tool/xmb.py

 #!/usr/bin/env python
 # Copyright (c) 2012 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 
 """The 'grit xmb' tool.
 """
 
 import getopt
 import os
@@ skipping 10 matching lines @@
 # Used to collapse presentable content to determine if
 # xml:space="preserve" is needed.
 _WHITESPACES_REGEX = lazy_re.compile(ur'\s\s*')
 
 
 # See XmlEscape below.
 _XML_QUOTE_ESCAPES = {
     u"'":  u''',
     u'"':  u'"',
 }
+# See http://www.w3.org/TR/xml/#charsets
 _XML_BAD_CHAR_REGEX = lazy_re.compile(u'[^\u0009\u000A\u000D'
-                                      u'\u0020-\uD7FF\uE000-\uFFFD]')
+                                      u'\u0020-\uD7FF\uE000-\uFFFD'
+                                      u'\U00010000-\U0010FFFF]')
 
 
 def _XmlEscape(s):
   """Returns text escaped for XML in a way compatible with Google's
   internal Translation Console tool.  May be used for attributes as
   well as for contents.
   """
   if not type(s) == unicode:
     s = unicode(s)
   result = saxutils.escape(s, _XML_QUOTE_ESCAPES)
-  return _XML_BAD_CHAR_REGEX.sub(u'', result).encode('utf-8')
+  illegal_chars = _XML_BAD_CHAR_REGEX.search(result)
+  if illegal_chars:

[newt (away), 2015/11/10 17:22:52]

All of Chrome's grd files pass this stricter error checking (as you'd hope!).
It's possible that some other project contains illegal characters but I think
they'd rather know about that just continue to get translations with missing
characters.

+    raise Exception('String contains characters disallowed in XML: %s' %
+                    repr(result))
+  return result.encode('utf-8')
 
 
 def _WriteAttribute(file, name, value):
   """Writes an XML attribute to the specified file.
 
     Args:
       file: file to write to
       name: name of the attribute
       value: (unescaped) value of the attribute
     """
@@ skipping 228 matching lines @@
     messages.sort(key=lambda x:x.GetId())
 
     if self.format == self.FORMAT_IDS_ONLY:
       # We just print the list of IDs to the output file.
       for msg in messages:
         output_file.write(msg.GetId())
         output_file.write('\n')
     else:
       assert self.format == self.FORMAT_XMB
       WriteXmbFile(output_file, messages)