# HG changeset patch
# User cmlenz
# Date 1160746958 0
# Node ID 38adb4aa7df5284f50a8e6ab5e890a343bc0d1e1
# Parent 6de057dfa90556aacf68240934e25ac6bd9470c6
Fix a bug in the XML parser, where attributes containing HTML entity references would get pulled out of the attribute value, and instead added as a text node just before the associated start tag. Thanks to Hamish Lawson for [http://groups.google.com/group/genshi/browse_thread/thread/c64eb48676b0ff96/0e6ce786e8820f3d pointing out the problem].
diff --git a/genshi/input.py b/genshi/input.py
--- a/genshi/input.py
+++ b/genshi/input.py
@@ -21,7 +21,7 @@
import htmlentitydefs
from StringIO import StringIO
-from genshi.core import Attrs, QName, Stream
+from genshi.core import Attrs, QName, Stream, stripentities
from genshi.core import DOCTYPE, START, END, START_NS, END_NS, TEXT, \
START_CDATA, END_CDATA, PI, COMMENT
@@ -71,6 +71,10 @@
END root
"""
+ _entitydefs = ['' % (name, value) for name, value in
+ htmlentitydefs.name2codepoint.items()]
+ _external_dtd = '\n'.join(_entitydefs)
+
def __init__(self, source, filename=None):
"""Initialize the parser for the given XML text.
@@ -100,7 +104,9 @@
# Tell Expat that we'll handle non-XML entities ourselves
# (in _handle_other)
parser.DefaultHandler = self._handle_other
+ parser.SetParamEntityParsing(expat.XML_PARAM_ENTITY_PARSING_ALWAYS)
parser.UseForeignDTD()
+ parser.ExternalEntityRefHandler = self._build_foreign
# Location reporting is only support in Python >= 2.4
if not hasattr(parser, 'CurrentLineNumber'):
@@ -141,6 +147,11 @@
def __iter__(self):
return iter(self.parse())
+ def _build_foreign(self, context, base, sysid, pubid):
+ parser = self.expat.ExternalEntityParserCreate(context)
+ parser.ParseFile(StringIO(self._external_dtd))
+ return 1
+
def _enqueue(self, kind, data=None, pos=None):
if pos is None:
pos = self._getpos()
@@ -292,7 +303,7 @@
for name, value in attrib: # Fixup minimized attributes
if value is None:
value = name
- fixed_attrib.append((name, unicode(value)))
+ fixed_attrib.append((name, unicode(stripentities(value))))
self._enqueue(START, (QName(tag), Attrs(fixed_attrib)))
if tag in self._EMPTY_ELEMS:
diff --git a/genshi/tests/input.py b/genshi/tests/input.py
--- a/genshi/tests/input.py
+++ b/genshi/tests/input.py
@@ -76,6 +76,15 @@
self.assertEqual(Stream.TEXT, kind)
self.assertEqual(u'\xa0', data)
+ def test_html_entity_in_attribute(self):
+ text = ''
+ events = list(XMLParser(StringIO(text)))
+ kind, data, pos = events[0]
+ self.assertEqual(Stream.START, kind)
+ self.assertEqual(u'\xa0', data[1].get('title'))
+ kind, data, pos = events[1]
+ self.assertEqual(Stream.END, kind)
+
def test_undefined_entity_with_dtd(self):
text = """
@@ -118,6 +127,22 @@
self.assertEqual(Stream.TEXT, kind)
self.assertEqual(u'\u2013', data)
+ def test_html_entity_in_attribute(self):
+ text = ''
+ events = list(HTMLParser(StringIO(text)))
+ kind, data, pos = events[0]
+ self.assertEqual(Stream.START, kind)
+ self.assertEqual(u'\xa0', data[1].get('title'))
+ kind, data, pos = events[1]
+ self.assertEqual(Stream.END, kind)
+
+ def test_html_entity_in_text(self):
+ text = '
'
+ events = list(HTMLParser(StringIO(text)))
+ kind, data, pos = events[1]
+ self.assertEqual(Stream.TEXT, kind)
+ self.assertEqual(u'\xa0', data)
+
def suite():
suite = unittest.TestSuite()