def handle_data(self,data):
if self.stack:
if '#PCDATA' not in HTML_DTD.get(self.stack[-1],[]) and not strip(data):
# this is probably ignorable whitespace
self._cont_handler.ignorableWhitespace(data)
else:
self._cont_handler.characters(to_xml_string(data,self._encoding))
def handle_data(self, data):
if self.stack:
if '#PCDATA' not in HTML_DTD.get(self.stack[-1],
[]) and not strip(data):
# this is probably ignorable whitespace
self._cont_handler.ignorableWhitespace(data)
else:
self._cont_handler.characters(
to_xml_string(data, self._encoding))
def finish_starttag(self, tag, attrs):
"""uses the HTML DTD to automatically generate events
for missing tags"""
# guess omitted close tags
while self.stack and \
upper(self.stack[-1]) in HTML_OPT_END and \
tag not in HTML_DTD.get(self.stack[-1],[]):
self.unknown_endtag(self.stack[-1])
del self.stack[-1]
if self.stack and tag not in HTML_DTD.get(self.stack[-1],[]) and self.verbose:
print 'Warning : trying to add %s as a child of %s'%\
(tag,self.stack[-1])
self.unknown_starttag(tag,attrs)
if upper(tag) in HTML_FORBIDDEN_END:
# close immediately tags for which we won't get an end
self.unknown_endtag(tag)
return 0
else:
self.stack.append(tag)
return 1
def finish_starttag(self, tag, attrs):
"""uses the HTML DTD to automatically generate events
for missing tags"""
# guess omitted close tags
while self.stack and \
upper(self.stack[-1]) in HTML_OPT_END and \
tag not in HTML_DTD.get(self.stack[-1],[]):
self.unknown_endtag(self.stack[-1])
del self.stack[-1]
if self.stack and tag not in HTML_DTD.get(self.stack[-1],
[]) and self.verbose:
print 'Warning : trying to add %s as a child of %s'%\
(tag,self.stack[-1])
self.unknown_starttag(tag, attrs)
if upper(tag) in HTML_FORBIDDEN_END:
# close immediately tags for which we won't get an end
self.unknown_endtag(tag)
return 0
else:
self.stack.append(tag)
return 1
def finish_starttag(self, tagname, attrs):
unicodeTagName = unicode(tagname, self._charset)
lowerTagName = string.lower(unicodeTagName)
if not HTML_DTD.has_key(lowerTagName):
# Skip any tags not defined in HTML 4.01
return
element = self._ownerDoc.createElementNS(EMPTY_NAMESPACE, unicodeTagName)
# Allows for multiple META tags in a document
if lowerTagName == 'meta':
lowered = map(lambda (name, value):
(string.lower(name), string.lower(value)),
attrs)
if ('http-equiv', 'content-type') in lowered:
for (name, value) in lowered:
if name == 'content':
match = g_reCharset.search(value)
if match:
self._charset = match.group('charset')
# Add any attributes to the tag
for (name, value) in attrs:
element.setAttributeNS(EMPTY_NAMESPACE, unicode(name, self._charset),
unicode(value, self._charset))
# Look for its parent
for i in range(1, len(self._stack)):
parent = self._stack[-i]
if lowerTagName in HTML_DTD[string.lower(parent.tagName)]:
parent.appendChild(element)
if i > 1:
self._stack = self._stack[:-i+1]
if HTML_DTD[lowerTagName]:
self._stack.append(element)
return
# no parent found
if not self._hasHtml and lowerTagName == 'html':
self._stack[0].appendChild(element)
self._stack.append(element)
self._hasHtml = 1
return
def _4dom_createHTMLElement(self, tagName):
lowered = string.lower(tagName)
if not HTML_DTD.has_key(lowered):
raise TypeError('Unknown HTML Element: %s' % tagName)
if lowered in NoClassTags:
from HTMLElement import HTMLElement
return HTMLElement(self, tagName)
#FIXME: capitalize() broken with unicode in Python 2.0
#normTagName = string.capitalize(tagName)
capitalized = string.upper(tagName[0]) + lowered[1:]
element = HTMLTagMap.get(capitalized, capitalized)
module = 'HTML%sElement' % element
if not self._html.has_key(module):
#Try to import it (should never fail)
__import__('pyxml.dom.html.%s' % module)
# Class and module have the same name
klass = getattr(self._html[module], module)
return klass(self, tagName)
def _4dom_createHTMLElement(self, tagName):
lowered = string.lower(tagName)
if not HTML_DTD.has_key(lowered):
raise TypeError('Unknown HTML Element: %s' % tagName)
if lowered in NoClassTags:
from HTMLElement import HTMLElement
return HTMLElement(self, tagName)
#FIXME: capitalize() broken with unicode in Python 2.0
#normTagName = string.capitalize(tagName)
capitalized = string.upper(tagName[0]) + lowered[1:]
element = HTMLTagMap.get(capitalized, capitalized)
module = 'HTML%sElement' % element
if not self._html.has_key(module):
#Try to import it (should never fail)
__import__('pyxml.dom.html.%s' % module)
# Class and module have the same name
klass = getattr(self._html[module], module)
return klass(self, tagName)
