def setAttribute(self, name, value):
dom_logging(log, "setAttribute", name, value)
if not isinstance(name, basestring):
name = str(name)
self.tag[name] = value
if name.lower() in ('src', 'archive'):
s = urlparse.urlsplit(value)
handler = getattr(log.SchemeHandler, 'handle_%s' % (s.scheme, ), None)
if handler:
handler(self.doc.window, value)
return
try:
response = self.doc.window._navigator.fetch(value, redirect_type = "element workaround")
except:
return
if response.status_code == 404:
return
ctype = response.headers.get('content-type', None)
if ctype is None:
return
handler = log.MIMEHandler.get_handler(ctype)
if handler:
handler(self.doc.window.url, response.content)
def setAttribute(self, name, value):
dom_logging(log, "setAttribute", name, value)
if not isinstance(name, basestring):
name = str(name)
self.tag[name] = value
if name.lower() in ('src', 'archive'):
s = urlparse.urlsplit(value)
handler = getattr(log.SchemeHandler, 'handle_%s' % (s.scheme, ),
None)
if handler:
handler(self.doc.window, value)
return
try:
response = self.doc.window._navigator.fetch(
value, redirect_type="element workaround")
except:
return
if response.status_code == 404:
return
ctype = response.headers.get('content-type', None)
if ctype is None:
return
handler = log.MIMEHandler.get_handler(ctype)
if handler:
handler(self.doc.window.url, response.content)
def write(self, html):
dom_logging(log, "document.write", html)
if self._html:
self._html.write(html)
return
tag = self.current
parent = tag.parent
pos = parent.contents.index(tag) + 1
if not isinstance(html, basestring):
html = str(html)
soup = BeautifulSoup.BeautifulSoup(html, "html5lib")
soup.html.unwrap()
soup.head.unwrap()
soup.body.unwrap()
for tag in soup:
parent.insert(pos, tag)
pos += 1
name = getattr(tag, "name", None)
if name in ("script", None):
continue
try:
handler = getattr(self._win.doc.DFT, "handle_%s" % (name,), None)
except:
handler = getattr(log.DFT, "handle_%s" % (name,), None)
if handler:
handler(tag)
def write(self, html):
dom_logging(log, "document.write", html)
if self._html:
self._html.write(html)
return
tag = self.current
parent = tag.parent
pos = parent.contents.index(tag) + 1
if not isinstance(html, basestring):
html = str(html)
soup = BeautifulSoup.BeautifulSoup(html, "html5lib")
soup.html.unwrap()
soup.head.unwrap()
soup.body.unwrap()
for tag in soup:
parent.insert(pos, tag)
pos += 1
name = getattr(tag, "name", None)
if name in ('script', None):
continue
try:
handler = getattr(self._win.doc.DFT, "handle_%s" % (name, ),
None)
except:
handler = getattr(log.DFT, "handle_%s" % (name, ), None)
if handler:
handler(tag)
def insertBefore(self, newChild, refChild):
dom_logging(log, 'insertBefore', str(newChild.tag))
if not newChild:
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
if not isinstance(newChild, Node):
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
# If refChild is null, insert newChild at the end of the list
# of children
if not refChild:
return self.appendChild(newChild)
if not isinstance(refChild, Node):
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
#index = self.findChild(refChild)
#if index < 0 and not self.is_text(refChild):
# raise DOMException(DOMException.NOT_FOUND_ERR)
# If the newChild is already in the tree, it is first removed
if getattr(newChild, 'tag', None) and newChild.tag in self.tag.contents:
newChildHash = hash(newChild.tag._node)
for p in self.tag.contents:
if getattr(p, '_node', None) is None:
continue
if newChildHash == hash(p._node):
p.extract()
#self.tag.contents.remove(newChild.tag)
index = self.findChild(refChild)
if index < 0 and not self.is_text(refChild):
raise DOMException(DOMException.NOT_FOUND_ERR)
if self.is_text(newChild):
self.tag.insert(index, newChild.data.output_ready(formatter = lambda x: x))
return newChild
if newChild.nodeType in (Node.DOCUMENT_FRAGMENT_NODE, ):
# self.tag.insert(index, newChild.tag.findChild())
node = None
for p in newChild.tag.find_all_next():
if node is None:
self.tag.insert(index, p)
else:
node.append(p)
node = p
return newChild
self.tag.insert(index, newChild.tag)
return newChild
def setter(self, text):
if tag == 'innerHTML':
dom_logging(log, "set innerHTML to", text)
#if self.tag.string:
# self.tag.contents[0] = BeautifulSoup.NavigableString(text)
#else:
# self.tag.append(text)
#
#self.tag.string = self.tag.contents[0]
self.tag.string = text
def setNamedItem(self, attr):
oldattr = self.parent.getAttributeNode(attr.name)
attr.parent = self.parent
dom_logging(log, "setNamedItem", attr.value)
self.parent.tag[attr.name] = attr.value
if oldattr:
oldattr.parent = None
return oldattr
def setNamedItem(self, attr):
oldattr = self.parent.getAttributeNode(attr.name)
attr.parent = self.parent
dom_logging(log, "setNamedItem", attr.value)
self.parent.tag[attr.name] = attr.value
if oldattr:
oldattr.parent = None
return oldattr
def appendChild(self, newChild):
dom_logging(log, "appendChild", str(newChild.tag))
# NO_MODIFICATION_ALLOWED_ERR: Raised if this node is readonly
if self.is_readonly(self):
raise DOMException(DOMException.NO_MODIFICATION_ALLOWED)
if self.is_text(self):
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
if not newChild:
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
if not isinstance(newChild, Node):
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
# If the newChild is already in the tree, it is first removed
if getattr(newChild, 'tag', None) and newChild.tag in self.tag.contents:
newChildHash = hash(newChild.tag._node)
for p in self.tag.contents:
if getattr(p, '_node', None) is None:
continue
if newChildHash == hash(p._node):
p.extract()
#self.tag.contents.remove(newChild.tag)
if self.is_text(newChild):
self.tag.append(newChild.data.output_ready(formatter = lambda x: x))
return newChild
if newChild.nodeType in (Node.DOCUMENT_FRAGMENT_NODE, ):
#self.tag.append(newChild.tag.findChild())
node = self.tag
for p in newChild.tag.find_all_next():
node.append(p)
node = p
return newChild
self.tag.append(newChild.tag)
return newChild
def replaceChild(self, newChild, oldChild):
dom_logging(log, "replaceChild", str(newChild.tag))
# NO_MODIFICATION_ALLOWED_ERR: Raised if this node or the parent of
# the new node is readonly.
if self.is_readonly(self):
raise DOMException(DOMException.NO_MODIFICATION_ALLOWED)
parent = getattr(newChild, 'parentNode', None)
if parent:
if self.is_readonly(parent):
raise DOMException(DOMException.NO_MODIFICATION_ALLOWED)
if not newChild or not oldChild:
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
if not isinstance(newChild, Node) or not isinstance(oldChild, Node):
raise DOMException(DOMException.HIERARCHY_REQUEST_ERR)
index = self.findChild(oldChild)
if index < 0 and not self.is_text(refChild):
raise DOMException(DOMException.NOT_FOUND_ERR)
if self.is_text(newChild):
self.tag.contents[index] = newChild.data.output_ready(formatter = lambda x: x)
return oldChild
if newChild.nodeType in (Node.DOCUMENT_FRAGMENT_NODE, ):
#self.tag.contents[index] = newChild.tag.findChild()
node = None
for p in newChild.tag.find_all_next():
if node is None:
self.tag.contents[index] = p
else:
node.append(p)
node = p
return oldChild
self.tag.contents[index] = newChild.tag
return oldChild
def evalScript(self, script, tag = None):
dom_logging(log, 'eval script', script)
result = 0
try:
log.JSClassifier.classify('[Local analysis]' if log.ThugOpts.local else self.url,
script)
except:
pass
if tag:
self.doc.current = tag
else:
try:
body = self.doc.body
except:
# This code is for when you are desperate :)
body = self.doc.getElementsByTagName('body')[0]
if body and body.tag.contents:
self.doc.current = body.tag.contents[-1]
else:
self.doc.current = self.doc.doc.contents[-1]
with self.context as ctxt:
try:
ast = AST(self, script)
except:
log.debug(traceback.format_exc())
return result
if log.ThugOpts.Personality.isIE():
cc = CCInterpreter()
script = cc.run(script)
shellcode = Shellcode.Shellcode(self, ctxt, ast, script)
result = shellcode.run()
return result
def evalScript(self, script, tag=None):
dom_logging(log, 'eval script', script)
result = 0
try:
log.JSClassifier.classify(
'[Local analysis]' if log.ThugOpts.local else self.url, script)
except:
pass
if tag:
self.doc.current = tag
else:
try:
body = self.doc.body
except:
# This code is for when you are desperate :)
body = self.doc.getElementsByTagName('body')[0]
if body and body.tag.contents:
self.doc.current = body.tag.contents[-1]
else:
self.doc.current = self.doc.doc.contents[-1]
with self.context as ctxt:
try:
ast = AST(self, script)
except:
log.debug(traceback.format_exc())
return result
if log.ThugOpts.Personality.isIE():
cc = CCInterpreter()
script = cc.run(script)
shellcode = Shellcode.Shellcode(self, ctxt, ast, script)
result = shellcode.run()
return result
def _navigate(self, location):
dom_logging(log, "Window.navigate {}".format(location))
self.location = location
return 0
def setLocation(self, location):
dom_logging(log, "Window.setLocation", location)
self._location.href = location
def setAttributeNode(self, attr):
dom_logging(log, "setAttributeNode", attr.name, attr.value)
self.tag[attr.name] = attr.value
def setAttributeNode(self, attr):
dom_logging(log, "setAttributeNode", attr.name, attr.value)
self.tag[attr.name] = attr.value
def setLocation(self, location):
dom_logging(log, "Window.setLocation", location)
self._location.href = location
def setNodeValue(self, value):
dom_logging(log, 'setNodeValue', str(value))
pass
def _navigate(self, location):
dom_logging(log, "Window.navigate {}".format(location))
self.location = location
return 0