def _process_html_tree(elt):
node_list = safe_dom.NodeList()
tail = elt.tail
if elt.tag in tag_bindings:
elt = tag_bindings[elt.tag]().render(elt, handler)
try:
if elt.tag.lower() == 'script':
out_elt = safe_dom.ScriptElement()
else:
out_elt = safe_dom.Element(elt.tag)
out_elt.add_attribute(**elt.attrib)
if elt.text:
out_elt.add_text(elt.text)
for child in elt:
out_elt.add_children(_process_html_tree(child))
except Exception as e: # pylint: disable-msg=broad-except
logging.error('Invalid HTML tag: %s. %s', elt, e)
out_elt = safe_dom.Element('span')
out_elt.add_attribute(className='gcb-error-tag')
out_elt.add_text(INVALID_HTML_TAG_MESSAGE)
node_list.append(out_elt)
if tail:
node_list.append(safe_dom.Text(tail))
return node_list
def _process_html_tree(elt):
"""Recursively parses an HTML tree into a safe_dom.NodeList()."""
# Return immediately with an error message if a duplicate instanceid is
# detected.
if 'instanceid' in elt.attrib:
if elt.attrib['instanceid'] in used_instance_ids:
return _generate_error_message_node_list(
elt, DUPLICATE_INSTANCE_ID_MESSAGE)
used_instance_ids.add(elt.attrib['instanceid'])
# Otherwise, attempt to parse this tag and all its child tags.
original_elt = elt
try:
if render_custom_tags and elt.tag in tag_bindings:
tag = tag_bindings[elt.tag]()
if isinstance(tag, ContextAwareTag):
# Get or initialize a environment dict for this type of tag.
# Each tag type gets a separate environment shared by all
# instances of that tag.
context = tag_contexts.get(elt.tag)
if context is None:
context = ContextAwareTag.Context(handler, {})
tag_contexts[elt.tag] = context
# Render the tag
elt = tag.render(elt, context)
else:
# Render the tag
elt = tag.render(elt, handler)
if elt.tag == cElementTree.Comment:
out_elt = safe_dom.Comment()
elif elt.tag.lower() == 'script':
out_elt = safe_dom.ScriptElement()
else:
out_elt = safe_dom.Element(_remove_namespace(elt.tag))
out_elt.add_attribute(**elt.attrib)
if elt.text:
out_elt.add_text(elt.text)
for child in elt:
out_elt.add_children(
_process_html_tree(child))
node_list = safe_dom.NodeList()
node_list.append(out_elt)
if original_elt.tail:
node_list.append(safe_dom.Text(original_elt.tail))
return node_list
except Exception as e: # pylint: disable=broad-except
logging.exception('Error handling tag: %s', elt.tag)
return _generate_error_message_node_list(
original_elt, '%s: %s' % (INVALID_HTML_TAG_MESSAGE, e))
def _process_html_tree(elt):
node_list = safe_dom.NodeList()
tail = elt.tail
if elt.tag in tag_bindings:
elt = tag_bindings[elt.tag]().render(elt)
if elt.tag.lower() == 'script':
out_elt = safe_dom.ScriptElement()
else:
out_elt = safe_dom.Element(elt.tag)
out_elt.add_attribute(**elt.attrib)
if elt.text:
out_elt.add_text(elt.text)
for child in elt:
out_elt.add_children(_process_html_tree(child))
node_list.append(out_elt)
if tail:
node_list.append(safe_dom.Text(tail))
return node_list
def _process_html_tree(elt, used_instance_ids):
# Return immediately with an error message if a duplicate instanceid is
# detected.
if 'instanceid' in elt.attrib:
if elt.attrib['instanceid'] in used_instance_ids:
return _generate_error_message_node_list(
elt, DUPLICATE_INSTANCE_ID_MESSAGE)
used_instance_ids.add(elt.attrib['instanceid'])
# Otherwise, attempt to parse this tag and all its child tags.
original_elt = elt
try:
if elt.tag in tag_bindings:
elt = tag_bindings[elt.tag]().render(elt, handler)
if elt.tag.lower() == 'script':
out_elt = safe_dom.ScriptElement()
else:
out_elt = safe_dom.Element(elt.tag)
out_elt.add_attribute(**elt.attrib)
if elt.text:
out_elt.add_text(elt.text)
for child in elt:
out_elt.add_children(
_process_html_tree(child, used_instance_ids))
node_list = safe_dom.NodeList()
node_list.append(out_elt)
if original_elt.tail:
node_list.append(safe_dom.Text(original_elt.tail))
return node_list
except Exception as e: # pylint: disable-msg=broad-except
return _generate_error_message_node_list(
original_elt, '%s: %s' % (INVALID_HTML_TAG_MESSAGE, e))