def comment(self, content):
if self._incl_comm is None:
return
if self._norm_sp is not None:
content = ' '.join(content.split())
if len(content) > 0:
self._n_elmt += 1
content = content.encode(self.encoding)
xpath = '%s/comment()' % self._xpath
_inc_xpath(self._h, xpath)
# nodes construction for comment
node = [NT_COMM, 'comment()', content, [], None, 0, self._h[xpath]]
link_node(self._p_stack[-1], node)
def characters(self, ch):
if self._norm_sp is not None:
ch = ' '.join(ch.split())
if len(ch) > 0 and ch != "\n" and ch != ' ':
ch = ch.encode(self.encoding)
parent = self._p_stack[-1]
# if sibling text nodes
if parent[N_CHILDS] and parent[N_CHILDS][-1][N_TYPE] == NT_TEXT:
n = parent[N_CHILDS][-1]
n[N_VALUE] = n[N_VALUE] + ch
else:
self._n_elmt += 1
xpath = '%s/text()' % self._xpath
_inc_xpath(self._h, xpath)
# nodes construction for text
node = [NT_TEXT, 'text()', ch, [],None,0, self._h[xpath]]
link_node(parent, node)
def startElement(self, name, attrs):
name = name.encode(self.encoding)
# process xpath
self._xpath = "%s%s%s" % (self._xpath, '/', name)
_inc_xpath(self._h, self._xpath)
# nodes construction for element
node = [NT_NODE, name, name, [], None, self._n_elmt+1,
self._h[self._xpath]]
self._n_elmt += 1
self._xpath = "%s%s%s%s" % (self._xpath, '[', self._h[self._xpath], ']')
# nodes construction for element's attributes
keys = attrs.keys()
# sort attributes to avoid further moves
keys.sort()
for key in keys:
key = key.encode(self.encoding)
self._n_elmt += 2
attr_node = [NT_ATTN, '@%sName'%key, key, [], None, 1, 0]
link_node(node, attr_node)
link_node(attr_node, [NT_ATTV, '@%s'%key,
attrs.get(key, '').encode(self.encoding),
[],None,0,0])
link_node(self._p_stack[-1], node)
# set current element on the top of the father stack
self._p_stack.append(node)