def _make_diff_tags(self, left_value, right_value, node, target=None):
if bool(self.normalize & WS_TEXT):
left_value = utils.cleanup_whitespace(left_value or u'').strip()
right_value = utils.cleanup_whitespace(right_value or u'').strip()
text_diff = diff_match_patch()
diff = text_diff.diff_main(left_value or '', right_value or '')
text_diff.diff_cleanupSemantic(diff)
diff = self._realign_placeholders(diff)
cur_child = None
if target is None:
target = node
else:
cur_child = node
for op, text in diff:
if op == 0:
if cur_child is None:
node.text = (node.text or u'') + text
else:
cur_child.tail = (cur_child.tail or u'') + text
continue
if op == -1:
action = 'delete'
elif op == 1:
action = 'insert'
if self.placeholderer.is_placeholder(text):
ph = self.placeholderer.mark_diff(text, action)
if cur_child is None:
node.text = (node.text or u'') + ph
else:
new_text = self.placeholderer.wrap_diff(text, action)
if cur_child is None:
node.text = (node.text or u'') + new_text
else:
cur_child.tail = (cur_child.tail or u'') + new_text
def node_text(self, node):
if node in self._text_cache:
return self._text_cache[node]
# Get the texts and the tag as a start
texts = node.xpath('text()')
# Then add attributes and values
for tag, value in sorted(node.attrib.items()):
if tag[0] == '{':
tag = tag.split('}', )[-1]
texts.append('%s:%s' % (tag, value))
# Finally make one string, useful to see how similar two nodes are
text = u' '.join(texts).strip()
result = utils.cleanup_whitespace(text)
self._text_cache[node] = result
return result
def node_text(self, node):
if node in self._text_cache:
return self._text_cache[node]
# Get the texts and the tag as a start
texts = node.xpath("text()")
# Then add attributes and values
for tag, value in sorted(self.node_attribs(node).items()):
if tag[0] == "{":
tag = tag.split("}", )[-1]
texts.append(f"{tag}:{value}")
# Finally make one string, useful to see how similar two nodes are
text = " ".join(texts).strip()
result = utils.cleanup_whitespace(text)
self._text_cache[node] = result
return result