def preprocess_xml(self, text):
document = Document()
root = ET.fromstring(text)
content = self.concat_elements(root, './text/p')
document.add_component(DocumentComponent('content', content))
document.add_metadata('title', self.concat_elements(root, './title'))
document.add_metadata('headline',
self.concat_elements(root, './headline'))
document.add_metadata('author', self.concat_elements(root, './byline'))
document.add_metadata('date', root.attrib['date'])
return document
return True
def retrieve_attribute(token):
s = deque([token.lemma_])
cur = token
while True:
compound = next(filter(lambda x: x.dep_ == 'compound', cur.children),
None)
if compound is None or not is_valid_attribute_token(compound):
break
else:
cur = compound
s.appendleft(compound.lemma_)
return " ".join(s)
if __name__ == '__main__':
vader = Vader()
extractor = RuleBasedExtractor(vader)
doc = Document()
doc.add_component(DocumentComponent(type="text", text=input().strip()))
doc = extractor.extract(doc)
for ent in doc.entities:
print(ent.text, [attr.text for attr in ent.attributes])
def preprocess_txt(self, text):
document = Document()
document.add_component(DocumentComponent('content', text))
return document
