def find_coincidences(cls, tokens):
# search for the site object which contains all sequence of tokens
sites = cls.objects.all()
coincidences = []
for site in sites:
map = site.map_link
for key, link in map.iteritems():
key = only_nouns(key) # remving all not necessary chars
if contains_any(tokens, key):
coincidences.append(link)
return coincidences
def tokens(self):
# Add spaces between puntuation signals
document = self.content.split('.')
document_filtered = []
for sentence in document:
sentence_filtered = sentence.replace(',', ' , ')
sentence_filtered = sentence_filtered.replace(';', ' ; ')
sentence_filtered = sentence_filtered.replace('.', ' . ')
sentence_filtered = sentence_filtered.replace(':', ' : ')
sentence_filtered = sentence_filtered.replace('"', ' " ')
sentence_filtered = sentence_filtered.replace("'", " ' ")
document_filtered.append(sentence_filtered)
# Extract nouns, adjectives
new_document = []
for sentence in document_filtered:
new_sentence = only_nouns(sentence)
if new_sentence:
new_document.append(new_sentence)
# Remove empty lines and duplicate words from sentence
texts = [list(set(sentence)) for sentence in new_document if sentence]
# TODO: Reduce the matrix dimension to 2 x 2
return texts
