def extend(self, annotated_sentence):
''' Extend an existing annotated sentence with types (from wordlists). '''
extended_sentence = []
for word in annotated_sentence.words:
# wordnet lemmatiser only accepts these 4 chars as pos-tags
lemma = self.stemmer.stem(word.lemma.lower())
if word.lemma == None or len(word.lemma) < 1:
postag = word.pos[0].lower()
postag = postag if postag in ['a','r','n','v'] else 'n'
lemma = self.stemmer.stem(self.lemmatiser.lemmatize(word.word.lower(), pos=postag))
# convert the current word types to a list
word_types = word.types.split(',') if word.types != '' else []
# for all the wordsets whose part of speech tags match the current word
for wordset in [s for s in self.wordsets if StringMatching.is_match(word.pos, s.pos)]:
if lemma in wordset.words:
word_types.append(wordset.label)
word.types = ','.join(word_types)
extended_sentence.append(word)
return AnnotatedSentence(extended_sentence)
def test_match_contains_pass(self):
word = 'Orange Fruit Mangoes'
strings = '*Fruit*,*Mangoe*'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_exact_fail(self):
word = 'Orange Fruit Mangoes'
strings = 'Orange Fruit Lemons'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_exact_pass(self):
word = 'Mangoes'
strings = 'Mangoes'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_ends_pass(self):
word = 'Apples'
strings = '*le,*les'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_starts_fail(self):
word = 'Apples'
strings = 'Apl*,Apple'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_case_fail2(self):
word = 'Orange Fruit Mangoes'
strings = '*fruit*'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings, False))
def test_match_surround_fail(self):
word = 'Orange Fruit Mangoes'
strings = 'A*Fruit Mangoes'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_case_pass3(self):
word = 'AbCdE'
strings = 'AbCdE'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings), True)
def test_match_case_pass1(self):
word = 'ABCdefGEH'
strings = 'ABCd*'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_mixed_fail2(self):
word = 'ABCDEFg'
strings = 'ABCDEF,*BCDEF'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_mixed_fail1(self):
word = 'Potatoes'
strings = 'Potatos*,*topato*,POTATO'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_mixed_pass2(self):
word = 'Orange Fruit Mangoes'
strings = 'Or*eg,Orange Fruit Mango,*ruit*'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_mixed_pass1(self):
word = 'Oranges'
strings = 'Fruit*,*papers,*ang*'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_contains_fail(self):
word = 'Orange Fruit Mangoes'
strings = '*Paper*'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings))
def test_match_surround_pass(self):
word = 'Orange Fruit Mangoes'
strings = 'Orange*oes'.split(',')
self.assertEqual(True, StringMatching.is_match(word, strings))
def test_match_case_fail1(self):
word = 'AbCdE'
strings = 'aBcde'.split(',')
self.assertEqual(False, StringMatching.is_match(word, strings, False))