def test_is_valid_false_if_two_emptynodes_has_the_same_sub_index(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
EmptyNode(main_index=0, sub_index=1),
Word(index=1),
])
self.assertFalse(sentence.is_valid())
def test_raw_tokens_on_sentence_wit_mixed_element(self):
expected = [
Multiword(first_index=1, last_index=2),
Word(index=3),
Multiword(first_index=4, last_index=6),
Word(index=7)
]
sentence = Sentence([
expected[0], # 1-2
Word(index=1),
Word(index=2),
expected[1], # 3
EmptyNode(main_index=3, sub_index=1),
EmptyNode(main_index=3, sub_index=2),
expected[2], # 4-6
Word(index=4),
EmptyNode(main_index=4, sub_index=1),
EmptyNode(main_index=4, sub_index=2),
Word(index=5),
Word(index=6),
expected[3] # 7
])
result = sentence.raw_tokens()
self.assertIsInstance(result, Generator)
self.assertEqual(expected, list(result))
def test_is_valid_false_if_an_emptynode_sub_index_is_skipped(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
# sub_index 2 missing
EmptyNode(main_index=0, sub_index=3),
Word(index=1),
])
self.assertFalse(sentence.is_valid())
def test_is_valid_false_if_an_emptynode_main_index_has_no_word_index(self):
sentence = Sentence([
Word(index=1),
# there is no word with index 2
EmptyNode(main_index=2, sub_index=1),
EmptyNode(main_index=2, sub_index=2)
])
self.assertFalse(sentence.is_valid())
def test_is_valid_false_if_consecutive_emptynodes_have_different_main_id(
self):
sentence = Sentence([
Word(index=1),
EmptyNode(main_index=1, sub_index=1),
# word with index 2 missing
EmptyNode(main_index=2, sub_index=2)
])
self.assertFalse(sentence.is_valid())
def test_to_conllu_with_deps_tuple(self):
emptynode = EmptyNode(
main_index=1,
sub_index=2,
deps=((1, 'Foo'), (2, 'Bar')))
self.assertEqual(
'1.2\t_\t_\t_\t_\t_\t_\t_\t1:Foo|2:Bar\t_',
emptynode.to_conllu())
def test_to_conllu_with_deps_str(self):
emptynode = EmptyNode(
main_index=1,
sub_index=2,
deps='1:Foo|2:Bar')
self.assertEqual(
'1.2\t_\t_\t_\t_\t_\t_\t_\t1:Foo|2:Bar\t_',
emptynode.to_conllu())
def test_to_conllu_with_feats_as_tuple(self):
emptynode = EmptyNode(
main_index=1,
sub_index=2,
feats=(('Foo', ('Bar',)), ('Baz', ('Qux', 'Zet'))))
self.assertEqual(
'1.2\t_\t_\t_\t_\tFoo=Bar|Baz=Qux,Zet\t_\t_\t_\t_',
emptynode.to_conllu())
def test_to_conllu_with_feats_as_str(self):
emptynode = EmptyNode(
main_index=1,
sub_index=2,
feats='Foo=Bar|Baz=Qux')
self.assertEqual(
'1.2\t_\t_\t_\t_\tFoo=Bar|Baz=Qux\t_\t_\t_\t_',
emptynode.to_conllu())
def test_is_valid_true_if_emptynodes_indexes_are_valid(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
EmptyNode(main_index=0, sub_index=2),
Word(index=1),
Word(index=2),
EmptyNode(main_index=2, sub_index=1),
EmptyNode(main_index=2, sub_index=2),
])
self.assertTrue(sentence.is_valid())
def test_is_valid_true_with_all_values_set(self):
element = EmptyNode(
main_index=1,
sub_index=2,
form='Form',
lemma='Lemma',
upos=UposTag.X,
xpos='XPOS',
feats={'foo': 'bar'},
deps={'baz': 'qux'},
misc='Misc'
)
self.assertTrue(element.is_valid())
def test_to_conllu_of_sentence_with_all_attributes(self):
emptynode = EmptyNode(
main_index=1,
sub_index=2,
form='Form',
lemma='Lemma',
upos=UposTag.X,
xpos='XPOS',
feats='Feat=Foo',
deps='0:Bar',
misc='Misc')
self.assertEqual(
'1.2\tForm\tLemma\tX\tXPOS\tFeat=Foo\t_\t_\t0:Bar\tMisc',
emptynode.to_conllu())
def test_is_valid_true_if_first_element_is_emptynode_with_index_0(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
# first word also included to prevent other validations to fail
Word(index=1)
])
self.assertTrue(sentence.is_valid())
def test_raw_tokens_on_sentence_without_word_and_multiwords(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
])
result = sentence.raw_tokens()
self.assertIsInstance(result, Generator)
self.assertEqual([], list(result))
def test_words_on_sentence_wit_mixed_element(self):
expected = [Word(index=1), Word(index=2), Word(index=3), Word(index=4)]
sentence = Sentence([
Multiword(first_index=1, last_index=2),
expected[0], # 1
expected[1], # 2
EmptyNode(main_index=2, sub_index=1),
EmptyNode(main_index=2, sub_index=2),
Multiword(first_index=3, last_index=4),
expected[2], # 3
expected[3], # 4
])
result = sentence.words()
self.assertIsInstance(result, Generator)
self.assertEqual(expected, list(result))
def test_words_on_sentence_without_word_elements(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
Multiword(first_index=1, last_index=2),
])
result = sentence.words()
self.assertIsInstance(result, Generator)
self.assertEqual([], list(result))
def test_to_conllu_with_many_elements_and_no_comments(self):
sentence = Sentence(elements=[
Multiword(first_index=1, last_index=2, form="Foobar"),
Word(index=1, form='Foo'),
Word(index=2, form='bar'),
EmptyNode(main_index=2, sub_index=1, form='Baz')
])
self.assertEqual(
'1-2\tFoobar\t_\t_\t_\t_\t_\t_\t_\t_\n'
'1\tFoo\t_\t_\t_\t_\t_\t_\t_\t_\n'
'2\tbar\t_\t_\t_\t_\t_\t_\t_\t_\n'
'2.1\tBaz\t_\t_\t_\t_\t_\t_\t_\t_\n'
'\n', sentence.to_conllu())
def p_wordline_emptynode(prod: YaccProduction) -> None:
'wordline : DECIMAL_ID TAB FORM TAB LEMMA TAB UPOS TAB XPOS TAB ' \
'FEATS TAB HEAD TAB DEPREL TAB DEPS TAB MISC NEWLINE'
if prod[13] is not None or prod[15] is not None:
raise IllegalEmptyNodeError(prod)
prod[0] = EmptyNode(main_index=prod[1][0],
sub_index=prod[1][1],
form=prod[3],
lemma=prod[5],
upos=UposTag[prod[7]] if prod[7] else None,
xpos=prod[9],
feats=prod[11],
deps=prod[17],
misc=prod[19])
def test_init_xpos(self):
element = EmptyNode(xpos='Foo')
self.assertEqual('Foo', element.xpos)
def test_init_feats(self):
element = EmptyNode(feats={'foo': 'bar'})
self.assertEqual({'foo': 'bar'}, element.feats)
def test_init_deps(self):
element = EmptyNode(deps={'foo': 'bar'})
self.assertEqual({'foo': 'bar'}, element.deps)
def test_init_misc(self):
element = EmptyNode(misc='Foo')
self.assertEqual('Foo', element.misc)
def test_is_valid_true_with_sub_index_greater_than_zero(self):
element = EmptyNode(main_index=42, sub_index=1)
self.assertTrue(element.is_valid())
def test_is_valid_true_with_main_index_equal_to_zero(self):
element = EmptyNode(main_index=0, sub_index=42)
self.assertTrue(element.is_valid())
def test_is_valid_false_if_first_element_is_emptynode_with_index_not_0(
self):
sentence = Sentence([EmptyNode(main_index=1, sub_index=1)])
self.assertFalse(sentence.is_valid())
def test_is_valid_false_on_sentence_without_word_elements(self):
sentence = Sentence([
EmptyNode(main_index=0, sub_index=1),
Multiword(first_index=1, last_index=2)
])
self.assertFalse(sentence.is_valid())
def test_is_valid_false_with_sub_index_is_not_set(self):
element = EmptyNode(main_index=42)
self.assertFalse(element.is_valid())
def test_is_valid_false_with_no_values_set(self):
element = EmptyNode()
self.assertFalse(element.is_valid())
def test_is_valid_false_with_sub_index_less_than_zero(self):
element = EmptyNode(main_index=42, sub_index=-1)
self.assertFalse(element.is_valid())
def test_is_valid_false_with_sub_index_equal_to_zero(self):
element = EmptyNode(main_index=42, sub_index=0)
self.assertFalse(element.is_valid())
