def test_en_sbd_serialization_projective(combined_all_model_fixture):
"""Test that before and after serialization, the sentence boundaries are
the same."""
text = "I bought a couch from IKEA It wasn't very comfortable."
transition = [
'L-nsubj', 'S', 'L-det', 'R-dobj', 'D', 'R-prep', 'R-pobj', 'B-ROOT',
'L-nsubj', 'R-neg', 'D', 'S', 'L-advmod', 'R-acomp', 'D', 'R-punct'
]
doc = combined_all_model_fixture.tokenizer(text)
apply_transition_sequence(en_with_combined_rule_tokenizer_fixture.parser,
doc, transition)
doc_serialized = Doc(
en_with_combined_rule_tokenizer_fixture.vocab).from_bytes(
doc.to_bytes())
assert doc.is_parsed == True
assert doc_serialized.is_parsed == True
assert doc.to_bytes() == doc_serialized.to_bytes()
assert [s.text
for s in doc.sents] == [s.text for s in doc_serialized.sents]
def test_en_sentence_breaks(combined_all_model_fixture, en_parser):
text = "This is a sentence . This is another one ."
heads = [1, 0, 1, -2, -3, 1, 0, 1, -2, -3]
deps = [
'nsubj', 'ROOT', 'det', 'attr', 'punct', 'nsubj', 'ROOT', 'det',
'attr', 'punct'
]
transition = [
'L-nsubj', 'S', 'L-det', 'R-attr', 'D', 'R-punct', 'B-ROOT', 'L-nsubj',
'S', 'L-attr', 'R-attr', 'D', 'R-punct'
]
tokens = combined_all_model_fixture(text)
doc = get_doc(tokens.vocab, [t.text for t in tokens],
heads=heads,
deps=deps)
apply_transition_sequence(en_parser, doc, transition)
assert len(list(doc.sents)) == 2
for token in doc:
assert token.dep != 0 or token.is_space
assert [token.head.i for token in doc] == [1, 1, 3, 1, 1, 6, 6, 8, 6, 6]