def test_patdict_bool_int(self, spacy_doc):
matches = list(extract.matches(spacy_doc, [{"IS_DIGIT": True}]))[:5]
assert matches
assert all(span[0].is_digit is True for span in matches)
matches = list(extract.matches(spacy_doc, [{"LENGTH": 5}]))[:5]
assert matches
assert all(len(span[0]) == 5 for span in matches)
def test_patstr_bool_int(self, spacy_doc):
matches = list(extract.matches(spacy_doc, "IS_DIGIT:bool(True)"))[:5]
assert matches
assert all(span[0].is_digit is True for span in matches)
matches = list(extract.matches(spacy_doc, "LENGTH:int(5)"))[:5]
assert matches
assert all(len(span[0]) == 5 for span in matches)
def test_patdict_op(self, spacy_doc):
matches = list(extract.matches(spacy_doc, [{
"POS": "NOUN",
"OP": "+"
}]))[:5]
assert matches
assert all(len(span) >= 1 for span in matches)
assert all(tok.pos_ == "NOUN" for span in matches for tok in span)
def test_pattern_types(self, spacy_doc):
all_patterns = [
"POS:NOUN",
["POS:NOUN", "POS:DET"],
[{"POS": "NOUN"}],
[[{"POS": "NOUN"}], [{"POS": "DET"}]],
]
for patterns in all_patterns:
matches = list(extract.matches(spacy_doc, patterns))[:5]
assert matches
assert all(isinstance(span, Span) for span in matches)
def test_patdict(self, spacy_doc):
matches = list(extract.matches(spacy_doc, [{"POS": "NOUN"}]))[:5]
assert matches
assert all(len(span) == 1 for span in matches)
assert all(span[0].pos_ == "NOUN" for span in matches)
def test_patstr_op(self, spacy_doc):
matches = list(extract.matches(spacy_doc, "POS:NOUN:+"))[:5]
assert matches
assert all(len(span) >= 1 for span in matches)
assert all(tok.pos_ == "NOUN" for span in matches for tok in span)
def test_patstr(self, spacy_doc):
matches = list(extract.matches(spacy_doc, "POS:NOUN"))[:5]
assert matches
assert all(len(span) == 1 for span in matches)
assert all(span[0].pos_ == "NOUN" for span in matches)
import test_spacy
from textacy.extract import matches
from scripting.knowledge_graph.HDSKG.helperfunctions import set_pos_exceptions, view_displacy
from scripting.knowledge_graph.HDSKG import patterns
nlp = test_spacy.load('nl_core_news_sm')
text = """Iedereen die in Nederland woont of werkt en die is verzekerd voor zorg vanuit de Wlz betaalt hiervoor premie.
Hieruit wordt de zorg betaald. Bijna altijd betaalt u daarnaast een eigen bijdrage."""
doc = nlp(text)
doc = set_pos_exceptions(doc)
verb_closed_results = matches(doc, patterns.verbs_closed)
verbs_closed = [span for span in verb_closed_results]
verbs_open_results = matches(doc, patterns.verbs_open)
verbs_open = [span for span in verbs_open_results]
verbs = [verbs_open, verbs_closed]
noun_results = matches(doc, patterns.nouns)
nouns = [span for span in noun_results]
# dependency tree
words = []
for word in doc:
if word.dep_ in 'nsubj':
words.append(word)
print(word.string)
