def test_shortcuts_spacy_inline():
spacy = pytest.importorskip("spacy", minversion="2.1")
nlp = spacy.load("en_core_web_sm")
rules = """
{WORD("TEST")}->MARK("TEST")
"""
setup_spacy(nlp, rules_string=rules)
def spacy_engine(rules, **kwargs):
spacy = pytest.importorskip("spacy", minversion="2.1")
nlp = spacy.load("en_core_web_sm")
setup_spacy(nlp, rules_string=rules, override_ents=True, **kwargs)
patterns = rita.compile_string(rules, **kwargs)
print(patterns)
def parse(text):
doc = nlp(text)
return list([(e.text, e.label_) for e in doc.ents])
return parse
def test_shortcuts_spacy_compiled():
spacy = pytest.importorskip("spacy", minversion="2.1")
nlp = spacy.load("en_core_web_sm")
tmp = tempfile.NamedTemporaryFile(mode="w",
encoding="UTF-8",
suffix=".jsonl",
delete=False)
patterns = rita.compile("examples/color-car.rita")
for pattern in patterns:
tmp.write(json.dumps(pattern) + "\n")
tmp.flush()
tmp.close()
setup_spacy(nlp, patterns=tmp.name)
os.unlink(tmp.name)
def test_shortcuts_spacy_giving_no_rules():
spacy = pytest.importorskip("spacy", minversion="2.1")
nlp = spacy.load("en_core_web_sm")
with pytest.raises(RuntimeError):
setup_spacy(nlp)
def test_shortcuts_spacy_file():
spacy = pytest.importorskip("spacy", minversion="2.1")
nlp = spacy.load("en_core_web_sm")
setup_spacy(nlp, rules_path="examples/color-car.rita")