explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-12 10:16:27 +03:00
Code Issues Packages Projects Releases Wiki Activity

Allow int values in token patterns (#4444)

Browse Source 
* Add missing int value option to top-level pattern validation in Matcher

* Adjust existing tests accordingly

* Add new test for valid pattern `{"LENGTH": int}`
This commit is contained in:
adrianeboyd 2019-10-16 13:40:18 +02:00 committed by Matthew Honnibal
parent 7d1efac4eb
commit 275c9ad872
2 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/matcher/matcher.pyx
View File

@ -658,7 +658,7 @@ def _get_attr_values(spec, string_store):
value = string_store.add(value) value = string_store.add(value)
elif isinstance(value, bool): elif isinstance(value, bool):
value = int(value) value = int(value)
elif isinstance(value, dict): elif isinstance(value, (dict, int)):
continue continue
else: else:
raise ValueError(Errors.E153.format(vtype=type(value).__name__)) raise ValueError(Errors.E153.format(vtype=type(value).__name__))

7
spacy/tests/matcher/test_pattern_validation.py
View File

@ -12,24 +12,25 @@ from spacy.util import get_json_validator, validate_json
TEST_PATTERNS = [ TEST_PATTERNS = [
# Bad patterns flagged in all cases # Bad patterns flagged in all cases
([{"XX": "foo"}], 1, 1), ([{"XX": "foo"}], 1, 1),
([{"LENGTH": "2", "TEXT": 2}, {"LOWER": "test"}], 2, 1),
([{"IS_ALPHA": {"==": True}}, {"LIKE_NUM": None}], 2, 1), ([{"IS_ALPHA": {"==": True}}, {"LIKE_NUM": None}], 2, 1),
([{"IS_PUNCT": True, "OP": "$"}], 1, 1), ([{"IS_PUNCT": True, "OP": "$"}], 1, 1),
([{"IS_DIGIT": -1}], 1, 1),
([{"ORTH": -1}], 1, 1),
([{"_": "foo"}], 1, 1), ([{"_": "foo"}], 1, 1),
('[{"TEXT": "foo"}, {"LOWER": "bar"}]', 1, 1), ('[{"TEXT": "foo"}, {"LOWER": "bar"}]', 1, 1),
([1, 2, 3], 3, 1), ([1, 2, 3], 3, 1),
# Bad patterns flagged outside of Matcher # Bad patterns flagged outside of Matcher
([{"_": {"foo": "bar", "baz": {"IN": "foo"}}}], 1, 0), ([{"_": {"foo": "bar", "baz": {"IN": "foo"}}}], 1, 0),
# Bad patterns not flagged with minimal checks # Bad patterns not flagged with minimal checks
([{"LENGTH": "2", "TEXT": 2}, {"LOWER": "test"}], 2, 0),
([{"LENGTH": {"IN": [1, 2, "3"]}}, {"POS": {"IN": "VERB"}}], 2, 0), ([{"LENGTH": {"IN": [1, 2, "3"]}}, {"POS": {"IN": "VERB"}}], 2, 0),
([{"LENGTH": {"VALUE": 5}}], 1, 0), ([{"LENGTH": {"VALUE": 5}}], 1, 0),
([{"TEXT": {"VALUE": "foo"}}], 1, 0), ([{"TEXT": {"VALUE": "foo"}}], 1, 0),
([{"IS_DIGIT": -1}], 1, 0),
([{"ORTH": -1}], 1, 0),
# Good patterns # Good patterns
([{"TEXT": "foo"}, {"LOWER": "bar"}], 0, 0), ([{"TEXT": "foo"}, {"LOWER": "bar"}], 0, 0),
([{"LEMMA": {"IN": ["love", "like"]}}, {"POS": "DET", "OP": "?"}], 0, 0), ([{"LEMMA": {"IN": ["love", "like"]}}, {"POS": "DET", "OP": "?"}], 0, 0),
([{"LIKE_NUM": True, "LENGTH": {">=": 5}}], 0, 0), ([{"LIKE_NUM": True, "LENGTH": {">=": 5}}], 0, 0),
([{"LENGTH": 2}], 0, 0),
([{"LOWER": {"REGEX": "^X", "NOT_IN": ["XXX", "XY"]}}], 0, 0), ([{"LOWER": {"REGEX": "^X", "NOT_IN": ["XXX", "XY"]}}], 0, 0),
([{"NORM": "a"}, {"POS": {"IN": ["NOUN"]}}], 0, 0), ([{"NORM": "a"}, {"POS": {"IN": ["NOUN"]}}], 0, 0),
([{"_": {"foo": {"NOT_IN": ["bar", "baz"]}, "a": 5, "b": {">": 10}}}], 0, 0), ([{"_": {"foo": {"NOT_IN": ["bar", "baz"]}, "a": 5, "b": {">": 10}}}], 0, 0),