explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-07-17 11:42:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Allow lookup of morphology by attribute name

Browse Source
This commit is contained in:
Matthew Honnibal 2019-03-08 01:38:15 +01:00
parent 3c32590243
commit 322b64dca0
2 changed files with 52 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
spacy/morphology.pxd
View File

@ -36,5 +36,6 @@ cdef class Morphology:
cdef int check_feature(const MorphAnalysisC* tag, attr_t feature) nogil cdef int check_feature(const MorphAnalysisC* tag, attr_t feature) nogil
cdef attr_t get_field(const MorphAnalysisC* tag, int field) nogil cdef attr_t get_field(const MorphAnalysisC* tag, int field) nogil
cdef list list_features(const MorphAnalysisC* tag) cdef list list_features(const MorphAnalysisC* tag)
cdef int attribute_to_field(unicode attribute)
cdef tag_to_json(const MorphAnalysisC* tag) cdef tag_to_json(const MorphAnalysisC* tag)

52
spacy/morphology.pyx
View File

@ -98,6 +98,10 @@ def parse_feature(feature):
return (field, offset) return (field, offset)
cdef int attribute_to_field(unicode attribute_name):
return LOWER_FIELDS[attribute_name]
def get_field_id(feature): def get_field_id(feature):
return FEATURE_FIELDS[feature] return FEATURE_FIELDS[feature]
@ -709,7 +713,6 @@ cdef int set_feature(MorphAnalysisC* tag,
raise ValueError("Unknown feature: %s (%d)" % (FEATURE_NAMES.get(feature), feature)) raise ValueError("Unknown feature: %s (%d)" % (FEATURE_NAMES.get(feature), feature))
FIELDS = { FIELDS = {
'Abbr': Field_Abbr, 'Abbr': Field_Abbr,
'AdpType': Field_AdpType, 'AdpType': Field_AdpType,
@ -756,6 +759,53 @@ FIELDS = {
'VerbType': Field_VerbType 'VerbType': Field_VerbType
} }
LOWER_FIELDS = {
'abbr': Field_Abbr,
'adp_type': Field_AdpType,
'adv_type': Field_AdvType,
'animacy': Field_Animacy,
'aspect': Field_Aspect,
'case': Field_Case,
'conj_type': Field_ConjType,
'connegative': Field_Connegative,
'definite': Field_Definite,
'degree': Field_Degree,
'derivation': Field_Derivation,
'echo': Field_Echo,
'foreign': Field_Foreign,
'gender': Field_Gender,
'hyph': Field_Hyph,
'inf_form': Field_InfForm,
'mood': Field_Mood,
'name_type': Field_NameType,
'negative': Field_Negative,
'noun_type': Field_NounType,
'number': Field_Number,
'num_form': Field_NumForm,
'num_type': Field_NumType,
'num_value': Field_NumValue,
'part_form': Field_PartForm,
'part_type': Field_PartType,
'person': Field_Person,
'polite': Field_Polite,
'polarity': Field_Polarity,
'poss': Field_Poss,
'prefix': Field_Prefix,
'prep_case': Field_PrepCase,
'pron_type': Field_PronType,
'punct_side': Field_PunctSide,
'punct_type': Field_PunctType,
'reflex': Field_Reflex,
'style': Field_Style,
'style_variant': Field_StyleVariant,
'tense': Field_Tense,
'typo': Field_Typo,
'verb_form': Field_VerbForm,
'voice': Field_Voice,
'verb_type': Field_VerbType
}
FEATURES = [ FEATURES = [
"Abbr_yes", "Abbr_yes",
"AdpType_circ", "AdpType_circ",