def sex_count(ent):
"""Enrich the match with data."""
data = {}
for token in ent:
label = token.ent_type_
value = token.lower_
if label == 'sex':
data['sex'] = value
elif (as_int := to_positive_int(value)) is not None:
data['count'] = as_int
def fraction(token):
"""Handle fractional values like 10 3/8 inches."""
trait = Trait(start=token.start, end=token.end)
trait.units = token.group.get("units")
trait.units_inferred = not bool(trait.units)
whole = to_positive_float(token.group.get("whole", "0"))
numerator = to_positive_int(token.group["numerator"])
denominator = to_positive_int(token.group["denominator"])
try:
trait.value = whole + Fraction(numerator, denominator)
except TypeError:
print(f"Fraction error: {numerator} / {denominator}")
return None
if trait.units:
trait.value = convert_units(trait.value, trait.units)
add_flags(token, trait)
return trait
def range_(span):
"""Get a count range."""
values = [to_positive_int(t.text) for t in span if t.is_digit]
data = {'low': min(values)}
if len(values) > 1:
data['high'] = max(values)
return data
def multiple_seta_count(ent):
"""Handle multiple seta in one match."""
data = {'body_part': 'seta'}
values = []
for token in ent:
label = token._.cached_label
if label == 'seta':
data['seta'] = REPLACE.get(token.lower_, token.lower_)
elif label == 'number_word':
values.append(to_positive_int(REPLACE.get(token.lower_)))
elif match := IS_INT.match(token.text):
value = to_positive_int(match.group(0))
values.append(value)
elif label == 'group':
data['group'] = token.lower_
def range_values(ent):
"""Extract values from the range and cached label."""
data = {}
range_ = [
e for e in ent.ents if e._.cached_label.split('.')[0] == 'range'
][0]
values = re.findall(FLOAT_RE, range_.text)
if not all([re.search(INT_TOKEN_RE, v) for v in values]):
raise RejectMatch
keys = range_.label_.split('.')[1:]
for key, value in zip(keys, values):
data[key] = to_positive_int(value)
range_._.data = data
range_._.new_label = 'count'
return range_
def seta_count(ent):
"""Enrich the match."""
data = {'body_part': 'seta'}
location = []
for token in ent:
label = token._.cached_label
if label == 'seta':
data['seta'] = REPLACE.get(token.lower_, token.lower_)
elif label == 'number_word':
data['count'] = int(REPLACE.get(token.lower_, -1))
elif token.lower_ in MISSING:
data['count'] = 0
elif label == 'group':
data['group'] = token.lower_
elif match := IS_INT.match(token.text):
data['count'] = to_positive_int(match.group(0))
def count_word(ent):
"""Enrich the match with data."""
ent._.new_label = 'count'
word = [e for e in ent.ents if e.label_ == 'count_word'][0]
word._.data = {'low': to_positive_int(REPLACE[word.text])}
word._.new_label = 'count'
def sample(token):
"""Convert the span into a single integer."""
match = re.search(INT_RE, token.text)
value = match.group(0)
return {'n': to_positive_int(value)}