def get_author_year_score_for_input_fields(result_record, hypothesis):
"""
returns evidences based on just author and year.
For most sources, you should rather use get_basic_score_for_input_fields
-- see there for more information.
:param result_record:
:param hypothesis:
:return:
"""
input_fields = hypothesis.get_detail('input_fields')
evidences = Evidences()
normalized_authors = hypothesis.get_detail('normalized_authors')
if normalized_authors is None:
normalized_authors = normalize_author_list(
input_fields.get('author', ''))
add_author_evidence(evidences,
normalized_authors,
result_record['author_norm'],
result_record['first_author_norm'],
has_etal=hypothesis.get_detail('has_etal'))
add_year_evidence(evidences, input_fields.get('year'),
result_record.get('year'))
return evidences
def get_author_year_score_for_input_fields(result_record, hypothesis):
"""
returns evidences based on just author and year.
:param result_record:
:param hypothesis:
:return:
"""
input_fields = hypothesis.get_detail('input_fields')
normalized_authors = hypothesis.get_detail('normalized_authors')
if normalized_authors is None:
normalized_authors = normalize_author_list(
input_fields.get('author', ''))
evidences = Evidences()
add_author_evidence(evidences,
normalized_authors,
result_record['author_norm'],
result_record['first_author_norm'],
has_etal=hypothesis.get_detail('has_etal'))
add_year_evidence(evidences, input_fields.get('year'),
result_record.get('year'))
return evidences
def get_score_for_baas_match(result_record, hypothesis):
"""
scores a BAAS->DDA match.
For these, volume and page are hidden deep inside pub_raw.
We also expect an expected_bibstem detail in the hypothesis, mainly
for robustness in case this gets used to score something else.
:param result_record:
:param hypothesis:
:return:
"""
evidences = Evidences()
if not re.match(r'....%s' % hypothesis.get_detail('expected_bibstem'),
result_record['bibcode']):
evidences.add_evidence(current_app.config['EVIDENCE_SCORE_RANGE'][0],
'no DDA bibcode')
return evidences
input_fields = hypothesis.get_detail('input_fields')
normalized_authors = hypothesis.get_detail('normalized_authors')
if normalized_authors is None:
normalized_authors = normalize_author_list(
input_fields.get('author', ''))
add_author_evidence(evidences, normalized_authors,
result_record['author_norm'],
result_record['first_author_norm'])
add_boolean_evidence(
evidences, 'Vol. %s' % input_fields['volume']
in result_record['pub_raw'], 'vol in pub_raw?')
add_boolean_evidence(
evidences,
re.search(r'p\.\s*%s\b' % input_fields['page'],
result_record['pub_raw']), 'page in pub_raw?')
return evidences