def test_search_query_with_spires_syntax_in_literature_search():
query = u"a First Middle Last"
expected_full_statement = inspire_query_parser.parse_query(query)
expected_must_statement = expected_full_statement['nested']['query']['bool']['filter']
search = LiteratureSearch().query_from_iq(query)
search_dict = search.to_dict()
must_statement = search_dict['query']['bool']['must']
assert len(must_statement) == 1
must_statement = must_statement[0]['nested']
assert must_statement['path'] == 'authors'
assert must_statement['query']['bool']['filter'][0] == expected_must_statement
def test_inspire_search_filter(app, app_client, user_info, total_count, es_filter):
"""Test default inspire search filter."""
if user_info:
login_user_via_view(app_client, email=user_info['email'],
password=user_info['password'],
login_url='/login/?local=1')
# Doing a client request creates a request context that allows the
# assert to correctly use the logged in user.
app_client.get('/search')
assert LiteratureSearch().to_dict()['query']['bool'][
'filter'] == es_filter
assert LiteratureSearch().count() == total_count
def get_literature_recids_for_orcid(orcid):
"""Return the Literature recids that were claimed by an ORCiD.
We record the fact that the Author record X has claimed the Literature
record Y by storing in Y an author object with a ``$ref`` pointing to X
and the key ``curated_relation`` set to ``True``. Therefore this method
first searches the DB for the Author records for the one containing the
given ORCiD, and then uses its recid to search in ES for the Literature
records that satisfy the above property.
Args:
orcid (str): the ORCiD.
Return:
list(int): the recids of the Literature records that were claimed
by that ORCiD.
"""
orcid_object = '[{"schema": "ORCID", "value": "%s"}]' % orcid
# this first query is written in a way that can use the index on (json -> ids)
author_rec_uuid = db.session.query(RecordMetadata.id)\
.filter(type_coerce(RecordMetadata.json, JSONB)['ids'].contains(orcid_object)).one().id
author_recid = db.session.query(PersistentIdentifier.pid_value).filter(
PersistentIdentifier.object_type == 'rec',
PersistentIdentifier.object_uuid == author_rec_uuid,
PersistentIdentifier.pid_type == 'aut',
).one().pid_value
query = Q('match', authors__curated_relation=True) & Q(
'match', authors__recid=author_recid)
search_by_curated_author = LiteratureSearch().query('nested', path='authors', query=query)\
.params(_source=['control_number'], size=9999)
return [el['control_number'] for el in search_by_curated_author]
def test_select_source_function_in_literature_search(request_mocked):
request_mocked.headers.get.return_value = 'application/vnd+inspire.record.ui+json'
expected_source = {
'includes': ['$schema', 'control_number', '_ui_display']
}
search = select_source(LiteratureSearch())
search_source = search.to_dict()['_source']
assert search_source == expected_source
def _process_record(pid, app):
if stop:
return
with app.app_context():
success = False
deleted = False
no_cits = False
db_cits = None
es_cits = None
es_citation_count_field = None
data = {}
rec = get_db_record('lit', pid.pid_value)
if rec.get('deleted'):
success = True
deleted = True
if not deleted:
try:
es_cits = get_citations_from_es(rec).total
search = LiteratureSearch().source(includes=['citation_count'])
results = search.get_record(rec.id).execute()
if not results.hits:
es_citation_count_field = None
else:
es_citation_count_field = results.hits[0]['citation_count']
db_cits = rec.get_citations_count()
except Exception as err:
click.echo("Cannot prepare data for %s record. %s",
pid.pid_value, err)
if not deleted and es_cits is not None and es_cits == db_cits == es_citation_count_field:
if es_cits == 0:
no_cits = True
success = True
else:
data = {
'pid_value': pid.pid_value,
'db_citations_count': db_cits,
'es_citations_count': es_cits,
'es_citations_field': es_citation_count_field
}
return (success, deleted, no_cits, data)
def _process_record(pid, app):
if stop:
return
with app.app_context():
success = False
deleted = False
no_cits = False
db_cits = None
es_cits = None
es_citation_count_field = None
data = {}
rec = get_db_record('lit', pid.pid_value)
if rec.get('deleted'):
success = True
deleted = True
if not deleted:
try:
es_cits = LiteratureSearch.citations(rec).total
search = LiteratureSearch().source(includes=['citation_count'])
results = search.get_record(rec.id).execute()
if not results.hits:
es_citation_count_field = None
else:
es_citation_count_field = results.hits[0]['citation_count']
db_cits = rec.get_citations_count()
except Exception as err:
click.echo("Cannot prepare data for %s record. %s",
pid.pid_value,
err)
if not deleted and es_cits is not None and es_cits == db_cits == es_citation_count_field:
if es_cits == 0:
no_cits = True
success = True
else:
data = {'pid_value': pid.pid_value,
'db_citations_count': db_cits,
'es_citations_count': es_cits,
'es_citations_field': es_citation_count_field}
return (success, deleted, no_cits, data)
def test_select_source_function_in_literature_search():
expected_source = {
'includes': [
'$schema', 'abstracts.value', 'arxiv_eprints.value',
'arxiv_eprints.categories', 'authors.affiliations',
'authors.full_name', 'authors.control_number', 'collaborations',
'control_number', 'citation_count', 'dois.value', 'earliest_date',
'inspire_categories', 'number_of_references', 'publication_info',
'report_numbers', 'titles.title'
]
}
search = select_source(LiteratureSearch())
search_source = search.to_dict()['_source']
assert search_source == expected_source
def get_signatures_matching_a_phonetic_encoding(phonetic_encoding):
"""Get all signatures matching a phonetic encoding from ES.
Args:
phonetic_encodings(str): a phonetic encoding.
Yields:
dict: a signature matching the phonetic encoding.
"""
query = Q('term', authors__signature_block__raw=phonetic_encoding)
search_by_phonetic_encoding = LiteratureSearch().query('nested', path='authors', query=query)\
.params(_source=SIGNATURE_FIELDS, size=9999)
for record in search_by_phonetic_encoding:
for signature in record.authors:
if signature.signature_block == phonetic_encoding:
yield signature.to_dict()
def get_signatures_matching_a_phonetic_encoding(phonetic_encoding):
"""Get all signatures matching a phonetic encoding from ES.
Args:
phonetic_encodings(str): a phonetic encoding.
Yields:
dict: a signature matching the phonetic encoding.
"""
query = Q('term', authors__signature_block__raw=phonetic_encoding)
search_by_phonetic_encoding = LiteratureSearch().query('nested', path='authors', query=query)\
.params(_source=SIGNATURE_FIELDS, size=9999)
for record in search_by_phonetic_encoding:
record = record.to_dict()
publication_id = record['control_number']
for author in record.get('authors', []):
if author.get('signature_block') == phonetic_encoding:
yield _build_signature(author, publication_id)
