def test_process_references_in_records_reindexes_conferences_when_pub_info_changes(
inspire_app, celery_app_with_context, celery_session_worker):
# disconnect this signal so records don't get indexed
models_committed.disconnect(index_after_commit)
conference_data = faker.record("con", with_control_number=True)
conference_record = InspireRecord.create(conference_data)
conference_control_number = conference_record["control_number"]
conf_ref = f"http://localhost:8000/api/conferences/{conference_control_number}"
data = faker.record("lit", with_control_number=True)
data["publication_info"] = [{"conference_record": {"$ref": conf_ref}}]
data["document_type"] = ["conference paper"]
record = InspireRecord.create(data)
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
uuids = [record.id]
task = process_references_in_records.delay(uuids)
result = task.get(timeout=5)
conference_record_es = InspireSearch.get_record_data_from_es(
conference_record)
expected_number_of_contributions = 1
assert (expected_number_of_contributions ==
conference_record_es["number_of_contributions"])
def test_process_references_in_records_reindexes_experiments_when_linked_experiments_change(
app, celery_app_with_context, celery_session_worker):
# disconnect this signal so records don't get indexed
models_committed.disconnect(index_after_commit)
experiment_data = faker.record("exp", with_control_number=True)
experiment = InspireRecord.create(experiment_data)
db.session.commit()
experiment_control_number = experiment["control_number"]
exp_ref = f"http://localhost:8000/api/experiments/{experiment_control_number}"
data = faker.record("lit", with_control_number=True)
data["accelerator_experiments"] = [{
"legacy_name": "LIGO",
"record": {
"$ref": exp_ref
}
}]
record = InspireRecord.create(data)
db.session.commit()
models_committed.connect(index_after_commit)
task = process_references_in_records.delay([record.id])
task.get(timeout=5)
experiment_record_es = InspireSearch.get_record_data_from_es(experiment)
expected_number_of_paper = 1
assert expected_number_of_paper == experiment_record_es["number_of_papers"]
def test_process_references_in_records_with_different_type_of_records_doesnt_throw_an_exception(
inspire_app, celery_app_with_context, celery_session_worker):
# disconnect this signal so records don't get indexed
models_committed.disconnect(index_after_commit)
cited_record_1 = LiteratureRecord.create(faker.record("lit"))
cited_record_2 = LiteratureRecord.create(faker.record("lit"))
data_citing_record_1 = faker.record(
"lit", literature_citations=[cited_record_1["control_number"]])
citing_record_1 = LiteratureRecord.create(data_citing_record_1)
data_citing_record_2 = faker.record(
"lit", literature_citations=[cited_record_2["control_number"]])
citing_record_2 = LiteratureRecord.create(data_citing_record_2)
db.session.commit()
records = [
create_record_async("aut"),
create_record_async("job"),
create_record_async("jou"),
create_record_async("exp"),
create_record_async("con"),
create_record_async("dat"),
create_record_async("ins"),
]
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
uuids = [record.id
for record in records] + [citing_record_1.id, citing_record_2.id]
task = process_references_in_records.delay(uuids)
results = task.get(timeout=5)
uuids = [str(uuid) for uuid in uuids]
assert results == uuids
result_cited_record_1 = InspireSearch.get_record_data_from_es(
cited_record_1)
expected_result_cited_record_1_citation_count = 1
assert (expected_result_cited_record_1_citation_count ==
result_cited_record_1["citation_count"])
result_cited_record_2 = InspireSearch.get_record_data_from_es(
cited_record_2)
expected_result_cited_record_2_citation_count = 1
assert (expected_result_cited_record_2_citation_count ==
result_cited_record_2["citation_count"])
def test_process_references_in_records_process_author_records(
mock_batch_index, inspire_app, clean_celery_session):
author_record = AuthorsRecord.create(faker.record("aut"))
lit_record = LiteratureRecord.create(
faker.record(
"lit",
data={
"authors": [{
"full_name": author_record["name"]["value"],
"record": author_record["self"],
}]
},
))
lit_record_2 = LiteratureRecord.create(
faker.record(
"lit",
data={
"authors": [{
"full_name": author_record["name"]["value"],
"record": author_record["self"],
}]
},
))
db.session.commit()
def assert_records_in_es():
lit_record_from_es = InspireSearch.get_record_data_from_es(lit_record)
lit_record_from_es_2 = InspireSearch.get_record_data_from_es(
lit_record_2)
aut_record_from_es = InspireSearch.get_record_data_from_es(
author_record)
assert lit_record_from_es and aut_record_from_es and lit_record_from_es_2
retry_until_pass(assert_records_in_es, retry_interval=5)
models_committed.disconnect(index_after_commit)
author_record["name"]["value"] = "Another Name"
author_record.update(dict(author_record))
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
task = process_references_in_records.delay([author_record.id])
task.get(timeout=5)
assert sorted(mock_batch_index.mock_calls[0][1][0]) == sorted(
[str(lit_record.id), str(lit_record_2.id)])
def test_process_references_in_records_process_conference_records(
mock_batch_index, inspire_app, clean_celery_session):
conf_record = ConferencesRecord.create(
faker.record("con", data={"titles": [{
"title": "Test conference"
}]}))
lit_data = {
"publication_info": [{
"conference_record": {
"$ref": conf_record["self"]["$ref"]
}
}],
"document_type": ["conference paper"],
}
lit_record = LiteratureRecord.create(faker.record("lit", data=lit_data))
lit_record_2 = LiteratureRecord.create(faker.record("lit", data=lit_data))
db.session.commit()
def assert_records_in_es():
lit_record_from_es = InspireSearch.get_record_data_from_es(lit_record)
lit_record_from_es_2 = InspireSearch.get_record_data_from_es(
lit_record_2)
aut_record_from_es = InspireSearch.get_record_data_from_es(conf_record)
assert lit_record_from_es and aut_record_from_es and lit_record_from_es_2
retry_until_pass(assert_records_in_es, retry_interval=5)
models_committed.disconnect(index_after_commit)
conf_record["titles"] = [{"title": "Southern California Strings Seminar "}]
conf_record.update(dict(conf_record))
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
task = process_references_in_records.delay([conf_record.id])
task.get(timeout=5)
assert sorted(mock_batch_index.mock_calls[0][1][0]) == sorted(
[lit_record.id, lit_record_2.id])
def test_process_references_in_records(inspire_app, celery_app_with_context,
celery_session_worker):
# disconnect this signal so records don't get indexed
models_committed.disconnect(index_after_commit)
cited_record_1 = LiteratureRecord.create(faker.record("lit"))
cited_record_2 = LiteratureRecord.create(faker.record("lit"))
data_citing_record_1 = faker.record(
"lit", literature_citations=[cited_record_1["control_number"]])
citing_record_1 = LiteratureRecord.create(data_citing_record_1)
data_citing_record_2 = faker.record(
"lit", literature_citations=[cited_record_2["control_number"]])
citing_record_2 = LiteratureRecord.create(data_citing_record_2)
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
uuids = [citing_record_1.id, citing_record_2.id]
task = process_references_in_records.delay(uuids)
result = task.get(timeout=5)
result_cited_record_1 = InspireSearch.get_record_data_from_es(
cited_record_1)
expected_result_cited_record_1_citation_count = 1
assert (expected_result_cited_record_1_citation_count ==
result_cited_record_1["citation_count"])
result_cited_record_2 = InspireSearch.get_record_data_from_es(
cited_record_2)
expected_result_cited_record_2_citation_count = 1
assert (expected_result_cited_record_2_citation_count ==
result_cited_record_2["citation_count"])
def test_process_references_in_records_reindexes_institutions_when_linked_institutions_change(
inspire_app, celery_app_with_context, celery_session_worker):
# disconnect this signal so records don't get indexed
models_committed.disconnect(index_after_commit)
institution_data = faker.record("ins", with_control_number=True)
institution = InspireRecord.create(institution_data)
institution_control_number = institution["control_number"]
inst_ref = f"http://localhost:8000/api/institutions/{institution_control_number}"
data = faker.record("lit", with_control_number=True)
data.update({
"authors": [{
"full_name":
"John Doe",
"affiliations": [{
"value": "Institution",
"record": {
"$ref": inst_ref
}
}],
}]
})
record_authors_aff = InspireRecord.create(data)
db.session.commit()
data = faker.record("lit", with_control_number=True)
data.update(
{"thesis_info": {
"institutions": [{
"record": {
"$ref": inst_ref
}
}]
}})
record_thesis_info = InspireRecord.create(data)
db.session.commit()
data = faker.record("lit", with_control_number=True)
data.update({
"record_affiliations": [{
"record": {
"$ref": inst_ref
},
"value": "Institution"
}]
})
record_affiliations = InspireRecord.create(data)
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
task = process_references_in_records.delay(
[record_authors_aff.id, record_thesis_info.id, record_affiliations.id])
task.get(timeout=5)
institution_record_es = InspireSearch.get_record_data_from_es(institution)
expected_number_of_paper = 3
assert expected_number_of_paper == institution_record_es[
"number_of_papers"]
def test_process_references_in_records_process_self_citations(
mock_batch_index, inspire_app, clean_celery_session,
enable_self_citations):
author_record = AuthorsRecord.create(
faker.record(
"aut",
data={
"name": {
"value": "'t Hooft, Gerardus",
"name_variants": ["'t Hooft, Gerard", "Hooft, Gerard T."],
"preferred_name": "Gerardus 't Hooft",
},
"ids": [
{
"value": "INSPIRE-00060582",
"schema": "INSPIRE ID"
},
{
"value": "G.tHooft.1",
"schema": "INSPIRE BAI"
},
],
},
))
author_record_2 = AuthorsRecord.create(
faker.record(
"aut",
data={
"name": {
"value": "'t Hooft, Gerardus Marcus",
"preferred_name": "Gerardus Marcus 't Hooft",
},
"ids": [
{
"value": "INSPIRE-00060583",
"schema": "INSPIRE ID"
},
{
"value": "G.tHooft.2",
"schema": "INSPIRE BAI"
},
],
},
))
lit_record = LiteratureRecord.create(
faker.record(
"lit",
data={
"authors": [{
"ids": [
{
"value": "INSPIRE-00060582",
"schema": "INSPIRE ID"
},
{
"value": "G.tHooft.1",
"schema": "INSPIRE BAI"
},
],
"full_name":
author_record["name"]["value"],
"record":
author_record["self"],
}]
},
))
lit_record_2 = LiteratureRecord.create(
faker.record(
"lit",
literature_citations=[lit_record["control_number"]],
data={
"authors": [{
"ids": [
{
"value": "INSPIRE-00060583",
"schema": "INSPIRE ID"
},
{
"value": "G.tHooft.2",
"schema": "INSPIRE BAI"
},
],
"full_name":
author_record_2["name"]["value"],
"record":
author_record_2["self"],
}]
},
))
db.session.commit()
def assert_records_in_es():
lit_record_from_es = InspireSearch.get_record_data_from_es(lit_record)
lit_record_from_es_2 = InspireSearch.get_record_data_from_es(
lit_record_2)
aut_record_from_es = InspireSearch.get_record_data_from_es(
author_record)
assert lit_record_from_es and aut_record_from_es and lit_record_from_es_2
retry_until_pass(assert_records_in_es, retry_interval=5)
models_committed.disconnect(index_after_commit)
lit_record["authors"].append({
"ids": [
{
"value": "INSPIRE-00060583",
"schema": "INSPIRE ID"
},
{
"value": "G.tHooft.2",
"schema": "INSPIRE BAI"
},
],
"full_name":
author_record_2["name"]["value"],
"record":
author_record_2["self"],
})
lit_record.update(dict(lit_record))
db.session.commit()
# reconnect signal before we call process_references_in_records
models_committed.connect(index_after_commit)
task = process_references_in_records.delay([lit_record.id])
task.get(timeout=5)
assert sorted(mock_batch_index.mock_calls[0][1][0]) == sorted(
[lit_record_2.id])