def test_revert_revision_works_correctly_and_runs_update(inspire_app):
cited_data = {
"$schema": "http://localhost:5000/schemas/records/hep.json",
"titles": [{
"title": "Test a valid record"
}],
"document_type": ["article"],
"_collections": ["Literature"],
}
cited_record = LiteratureRecord.create(cited_data)
citing_data = {
"$schema":
"http://localhost:5000/schemas/records/hep.json",
"titles": [{
"title": "My title"
}],
"document_type": ["article"],
"_collections": ["Literature"],
"preprint_date":
"2019-06-28",
"references": [{
"record": {
"$ref":
f"http://localhost:5000/api/literature/{cited_record['control_number']}"
}
}],
}
citing_record = LiteratureRecord.create(citing_data)
db.session.commit()
assert len(citing_record.model.references) == 1
assert len(cited_record.model.citations) == 1
assert citing_record.revision_id == 2
citing_record = LiteratureRecord.get_record(citing_record.id)
data = dict(citing_record)
del data["references"]
citing_record.update(data)
db.session.commit()
citing_record = LiteratureRecord.get_record(citing_record.id)
assert len(citing_record.model.references) == 0
assert len(cited_record.model.citations) == 0
assert citing_record.revision_id == 4
citing_record.revert(2)
db.session.commit()
citing_record = LiteratureRecord.get_record(citing_record.id)
assert len(citing_record.model.references) == 1
assert len(cited_record.model.citations) == 1
# Reverted to revision 2 but added as next revision
# so it will be revision 6
assert citing_record.revision_id == 6
assert dict(citing_record.revisions[2]) == dict(citing_record)
def test_match_references(inspire_app, cli, celery_app_with_context,
celery_session_worker):
cited_data = {
"document_type": ["article"],
"dois": [{
"value": "10.1371/journal.pone.0188398"
}],
}
cited_record = create_record_async("lit", data=cited_data)
cited_record.index(
delay=False) # reference-matcher requires cited record to be indexed
citer_data = {
"references": [{
"reference": {
"dois": ["10.1371/journal.pone.0188398"]
}
}]
}
citer_record_1 = create_record_async("lit", data=citer_data)
citer_record_2 = create_record_async("lit", data=citer_data)
citer_record_3 = create_record_async("lit", data=citer_data)
citer_ids = [citer_record_1.id, citer_record_2.id, citer_record_3.id]
create_record_async("dat")
result = cli.invoke(["match", "references", "-bs", 2])
assert result.exit_code == 0
for citer_id in citer_ids:
updated_citer_record = LiteratureRecord.get_record(citer_id)
assert (get_value(updated_citer_record,
"references[0].record") == cited_record["self"])
def test_match_references_by_uuids_dedupes_references_after_matching(
inspire_app):
citer_data = {
"references": [
{
"reference": {
"report_numbers": ["AMBIGUOUS-42"]
},
"record": {
"$ref": "https://inspirehep.net/api/literature/1234"
},
},
{
"reference": {
"report_numbers": ["AMBIGUOUS-42"]
},
"record": {
"$ref": "https://inspirehep.net/api/literature/5678"
},
},
]
}
citer_record = create_record("lit", data=citer_data)
match_references_by_uuids([str(citer_record.id)])
updated_citer_record = LiteratureRecord.get_record(citer_record.id)
expected_references = [{"reference": {"report_numbers": ["AMBIGUOUS-42"]}}]
assert updated_citer_record["references"] == expected_references
def test_match_references_by_uuids(inspire_app):
cited_data = {
"document_type": ["article"],
"dois": [{
"value": "10.1371/journal.pone.0188398"
}],
}
cited_record = create_record("lit", data=cited_data)
citer_data = {
"references": [{
"reference": {
"dois": ["10.1371/journal.pone.0188398"]
}
}]
}
citer_record = create_record("lit", data=citer_data)
excluded_citer_record = create_record("lit",
data=citer_data) # won't be passed
deleted_record = create_record("lit",
data={
"deleted": True,
**citer_data
},
with_control_number=True)
record_without_references = create_record("lit")
match_references_by_uuids([
str(citer_record.id),
str(record_without_references.id),
str(deleted_record.id),
])
updated_citer_record = LiteratureRecord.get_record(citer_record.id)
excluded_citer_record = LiteratureRecord.get_record(
excluded_citer_record.id)
with pytest.raises(NoResultFound):
LiteratureRecord.get_record(deleted_record.id)
deleted_record = LiteratureRecord.get_record(deleted_record.id,
with_deleted=True)
assert (get_value(updated_citer_record,
"references[0].record") == cited_record["self"])
assert "record" not in get_value(excluded_citer_record, "references[0]")
assert "record" not in get_value(deleted_record, "references[0]")
def test_match_references_by_uuids(inspire_app):
cited_data = {
"document_type": ["article"],
"dois": [{"value": "10.1371/journal.pone.0188398"}],
}
cited_record = create_record("lit", data=cited_data)
citer_data = {
"references": [{"reference": {"dois": ["10.1371/journal.pone.0188398"]}}]
}
citer_record = create_record("lit", data=citer_data)
excluded_citer_record = create_record("lit", data=citer_data) # won't be passed
record_without_references = create_record("lit")
match_references_by_uuids([str(citer_record.id), str(record_without_references.id)])
updated_citer_record = LiteratureRecord.get_record(citer_record.id)
excluded_citer_record = LiteratureRecord.get_record(excluded_citer_record.id)
assert (
get_value(updated_citer_record, "references[0].record") == cited_record["self"]
)
assert "record" not in get_value(excluded_citer_record, "references[0]")
def test_match_references(inspire_app, cli, clean_celery_session):
cited_data = {
"document_type": ["article"],
"dois": [{
"value": "10.1371/journal.pone.0188398"
}],
}
cited_record = create_record_async("lit", data=cited_data)
cited_record.index(
delay=False) # reference-matcher requires cited record to be indexed
citer_data = {
"references": [{
"reference": {
"dois": ["10.1371/journal.pone.0188398"]
}
}]
}
citer_record_1 = create_record_async("lit", data=citer_data)
citer_record_2 = create_record_async("lit", data=citer_data)
citer_record_3 = create_record_async("lit", data=citer_data)
citer_ids = [citer_record_1.id, citer_record_2.id, citer_record_3.id]
record_data = create_record_async("dat")
record_data_uuids = record_data.id
def assert_all_records_are_indexed():
current_search.flush_and_refresh("*")
result = es_search("records-hep")
uuids = get_value(result, "hits.hits._id")
for uuid in citer_ids:
assert str(uuid) in uuids
result = es_search("records-data")
uuids = get_value(result, "hits.hits._id")
assert str(record_data_uuids) in uuids
retry_until_pass(assert_all_records_are_indexed)
result = cli.invoke(["match", "references", "-bs", 2])
assert result.exit_code == 0
for citer_id in citer_ids:
updated_citer_record = LiteratureRecord.get_record(citer_id)
assert (get_value(updated_citer_record,
"references[0].record") == cited_record["self"])
def test_redirect_and_delete_record_from_deleted_records_field(inspire_app):
record_to_delete = create_record("lit")
record = create_record("lit")
data = dict(record)
data["deleted_records"] = [record_to_delete["self"]]
record.update(data)
deleted_records = record["deleted_records"]
assert len(deleted_records) == 1
old_pid = record_to_delete.control_number_pid
assert old_pid.is_redirected()
record_redirected = LiteratureRecord.get_record_by_pid_value(
record_to_delete.control_number)
assert record_redirected.id == record.id
original_record = LiteratureRecord.get_record(record_to_delete.id,
with_deleted=True)
assert original_record["deleted"] is True
def test_fulltext_indexer_updates_documents_when_record_changed(
inspire_app, clean_celery_session, override_config):
with override_config(FEATURE_FLAG_ENABLE_FULLTEXT=True):
data = faker.record("lit")
data.update({
"arxiv_eprints": [{
"categories": ["hep-ph"],
"value": "hep-ph/9404247"
}],
"documents": [{
"source":
"arxiv",
"fulltext":
True,
"filename":
"arXiv:nucl-th_9310030.pdf",
"key":
"arXiv:nucl-th_9310030.pdf",
"url":
"https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf",
}],
})
record = LiteratureRecord.create(data)
db.session.commit()
def assert_record_in_es():
current_search.flush_and_refresh("*")
record_lit_es = (LiteratureSearch().get_record(str(
record.id)).execute().hits.hits[0])
assert "attachment" in record_lit_es._source["documents"][0]
retry_until_pass(assert_record_in_es, timeout=90, retry_interval=20)
record_first_attachment = (LiteratureSearch().get_record(str(
record.id)).execute().hits.hits[0]["_source"]["documents"][0]
["attachment"])
db.session.expire_all()
record = LiteratureRecord.get_record(record.id)
record["documents"] = [{
"source":
"arxiv",
"fulltext":
True,
"filename":
"new_doc.pdf",
"key":
"new_doc.pdf",
"url":
"http://www.africau.edu/images/default/sample.pdf",
}]
record.update(dict(record))
db.session.commit()
def assert_update_in_es():
current_search.flush_and_refresh("*")
record_lit_es = (LiteratureSearch().get_record(str(
record.id)).execute().hits.hits[0])
assert "new_doc.pdf" == record_lit_es._source["documents"][0][
"key"]
assert (record_first_attachment !=
record_lit_es._source["documents"][0]["attachment"])
retry_until_pass(assert_update_in_es, timeout=90, retry_interval=20)
