def test_opinionscited_creation(self) -> None:
"""Make sure that found citations are stored in the database as
OpinionsCited objects with the appropriate references and depth.
"""
# Opinion fixture info:
# pk=10 is our mock citing opinion, containing a number of references
# to other mocked opinions, mixed about. It's hard to exhaustively
# test all combinations, but this test case is made to be deliberately
# complex, in an effort to "trick" the algorithm. Cited opinions:
# pk=7: 1 FullCaseCitation, 1 ShortCaseCitation, 1 SupraCitation (depth=3)
# pk=8: 1 FullCaseCitation, 2 IdCitation (one Id. and one Ibid.),
# 1 ShortCaseCitation, 2 SupraCitation (depth=6)
# pk=9: 1 FullCaseCitation, 1 ShortCaseCitation (depth=2)
remove_citations_from_imported_fixtures()
citing = Opinion.objects.get(pk=10)
find_citations_for_opinion_by_pks.delay([10])
test_pairs = [
(Opinion.objects.get(pk=7), 3),
(Opinion.objects.get(pk=8), 6),
(Opinion.objects.get(pk=9), 2),
]
for cited, depth in test_pairs:
with self.subTest(
f"Testing OpinionsCited creation for {cited}...",
cited=cited,
depth=depth,
):
self.assertEqual(
OpinionsCited.objects.get(citing_opinion=citing,
cited_opinion=cited).depth,
depth,
)
def test_citation_increment(self) -> None:
"""Make sure that found citations update the increment on the cited
opinion's citation count"""
remove_citations_from_imported_fixtures()
# Updates d1's citation count in a Celery task
find_citations_for_opinion_by_pks.delay([3])
cited = Opinion.objects.get(pk=2)
expected_count = 1
self.assertEqual(
cited.cluster.citation_count,
expected_count,
msg="'cited' was not updated by a citation found in 'citing', or "
"the citation was not found. Count was: %s instead of %s" %
(cited.cluster.citation_count, expected_count),
)
def test_citation_matching(self):
"""Creates a few documents that contain specific citations, then
attempts to find and match those citations.
This becomes a bit of an integration test, which is fine.
"""
remove_citations_from_imported_fixtures()
# Updates d1's citation count in a Celery task
find_citations_for_opinion_by_pks.delay([3])
cited = Opinion.objects.get(pk=2)
expected_count = 1
self.assertEqual(
cited.cluster.citation_count,
expected_count,
msg=u"'cited' was not updated by a citation found in 'citing', or "
u"the citation was not found. Count was: %s instead of %s"
% (cited.cluster.citation_count, expected_count)
)
def update_documents(self, opinion_pks, count):
sys.stdout.write('Graph size is {0:d} nodes.\n'.format(self.count))
sys.stdout.flush()
index_during_subtask = False
if self.index == 'concurrently':
index_during_subtask = True
chunk = []
chunk_size = 100
processed_count = 0
throttle = CeleryThrottle(min_items=500)
for opinion_pk in opinion_pks:
processed_count += 1
last_item = (count == processed_count)
chunk.append(opinion_pk)
if processed_count % chunk_size == 0 or last_item:
throttle.maybe_wait()
find_citations_for_opinion_by_pks.delay(
chunk, index_during_subtask)
chunk = []
self.log_progress(processed_count, opinion_pk)