def test_regression_3(self):
cluster.objects.create(id=1, name="lorem ipsum dolor")
result = [
obj.name for obj in search_title(
"lorem ipsum dolores", threshold=0.5, limit=5)
]
self.assertEqual(len(result), 1)
def test_regression_2(self):
cluster.objects.create(
id=1,
name=
"high temperature bonding solutions enabling thin wafer process and handling on 3dic manufacturing"
)
result = search_title("TSV process solution for 3D-IC.")
print(result)
def test_match2_no_results(self):
cluster.objects.create(id=1, name="lorem ipsum dolor title")
cluster.objects.create(id=2, name="lorem hababhu maklu title")
cluster.objects.create(id=3, name="dolor maraku rtrua title")
cluster.objects.create(id=4, name="dolor maraku rtrua title lorem")
cluster.objects.create(id=5, name="lorem ipsum dor")
cluster.objects.create(id=6, name="lorem ipsum dolor")
result = [obj.name for obj in search_title("funny bunny")]
self.assertEqual(result, [])
def test_regression_4(self):
title = normalize_title(
"Beneficial effects of quercetin on sperm parameters in streptozotocin induced diabetic male rats"
)
cluster.objects.create(id=1, name=title)
result = search_title(
"Beneficial effects of quercetin on sperm parameters in streptozotocin‐induced diabetic male rats"
)
# test fails so far
self.assertEqual(len(result), 1)
def test_regression_3(self):
title = normalize_title(
"Physical activity 153 and incidence of coronary heart disease in middle-aged women Relative weight gain and obesity as a child predict metabolic and men"
)
cluster.objects.create(id=1, name=title)
result = search_title(
"Physical activity and incidence of coronary heart disease in middle-aged women and men."
)
# test fails so far
self.assertEqual(len(result), 1)
def test_match_limit(self):
cluster.objects.create(id=1, name="lorem ipsum dolor title")
cluster.objects.create(id=2, name="lorem ipsum title")
cluster.objects.create(id=3, name="dolor lorem")
cluster.objects.create(id=4, name="dolor lorem")
cluster.objects.create(id=5, name="lorem ipsum dor")
cluster.objects.create(id=6, name="lorem ipsum dolor")
result = [
obj.name for obj in search_title("lorem", threshold=0, limit=5)
]
self.assertEqual(len(result), 5)
def test_regression(self):
cluster.objects.create(
id=1, name="shrec10 track nonrigid 3d shape retrieval")
cluster.objects.create(
id=2,
name=
"shrec 11 track shape retrieval on nonrigid 3d watertight meshes")
cluster.objects.create(
id=3,
name=
"retrieval of nonrigid textured shapes using low quality 3d models"
)
cluster.objects.create(
id=4, name="canonical forms for nonrigid 3d shape retrieval")
result = [
obj.name for obj in search_title("Non-rigid 3D Shape Retrieval.")
if obj is not None
]
self.assertEqual(len(result), 2)
def run(self, test_mode=False):
today = date.today()
compare_date = date.today(
) if test_mode is False else date.today() + datetime.timedelta(days=1)
reference_list = []
# fetch all single_references of an open_ref
# fetch open_references to gather all ingester url_objects
openreference_list = OpenReferences.objects.filter(
last_updated__lt=compare_date).select_related('ingester_key').only(
'ingester_key')[:self.limit]
for open_ref in openreference_list:
open_ref.last_updated = today
open_ref.save()
single_ref_list = SingleReference.objects.filter(
source=open_ref).filter(Q(status='OP') | Q(status='INC')).only(
'tries', 'title', 'status').all()
for single_ref in single_ref_list:
title_matches = search_title(single_ref.title, threshold=0.8)
if len(title_matches) == 1:
# single match: create reference for only match
PubReference.objects.get_or_create(
reference=title_matches[0],
source=open_ref.ingester_key,
defaults={
'original_title': single_ref.title,
'original_key': single_ref.id
})
#reference_list.append(PubReference(reference=title_matches[0], source=open_ref.ingester_key, original_title=single_ref.title))
single_ref.status = 'FIN'
elif len(title_matches) == 0:
# no match: increment tries and set as incomplete
single_ref.tries += 1
single_ref.status = 'INC' if single_ref.tries < LIMBO_LIMIT else 'LIM'
else:
# multi match:
normal_title = normalize_title(single_ref.title)
single_ref.status = 'LIM'
for title in title_matches:
if title.name == normal_title:
single_ref.status = 'FIN'
PubReference.objects.get_or_create(
reference=title,
source=open_ref.ingester_key,
defaults={
'original_title': single_ref.title,
'original_key': single_ref.id
})
#reference_list.append(PubReference(reference=title, source=open_ref.ingester_key, original_title=single_ref.title))
single_ref.save()
try:
PubReference.objects.bulk_create(reference_list)
except IntegrityError:
# duplicate key on insert, insert everything manually
for element in reference_list:
print("Rollback")
PubReference.objects.get_or_create(
reference=element.reference,
source=element.source,
original_title=single_ref.title,
original_key=single_ref.id)