def test_add_multiple_sources(self, celery_app):
source1 = factories.SourceFactory()
source2 = factories.SourceFactory()
work = factories.AbstractCreativeWorkFactory(title='All about Canada')
data = {
'@id': IDObfuscator.encode(work),
'@type': 'creativework',
'title': 'All aboot Canada'
}
nd1 = NormalizedData.objects.create(source=source1.user,
data={'@graph': [data]})
nd2 = NormalizedData.objects.create(source=source2.user,
data={'@graph': [data]})
assert work.sources.count() == 0
celery_app.tasks['share.tasks.disambiguate'](nd1.id)
work.refresh_from_db()
assert work.title == 'All aboot Canada'
assert work.sources.count() == 1
celery_app.tasks['share.tasks.disambiguate'](nd2.id)
work.refresh_from_db()
assert work.title == 'All aboot Canada'
assert work.sources.count() == 2
def test_canonical(self, Graph, ingest, first_canonical, second_canonical,
change):
first_source = factories.SourceFactory(canonical=first_canonical)
second_source = factories.SourceFactory(canonical=second_canonical)
first_graph = Graph(
Preprint(
id=1,
title='The first title',
identifiers=[WorkIdentifier(1)],
))
cs = ingest(first_graph, user=first_source.user)
cw = cs.changes.first().target
assert cw.type == 'share.preprint'
assert cw.title == 'The first title'
second_graph = Graph(
Article(
id=1,
title='The Second Title',
identifiers=[WorkIdentifier(1)],
))
ingest(second_graph, user=second_source.user)
cw = models.AbstractCreativeWork.objects.get(id=cw.id)
assert cw.type == change.get('type', 'share.preprint')
assert cw.title == change.get('title', 'The first title')
def share_user():
user = ShareUser(username='******', )
user.save()
# add source
factories.SourceFactory(user=user),
return user
def test_no_icon(self, schema):
x = factories.AbstractCreativeWorkFactory()
source = factories.SourceFactory(icon='')
x.sources.add(source.user)
# Have to use % formats because of {}s everywhere
result = schema.execute('''
query {
creativeWork(id: "%s") {
id,
title,
description,
sources {
title
}
}
}
''' % (IDObfuscator.encode(x), ))
assert not result.errors
assert result.data == OrderedDict([
('creativeWork', OrderedDict([
('id', IDObfuscator.encode(x)),
('title', x.title),
('description', x.description),
('sources', [])
]))
])
def __init__(self,
elastic,
index=False,
num_identifiers=1,
num_sources=1,
date=None):
self.elastic = elastic
if date is None:
self.work = factories.AbstractCreativeWorkFactory()
else:
models.AbstractCreativeWork._meta.get_field(
'date_created').auto_now_add = False
self.work = factories.AbstractCreativeWorkFactory(
date_created=date,
date_modified=date,
)
models.AbstractCreativeWork._meta.get_field(
'date_created').auto_now_add = True
self.sources = [factories.SourceFactory() for _ in range(num_sources)]
self.work.sources.add(*[s.user for s in self.sources])
for i in range(num_identifiers):
factories.WorkIdentifierFactory(
uri='http://example.com/{}/{}'.format(self.work.id, i),
creative_work=self.work)
if index:
index_helpers(self)
def test_index_no_icon(self, elastic):
source = factories.SourceFactory(icon=None)
tasks.index_sources()
with pytest.raises(NotFoundError):
elastic.es_client.get(index=elastic.es_index,
doc_type='sources',
id=source.name)
def test_add_multiple_sources(self):
source1 = factories.SourceFactory()
source2 = factories.SourceFactory()
work = factories.AbstractCreativeWorkFactory(title='All about Canada')
data = [{'@id': IDObfuscator.encode(work), '@type': 'creativework', 'title': 'All aboot Canada'}]
assert work.sources.count() == 0
Ingester(data).as_user(source1.user).ingest(index=False)
work.refresh_from_db()
assert work.title == 'All aboot Canada'
assert work.sources.count() == 1
Ingester(data).as_user(source2.user).ingest(index=False)
work.refresh_from_db()
assert work.title == 'All aboot Canada'
assert work.sources.count() == 2
def test_index(self, elastic):
source = factories.SourceFactory()
tasks.index_sources()
doc = elastic.es_client.get(index=elastic.es_index,
doc_type='sources',
id=source.name)
assert doc['_id'] == source.name
assert doc['_source']['name'] == source.long_title
assert doc['_source']['short_name'] == source.name
def test_canonical(self, Graph, first_canonical, second_canonical, change):
first_source = factories.SourceFactory(canonical=first_canonical)
second_source = factories.SourceFactory(canonical=second_canonical)
first_graph = ChangeGraph(Graph(
Preprint(
id=1,
title='The first title',
identifiers=[WorkIdentifier(1)],
)),
namespace=first_source.user.username)
second_graph = ChangeGraph(Graph(
Article(
id=1,
title='The Second Title',
identifiers=[WorkIdentifier(1)],
)),
namespace=second_source.user.username)
first_graph.process()
(cw, _) = ChangeSet.objects.from_graph(
first_graph,
NormalizedDataFactory(source=first_source.user).id).accept()
assert cw.type == 'share.preprint'
assert cw.title == 'The first title'
second_graph.process()
ChangeSet.objects.from_graph(
second_graph,
NormalizedDataFactory(source=second_source.user).id).accept()
cw = models.AbstractCreativeWork.objects.get(id=cw.id)
assert second_graph.nodes[0].change == change
assert cw.type == change.get('type', 'share.preprint')
assert cw.title == change.get('title', 'The first title')
def make_source_config(context, label, name=None, interval=None, time=None):
kwargs = {'label': label}
if name is None:
kwargs['source'] = factories.SourceFactory()
else:
kwargs['source'] = models.Source.objects.get(name=name)
if interval is not None:
kwargs['harvest_interval'] = {
'daily': '1 day',
'weekly': '1 week',
'fortnightly': '2 weeks',
'yearly': '1 year',
'monthly': '1 month',
}[interval]
if time is not None:
kwargs['harvest_after'] = time
factories.SourceConfigFactory(**kwargs)
def test_creativework_fetcher(self):
works = [
factories.AbstractCreativeWorkFactory(),
factories.AbstractCreativeWorkFactory(is_deleted=True),
factories.AbstractCreativeWorkFactory(),
factories.AbstractCreativeWorkFactory(),
]
for work in works[:-1]:
factories.WorkIdentifierFactory(creative_work=work)
factories.WorkIdentifierFactory.create_batch(5, creative_work=works[0])
source = factories.SourceFactory()
works[1].sources.add(source.user)
# Trim trailing zeros
def iso(x):
return re.sub(r'(\.\d+?)0*\+', r'\1+', x.isoformat())
fetched = list(fetchers.CreativeWorkFetcher()(work.id
for work in works))
# TODO add more variance
assert fetched == [{
'id':
util.IDObfuscator.encode(work),
'type':
work._meta.verbose_name,
'types': [
cls._meta.verbose_name for cls in type(work).mro()
if hasattr(cls, '_meta') and cls._meta.proxy
],
'title':
work.title,
'description':
work.description,
'date':
iso(work.date_published),
'date_created':
iso(work.date_created),
'date_modified':
iso(work.date_modified),
'date_published':
iso(work.date_published),
'date_updated':
iso(work.date_updated),
'is_deleted':
work.is_deleted or not work.identifiers.exists(),
'justification':
getattr(work, 'justification', None),
'language':
work.language,
'registration_type':
getattr(work, 'registration_type', None),
'retracted':
work.outgoing_creative_work_relations.filter(
type='share.retracted').exists(),
'withdrawn':
getattr(work, 'withdrawn', None),
'identifiers':
list(work.identifiers.values_list('uri', flat=True)),
'sources': [user.source.long_title for user in work.sources.all()],
'subjects': [],
'subject_synonyms': [],
'tags': [],
'lists': {},
} for work in works]
def make_source(context, name):
if not hasattr(context, 'source'):
context.sources = {}
context.sources[name] = factories.SourceFactory(name=name)
context.subject = ('sources', context.sources[name])