def test_infinite_regulate(self):
reg = Regulator()
reg._default_steps.graph_steps = (InfiniteGraphStep(),)
graph = MutableGraph()
graph.add_node(None, 'agent', {'name': 'Agent Agent'})
with pytest.raises(InfiniteRegulationError):
reg.regulate(graph)
def test_blocks(self, schemes, authorities, expected_identifiers):
identifiers = [
# (uri, scheme, authority)
('http://osf.io/mst3k/', 'http', 'osf.io'),
('mailto:[email protected]', 'mailto', 'example.com'),
('2049-3630', 'urn', 'issn'),
('0000-0002-1825-0097', 'http', 'orcid.org'),
]
step = NormalizeIRIs(blocked_schemes=schemes,
blocked_authorities=authorities)
graph = MutableGraph()
for uri, scheme, authority in identifiers:
node = graph.add_node('id_{}'.format(authority), 'workidentifier',
{'uri': uri})
assert node['scheme'] is None
assert node['host'] is None
step.regulate_node(node)
if scheme not in schemes and authority not in authorities:
assert node['scheme'] == scheme
assert node['host'] == authority
assert len(graph.filter_type('workidentifier')) == expected_identifiers
def test_calls_run(self, mock_steps, num_nodes):
graph = MutableGraph()
for i in range(num_nodes):
graph.add_node(i, 'creativework')
Regulator(regulator_config={'not': 'empty'}).regulate(graph)
assert all(s.run.call_count == 1 for st in mock_steps.values()
for s in st)
def test_infinite_regulate(self):
reg = Regulator()
reg._default_steps.graph_steps = (InfiniteGraphStep(), )
graph = MutableGraph()
graph.add_node(None, 'agent', {'name': 'Agent Agent'})
with pytest.raises(InfiniteRegulationError):
reg.regulate(graph)
def test_can_delete_work(self, john_doe, normalized_data):
graph = MutableGraph.from_jsonld([{
'@id': '_:abc',
'@type': 'workidentifier',
'uri': 'http://osf.io/faq',
'creative_work': {'@id': '_:789', '@type': 'preprint'}
}, {
'@id': '_:789',
'@type': 'preprint',
'title': 'All About Cats',
}])
change_set = ChangeSetBuilder(graph, normalized_data, disambiguate=True).build_change_set()
preprint, identifier = change_set.accept()
assert preprint.is_deleted is False
graph = MutableGraph.from_jsonld([{
'@id': '_:abc',
'@type': 'workidentifier',
'uri': 'http://osf.io/faq',
'creative_work': {'@id': '_:789', '@type': 'preprint'}
}, {
'@id': '_:789',
'is_deleted': True,
'@type': 'preprint',
}])
ChangeSetBuilder(graph, normalized_data, disambiguate=True).build_change_set().accept()
preprint.refresh_from_db()
assert preprint.is_deleted is True
def test_normalize_agentidentifier(self, input, output):
graph = MutableGraph()
node = graph.add_node('1', 'agentidentifier', uri=input)
NormalizeIRIs().regulate_node(node)
if output:
assert node['uri'] == output
else:
assert len(graph) == 0
def test_normalize_agentidentifier(self, input, output):
graph = MutableGraph()
node = graph.add_node('1', 'agentidentifier', {'uri': input})
NormalizeIRIs().regulate_node(node)
if output:
assert node['uri'] == output
else:
assert len(graph) == 0
def test_add_relation_related(self, normalized_data):
'''
A work exists. Add a second work with a relation to the first work.
The first work should have the appropriate inverse relation to the
second work.
'''
uri = 'http://osf.io/special-snowflake'
ChangeSetBuilder(MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'article',
'title': 'All About Cats',
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'article'}
}]), normalized_data).build_change_set().accept()
assert models.Article.objects.count() == 1
graph = MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'preprint',
'title': 'Dogs are okay too',
'related_works': [{'@id': '_:foo', '@type': 'cites'}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {'@id': '_:1234', '@type': 'preprint'},
'related': {'@id': '_:2345', '@type': 'creativework'},
}, {
'@id': '_:2345',
'@type': 'creativework',
'identifiers': [{'@id': '_:4567', '@type': 'workidentifier'}]
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:2345', '@type': 'creativework'}
}])
change_set = ChangeSetBuilder(graph, normalized_data, disambiguate=True).build_change_set()
change_set.accept()
assert models.Article.objects.count() == 1
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 0
cat = models.Article.objects.first()
dog = models.Preprint.objects.first()
assert dog.outgoing_creative_work_relations.count() == 1
assert dog.outgoing_creative_work_relations.first()._meta.model_name == 'cites'
assert dog.outgoing_creative_work_relations.first().related == cat
assert cat.incoming_creative_work_relations.count() == 1
assert cat.incoming_creative_work_relations.first()._meta.model_name == 'cites'
assert cat.incoming_creative_work_relations.first().subject == dog
def test_normalize_workidentifier(self, input, output):
graph = MutableGraph()
node = graph.add_node('1', 'workidentifier', {'uri': input})
step = NormalizeIRIs(blocked_schemes=['mailto'], blocked_authorities=['issn', 'orcid.org'])
step.regulate_node(node)
if output:
assert node['uri'] == output
else:
assert len(graph) == 0
def test_calls_steps(self, steps, num_nodes):
graph = MutableGraph()
for i in range(num_nodes):
graph.add_node(i, 'creativework')
Regulator().regulate(graph)
assert all(m.should_regulate.call_count == num_nodes for m in steps['node'])
assert all(m.regulate_node.call_count == num_nodes for m in steps['node'])
assert all(m.regulate_graph.call_count == 1 for m in steps['graph'])
assert all(m.validate_graph.call_count == 1 for m in steps['validation'])
def test_normalize_workidentifier(self, input, output):
graph = MutableGraph()
node = graph.add_node('1', 'workidentifier', uri=input)
step = NormalizeIRIs(blocked_schemes=['mailto'], blocked_authorities=['issn', 'orcid.org'])
step.regulate_node(node)
if output:
assert node['uri'] == output
else:
assert len(graph) == 0
def test_change_work_type(self, normalized_data):
'''
A CreativeWork with an Identifier exists. Accept a new changeset
with a Preprint with the same Identifier. The preprint should
disambiguate to the existing work, and the work's type should be
updated to Preprint
'''
title = 'Ambiguous Earthquakes'
uri = 'http://osf.io/special-snowflake'
cg = MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'project',
'title': title,
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'project'}
}])
original_change_set = ChangeSetBuilder(cg, normalized_data, disambiguate=True).build_change_set()
work, identifier = original_change_set.accept()
id = work.id
assert identifier.uri == uri
assert models.Project.objects.count() == 1
assert models.Preprint.objects.count() == 0
assert models.CreativeWork.objects.count() == 1
assert models.Project.objects.all()[0].changes.count() == 1
cg = MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'preprint',
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'preprint'}
}])
change_set = ChangeSetBuilder(cg, normalized_data, disambiguate=True).build_change_set()
change_set.accept()
assert models.Project.objects.count() == 0
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.count() == 1
assert models.Preprint.objects.get(id=id).title == title
assert models.Preprint.objects.all()[0].changes.count() == 2
def test_generic_creative_work(self, normalized_data):
'''
A Preprint with an Identifier exists. Accept a changeset with a
CreativeWork with the same Identifier and a different title.
The Preprint's title should be updated to the new value, but its type
should remain the same.
'''
old_title = 'Ambiguous Earthquakes'
uri = 'http://osf.io/special-snowflake'
original_change_set = ChangeSetBuilder(MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'preprint',
'title': old_title,
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'preprint'}
}]), normalized_data).build_change_set()
preprint, identifier = original_change_set.accept()
id = preprint.id
assert identifier.uri == uri
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 0
assert models.Preprint.objects.get(id=id).title == old_title
new_title = 'Ambidextrous Earthquakes'
graph = MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'creativework',
'title': new_title,
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'creativework'}
}])
change_set = ChangeSetBuilder(graph, normalized_data, disambiguate=True).build_change_set()
change_set.accept()
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 0
assert models.Preprint.objects.get(id=id).title == new_title
def change_node():
return next(n for n in MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'person',
'given_name': 'No',
'family_name': 'Matter',
}]))
def create_graph():
return MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'person',
'given_name': 'Jane',
'family_name': 'Doe',
}])
def change_node():
return next(n for n in MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'person',
'given_name': 'No',
'family_name': 'Matter',
}]))
def test_update_dependencies_accept(self, john_doe, normalized_data):
john_doe_id = IDObfuscator.encode(john_doe)
graph = MutableGraph.from_jsonld([{
'@id': john_doe_id,
'@type': 'person',
'given_name': 'Jane',
}, {
'@id': '_:456',
'@type': 'Creator',
'agent': {'@id': john_doe_id, '@type': 'person'},
'creative_work': {'@id': '_:789', '@type': 'preprint'},
}, {
'@id': '_:789',
'@type': 'preprint',
'title': 'All About Cats',
}])
change_set = ChangeSetBuilder(graph, normalized_data, matches={
john_doe_id: john_doe,
}).build_change_set()
change_set.accept()
john_doe.refresh_from_db()
assert john_doe.given_name == 'Jane'
assert models.Preprint.objects.filter(agent_relations__agent=john_doe).count() == 1
assert models.Preprint.objects.filter(agent_relations__agent=john_doe).first().title == 'All About Cats'
def create_graph():
return MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'person',
'given_name': 'Jane',
'family_name': 'Doe',
}])
def transform(self, datum, **kwargs):
"""Transform a RawDatum
Args:
datum: RawDatum to transform
**kwargs: Forwared to do_transform. Overrides values in the source config's transformer_kwargs
Returns a MutableGraph
"""
source_id = None
if not isinstance(datum, (str, bytes)):
source_id = datum.suid.identifier
datum = datum.datum
if isinstance(datum, bytes):
datum = datum.decode()
jsonld, root_ref = self.do_transform(datum, **self._get_kwargs(**kwargs))
if not jsonld:
return None
if source_id and jsonld and root_ref:
self.add_source_identifier(source_id, jsonld, root_ref)
# TODO return a MutableGraph from do_transform, maybe build it directly in Parser?
return MutableGraph.from_jsonld(jsonld)
def _consume_job(self, job, superfluous, force, apply_changes=True, index=True, urgent=False):
datum = None
# Check whether we've already done transform/regulate
if not superfluous:
datum = job.ingested_normalized_data.order_by('-created_at').first()
if superfluous or datum is None:
graph = self._transform(job)
if not graph:
return
graph = self._regulate(job, graph)
if not graph:
return
datum = NormalizedData.objects.create(
data={'@graph': graph.to_jsonld()},
source=job.suid.source_config.source.user,
raw=job.raw,
)
job.ingested_normalized_data.add(datum)
else:
graph = MutableGraph.from_jsonld(datum.data)
if apply_changes:
updated_work_ids = self._apply_changes(job, graph, datum)
if index and updated_work_ids:
self._update_index(updated_work_ids, urgent)
def accept_changes(context, nodes, username):
user = models.ShareUser.objects.get(username=username)
graph = MutableGraph.from_jsonld(nodes)
Regulator().regulate(graph)
nd = NormalizedDataFactory(source=user)
change_set = ChangeSetBuilder(graph, nd, disambiguate=True).build_change_set()
return change_set.accept() if change_set else None
def create_graph_dependencies():
return MutableGraph.from_jsonld([{
'@id': '_:123',
'@type': 'person',
'given_name': 'Jane',
'family_name': 'Doe',
}, {
'@id': '_:456',
'@type': 'Creator',
'agent': {
'@id': '_:123',
'@type': 'person'
},
'creative_work': {
'@id': '_:789',
'@type': 'preprint'
},
}, {
'@id':
'_:789',
'@type':
'preprint',
'title':
'All About Cats',
'related_agents': [{
'@id': '_:456',
'@type': 'Creator'
}]
}])
def _consume_job(self,
job,
superfluous,
force,
apply_changes=True,
index=True,
urgent=False):
datum = None
# Check whether we've already done transform/regulate
if not superfluous:
datum = job.ingested_normalized_data.order_by(
'-created_at').first()
if superfluous or datum is None:
graph = self._transform(job)
if not graph:
return
graph = self._regulate(job, graph)
if not graph:
return
datum = NormalizedData.objects.create(
data={'@graph': graph.to_jsonld()},
source=job.suid.source_config.source.user,
raw=job.raw,
)
job.ingested_normalized_data.add(datum)
else:
graph = MutableGraph.from_jsonld(datum.data)
if apply_changes:
updated_work_ids = self._apply_changes(job, graph, datum)
if index and updated_work_ids:
self._update_index(updated_work_ids, urgent)
def graph(self):
g = MutableGraph()
g.add_node(1, 'creativework', title='A work!', extra={
'foo': 'flooby',
'bah': 'hab',
})
g.add_node(2, 'creativework', title='Another work!', extra={
'extra': 'extra',
'bah': 'hab',
})
g.add_node(3, 'creativework', title='No extra :(')
return g
def test_can_delete_work(self, john_doe, normalized_data):
graph = MutableGraph.from_jsonld([{
'@id': '_:abc',
'@type': 'workidentifier',
'uri': 'http://osf.io/faq',
'creative_work': {
'@id': '_:789',
'@type': 'preprint'
}
}, {
'@id': '_:789',
'@type': 'preprint',
'title': 'All About Cats',
}])
change_set = ChangeSetBuilder(graph,
normalized_data,
disambiguate=True).build_change_set()
preprint, identifier = change_set.accept()
assert preprint.is_deleted is False
graph = MutableGraph.from_jsonld([{
'@id': '_:abc',
'@type': 'workidentifier',
'uri': 'http://osf.io/faq',
'creative_work': {
'@id': '_:789',
'@type': 'preprint'
}
}, {
'@id': '_:789',
'is_deleted': True,
'@type': 'preprint',
}])
ChangeSetBuilder(graph, normalized_data,
disambiguate=True).build_change_set().accept()
preprint.refresh_from_db()
assert preprint.is_deleted is True
def test_legacy_pipeline(self, legacy, monkeypatch):
mock_apply_changes = mock.Mock(return_value=[])
monkeypatch.setattr(
'share.tasks.jobs.IngestJobConsumer._apply_changes',
mock_apply_changes)
monkeypatch.setattr('django.conf.settings.SHARE_LEGACY_PIPELINE',
legacy)
g = MutableGraph()
g.add_node('_:id', 'creativework', title='This is a title')
job = factories.IngestJobFactory(
raw__datum=json.dumps({'@graph': g.to_jsonld(in_edges=False)}))
ingest.apply(kwargs={'job_id': job.id}, throw=True)
if legacy:
assert NormalizedData.objects.count() == 1
assert mock_apply_changes.call_count == 1
else:
assert NormalizedData.objects.count() == 0
assert not mock_apply_changes.called
def test_blocks(self, schemes, authorities, expected_identifiers):
identifiers = [
# (uri, scheme, authority)
('http://osf.io/mst3k/', 'http', 'osf.io'),
('mailto:[email protected]', 'mailto', 'example.com'),
('2049-3630', 'urn', 'issn'),
('0000-0002-1825-0097', 'http', 'orcid.org'),
]
step = NormalizeIRIs(blocked_schemes=schemes, blocked_authorities=authorities)
graph = MutableGraph()
for uri, scheme, authority in identifiers:
node = graph.add_node('id_{}'.format(authority), 'workidentifier', {'uri': uri})
assert node['scheme'] is None
assert node['host'] is None
step.regulate_node(node)
if scheme not in schemes and authority not in authorities:
assert node['scheme'] == scheme
assert node['host'] == authority
assert len(graph.filter_type('workidentifier')) == expected_identifiers
def create_graph_dependencies():
return MutableGraph.from_jsonld([{
'@id': '_:123',
'@type': 'person',
'given_name': 'Jane',
'family_name': 'Doe',
}, {
'@id': '_:456',
'@type': 'Creator',
'agent': {'@id': '_:123', '@type': 'person'},
'creative_work': {'@id': '_:789', '@type': 'preprint'},
}, {
'@id': '_:789',
'@type': 'preprint',
'title': 'All About Cats',
'related_agents': [{'@id': '_:456', '@type': 'Creator'}]
}])
def test_related_works(self, normalized_data):
'''
Create two works with a relation between them.
'''
uri = 'http://osf.io/special-snowflake'
change_set = ChangeSetBuilder(MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'preprint',
'title': 'Dogs are okay too',
'related_works': [{'@id': '_:foo', '@type': 'cites'}]
}, {
'@id': '_:2345',
'@type': 'creativework',
'title': 'Cats, tho',
'identifiers': [{'@id': '_:4567', '@type': 'workidentifier'}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {'@id': '_:1234', '@type': 'preprint'},
'related': {'@id': '_:2345', '@type': 'creativework'},
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:2345', '@type': 'creativework'}
}]), normalized_data).build_change_set()
change_set.accept()
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 1
p = models.Preprint.objects.first()
c = models.AbstractCreativeWork.objects.get(title='Cats, tho')
assert p.related_works.count() == 1
assert p.related_works.first() == c
assert p.outgoing_creative_work_relations.count() == 1
assert p.outgoing_creative_work_relations.first()._meta.model_name == 'cites'
assert p.outgoing_creative_work_relations.first().related == c
assert c.incoming_creative_work_relations.count() == 1
assert c.incoming_creative_work_relations.first()._meta.model_name == 'cites'
assert c.incoming_creative_work_relations.first().subject == p
def graph(self):
g = MutableGraph()
g.add_node(1, 'creativework', {
'title': 'A work!',
'extra': {
'foo': 'flooby',
'bah': 'hab',
},
})
g.add_node(2, 'creativework', {
'title': 'Another work!',
'extra': {
'extra': 'extra',
'bah': 'hab',
},
})
g.add_node(3, 'creativework', {'title': 'No extra :('})
return g
def test_update_dependencies_accept(self, john_doe, normalized_data):
john_doe_id = IDObfuscator.encode(john_doe)
graph = MutableGraph.from_jsonld([{
'@id': john_doe_id,
'@type': 'person',
'given_name': 'Jane',
}, {
'@id': '_:456',
'@type': 'Creator',
'agent': {
'@id': john_doe_id,
'@type': 'person'
},
'creative_work': {
'@id': '_:789',
'@type': 'preprint'
},
}, {
'@id': '_:789',
'@type': 'preprint',
'title': 'All About Cats',
}])
change_set = ChangeSetBuilder(graph,
normalized_data,
matches={
john_doe_id: john_doe,
}).build_change_set()
change_set.accept()
john_doe.refresh_from_db()
assert john_doe.given_name == 'Jane'
assert models.Preprint.objects.filter(
agent_relations__agent=john_doe).count() == 1
assert models.Preprint.objects.filter(
agent_relations__agent=john_doe).first().title == 'All About Cats'
def test_change_work_type(self, normalized_data):
'''
A CreativeWork with an Identifier exists. Accept a new changeset
with a Preprint with the same Identifier. The preprint should
disambiguate to the existing work, and the work's type should be
updated to Preprint
'''
title = 'Ambiguous Earthquakes'
uri = 'http://osf.io/special-snowflake'
cg = MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'project',
'title':
title,
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'project'
}
}])
original_change_set = ChangeSetBuilder(
cg, normalized_data, disambiguate=True).build_change_set()
work, identifier = original_change_set.accept()
id = work.id
assert identifier.uri == uri
assert models.Project.objects.count() == 1
assert models.Preprint.objects.count() == 0
assert models.CreativeWork.objects.count() == 1
assert models.Project.objects.all()[0].changes.count() == 1
cg = MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'preprint',
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'preprint'
}
}])
change_set = ChangeSetBuilder(cg, normalized_data,
disambiguate=True).build_change_set()
change_set.accept()
assert models.Project.objects.count() == 0
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.count() == 1
assert models.Preprint.objects.get(id=id).title == title
assert models.Preprint.objects.all()[0].changes.count() == 2
def test_calls_run(self, mock_steps, num_nodes):
graph = MutableGraph()
for i in range(num_nodes):
graph.add_node(i, 'creativework')
Regulator(regulator_config={'not': 'empty'}).regulate(graph)
assert all(s.run.call_count == 1 for st in mock_steps.values() for s in st)
def from_graph(self, jsonld, disambiguate=False):
nd = NormalizedData.objects.create(data=jsonld, source=share_user)
graph = MutableGraph.from_jsonld(jsonld)
return ChangeSetBuilder(graph, nd, disambiguate=disambiguate).build_change_set()
def update_graph(jane_doe):
return MutableGraph.from_jsonld([{
'@id': IDObfuscator.encode(jane_doe),
'@type': 'person',
'family_name': 'Dough',
}])
def test_add_work_with_existing_relation(self, normalized_data):
'''
Harvest a work that has a relation to some work identified by a DOI.
The related work should be a CreativeWork with no information except
the one Identifier.
Then harvest a work with the same DOI. It should update the
CreativeWork's type and attributes instead of creating a new work.
'''
uri = 'http://osf.io/special-snowflake'
ChangeSetBuilder(
MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'preprint',
'title':
'Dogs are okay',
'related_works': [{
'@id': '_:foo',
'@type': 'cites'
}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {
'@id': '_:1234',
'@type': 'preprint'
},
'related': {
'@id': '_:2345',
'@type': 'creativework'
},
}, {
'@id':
'_:2345',
'@type':
'creativework',
'identifiers': [{
'@id': '_:4567',
'@type': 'workidentifier'
}]
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:2345',
'@type': 'creativework'
}
}]), normalized_data).build_change_set().accept()
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 1
assert models.Preprint.objects.count() == 1
assert models.Article.objects.count() == 0
change = MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'article',
'title':
'All About Cats',
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'article'
}
}])
ChangeSetBuilder(change, normalized_data,
disambiguate=True).build_change_set().accept()
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 0
assert models.Article.objects.count() == 1
assert models.Preprint.objects.count() == 1
cat = models.Article.objects.first()
dog = models.Preprint.objects.first()
assert dog.outgoing_creative_work_relations.count() == 1
assert dog.outgoing_creative_work_relations.first(
)._meta.model_name == 'cites'
assert dog.outgoing_creative_work_relations.first().related == cat
assert cat.incoming_creative_work_relations.count() == 1
assert cat.incoming_creative_work_relations.first(
)._meta.model_name == 'cites'
assert cat.incoming_creative_work_relations.first().subject == dog
def update_graph(jane_doe):
return MutableGraph.from_jsonld([{
'@id': IDObfuscator.encode(jane_doe),
'@type': 'person',
'family_name': 'Dough',
}])
def test_add_relation_related(self, normalized_data):
'''
A work exists. Add a second work with a relation to the first work.
The first work should have the appropriate inverse relation to the
second work.
'''
uri = 'http://osf.io/special-snowflake'
ChangeSetBuilder(
MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'article',
'title':
'All About Cats',
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'article'
}
}]), normalized_data).build_change_set().accept()
assert models.Article.objects.count() == 1
graph = MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'preprint',
'title':
'Dogs are okay too',
'related_works': [{
'@id': '_:foo',
'@type': 'cites'
}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {
'@id': '_:1234',
'@type': 'preprint'
},
'related': {
'@id': '_:2345',
'@type': 'creativework'
},
}, {
'@id':
'_:2345',
'@type':
'creativework',
'identifiers': [{
'@id': '_:4567',
'@type': 'workidentifier'
}]
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:2345',
'@type': 'creativework'
}
}])
change_set = ChangeSetBuilder(graph,
normalized_data,
disambiguate=True).build_change_set()
change_set.accept()
assert models.Article.objects.count() == 1
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 0
cat = models.Article.objects.first()
dog = models.Preprint.objects.first()
assert dog.outgoing_creative_work_relations.count() == 1
assert dog.outgoing_creative_work_relations.first(
)._meta.model_name == 'cites'
assert dog.outgoing_creative_work_relations.first().related == cat
assert cat.incoming_creative_work_relations.count() == 1
assert cat.incoming_creative_work_relations.first(
)._meta.model_name == 'cites'
assert cat.incoming_creative_work_relations.first().subject == dog
def test_add_work_with_existing_relation(self, normalized_data):
'''
Harvest a work that has a relation to some work identified by a DOI.
The related work should be a CreativeWork with no information except
the one Identifier.
Then harvest a work with the same DOI. It should update the
CreativeWork's type and attributes instead of creating a new work.
'''
uri = 'http://osf.io/special-snowflake'
ChangeSetBuilder(MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'preprint',
'title': 'Dogs are okay',
'related_works': [{'@id': '_:foo', '@type': 'cites'}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {'@id': '_:1234', '@type': 'preprint'},
'related': {'@id': '_:2345', '@type': 'creativework'},
}, {
'@id': '_:2345',
'@type': 'creativework',
'identifiers': [{'@id': '_:4567', '@type': 'workidentifier'}]
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:2345', '@type': 'creativework'}
}]), normalized_data).build_change_set().accept()
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 1
assert models.Preprint.objects.count() == 1
assert models.Article.objects.count() == 0
change = MutableGraph.from_jsonld([{
'@id': '_:1234',
'@type': 'article',
'title': 'All About Cats',
'identifiers': [{'@id': '_:2345', '@type': 'workidentifier'}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {'@id': '_:1234', '@type': 'article'}
}])
ChangeSetBuilder(change, normalized_data, disambiguate=True).build_change_set().accept()
assert models.CreativeWork.objects.filter(type='share.creativework').count() == 0
assert models.Article.objects.count() == 1
assert models.Preprint.objects.count() == 1
cat = models.Article.objects.first()
dog = models.Preprint.objects.first()
assert dog.outgoing_creative_work_relations.count() == 1
assert dog.outgoing_creative_work_relations.first()._meta.model_name == 'cites'
assert dog.outgoing_creative_work_relations.first().related == cat
assert cat.incoming_creative_work_relations.count() == 1
assert cat.incoming_creative_work_relations.first()._meta.model_name == 'cites'
assert cat.incoming_creative_work_relations.first().subject == dog
def mutable_graph(mutable_graph_nodes):
return MutableGraph.from_jsonld(mutable_graph_nodes)
def from_graph(self, jsonld, disambiguate=False):
nd = NormalizedData.objects.create(data=jsonld, source=share_user)
graph = MutableGraph.from_jsonld(jsonld)
return ChangeSetBuilder(
graph, nd, disambiguate=disambiguate).build_change_set()
def test_generic_creative_work(self, normalized_data):
'''
A Preprint with an Identifier exists. Accept a changeset with a
CreativeWork with the same Identifier and a different title.
The Preprint's title should be updated to the new value, but its type
should remain the same.
'''
old_title = 'Ambiguous Earthquakes'
uri = 'http://osf.io/special-snowflake'
original_change_set = ChangeSetBuilder(
MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'preprint',
'title':
old_title,
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'preprint'
}
}]), normalized_data).build_change_set()
preprint, identifier = original_change_set.accept()
id = preprint.id
assert identifier.uri == uri
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 0
assert models.Preprint.objects.get(id=id).title == old_title
new_title = 'Ambidextrous Earthquakes'
graph = MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'creativework',
'title':
new_title,
'identifiers': [{
'@id': '_:2345',
'@type': 'workidentifier'
}]
}, {
'@id': '_:2345',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:1234',
'@type': 'creativework'
}
}])
change_set = ChangeSetBuilder(graph,
normalized_data,
disambiguate=True).build_change_set()
change_set.accept()
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 0
assert models.Preprint.objects.get(id=id).title == new_title
def mutable_graph(mutable_graph_nodes):
return MutableGraph.from_jsonld(mutable_graph_nodes)
def test_related_works(self, normalized_data):
'''
Create two works with a relation between them.
'''
uri = 'http://osf.io/special-snowflake'
change_set = ChangeSetBuilder(
MutableGraph.from_jsonld([{
'@id':
'_:1234',
'@type':
'preprint',
'title':
'Dogs are okay too',
'related_works': [{
'@id': '_:foo',
'@type': 'cites'
}]
}, {
'@id':
'_:2345',
'@type':
'creativework',
'title':
'Cats, tho',
'identifiers': [{
'@id': '_:4567',
'@type': 'workidentifier'
}]
}, {
'@id': '_:foo',
'@type': 'cites',
'subject': {
'@id': '_:1234',
'@type': 'preprint'
},
'related': {
'@id': '_:2345',
'@type': 'creativework'
},
}, {
'@id': '_:4567',
'@type': 'workidentifier',
'uri': uri,
'creative_work': {
'@id': '_:2345',
'@type': 'creativework'
}
}]), normalized_data).build_change_set()
change_set.accept()
assert models.Preprint.objects.count() == 1
assert models.CreativeWork.objects.filter(
type='share.creativework').count() == 1
p = models.Preprint.objects.first()
c = models.AbstractCreativeWork.objects.get(title='Cats, tho')
assert p.related_works.count() == 1
assert p.related_works.first() == c
assert p.outgoing_creative_work_relations.count() == 1
assert p.outgoing_creative_work_relations.first(
)._meta.model_name == 'cites'
assert p.outgoing_creative_work_relations.first().related == c
assert c.incoming_creative_work_relations.count() == 1
assert c.incoming_creative_work_relations.first(
)._meta.model_name == 'cites'
assert c.incoming_creative_work_relations.first().subject == p