def test_all_dataset_fields(self):
resources = ResourceFactory.build_batch(3)
dataset = DatasetFactory(tags=faker.words(nb=3), resources=resources,
frequency='daily', acronym='acro')
d = dataset_to_rdf(dataset)
g = d.graph
assert isinstance(d, RdfResource)
assert len(list(g.subjects(RDF.type, DCAT.Dataset))) is 1
assert g.value(d.identifier, RDF.type) == DCAT.Dataset
assert isinstance(d.identifier, URIRef)
uri = url_for('datasets.show_redirect',
dataset=dataset.id, _external=True)
assert str(d.identifier) == uri
assert d.value(DCT.identifier) == Literal(dataset.id)
assert d.value(DCT.title) == Literal(dataset.title)
assert d.value(SKOS.altLabel) == Literal(dataset.acronym)
assert d.value(DCT.description) == Literal(dataset.description)
assert d.value(DCT.issued) == Literal(dataset.created_at)
assert d.value(DCT.modified) == Literal(dataset.last_modified)
assert d.value(DCT.accrualPeriodicity).identifier == FREQ.daily
expected_tags = set(Literal(t) for t in dataset.tags)
assert set(d.objects(DCAT.keyword)) == expected_tags
assert len(list(d.objects(DCAT.distribution))) == len(resources)
def build_catalog(site, datasets, format=None):
'''Build the DCAT catalog for this site'''
site_url = url_for('site.home_redirect', _external=True)
catalog_url = url_for('site.rdf_catalog', _external=True)
graph = Graph(namespace_manager=namespace_manager)
catalog = graph.resource(URIRef(catalog_url))
catalog.set(RDF.type, DCAT.Catalog)
catalog.set(DCT.title, Literal(site.title))
catalog.set(DCT.language, Literal(current_app.config['DEFAULT_LANGUAGE']))
catalog.set(FOAF.homepage, URIRef(site_url))
publisher = graph.resource(BNode())
publisher.set(RDF.type, FOAF.Organization)
publisher.set(FOAF.name, Literal(current_app.config['SITE_AUTHOR']))
catalog.set(DCT.publisher, publisher)
for dataset in datasets:
rdf_dataset = dataset_to_rdf(dataset, graph)
if dataset.owner:
rdf_dataset.add(DCT.publisher, user_to_rdf(dataset.owner, graph))
elif dataset.organization:
rdf_dataset.add(DCT.publisher,
organization_to_rdf(dataset.organization, graph))
catalog.add(DCAT.dataset, rdf_dataset)
if isinstance(datasets, Paginable):
paginate_catalog(catalog, graph, datasets, format,
'site.rdf_catalog_format')
return catalog
def test_all_dataset_fields(self):
resources = ResourceFactory.build_batch(3)
dataset = DatasetFactory(tags=faker.words(nb=3), resources=resources,
frequency='daily')
d = dataset_to_rdf(dataset)
g = d.graph
self.assertIsInstance(d, RdfResource)
self.assertEqual(len(list(g.subjects(RDF.type, DCAT.Dataset))), 1)
self.assertEqual(g.value(d.identifier, RDF.type), DCAT.Dataset)
self.assertIsInstance(d.identifier, URIRef)
uri = url_for('datasets.show_redirect',
dataset=dataset.id, _external=True)
self.assertEqual(str(d.identifier), uri)
self.assertEqual(d.value(DCT.identifier), Literal(dataset.id))
self.assertEqual(d.value(DCT.title), Literal(dataset.title))
self.assertEqual(d.value(DCT.description),
Literal(dataset.description))
self.assertEqual(d.value(DCT.issued), Literal(dataset.created_at))
self.assertEqual(d.value(DCT.modified),
Literal(dataset.last_modified))
self.assertEqual(d.value(DCT.accrualPeriodicity).identifier,
FREQ.daily)
expected_tags = set(Literal(t) for t in dataset.tags)
self.assertEqual(set(d.objects(DCAT.keyword)), expected_tags)
self.assertEqual(len(list(d.objects(DCAT.distribution))),
len(resources))
def build_catalog(site, datasets, format=None):
'''Build the DCAT catalog for this site'''
site_url = url_for('site.home_redirect', _external=True)
catalog_url = url_for('site.rdf_catalog', _external=True)
graph = Graph(namespace_manager=namespace_manager)
catalog = graph.resource(URIRef(catalog_url))
catalog.set(RDF.type, DCAT.Catalog)
catalog.set(DCT.title, Literal(site.title))
catalog.set(DCT.language, Literal(current_app.config['DEFAULT_LANGUAGE']))
catalog.set(FOAF.homepage, URIRef(site_url))
publisher = graph.resource(BNode())
publisher.set(RDF.type, FOAF.Organization)
publisher.set(FOAF.name, Literal(current_app.config['SITE_AUTHOR']))
catalog.set(DCT.publisher, publisher)
for dataset in datasets:
catalog.add(DCAT.dataset, dataset_to_rdf(dataset, graph))
if isinstance(datasets, Paginable):
if not format:
raise ValueError('Pagination requires format')
catalog.add(RDF.type, HYDRA.Collection)
catalog.set(HYDRA.totalItems, Literal(datasets.total))
kwargs = {
'format': format,
'page_size': datasets.page_size,
'_external': True,
}
first_url = url_for('site.rdf_catalog_format', page=1, **kwargs)
page_url = url_for('site.rdf_catalog_format',
page=datasets.page,
**kwargs)
last_url = url_for('site.rdf_catalog_format',
page=datasets.pages,
**kwargs)
pagination = graph.resource(URIRef(page_url))
pagination.set(RDF.type, HYDRA.PartialCollectionView)
pagination.set(HYDRA.first, URIRef(first_url))
pagination.set(HYDRA.last, URIRef(last_url))
if datasets.has_next:
next_url = url_for('site.rdf_catalog_format',
page=datasets.page + 1,
**kwargs)
pagination.set(HYDRA.next, URIRef(next_url))
if datasets.has_prev:
prev_url = url_for('site.rdf_catalog_format',
page=datasets.page - 1,
**kwargs)
pagination.set(HYDRA.previous, URIRef(prev_url))
catalog.set(HYDRA.view, pagination)
return catalog
def test_from_external_repository(self):
dataset = DatasetFactory(extras={
'dct:identifier': 'an-identifier',
'uri': 'https://somewhere.org/dataset',
})
d = dataset_to_rdf(dataset)
assert isinstance(d.identifier, URIRef)
assert str(d.identifier) == 'https://somewhere.org/dataset'
assert d.value(DCT.identifier) == Literal('an-identifier')
def test_from_external_repository(self):
dataset = DatasetFactory(extras={
'dct:identifier': 'an-identifier',
'uri': 'https://somewhere.org/dataset',
})
d = dataset_to_rdf(dataset)
self.assertIsInstance(d.identifier, URIRef)
self.assertEqual(str(d.identifier), 'https://somewhere.org/dataset')
self.assertEqual(d.value(DCT.identifier), Literal('an-identifier'))
def test_temporal_coverage(self):
start = faker.past_date(start_date='-30d')
end = faker.future_date(end_date='+30d')
temporal_coverage = db.DateRange(start=start, end=end)
dataset = DatasetFactory(temporal_coverage=temporal_coverage)
d = dataset_to_rdf(dataset)
pot = d.value(DCT.temporal)
assert pot.value(RDF.type).identifier == DCT.PeriodOfTime
assert pot.value(SCHEMA.startDate).toPython() == start
assert pot.value(SCHEMA.endDate).toPython() == end
def test_with_owner(self):
user = UserFactory()
dataset = DatasetFactory(owner=user)
d = dataset_to_rdf(dataset)
g = d.graph
assert isinstance(d, RdfResource)
datasets = g.subjects(RDF.type, DCAT.Dataset)
users = g.subjects(RDF.type, FOAF.Person)
assert len(list(datasets)) == 1
assert len(list(users)) == 1
publisher = d.value(DCT.publisher)
assert publisher.value(RDF.type).identifier == FOAF.Person
def test_with_org(self):
org = OrganizationFactory()
dataset = DatasetFactory(organization=org)
d = dataset_to_rdf(dataset)
g = d.graph
assert isinstance(d, RdfResource)
datasets = g.subjects(RDF.type, DCAT.Dataset)
organizations = g.subjects(RDF.type, FOAF.Organization)
assert len(list(datasets)) == 1
assert len(list(organizations)) == 1
publisher = d.value(DCT.publisher)
assert publisher.value(RDF.type).identifier == FOAF.Organization
def test_with_owner(self):
user = UserFactory()
dataset = DatasetFactory(owner=user)
d = dataset_to_rdf(dataset)
g = d.graph
self.assertIsInstance(d, RdfResource)
datasets = g.subjects(RDF.type, DCAT.Dataset)
users = g.subjects(RDF.type, FOAF.Person)
self.assertEqual(len(list(datasets)), 1)
self.assertEqual(len(list(users)), 1)
publisher = d.value(DCT.publisher)
self.assertEqual(publisher.value(RDF.type).identifier, FOAF.Person)
def test_minimal(self):
dataset = DatasetFactory.build() # Does not have an URL
d = dataset_to_rdf(dataset)
g = d.graph
assert isinstance(d, RdfResource)
assert len(list(g.subjects(RDF.type, DCAT.Dataset))) is 1
assert g.value(d.identifier, RDF.type) == DCAT.Dataset
assert isinstance(d.identifier, BNode)
assert d.value(DCT.identifier) == Literal(dataset.id)
assert d.value(DCT.title) == Literal(dataset.title)
assert d.value(DCT.issued) == Literal(dataset.created_at)
assert d.value(DCT.modified) == Literal(dataset.last_modified)
def test_with_org(self):
org = OrganizationFactory()
dataset = DatasetFactory(organization=org)
d = dataset_to_rdf(dataset)
g = d.graph
self.assertIsInstance(d, RdfResource)
datasets = g.subjects(RDF.type, DCAT.Dataset)
organizations = g.subjects(RDF.type, FOAF.Organization)
self.assertEqual(len(list(datasets)), 1)
self.assertEqual(len(list(organizations)), 1)
publisher = d.value(DCT.publisher)
self.assertEqual(publisher.value(RDF.type).identifier,
FOAF.Organization)
def test_minimal(self):
dataset = DatasetFactory.build() # Does not have an URL
d = dataset_to_rdf(dataset)
g = d.graph
self.assertIsInstance(d, RdfResource)
self.assertEqual(len(list(g.subjects(RDF.type, DCAT.Dataset))), 1)
self.assertEqual(g.value(d.identifier, RDF.type), DCAT.Dataset)
self.assertIsInstance(d.identifier, BNode)
self.assertEqual(d.value(DCT.identifier), Literal(dataset.id))
self.assertEqual(d.value(DCT.title), Literal(dataset.title))
self.assertEqual(d.value(DCT.issued), Literal(dataset.created_at))
self.assertEqual(d.value(DCT.modified), Literal(dataset.last_modified))
def build_org_catalog(org, datasets, format=None):
graph = Graph(namespace_manager=namespace_manager)
org_catalog_url = url_for('organizations.rdf_catalog', org=org.id, _external=True)
catalog = graph.resource(URIRef(org_catalog_url))
catalog.set(RDF.type, DCAT.Catalog)
catalog.set(DCT.publisher, organization_to_rdf(org, graph))
for dataset in datasets:
catalog.add(DCAT.dataset, dataset_to_rdf(dataset, graph))
values = {'org': org.id}
if isinstance(datasets, Paginable):
paginate_catalog(catalog, graph, datasets, format, 'organizations.rdf_catalog_format', **values)
return catalog
