def test_extracting_identifiers_from_urls():
# Returns None when it should
assert dataone.extractIdentifierFromFullURL('asdf') is None
assert dataone.extractIdentifierFromFullURL(1) is None
assert dataone.extractIdentifierFromFullURL('1') is None
assert dataone.extractIdentifierFromFullURL('http://google.com') is None
# Extracts the right thing
assert dataone.extractIdentifierFromFullURL('https://cn.dataone.org/cn/v1/meta/some_pid') == 'some_pid'
assert dataone.extractIdentifierFromFullURL('https://cn.dataone.org/cn/v1/meta/kgordon.23.30') == 'kgordon.23.30'
assert dataone.extractIdentifierFromFullURL('https://cn.dataone.org/cn/v1/resolve/kgordon.23.30') == 'kgordon.23.30'
assert dataone.extractIdentifierFromFullURL('https://cn.dataone.org/cn/v1/object/kgordon.23.30') == 'kgordon.23.30'
assert dataone.extractIdentifierFromFullURL('https://cn.dataone.org/cn/v2/object/kgordon.23.30') == 'kgordon.23.30'
def addDatasetTriples(self, dataset_node, doc):
if self.model is None:
raise Exception("Model not found.")
identifier = dataone.extractDocumentIdentifier(doc)
identifier_esc = urllib.quote_plus(identifier)
# type Dataset
self.add(dataset_node, 'rdf:type', 'geolink:Dataset')
# Title
title_element = doc.find("./str[@name='title']")
if title_element is not None:
self.add(dataset_node, 'rdfs:label', RDF.Node(title_element.text))
# Add geolink:Identifier
self.addIdentifierTriples(dataset_node, identifier)
# Abstract
abstract_element = doc.find("./str[@name='abstract']")
if (abstract_element is not None):
self.add(dataset_node, 'geolink:description', RDF.Node(abstract_element.text))
# Spatial Coverage
bound_north = doc.find("./float[@name='northBoundCoord']")
bound_east = doc.find("./float[@name='eastBoundCoord']")
bound_south = doc.find("./float[@name='southBoundCoord']")
bound_west = doc.find("./float[@name='westBoundCoord']")
if all(ele is not None for ele in [bound_north, bound_east, bound_south, bound_west]):
if bound_north.text == bound_south.text and bound_west.text == bound_east.text:
wktliteral = "POINT (%s %s)" % (bound_north.text, bound_east.text)
else:
wktliteral = "POLYGON ((%s %s, %s %s, %s %s, %s, %s))" % (bound_west.text, bound_north.text, bound_east.text, bound_north.text, bound_east.text, bound_south.text, bound_west.text, bound_south.text)
self.add(dataset_node, 'geolink:hasGeometryAsWktLiteral', RDF.Node(wktliteral))
# Temporal Coverage
begin_date = doc.find("./date[@name='beginDate']")
end_date = doc.find("./date[@name='endDate']")
if begin_date is not None:
self.add(dataset_node, 'geolink:hasStartDate', RDF.Node(begin_date.text))
if end_date is not None:
self.add(dataset_node, 'geolink:hasEndDate', RDF.Node(end_date.text))
# Obsoletes as PROV#wasRevisionOf
obsoletes_node = doc.find("./str[@name='obsoletes']")
if obsoletes_node is not None:
other_document_esc = urllib.quote_plus(obsoletes_node.text)
self.add(dataset_node, 'prov:wasRevisionOf', RDF.Uri(self.repository.ns['d1dataset'] + other_document_esc))
# Landing page
self.add(dataset_node, 'geolink:hasLandingPage', RDF.Uri("https://search.dataone.org/#view/" + identifier_esc))
# Digital Objects
# If this document has a resource map, get digital objects from there
# Otherwise, use the cito:documents field in Solr
resource_map_identifiers = doc.findall("./arr[@name='resourceMap']/str")
if len(resource_map_identifiers) > 0:
for resource_map_node in resource_map_identifiers:
resource_map_identifier = resource_map_node.text
digital_objects = dataone.getAggregatedIdentifiers(resource_map_identifier)
for digital_object in digital_objects:
digital_object_identifier = urllib.unquote_plus(digital_object)
self.addDigitalObject(identifier, digital_object_identifier)
else:
# If no resourceMap or documents field, at least add the metadata
# file as a digital object
# dataUrl e.g. https://cn.dataone.org/cn/v1/resolve/doi%3A10.6073%2FAA%2Fknb-lter-cdr.70061.123
data_url_node = doc.find("./str[@name='dataUrl']")
if data_url_node is not None:
data_url = data_url_node.text
digital_object = dataone.extractIdentifierFromFullURL(data_url)
digital_object = urllib.unquote_plus(digital_object)
self.addDigitalObject(identifier, digital_object)
