def handle(self, *args, **options):
if not args:
raise CommandError("No file or resource specified.")
info = options.get('verbosity') >= 2
store_id = options.get('store')
context_id = options.get('context')
source = args[0]
if info:
print("Parsing {}".format(source))
intermediate = Graph()
try:
intermediate.parse(source, format=options.get('format'))
except Exception as e:
raise CommandError(e)
if info:
print("Parsed {} triples".format(len(intermediate)))
identifier = URIRef(context_id) if context_id else BNode()
graph = utils.get_named_graph(identifier, store_id=store_id)
if info:
print("Storing {} triples".format(len(intermediate)))
for triple in intermediate:
graph.add(triple)
if info:
print("Done")
def handle(self, *args, **options):
if not args:
raise CommandError("No file or resource specified.")
info = options.get('verbosity') >= 2
store_id = options.get('store')
context_id = options.get('context')
source = args[0]
if info:
print "Parsing {0}".format(source)
intermediate = Graph()
try:
intermediate.parse(source, format=options.get('format'))
except Exception as e:
raise CommandError(e)
if info:
print "Parsed {0} triples".format(len(intermediate))
identifier = URIRef(context_id) if context_id else BNode()
graph = utils.get_named_graph(identifier, store_id=store_id)
if info:
print "Storing {0} triples".format(len(intermediate))
for triple in intermediate:
graph.add(triple)
if info:
print "Done"
def lineages(self, request, graph_identifier=None):
if graph_identifier:
graph = get_named_graph(graph_identifier)
else:
graph = get_conjunctive_graph()
object = self.node()
assert object
MAX_DISTANCE = 5
MAX_ITERATIONS = 5
history = get_history(request)
n_nodes = len(history)
out_paths = []
paths = [[[self, None]]] # start with only 1 path containing only a pseudo-edge
while paths:
path = paths[0] # 1st path
object = path[0][0].node() # left element (subject) of first edge becomes target object of previous triples
triples = list(graph.triples((None, None, object)))
triples += get_reified_triples(object, graph)
n_triples = len(triples)
i = 0
while n_triples and node_id(object) not in settings.EU_COUNTRY_KEYS:
i += 1
if i > MAX_ITERATIONS: # endless loops could result from bugs or wrong network
exit
best_k = 0
if n_triples > 1:
if history:
min_distance = MAX_DISTANCE
k_range = range(n_triples)
for k in k_range:
s, p, o = triples[k]
if node_id(s) in history:
distance = n_nodes - history.index(node_id(s))
if distance < min_distance:
best_k = k
s, p, o = triples[best_k]
best_edge = [make_item(s), Predicate(uriref=p)]
del triples[best_k]
if not history:
for triple in triples:
edge = [make_item(triple[0]), Predicate(uriref=triple[1])]
# paths.append(deepcopy([edge]) + deepcopy(path))
paths.append([edge] + path)
# path = deepcopy([best_edge]) + deepcopy(path)
path = [best_edge] + path
object = s
triples = list(graph.triples((None, None, object)))
triples += get_reified_triples(object, graph)
n_triples = len(triples)
out_paths.append(path[:-1])
paths = paths[1:]
# print_paths(out_paths)
return out_paths
def test_get_single_tag():
from rdflib_django import utils
graph = utils.get_named_graph('http://dbpedia.org/resource/Scotland')
graph.add((URIRef('http://dbpedia.org/resource/Scotland'), TOPICA.tag, URIRef('http://dbpedia.org/resource/Robert_Burns')))
graph.add((URIRef('http://dbpedia.org/resource/Robert_Burns'), RDFS.label, Literal('Robert Burns')))
item = Item(iri='http://dbpedia.org/resource/Scotland')
item.save()
assert len(item.get_tags()) == 1
assert Tag('http://dbpedia.org/resource/Robert_Burns', 'Robert Burns') in item.get_tags()
def handle(self, *args, **options):
store_id = options.get('store')
context_id = options.get('context')
target = args[0] if args else sys.stdout
if context_id:
graph = utils.get_named_graph(URIRef(context_id), store_id=store_id)
else:
graph = utils.get_conjunctive_graph(store_id)
#noinspection PyUnresolvedReferences
graph.serialize(target, format=options.get('format'))
def test_distance_between_overlapping_items():
from rdflib_django import utils
g1 = utils.get_named_graph('http://dbpedia.org/resource/Scotland')
g1.add((URIRef('http://dbpedia.org/resource/Scotland'), TOPICA.tag, URIRef('http://dbpedia.org/resource/Robert_Burns')))
g1.add((URIRef('http://dbpedia.org/resource/Robert_Burns'), RDFS.label, Literal('Robert Burns')))
g1.add((URIRef('http://dbpedia.org/resource/Scotland'), TOPICA.tag, URIRef('http://dbpedia.org/resource/United_Kingdom')))
g1.add((URIRef('http://dbpedia.org/resource/United_Kingdom'), RDFS.label, Literal('United Kingdom')))
g2 = utils.get_named_graph('http://dbpedia.org/resource/England')
g2.add((URIRef('http://dbpedia.org/resource/England'), TOPICA.tag, URIRef('http://dbpedia.org/resource/United_Kingdom')))
g2.add((URIRef('http://dbpedia.org/resource/United_Kingdom'), RDFS.label, Literal('United Kingdom')))
g2.add((URIRef('http://dbpedia.org/resource/England'), TOPICA.tag, URIRef('http://dbpedia.org/resource/William_Shakespeare')))
g2.add((URIRef('http://dbpedia.org/resource/William_Shakespeare'), RDFS.label, Literal('William Shakespeare')))
item1 = Item(iri='http://dbpedia.org/resource/Scotland')
item1.save()
item2 = Item(iri='http://dbpedia.org/resource/England')
item2.save()
assert item1.get_distance(item2) == 2.0/3.0
assert item2.get_distance(item1) == 2.0/3.0
def test_distance_between_identical_items_is_zero():
from rdflib_django import utils
g1 = utils.get_named_graph('http://dbpedia.org/resource/Scotland')
g1.add((URIRef('http://dbpedia.org/resource/Scotland'), TOPICA.tag, URIRef('http://dbpedia.org/resource/Robert_Burns')))
g1.add((URIRef('http://dbpedia.org/resource/Robert_Burns'), RDFS.label, Literal('Robert Burns')))
g1.add((URIRef('http://dbpedia.org/resource/Scotland'), TOPICA.tag, URIRef('http://dbpedia.org/resource/William_Wallace')))
g1.add((URIRef('http://dbpedia.org/resource/William_Wallace'), RDFS.label, Literal('William Wallace')))
g2 = utils.get_named_graph('http://dbpedia.org/resource/Alba')
g2.add((URIRef('http://dbpedia.org/resource/Alba'), TOPICA.tag, URIRef('http://dbpedia.org/resource/Robert_Burns')))
g2.add((URIRef('http://dbpedia.org/resource/Robert_Burns'), RDFS.label, Literal('Robert Burns')))
g2.add((URIRef('http://dbpedia.org/resource/Alba'), TOPICA.tag, URIRef('http://dbpedia.org/resource/William_Wallace')))
g2.add((URIRef('http://dbpedia.org/resource/William_Wallace'), RDFS.label, Literal('William Wallace')))
item1 = Item(iri='http://dbpedia.org/resource/Scotland')
item1.save()
item2 = Item(iri='http://dbpedia.org/resource/Alba')
item2.save()
assert item1.get_distance(item2) == 0.0
assert item2.get_distance(item1) == 0.0
def find_holes():
graph = get_named_graph(make_uriref('http://www.wikidata.org'))
for country_code in settings.EU_COUNTRY_KEYS:
country = Country(id=country_code)
holes = []
for prop_code in settings.EU_COUNTRY_PROPERTIES:
predicate = Predicate(id=prop_code)
objects = graph.objects(subject=country.uriref,
predicate=predicate.uriref)
if not list(objects):
holes.append(predicate.label())
print('- {}: {}'.format(country.label(), ', '.join(holes)))
def handle(self, *args, **options):
store_id = options.get('store')
context_id = options.get('context')
target = args[0] if args else sys.stdout
if context_id:
graph = utils.get_named_graph(URIRef(context_id),
store_id=store_id)
else:
graph = utils.get_conjunctive_graph(store_id)
# noinspection PyUnresolvedReferences
graph.serialize(target, format=options.get('format'))
def test_can_access_graph_via_item():
# Given
iri = 'http://dbpedia.org/resource/Kevin_Bacon'
from rdflib_django import utils
graph = utils.get_named_graph(iri)
graph.add((URIRef('http://dbpedia.org/resource/Kevin_Bacon'), RDF.type, Literal('Person')))
# When
item = Item(iri=iri)
item.save()
# Then
assert (URIRef('http://dbpedia.org/resource/Kevin_Bacon'), RDF.type, Literal('Person')) in item.graph
def list_statements(request, graph_identifier=None):
lang = get_language()[:2]
if graph_identifier:
graph = get_named_graph(graph_identifier)
else:
graph = get_conjunctive_graph()
statement_dicts = [{
'graph': 'wikidata',
'subject': friend_uri(s, lang=lang),
'predicate': friend_uri(p, lang=lang),
'object': friend_uri(o, lang=lang)
} for s, p, o in graph]
return render(request, 'list_statements.html',
{'statement_dicts': statement_dicts})
def removeProperty(request, item_code, graph_identifier=None):
if graph_identifier:
graph = get_named_graph(graph_identifier)
else:
graph = get_conjunctive_graph()
subject = make_node(item_code)
country_id = request.GET.get('c')
property = make_node(request.GET.get('p'))
datatype_id = request.GET.get('dt')
object_value_or_id = request.GET.get('o')
language = request.GET.get('lang', None)
triples = graph.triples((subject, property, None))
for triple in triples:
object = triple[2]
if isinstance(object, Literal):
if datatype_id:
if datatype_id == object.datatype_id(
) and object_value_or_id == str(object.value):
graph.remove(triple)
elif language and object.language and (
language == object.language) and (object_value_or_id
== object.value):
graph.remove(triple)
elif language and not object.language and (object_value_or_id
== object.value):
graph.remove(triple)
elif not language and not object.language and (object_value_or_id
== object.value):
graph.remove(triple)
elif object_value_or_id == str(object.value):
graph.remove(triple)
""" MMR 200723
elif language and language == object.language:
graph.remove(triple)
elif not language and not object.language:
graph.remove(triple)
"""
else:
if isinstance(object, BNode):
if object_value_or_id == str(object):
remove_node(object, graph)
graph.remove(triple)
elif isinstance(object, URIRef):
if id_from_uriref(object) == object_value_or_id:
remove_node(object, graph)
graph.remove(triple)
return HttpResponseRedirect(request.META.get('HTTP_REFERER'))
def remove_item(request, item_code, graph_identifier=None):
if graph_identifier:
graph = get_named_graph(graph_identifier)
else:
graph = get_conjunctive_graph()
item = item_from_id(item_code)
lineages = item.lineages(request)
parent, predicate = lineages[0][-1]
parent_code = parent.id
node = item.uriref
in_triple = next(graph.triples((parent.uriref, predicate.uriref, node)))
graph.remove(in_triple)
remove_node(node, graph)
if parent_code in settings.EU_COUNTRY_KEYS:
return HttpResponseRedirect('/country/{}/'.format(parent_code))
else:
return HttpResponseRedirect('/item/{}/'.format(parent_code))
def __init__(self, uriref=None, id=None, bnode=None, graph=None, graph_identifier=None, in_predicate=None):
assert uriref or id or bnode
self.uriref = self.id = self.bnode = None
if uriref:
self.uriref = uriref
self.id = id_from_uriref(uriref)
elif id:
self.id = id
self.uriref = make_uriref(id)
else:
self.bnode = bnode
self.id = str(bnode) # added 200708
if graph_identifier:
graph = get_named_graph(graph_identifier)
elif not graph:
graph = get_conjunctive_graph()
self.graph = graph
self.in_predicate = in_predicate
def __init__(self, *args, **kwargs):
super(Item, self).__init__(*args, **kwargs)
self.graph = utils.get_named_graph(self.iri)
def export_countries(format="rdf"):
target = "/tmp/countries.{}".format(format)
graph = get_named_graph(make_uriref('http://www.wikidata.org'))
graph.serialize(target, format=format)
