def main():
store = SPARQLStore(SPARQL_ENDPOINT)
g = ConjunctiveGraph(store=store)
# g.bind("sg", "http://www.springernature.com/scigraph/ontologies/core/")
# get a few articles
q1 = g.query(ALL_ARTICLES_IDS_SAMPLE)
for row in q1:
print("Article URI:", str(row[0]))
# extract more article info
for row in q1:
try:
with time_limit(MAX_TIMEOUT):
raw = g.query(ARTICLE_INFO_QUERY % str(row[0]))
g1 = ConjunctiveGraph()
g1.parse(data=raw.serialize())
# create JSON-LD
context = {
"@vocab": "http://elastic-index.scigraph.com/",
"@language": "en"
}
print(g1.serialize(format='json-ld', context=context,
indent=4))
print("======")
except TimeoutException, msg:
error = "Timed out!"
print(error)
except Exception, e:
error = "Exception: %s" % e
print(error)
def test_exclude_xhtml(self):
ns = "http://www.w3.org/1999/xhtml/vocab#"
kg = ConjunctiveGraph()
kg.add((
BNode(),
URIRef("http://www.w3.org/1999/xhtml/vocab#role"),
URIRef("http://www.w3.org/1999/xhtml/vocab#button"),
))
print(kg.serialize(format="turtle"))
q_xhtml = ('SELECT * WHERE { ?s ?p ?o . FILTER (strstarts(str(?p), "' +
ns + '"))}')
print(q_xhtml)
res = kg.query(q_xhtml)
self.assertEquals(len(res), 1)
q_del = (
'DELETE {?s ?p ?o} WHERE { ?s ?p ?o . FILTER (strstarts(str(?p), "'
+ ns + '"))}')
kg.update(q_del)
print(kg.serialize(format="turtle"))
res = kg.query(q_xhtml)
self.assertEquals(len(res), 0)
def validate_sparql_endpoint(form, field):
try:
g = ConjunctiveGraph('SPARQLStore')
g.open(field.data)
g.query('SELECT * WHERE { ?s ?p ?o } LIMIT 1')
except:
raise ValidationError('This is not a valid SPARQL endpoint.')
def test_OLS(self):
print(util.cache_OLS.values)
turtle_edam = self.turtle_edam
kg = ConjunctiveGraph()
kg.parse(data=turtle_edam, format="turtle")
query_classes = self.query_classes
query_properties = self.query_properties
table_content = {"classes": [], "properties": []}
qres = kg.query(query_classes)
for row in qres:
table_content["classes"].append({"name": row["class"], "tag": []})
qres = kg.query(query_properties)
for row in qres:
table_content["properties"].append({
"name": row["prop"],
"tag": []
})
start = datetime.now().timestamp()
class_or_property_found = False
for c in table_content["classes"]:
if util.ask_OLS(c["name"]):
c["tag"].append("OLS")
class_or_property_found = True
print(c)
for p in table_content["properties"]:
if util.ask_OLS(p["name"]):
p["tag"].append("OLS")
class_or_property_found = True
print(p)
end = datetime.now().timestamp()
delta = end - start
print(f"OLS check done in {delta}")
self.assertTrue(class_or_property_found, True)
self.assertGreaterEqual(delta, 2)
print(util.cache_OLS.values)
# check that cache is working --> fast answers
start = datetime.now().timestamp()
for c in table_content["classes"]:
if util.ask_OLS(c["name"]):
c["tag"].append("OLS")
class_or_property_found = True
print(c)
for p in table_content["properties"]:
if util.ask_OLS(p["name"]):
p["tag"].append("OLS")
class_or_property_found = True
print(p)
end = datetime.now().timestamp()
delta = end - start
print(f"OLS check done in {delta}")
self.assertLessEqual(delta, 0.1)
print(util.cache_OLS.values)
def test_LOV(self):
turtle_edam = self.turtle_edam
kg = ConjunctiveGraph()
kg.parse(data=turtle_edam, format="turtle")
query_classes = self.query_classes
query_properties = self.query_properties
table_content = {"classes": [], "properties": []}
qres = kg.query(query_classes)
for row in qres:
table_content["classes"].append({"name": row["class"], "tag": []})
print(f'{row["class"]}')
qres = kg.query(query_properties)
print("Prop")
for row in qres:
table_content["properties"].append({
"name": row["prop"],
"tag": []
})
print(f'{row["prop"]}')
class_or_property_found = False
for c in table_content["classes"]:
if util.ask_LOV(c["name"]):
c["tag"].append("LOV")
class_or_property_found = True
for p in table_content["properties"]:
if util.ask_LOV(p["name"]):
p["tag"].append("LOV")
class_or_property_found = True
self.assertTrue(class_or_property_found, True)
def test_BioPortal(self):
turtle_edam = self.turtle_edam
kg = ConjunctiveGraph()
kg.parse(data=turtle_edam, format="turtle")
query_classes = self.query_classes
query_properties = self.query_properties
table_content = {"classes": [], "properties": []}
qres = kg.query(query_classes)
print("Class")
for row in qres:
table_content["classes"].append({"name": row["class"], "tag": []})
print(f'{row["class"]}')
qres = kg.query(query_properties)
print("Prop")
for row in qres:
table_content["properties"].append({
"name": row["prop"],
"tag": []
})
print(f'{row["prop"]}')
for c in table_content["classes"]:
if util.ask_BioPortal(c["name"], "class"):
c["tag"].append("BioPortal")
for p in table_content["properties"]:
if util.ask_BioPortal(c["name"], "property"):
c["tag"].append("BioPortal")
def validate_sparql_endpoint(form, field):
try:
g = ConjunctiveGraph('SPARQLStore')
g.open(field.data)
g.query('SELECT * WHERE { ?s ?p ?o } LIMIT 1')
except:
raise ValidationError('This is not a valid SPARQL endpoint.')
def __init__(self, raw_result):
self._detect_fails(raw_result)
rdf = Namespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
c = Namespace('http://s.opencalais.com/1/pred/')
g = Graph()
self.graph = g
g.parse(StringIO(raw_result.decode('utf-8').encode('utf-8')))
self.categories = [row for row in g.query(CATEGORY_QUERY['SPARQL'])]
self.entities = [row for row in g.query(ENTITY_QUERY['SPARQL'])]
def __init__(self, raw_result):
rdf = Namespace("http://www.w3.org/1999/02/22-rdf-syntax-ns#")
c = Namespace("http://s.opencalais.com/1/pred/")
g = Graph()
self.graph = g
g.parse(StringIO(raw_result.decode('utf-8').encode('utf-8')))
self.categories = [row for row in g.query(CATEGORY_QUERY["SPARQL"])]
self.entities = [row for row in g.query(ENTITY_QUERY["SPARQL"])]
def test_import_jsonld_into_named_graph(flatten_before_import: bool):
"""Test named graphs we use."""
graph = ConjunctiveGraph()
jsonld_document = JSONLD_DOCUMENT
if flatten_before_import:
jsonld_document = jsonld.flatten(jsonld_document)
serialized_document = json.dumps(jsonld_document, indent=4)
graph.parse(
data=serialized_document,
format='json-ld',
publicID=PUBLIC_ID,
)
print(graph.serialize(format='n3').decode('utf-8'))
# Make sure only one NAMED GRAPH is created on import.
assert list(map(
operator.itemgetter(Variable('g')),
graph.query(
'SELECT DISTINCT ?g WHERE { GRAPH ?g { ?s ?p ?o . } } ORDER BY ?g',
).bindings,
)) == [
URIRef('https://myblog.net/rdf/'),
]
# The information in @included section was properly parsed.
assert graph.query('''
SELECT * WHERE {
GRAPH ?g {
ex:Rover rdfs:subClassOf ex:Robot .
}
}
''', initNs=NAMESPACES).bindings == [{
Variable('g'): PUBLIC_ID,
}]
# The information in the root was properly parsed.
assert graph.query('''
SELECT * WHERE {
GRAPH ?g {
blog:JSONLD-and-named-graphs a schema:blogPost .
}
}
''', initNs=NAMESPACES).bindings == [{
Variable('g'): PUBLIC_ID,
}]
def test_import_jsonld_into_named_graph():
"""Test named graphs we use."""
graph = ConjunctiveGraph()
serialized_document = json.dumps(JSONLD_DOCUMENT, indent=4)
graph.parse(
data=serialized_document,
format='json-ld',
# All the semantic data about my blog is stored in a particular
# named graph.
publicID=PUBLIC_ID,
)
assert list(map(
operator.itemgetter(Variable('g')),
graph.query(
'SELECT DISTINCT ?g WHERE { GRAPH ?g { ?s ?p ?o . } } ORDER BY ?g',
).bindings,
)) == [
URIRef('https://blog.me/JSONLD-and-named-graphs'),
URIRef('https://myblog.net/rdf/'),
]
# I wanted **one** named graph but got two of them. That's because, when
# the system was importing contents of @graph block, it was using @id
# as name for this graph.
assert graph.query('''
SELECT * WHERE {
GRAPH ?g {
ex:Rover rdfs:subClassOf ex:Robot .
}
}
''', initNs=NAMESPACES).bindings == [{
Variable('g'): URIRef('https://blog.me/JSONLD-and-named-graphs'),
}]
# `publicID` was used for the part of data which was on the top level
# of the document.
assert graph.query('''
SELECT * WHERE {
GRAPH ?g {
blog:JSONLD-and-named-graphs a schema:blogPost .
}
}
''', initNs=NAMESPACES).bindings == [{
Variable('g'): PUBLIC_ID,
}]
def query_handler(source_id, query, context):
# Attempt to cnonicalize this object by round-tripping decoding and re-encoding
query = json.loads(json.dumps(query, sort_keys=True))
rdf_store = context["rdf_store"]
graph = ConjunctiveGraph(rdf_store)
graph.open(DB_URI)
results = graph.query(
"select ?subject ?predicate ?object where {?subject ?predicate ?object}"
)
# Inelegant, but works: serialize to JSON string, then re-parse
statements = json.loads(results.serialize(format="json"))
graph.close()
context = {
"@context": {
"@vocab": "http://facta.kendra.io/vocab#",
"kv": "http://facta.kendra.io/vocab#",
"kendra": "http://kendra.io/types#",
"kuid": "http://kendra.io/uuid#",
"schema": "http://schema.org/",
"xsd": "http://www.w3.org/2001/XMLSchema#"
}
}
compacted_graph, contained = s2j.result_data_to_jsonld(
statements, context)
result = s2j.extract_salient_results(
compacted_graph, contained,
["kendra:InclusionRelationship", "kendra:TextSelection"])
result = s2j.quicksearch(result, query)
return {"result": result}
class Config(object):
def __init__(self, masterGraph, hubHost):
self.etcd = etcd3.client(host=hubHost, port=9022)
self.masterGraph = masterGraph
self.hubHost = hubHost
self.configGraph = ConjunctiveGraph()
self.boards = []
self.etcPrefix = 'pi/'
self.reread()
deferToThread(self.watchEtcd)
def watchEtcd(self):
events, cancel = self.etcd.watch_prefix(self.etcPrefix)
reactor.addSystemEventTrigger('before', 'shutdown', cancel)
for ev in events:
log.info('%s changed', ev.key)
reactor.callFromThread(self.configChanged)
def configChanged(self):
self.cancelRead()
self.rereadLater = reactor.callLater(.1, self.reread)
def cancelRead(self):
if getattr(self, 'rereadLater', None):
self.rereadLater.cancel()
self.rereadLater = None
@STATS.configReread.time()
def reread(self):
self.rereadLater = None
log.info('read config')
self.configGraph = ConjunctiveGraph()
for v, md in self.etcd.get_prefix(self.etcPrefix):
log.info(' read file %r', md.key)
self.configGraph.parse(StringInputSource(v), format='n3')
self.configGraph.bind('', ROOM)
self.configGraph.bind('rdf', RDF)
# config graph is too noisy; maybe make it a separate resource
#masterGraph.patch(Patch(addGraph=self.configGraph))
self.setupBoards()
def setupBoards(self):
thisHost = Literal(hostname)
for row in self.configGraph.query(
'SELECT ?board WHERE { ?board a :PiBoard; :hostname ?h }',
initBindings=dict(h=thisHost)):
thisBoard = row.board
break
else:
log.warn("config had no board for :hostname %s. Waiting for config update." %
thisHost)
self.boards = []
return
log.info("found config for board %r" % thisBoard)
self.boards = [Board(self.configGraph, self.masterGraph, thisBoard, self.hubHost)]
self.boards[0].startPolling()
def test_dSet_parsed_as_context_returns_results(self):
querystr = """SELECT DISTINCT ?s FROM <http://test/> { ?s ?p ?o }"""
graph = ConjunctiveGraph()
graph.get_context(URIRef('http://test/')
).parse("http://www.w3.org/People/Berners-Lee/card.rdf")
r = graph.query(querystr, loadContexts=True)
self.assert_(len(r.bindings) is not 0)
def _construct(compiler, sources, query=None):
dataset = ConjunctiveGraph()
if not isinstance(sources, list):
sources = [sources]
for sourcedfn in sources:
source = sourcedfn['source']
graph = dataset.get_context(URIRef(sourcedfn.get('dataset') or source))
if isinstance(source, (dict, list)):
context_data = sourcedfn['context']
if not isinstance(context_data, list):
context_data = compiler.load_json(context_data )['@context']
context_data = [compiler.load_json(ctx)['@context']
if isinstance(ctx, unicode) else ctx
for ctx in context_data]
to_rdf(source, graph, context_data=context_data)
elif isinstance(source, Graph):
graph += source
else:
graph += compiler.cached_rdf(source)
if not query:
return graph
with compiler.path(query).open() as fp:
result = dataset.query(fp.read())
g = Graph()
for spo in result:
g.add(spo)
return g
def get_entities_from_text(text):
"""
This function extract RDF graph for textual entities, their YAGO type, beginindex and endindex, processed by PIKES tool.
input:
text – Image caption of Flickr30k entities dataset
output:
sparql query result – stored textual entities recognized and linked by PIKES
"""
pikes_answer = pikes_text2rdf(text.lower())
g = ConjunctiveGraph()
g.parse(data = pikes_answer.content.decode('utf-8'),format="trig")
#Sparql query for entities information extraction
sparql_query = """SELECT ?TED ?TEM ?TET ?anchor ?beginindex ?endindex
WHERE {
GRAPH ?g1 {?TED <http://groundedannotationframework.org/gaf#denotedBy> ?TEM}
GRAPH ?g2 {?TED a ?TET}
GRAPH ?g3 {?TEM nif:anchorOf ?anchor}
GRAPH ?g4 {?TEM nif:beginIndex ?beginindex}
GRAPH ?g5 {?TEM nif:endIndex ?endindex}
}"""
return g.query(sparql_query)
def calculate_edges((offset_limit, config)):
g = ConjunctiveGraph(config['graph_store'], config['graph_identifier'])
g.open(config['db_configstring'], create=False)
engine = create_engine(config['alchemy_configstring'])
Session = sessionmaker(bind=engine)
session = Session()
results = session.query(Node).filter_by(node_type="subject").order_by(
Node.node_id.desc()).offset(offset_limit[0]).limit(
offset_limit[1]).all()
for result in results:
query = "SELECT ?p ?o WHERE { <%s> ?p ?o }" % result.node_uri
items = g.query(query)
for item in items:
if str(item[0]
) != "http://www.w3.org/1999/02/22-rdf-syntax-ns#type":
neighbors = session.query(Node).filter_by(
node_uri=item[1]).all()
if len(neighbors) > 0:
for neighbor in neighbors:
result.add_neighbors(str(item[0]), neighbor)
session.commit()
g.close()
session.close()
def test_flowcells_index_rdfa(self):
model = ConjunctiveGraph()
response = self.client.get(reverse('flowcell_index'))
self.assertEqual(response.status_code, 200)
model.parse(data=smart_text(response.content), format='rdfa')
add_default_schemas(model)
inference = Infer(model)
errmsgs = list(inference.run_validation())
self.assertEqual(len(errmsgs), 0, errmsgs)
body = """prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
select ?flowcell
where {
?flowcell a libns:IlluminaFlowcell .
}"""
bindings = set(['flowcell'])
count = 0
for r in model.query(body):
count += 1
self.assertEqual(count, len(FlowCell.objects.all()))
class FOAF(callbacks.Privmsg):
DATAFILE = '/var/www/rc98.net/zoia.rdf'
def __init__(self, irc):
self.g = Graph()
# self.g.parse('http://rc98.net/zoia.rdf')
self.g.parse(self.DATAFILE, format="xml")
self.uri = rdflib.URIRef('http://www.code4lib.org/id/zoia')
self.FOAF = Namespace('http://xmlns.com/foaf/0.1/')
super(callbacks.Plugin, self).__init__(irc)
def _uri_of_user(self, nick):
result = self.g.query("""
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?uri WHERE
{<http://www.code4lib.org/id/zoia> foaf:knows ?uri . ?uri foaf:nick ?nick .}
""",
initBindings={'nick': nick})
if len(result) > 0:
userURI = list(result)[0][0]
return (userURI)
else:
return (None)
def _user_graph(self, uri):
userGraph = Graph()
try:
userGraph.parse(uri)
except Exception, e:
u = "http://www.w3.org/2007/08/pyRdfa/extract?space-preserve=true&uri=" + uri
userGraph.parse(u, identifier=uri)
return userGraph
class FOAF(callbacks.Privmsg):
DATAFILE = "/var/www/rc98.net/zoia.rdf"
def __init__(self, irc):
self.g = Graph()
# self.g.parse('http://rc98.net/zoia.rdf')
self.g.parse(self.DATAFILE, format="xml")
self.uri = rdflib.URIRef("http://www.code4lib.org/id/zoia")
self.FOAF = Namespace("http://xmlns.com/foaf/0.1/")
super(callbacks.Plugin, self).__init__(irc)
def _uri_of_user(self, nick):
result = self.g.query(
"""
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?uri WHERE
{<http://www.code4lib.org/id/zoia> foaf:knows ?uri . ?uri foaf:nick ?nick .}
""",
initBindings={"nick": nick},
)
if len(result) > 0:
userURI = list(result)[0][0]
return userURI
else:
return None
def _user_graph(self, uri):
userGraph = Graph()
try:
userGraph.parse(uri)
except Exception, e:
u = "http://www.w3.org/2007/08/pyRdfa/extract?space-preserve=true&uri=" + uri
userGraph.parse(u, identifier=uri)
return userGraph
def get_all_measurement_types(ontology_file):
graph = ConjunctiveGraph()
graph.load(ontology_file, format="n3")
query_str = '''SELECT DISTINCT ?mt ?label ?comment ?defn
WHERE {
?mt rdfs:label ?label .
?mt rdfs:subClassOf <%s> .
?mt rdfs:subClassOf ?r1 .
?r1 owl:onProperty oboe:measuresEntity ; owl:someValuesFrom ?ent .
?mt rdfs:subClassOf ?r2 .
?r2 owl:onProperty oboe:measuresCharacteristic ; owl:someValuesFrom ?char .
OPTIONAL { ?mt rdfs:comment ?comment }
OPTIONAL { ?mt skos:definition ?defn }
}''' % (MeasurementType)
qres = list(graph.query(query_str, initNs=dict(oboe=URIRef("http://ecoinformatics.org/oboe/oboe.1.2/oboe-core.owl#"),
owl=OWL,rdfs=RDFS,skos=SKOS)))
if len(qres) > 0:
qres.sort(key=lambda x: x[0], reverse=True)
result = dict()
i = 0
for row in qres:
result[i] = {'uri' : row[0], 'label' : row[1], 'comment' : row[2], 'defn' : row[3]}
i = i + 1
print "Sparql query finished!"
return result
return None
def run_query(source, query_file, context=None):
graph = ConjunctiveGraph()
fmt = ('json-ld' if source.endswith('.jsonld') else guess_format(source))
graph.parse(source, format=fmt, context=context)
with open(query_file) as fp:
query_text = fp.read().decode('utf-8')
result = graph.query(query_text)
for row in result:
print('\t'.join(t.n3() if t else 'UNDEF' for t in row).encode('utf-8'))
class TestSparqlJsonResults(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.parse(StringIO(test_data), format="n3")
def test_base_ref(self):
rt = self.graph.query(test_query).serialize("python")
self.failUnless(rt[0] == Literal("Alice"),
"Expected:\n 'Alice' \nGot:\n %s" % rt)
class TestSparqlJsonResults(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.parse(StringIO(test_data), format="n3")
def test_base_ref(self):
rt=self.graph.query(test_query).serialize("python")
self.failUnless(rt[0] == Literal("Alice"),"Expected:\n 'Alice' \nGot:\n %s" % rt)
def testOrderBy(self):
graph = ConjunctiveGraph(plugin.get('IOMemory', Store)())
graph.parse(StringIO(test_data), format="n3")
results = graph.query(test_query)
self.failUnless(False not in [
r[0] == a
for r, a in zip(results, ['Alice', 'Bob', 'Charlie', 'Dave'])
])
def runTest(self):
"""Run the specified task force test."""
import rdfadict
import rdfadict.sink.graph
print self._uri
# set up our target sink
g = Graph()
sink = rdfadict.sink.graph.GraphSink(g)
# run rdfadict over the input
parser = rdfadict.RdfaParser()
parser.parseurl(self._source, sink)
# execute the test SPARQL
g.query(self._result)
print g.selected
def runTest(self):
"""Run the specified task force test."""
import rdfadict
import rdfadict.sink.graph
print self._uri
# set up our target sink
g = Graph()
sink = rdfadict.sink.graph.GraphSink(g)
# run rdfadict over the input
parser = rdfadict.RdfaParser()
parser.parseurl(self._source, sink)
# execute the test SPARQL
g.query(self._result)
print g.selected
class TestSparqlOPT_FILTER2(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.load(StringIO(testContent), format='n3')
def test_OPT_FILTER(self):
results = self.graph.query(QUERY,
DEBUG=False)
results = list(results)
self.failUnless(
results == [(doc1,)],
"expecting : %s . Got: %s"%([(doc1,)],repr(results)))
def sparql(
instance: rdflib.ConjunctiveGraph,
query: str,
**kwargs: str,
) -> SPARQLResult:
bindings = {
argument_name: argument_value
for argument_name, argument_value in kwargs.items()
}
return instance.query(query, initBindings=bindings)
class TestSparqlOPT_FILTER2(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.load(StringIO(testContent), format='n3')
def test_OPT_FILTER(self):
results = self.graph.query(QUERY, DEBUG=False)
results = list(results)
self.failUnless(
results == [(doc1, )],
"expecting : %s . Got: %s" % ([(doc1, )], repr(results)))
def test_simple_recursion(self):
graph = ConjunctiveGraph()
graph.load(StringIO(BASIC_KNOWS_DATA), format='n3')
results = graph.query(KNOWS_QUERY, processor="sparql", DEBUG=False)
results = set(results)
person1 = URIRef('ex:person.1')
person2 = URIRef('ex:person.2')
nose.tools.assert_equal(
results,
set([(person1, None), (person1, Literal('person 3')),
(person2, Literal('person 3'))]))
def test_secondary_recursion(self):
graph = ConjunctiveGraph()
graph.load(StringIO(SUBCLASS_DATA), format='n3')
results = graph.query(SUBCLASS_QUERY, processor="sparql", DEBUG=False)
results = set(results)
ob = URIRef('ex:ob')
class1 = URIRef('ex:class.1')
class2 = URIRef('ex:class.2')
class3 = URIRef('ex:class.3')
nose.tools.assert_equal(
results, set([(ob, class1), (ob, class2), (ob, class3)]))
class JSON(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph(plugin.get('IOMemory', Store)())
self.graph.parse(StringIO(test_data), format="n3")
def testComma(self):
"""
Verify the serialisation of the data as json contains an exact
substring, with the comma in the correct place.
"""
results = self.graph.query(test_query)
result_json = results.serialize(format='json')
self.failUnless(result_json.find(correct) > 0)
def testHeader(self):
"""
Verify that the "x", substring is omitted from the serialised output.
"""
results = self.graph.query(test_header_query)
result_json = results.serialize(format='json')
self.failUnless(result_json.find('"x",') == -1)
def test_named_filter_graph_query():
g = ConjunctiveGraph()
g.namespace_manager.bind('rdf', RDF)
g.namespace_manager.bind('rdfs', RDFS)
ex = Namespace('https://ex.com/')
g.namespace_manager.bind('ex', ex)
g.get_context(ex.g1).parse(format="turtle", data=f"""
PREFIX ex: <{str(ex)}>
PREFIX rdfs: <{str(RDFS)}>
ex:Boris rdfs:label "Boris" .
ex:Susan rdfs:label "Susan" .
""")
g.get_context(ex.g2).parse(format="turtle", data=f"""
PREFIX ex: <{str(ex)}>
ex:Boris a ex:Person .
""")
assert list(g.query("SELECT ?l WHERE { GRAPH ex:g1 { ?a rdfs:label ?l } ?a a ?type }",
initNs={'ex': ex})) == [(Literal('Boris'),)]
assert list(g.query("SELECT ?l WHERE { GRAPH ex:g1 { ?a rdfs:label ?l } FILTER EXISTS { ?a a ?type }}",
initNs={'ex': ex})) == [(Literal('Boris'),)]
assert list(g.query("SELECT ?l WHERE { GRAPH ex:g1 { ?a rdfs:label ?l } FILTER NOT EXISTS { ?a a ?type }}",
initNs={'ex': ex})) == [(Literal('Susan'),)]
assert list(g.query("SELECT ?l WHERE { GRAPH ?g { ?a rdfs:label ?l } ?a a ?type }",
initNs={'ex': ex})) == [(Literal('Boris'),)]
assert list(g.query("SELECT ?l WHERE { GRAPH ?g { ?a rdfs:label ?l } FILTER EXISTS { ?a a ?type }}",
initNs={'ex': ex})) == [(Literal('Boris'),)]
assert list(g.query("SELECT ?l WHERE { GRAPH ?g { ?a rdfs:label ?l } FILTER NOT EXISTS { ?a a ?type }}",
initNs={'ex': ex})) == [(Literal('Susan'),)]
def test_simple_recursion(self):
graph = ConjunctiveGraph()
graph.load(StringIO(BASIC_KNOWS_DATA), format='n3')
results = graph.query(KNOWS_QUERY,
processor="sparql",
DEBUG=False)
results = set(results)
person1 = URIRef('ex:person.1')
person2 = URIRef('ex:person.2')
nose.tools.assert_equal(
results,
set([(person1, None), (person1, Literal('person 3')),
(person2, Literal('person 3'))]))
class JSON(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph(plugin.get('IOMemory',Store)())
self.graph.parse(StringIO(test_data), format="n3")
def testComma(self):
"""
Verify the serialisation of the data as json contains an exact
substring, with the comma in the correct place.
"""
results = self.graph.query(test_query)
result_json = results.serialize(format='json')
self.failUnless(result_json.find(correct) > 0)
def testHeader(self):
"""
Verify that the "x", substring is omitted from the serialised output.
"""
results = self.graph.query(test_header_query)
result_json = results.serialize(format='json')
self.failUnless(result_json.find('"x",') == -1)
class TestSPARQLFilters(unittest.TestCase):
debug = False
sparql = True
def setUp(self):
NS = u"http://example.org/"
self.graph = ConjunctiveGraph()
self.graph.parse(data=testgraph, format="n3", publicID=NS)
def testSPARQLNotEquals(self):
rt = self.graph.query(testquery, initNs={'rdf': RDF.uri}, DEBUG=False)
for row in rt:
assert str(row[0]) == "http://example.org/bar" #, "unexpected item of '%s'" % repr(row[0])
class _DBPediaMixin(object):
def __init__(self):
self.graph = ConjunctiveGraph('SPARQLStore')
def _get_media_api_host(self, language):
return language + u'.wikipedia.org'
def _get_dbpedia_url(self, language):
return u'http://' +('' if language == u'en' else language + u'.') + u'dbpedia.org'
def _fetch_dbpedia_query_result(self, query):
result = self.graph.query( unicode(query) )
return_fetch_results = [ [ unicode(e.n3()) if (not e is None) else '' for e in fetch_result] for fetch_result in result ]
return return_fetch_results
def test_secondary_recursion(self):
graph = ConjunctiveGraph()
graph.load(StringIO(SUBCLASS_DATA), format='n3')
results = graph.query(SUBCLASS_QUERY,
processor="sparql",
DEBUG=False)
results = set(results)
ob = URIRef('ex:ob')
class1 = URIRef('ex:class.1')
class2 = URIRef('ex:class.2')
class3 = URIRef('ex:class.3')
nose.tools.assert_equal(
results,
set([(ob, class1), (ob, class2), (ob, class3)]))
class TestSPARQLFilters(unittest.TestCase):
debug = False
sparql = True
def setUp(self):
NS = u"http://example.org/"
self.graph = ConjunctiveGraph()
self.graph.parse(data=testgraph, format="n3", publicID=NS)
def testSPARQLNotEquals(self):
rt = self.graph.query(testquery, initNs={'rdf': RDF.uri}, DEBUG=False)
for row in rt:
assert str(
row[0]
) == "http://example.org/bar" #, "unexpected item of '%s'" % repr(row[0])
def url(
resource: rdflib.URIRef,
graph: rdflib.ConjunctiveGraph
) -> Optional[str]:
"""Convert a URIRef to a clickable URL."""
bindings = graph.query(
'SELECT ?url WHERE { ?resource octa:subjectOf/octa:url ?url . } ',
initBindings={
'resource': resource,
}
).bindings
if not bindings:
return None
return '/' + bindings[0][rdflib.Variable('url')].value
class TestSparqlXmlResults(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.parse(StringIO(test_data), format="n3")
def testSimple(self):
self._query_result_contains(query, expected_fragments)
def _query_result_contains(self, query, fragments):
results = self.graph.query(query)
result_xml = results.serialize(format='xml')
result_xml = normalize(result_xml) # TODO: poor mans c14n..
print result_xml
for frag in fragments:
print frag
self.failUnless(frag in result_xml)
def label(
resource: rdflib.URIRef,
graph: rdflib.ConjunctiveGraph
) -> Optional[str]:
"""Convert a URIRef to a clickable URL."""
bindings = graph.query(
'SELECT ?label WHERE { ?resource rdfs:label ?label . } ',
initBindings={
'resource': resource,
}
).bindings
if not bindings:
return None
return bindings[0][rdflib.Variable('label')].value
class TripleStore(object):
""" A root object to supply store connectivity to view code """
def __init__(self, request):
self.request = request
q = self.request.registry.queryUtility
self.logger = q(IDebugLogger)
if self.request.registry.settings:
if (self.request.registry.settings.get('store_type',None) == 'talis'):
# Configure a connection to the talis store
self.store = Talis(self.request.registry.settings['talis_store'],
self.request.registry.settings['talis_user'],
self.request.registry.settings['talis_password'])
self.debug_sparql = self.request.registry.settings['debug_sparql']
else:
self.store = Graph()
self.debug_sparql = False
else:
self.store = Graph()
self.debug_sparql = False
def query(self, sparql, format='json'):
if self.debug_sparql:
msg = (sparql)
self.logger and self.logger.debug(msg)
#if isinstance(self.store, Graph) and "DESCRIBE" in sparql:
# """ rdflib currently requires a hack for describe queries involving literals """
# res = [x for x in self.store if not 'http://plings.net/' in str(x[0])]
## import pdb
# pdb.set_trace()
#else:
res = self.store.query(sparql)
# If we're working with a local rdflib graph we'll need to delve to get the results
if isinstance(res, SPARQLQueryResult):
return res.result
else:
return res
def sync(self):
if self.request.registry.settings['store_type'] == 'talis':
self.store.sync()
else:
pass
class TestSparqlXmlResults(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.parse(StringIO(test_data), format="n3")
def testSimple(self):
self._query_result_contains(query, expected_fragments)
def _query_result_contains(self, query, fragments):
results = self.graph.query(query)
result_xml = results.serialize(format='xml')
result_xml = normalize(result_xml) # TODO: poor mans c14n..
print result_xml
for frag in fragments:
print frag
self.failUnless(frag in result_xml)
class DateFilterTest(unittest.TestCase):
# debug = True
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.load(StringIO(testContent), format='n3')
def test_DATE_FILTER1(self):
for query in [QUERY1,QUERY2,QUERY3]:
# print query
#pQuery = Parse(query)
#print RenderSPARQLAlgebra(pQuery)
results = self.graph.query(query,
processor="sparql",
DEBUG=False)
results = list(results)
self.failUnless(
len(results) and results == [(ANSWER1,)],
"expecting : %s . Got: %s"%([(ANSWER1,)],repr(results)))
def test_library_rdf(self):
library = LibraryFactory.create()
model = ConjunctiveGraph()
response = self.client.get(library.get_absolute_url())
self.assertEqual(response.status_code, 200)
model.parse(data=smart_text(response.content), format='rdfa')
body = """prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
select ?library ?name ?library_id ?gel_cut ?made_by
where {
?library a libns:Library ;
libns:name ?name ;
libns:library_id ?library_id ;
libns:gel_cut ?gel_cut ;
libns:made_by ?made_by
}"""
count = 0
for r in model.query(body):
self.assertEqual(r['library_id'].toPython(),
library.id)
self.assertEqual(r['name'].toPython(),
library.library_name)
self.assertEqual(r['gel_cut'].toPython(),
library.gel_cut_size)
self.assertEqual(r['made_by'].toPython(),
library.made_by)
count += 1
# make sure there was actually data
self.assertGreater(count, 0)
state = validate_xhtml(smart_bytes(response.content))
if state is not None:
self.assertTrue(state)
# validate a library page.
add_default_schemas(model)
inference = Infer(model)
errmsgs = list(inference.run_validation())
self.assertEqual(len(errmsgs), 0)
def parse_lastfm(username):
''' return an array of zitgist uris for scrobbled artists '''
graph = ConjunctiveGraph()
print 'parsing last.fm data from dbtune for %s' % username
graph.parse('http://dbtune.org/last-fm/'+username)
query = '''
PREFIX mo: <http://purl.org/ontology/mo/>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
SELECT DISTINCT ?same WHERE {
?artist a mo:MusicArtist ;
owl:sameAs ?same .
}
'''
same = []
for row in graph.query(query):
same.append(row[0])
return same
def get_test(uri):
"""Return a single test case for the specified URI."""
for manifest in MANIFESTS:
LOG.debug("Retrieving manifest %s" % manifest)
test_manifest = Graph()
test_manifest.parse(manifest)
# find all the approved tests'
for result in test_manifest.query(APPROVED_TESTS):
if str(result[0]) == uri:
return new.classobj(py_name(str(result[3])), (TaskForceTest, ),
{})(*result)
return None
class Config(object):
def __init__(self, masterGraph):
self.graph = ConjunctiveGraph()
log.info('read config')
for f in os.listdir('config'):
if f.startswith('.'): continue
self.graph.parse('config/%s' % f, format='n3')
self.graph.bind('', ROOM) # not working
self.graph.bind('rdf', RDF)
# config graph is too noisy; maybe make it a separate resource
#masterGraph.patch(Patch(addGraph=self.graph))
def serialDevices(self):
return dict([(row.dev, row.board) for row in self.graph.query(
"""SELECT ?board ?dev WHERE {
?board :device ?dev;
a :ArduinoBoard .
}""", initNs={'': ROOM})])
def get_test(uri):
"""Return a single test case for the specified URI."""
for manifest in MANIFESTS:
LOG.debug("Retrieving manifest %s" % manifest)
test_manifest = Graph()
test_manifest.parse(manifest)
# find all the approved tests'
for result in test_manifest.query(APPROVED_TESTS):
if str(result[0]) == uri:
return new.classobj(py_name(str(result[3])),
(TaskForceTest,),{})(*result)
return None
def testSPARQLNotEquals():
NS = u"http://example.org/"
graph = ConjunctiveGraph()
graph.parse(StringInputSource(b("""
@prefix : <http://example.org/> .
@prefix rdf: <%s> .
:foo rdf:value 1.
:bar rdf:value 2.""") % RDF.uri), format="n3")
rt = graph.query(b("""SELECT ?node
WHERE {
?node rdf:value ?val.
FILTER (?val != 1)
}"""),
initNs={b('rdf'): RDF.uri},
DEBUG=False)
for row in rt:
item = row[0]
assert item == URIRef("http://example.org/bar"), "unexpected item of '%s'" % repr(item)
def find_long_lat(uri):
'''return tuple of the longitude and latitude (long,lat) for a geonames uri and the name'''
graph = ConjunctiveGraph()
graph.parse(uri)
query = '''
PREFIX wgs: <http://www.w3.org/2003/01/geo/wgs84_pos#>
PREFIX geo: <http://www.geonames.org/ontology#>
SELECT ?long ?lat ?name WHERE
{ ?x wgs:long ?long;
wgs:lat ?lat;
geo:name ?name.
}
'''
long_lat_name = None
print 'finding coordinates for %s ' % uri
for row in graph.query(query):
long_lat_name = (row[0], row[1], row[2])
return long_lat_name
class DeepGraphStore():
store_name = 'SQLite'
def __init__(self, create=False, parse=None):
self.parse = parse
self.create = create
self.graph = None
def setUp(self):
self.path = "" + random_file_generating()
self.graph = Graph(store=self.store_name)
self.graph.open(self.path, create=self.create)
if self.create:
if not self.parse:
self.graph.parse("http://njh.me/foaf.rdf", format='xml')
else:
self.graph.parse(self.parse)
self.graph.commit()
def open(self, path):
self.graph = ConjunctiveGraph(self.store_name)
self.path = path
self.graph.open(self.path, create=False)
def query(self, sparql_query):
return self.graph.query(sparql_query)
def parse(self, path_to_file_):
self.graph.parse(path_to_file_)
def load(self, triples):
self.graph.load(triples)
def close(self):
self.graph.close()
def size(self):
size = self.graph.__len__()
size = len(self.graph)
# self.close()
return size
def calculate_edges((offset_limit, config)):
g = ConjunctiveGraph(config['graph_store'], config['graph_identifier'])
g.open(config['db_configstring'], create=False)
engine = create_engine(config['alchemy_configstring'])
Session = sessionmaker(bind=engine)
session = Session()
results = session.query(Node).filter_by(node_type="subject").order_by(Node.node_id.desc()).offset(offset_limit[0]).limit(offset_limit[1]).all()
for result in results:
query = "SELECT ?p ?o WHERE { <%s> ?p ?o }" % result.node_uri
items = g.query(query)
for item in items:
if str(item[0]) != "http://www.w3.org/1999/02/22-rdf-syntax-ns#type":
neighbors = session.query(Node).filter_by(node_uri=item[1]).all()
if len(neighbors) > 0:
for neighbor in neighbors:
result.add_neighbors(str(item[0]), neighbor)
session.commit()
g.close()
session.close()
def test_0_not_run(self):
"""Create a library that was run, so it won't show up with not_run filter
"""
lane = LaneFactory()
library = lane.library
url = reverse('library_not_run')
response = self.client.get(url)
model = ConjunctiveGraph()
model.parse(data=smart_text(response.content), format='rdfa')
body = """prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
prefix libns: <http://jumpgate.caltech.edu/wiki/LibraryOntology#>
select ?library ?library_id
where {
?library a libns:Library ;
libns:library_id ?library_id .
}"""
results = []
for r in model.query(body):
results.append(r['library_id'])
self.assertEqual(len(results), 0)
class TestSparqlJsonResults(unittest.TestCase):
def setUp(self):
self.graph = ConjunctiveGraph()
self.graph.parse(StringIO(test_data), format="n3")
def _query_result_contains(self, query, correct):
results = self.graph.query(query)
result_json = results.serialize(format='json')
self.failUnless(result_json.find(correct) >= 0,
"Expected:\n %s \n- to contain:\n%s" % (result_json, correct))
testOptional = make_method('optional')
testWildcard = make_method('wildcard')
testUnion = make_method('union')
testUnion3 = make_method('union3')
testSelectVars = make_method('select_vars')
testWildcardVars = make_method('wildcard_vars')
def test_suite():
"""Generate a test suite of all approved task force tests."""
test_suite = unittest.TestSuite()
return test_suite
for manifest in MANIFESTS:
LOG.debug("Retrieving manifest %s" % manifest)
test_manifest = Graph()
test_manifest.parse(manifest)
# find all the approved tests'
for result in test_manifest.query(APPROVED_TESTS):
LOG.debug("Adding test case: %s" % str(result[3]))
test_suite.addTest(
new.classobj(py_name(str(result[3])),
(TaskForceTest,),{})(*result))
return test_suite
