def test_result_fragments(self):
rdfXml = serialize(self.sourceGraph, self.serializer)
assert b('<Test rdf:about="http://example.org/data/a">') in rdfXml
assert b('<rdf:Description rdf:about="http://example.org/data/b">') in rdfXml
assert b('<name xml:lang="en">Bee</name>') in rdfXml
assert b('<value rdf:datatype="http://www.w3.org/2001/XMLSchema#integer">3</value>') in rdfXml
assert b('<BNode rdf:nodeID="') in rdfXml, "expected one identified bnode in serialized graph"
def to_key(triple, context):
"Takes a string; returns key"
return b("^").join(
(context,
triple[i % 3],
triple[(i + 1) % 3],
triple[(i + 2) % 3], b(""))) # "" to tac on the trailing ^
def to_key(triple, context):
"Takes a string; returns key"
return b("^").join(
(context,
triple[i % 3],
triple[(i + 1) % 3],
triple[(i + 2) % 3], b(""))) # "" to tac on the trailing ^
def testFinalNewline():
"""
http://code.google.com/p/rdflib/issues/detail?id=5
"""
import sys
import platform
if getattr(sys, 'pypy_version_info', None) or platform.system() == 'Java':
from nose import SkipTest
raise SkipTest(
'Testing under pypy and Jython2.5 fails to detect that ' + \
'IOMemory is a context_aware store')
graph=Graph()
graph.add((URIRef("http://ex.org/a"),
URIRef("http://ex.org/b"),
URIRef("http://ex.org/c")))
failed = set()
for p in rdflib.plugin.plugins(None, rdflib.plugin.Serializer):
if p.name not in ( 'nquads', 'trix' ):
v = graph.serialize(format=p.name)
lines = v.split(b("\n"))
if b("\n") not in v or (lines[-1]!=b('')):
failed.add(p.name)
assert len(failed)==0, "No final newline for formats: '%s'" % failed
def test_validating_unquote_raises(self):
ntriples.validate = True
uniquot = b("""<http://www.w3.org/People/Berners-Lee/card#cm> <http://xmlns.com/foaf/0.1/name> "R\\u00E4ksm\\u00F6rg\\u00E5s" <http://www.w3.org/People/Berners-Lee/card> .""")
self.assertRaises(ntriples.ParseError, ntriples.unquote, uniquot)
uniquot = b("""<http://www.w3.org/People/Berners-Lee/card#cm> <http://xmlns.com/foaf/0.1/name> "R\\\\u00E4ksm\\u00F6rg\\u00E5s" <http://www.w3.org/People/Berners-Lee/card> .""")
self.assertRaises(ntriples.ParseError, ntriples.unquote, uniquot)
# revert to default
ntriples.validate = False
def test_result_fragments_with_base(self):
rdfXml = serialize(self.sourceGraph, self.serializer,
extra_args={'base':"http://example.org/", 'xml_base':"http://example.org/"})
assert b('xml:base="http://example.org/"') in rdfXml
assert b('<Test rdf:about="data/a">') in rdfXml
assert b('<rdf:Description rdf:about="data/b">') in rdfXml
assert b('<value rdf:datatype="http://www.w3.org/2001/XMLSchema#integer">3</value>') in rdfXml
assert b('<BNode rdf:nodeID="') in rdfXml, "expected one identified bnode in serialized graph"
def remove(self, triple, context):
(subject, predicate, object) = triple
assert self.__open, "The Store must be open."
Store.remove(self, (subject, predicate, object), context)
_to_string = self._to_string
if context is not None:
if context == self:
context = None
if subject is not None \
and predicate is not None \
and object is not None \
and context is not None:
s = _to_string(subject)
p = _to_string(predicate)
o = _to_string(object)
c = _to_string(context)
value = self.__indices[0].get(bb("%s^%s^%s^%s^" % (c, s, p, o)))
if value is not None:
self.__remove((bb(s), bb(p), bb(o)), bb(c))
self.__needs_sync = True
else:
cspo, cpos, cosp = self.__indices
index, prefix, from_key, results_from_key = self.__lookup(
(subject, predicate, object), context)
needs_sync = False
for key in index.match_prefix(prefix):
c, s, p, o = from_key(key)
if context is None:
contexts_value = index.get(key) or b("")
# remove triple from all non quoted contexts
contexts = set(contexts_value.split(b("^")))
contexts.add(b("")) # and from the conjunctive index
for c in contexts:
for i, _to_key, _ in self.__indices_info:
i.remove(_to_key((s, p, o), c))
else:
self.__remove((s, p, o), c)
needs_sync = True
if context is not None:
if subject is None and predicate is None and object is None:
# TODO: also if context becomes empty and not just on
# remove((None, None, None), c)
try:
self.__contexts.remove(bb(_to_string(context)))
# except db.DBNotFoundError, e:
# pass
except Exception as e: # pragma: NO COVER
print("%s, Failed to delete %s" % (
e, context)) # pragma: NO COVER
pass # pragma: NO COVER
self.__needs_sync = needs_sync
def test_validating_unquote_raises(self):
ntriples.validate = True
uniquot = b(
"""<http://www.w3.org/People/Berners-Lee/card#cm> <http://xmlns.com/foaf/0.1/name> "R\\u00E4ksm\\u00F6rg\\u00E5s" <http://www.w3.org/People/Berners-Lee/card> ."""
)
self.assertRaises(ntriples.ParseError, ntriples.unquote, uniquot)
uniquot = b(
"""<http://www.w3.org/People/Berners-Lee/card#cm> <http://xmlns.com/foaf/0.1/name> "R\\\\u00E4ksm\\u00F6rg\\u00E5s" <http://www.w3.org/People/Berners-Lee/card> ."""
)
self.assertRaises(ntriples.ParseError, ntriples.unquote, uniquot)
# revert to default
ntriples.validate = False
def __len__(self, context=None):
assert self.__open, "The Store must be open."
if context is not None:
if context == self:
context = None
if context is None:
prefix = b("^")
else:
prefix = bb("%s^" % self._to_string(context))
index = self.__indicies[0]
cursor = index.cursor()
current = cursor.set_range(prefix)
count = 0
while current:
key, value = current
if key.startswith(prefix):
count += 1
# Hack to stop 2to3 converting this to next(cursor)
current = getattr(cursor, 'next')()
else:
break
cursor.close()
return count
def contexts(self, triple=None):
_from_string = self._from_string
_to_string = self._to_string
if triple:
s, p, o = triple
s = _to_string(s)
p = _to_string(p)
o = _to_string(o)
contexts = self.__indicies[0].get(
bb("%s^%s^%s^%s^" % ("", s, p, o)))
if contexts:
for c in contexts.split(b("^")):
if c:
yield _from_string(c)
else:
index = self.__contexts
cursor = index.cursor()
current = cursor.first()
cursor.close()
while current:
key, value = current
context = _from_string(key)
yield context
cursor = index.cursor()
try:
cursor.set_range(key)
# Hack to stop 2to3 converting this to next(cursor)
current = getattr(cursor, 'next')()
except db.DBNotFoundError:
current = None
cursor.close()
def contexts(self, triple=None):
_from_string = self._from_string
_to_string = self._to_string
if triple:
s, p, o = triple
s = _to_string(s)
p = _to_string(p)
o = _to_string(o)
contexts = self.__indicies[0].get(bb(
"%s^%s^%s^%s^" % ("", s, p, o)))
if contexts:
for c in contexts.split(b("^")):
if c:
yield _from_string(c)
else:
index = self.__contexts
cursor = index.cursor()
current = cursor.first()
cursor.close()
while current:
key, value = current
context = _from_string(key)
yield context
cursor = index.cursor()
try:
cursor.set_range(key)
# Hack to stop 2to3 converting this to next(cursor)
current = getattr(cursor, 'next')()
except db.DBNotFoundError:
current = None
cursor.close()
def test_nonvalidating_unquote(self):
safe = b(
"""<http://example.org/alice/foaf.rdf#me> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> <http://example.org/alice/foaf1.rdf> ."""
)
ntriples.validate = False
res = ntriples.unquote(safe)
self.assert_(isinstance(res, unicode))
def uriref(self):
if self.peek(b('<')):
uri = self.eat(r_uriref).group(1)
uri = unquote(uri)
uri = uriquote(uri)
return URI(uri)
return False
def unquote(s):
"""Unquote an N-Triples string."""
if not validate:
return s.decode('unicode-escape')
else:
result = []
while s:
m = r_safe.match(s)
if m:
s = s[m.end():]
result.append(m.group(1).decode('ascii'))
continue
m = r_quot.match(s)
if m:
s = s[2:]
result.append(quot[m.group(1)])
continue
m = r_uniquot.match(s)
if m:
s = s[m.end():]
u, U = m.groups()
codepoint = int(u or U, 16)
if codepoint > 0x10FFFF:
raise ParseError("Disallowed codepoint: %08X" % codepoint)
result.append(unichr(codepoint))
elif s.startswith(b('\\')):
raise ParseError("Illegal escape at: %s..." % s[:10])
else:
raise ParseError("Illegal literal character: %r" % s[0])
return u''.join(result)
def unquote(s):
"""Unquote an N-Triples string."""
if not validate:
return s.decode('unicode-escape')
else:
result = []
while s:
m = r_safe.match(s)
if m:
s = s[m.end():]
result.append(m.group(1).decode('ascii'))
continue
m = r_quot.match(s)
if m:
s = s[2:]
result.append(quot[m.group(1)])
continue
m = r_uniquot.match(s)
if m:
s = s[m.end():]
u, U = m.groups()
codepoint = int(u or U, 16)
if codepoint > 0x10FFFF:
raise ParseError("Disallowed codepoint: %08X" % codepoint)
result.append(unichr(codepoint))
elif s.startswith(b('\\')):
raise ParseError("Illegal escape at: %s..." % s[:10])
else: raise ParseError("Illegal literal character: %r" % s[0])
return u''.join(result)
def test_issue_130():
g = rdflib.Graph()
try:
g.parse(location="http://linked-data.ru/example")
except:
raise SkipTest('Test data URL unparseable')
assert b('rdf:about="http://semanticfuture.net/linked-data/example/#company"') in g.serialize(), g.serialize()
def uriref(self):
if self.peek(b('<')):
uri = self.eat(r_uriref).group(1)
uri = unquote(uri)
uri = uriquote(uri)
return URI(uri)
return False
def testIssue78(self):
g = Graph()
g.add((URIRef("foo"), URIRef("foo"),
Literal(u"R\u00E4ksm\u00F6rg\u00E5s")))
s = g.serialize(format='nt')
self.assertEquals(type(s), bytestype)
self.assertTrue(b(r"R\u00E4ksm\u00F6rg\u00E5s") in s)
def __len__(self, context=None):
assert self.__open, "The Store must be open."
if context is not None:
if context == self:
context = None
if context is None:
prefix = b("^")
else:
prefix = bb("%s^" % self._to_string(context))
index = self.__indicies[0]
cursor = index.cursor()
current = cursor.set_range(prefix)
count = 0
while current:
key, value = current
if key.startswith(prefix):
count += 1
# Hack to stop 2to3 converting this to next(cursor)
current = getattr(cursor, 'next')()
else:
break
cursor.close()
return count
def test_turtle_namespace_prefixes(self):
g = ConjunctiveGraph()
n3 = \
"""
@prefix _9: <http://data.linkedmdb.org/resource/movie/> .
@prefix p_9: <urn:test:> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
p_9:a p_9:b p_9:c .
<http://data.linkedmdb.org/resource/director/1> a
<http://data.linkedmdb.org/resource/movie/director>;
rdfs:label "Cecil B. DeMille (Director)";
_9:director_name "Cecil B. DeMille" ."""
g.parse(data=n3, format='n3')
turtle = g.serialize(format="turtle")
# Check round-tripping, just for kicks.
g = ConjunctiveGraph()
g.parse(data=turtle, format='turtle')
# Shouldn't have got to here
s=g.serialize(format="turtle")
self.assert_(b('@prefix _9') not in s)
def test_validating_unquote(self):
quot = b("""<http://example.org/alice/foaf.rdf#me> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> <http://example.org/alice/foaf1.rdf> .""")
ntriples.validate = True
res = ntriples.unquote(quot)
# revert to default
ntriples.validate = False
log.debug("restype %s" % type(res))
def test_turtle_namespace_prefixes(self):
g = ConjunctiveGraph()
n3 = \
"""
@prefix _9: <http://data.linkedmdb.org/resource/movie/> .
@prefix p_9: <urn:test:> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
p_9:a p_9:b p_9:c .
<http://data.linkedmdb.org/resource/director/1> a
<http://data.linkedmdb.org/resource/movie/director>;
rdfs:label "Cecil B. DeMille (Director)";
_9:director_name "Cecil B. DeMille" ."""
g.parse(data=n3, format='n3')
turtle = g.serialize(format="turtle")
# Check round-tripping, just for kicks.
g = ConjunctiveGraph()
g.parse(data=turtle, format='turtle')
# Shouldn't have got to here
s = g.serialize(format="turtle")
self.assertTrue(b('@prefix _9') not in s)
def test_nonvalidating_unquote(self):
safe = b(
"""<http://example.org/alice/foaf.rdf#me> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> <http://example.org/alice/foaf1.rdf> ."""
)
ntriples.validate = False
res = ntriples.unquote(safe)
self.assert_(isinstance(res, unicode))
def from_key(key):
"Takes a key; returns string"
parts = key.split(b("^"))
return \
parts[0], \
parts[(3 - i + 0) % 3 + 1], \
parts[(3 - i + 1) % 3 + 1], \
parts[(3 - i + 2) % 3 + 1]
def from_key(key):
"Takes a key; returns string"
parts = key.split(b("^"))
return \
parts[0], \
parts[(3 - i + 0) % 3 + 1], \
parts[(3 - i + 1) % 3 + 1], \
parts[(3 - i + 2) % 3 + 1]
def testBaseSerialize(self):
g = Graph()
g.add((URIRef('http://example.com/people/Bob'), URIRef('urn:knows'),
URIRef('http://example.com/people/Linda')))
s = g.serialize(base='http://example.com/', format='n3')
self.assertTrue(b('<people/Bob>') in s)
g2 = ConjunctiveGraph()
g2.parse(data=s, publicID='http://example.com/', format='n3')
self.assertEqual(list(g), list(g2))
def test_n32(self):
# this test not generating prefixes for subjects/objects
g = Graph()
g.add((URIRef("http://example1.com/foo"),
URIRef("http://example2.com/bar"),
URIRef("http://example3.com/baz")))
n3 = g.serialize(format="n3")
self.assertTrue(b("<http://example1.com/foo> ns1:bar <http://example3.com/baz> .") in n3)
def setUp(self):
NS = u"http://example.org/"
self.graph = Graph(store)
self.graph.parse(StringInputSource(b("""
@prefix : <http://example.org/> .
@prefix rdf: <%s> .
@prefix rdfs: <%s> .
[ :prop :val ].
[ a rdfs:Class ]."""%(RDF,RDFS))), format="n3")
def serialize(self, stream, base=None, encoding=None, **args):
if base is not None:
warnings.warn("NTSerializer does not support base.")
if encoding is not None:
warnings.warn("NTSerializer does not use custom encoding.")
encoding = self.encoding
for triple in self.store:
stream.write(_nt_row(triple).encode(encoding, "replace"))
stream.write(b("\n"))
def test_validating_unquote(self):
quot = b(
"""<http://example.org/alice/foaf.rdf#me> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://xmlns.com/foaf/0.1/Person> <http://example.org/alice/foaf1.rdf> ."""
)
ntriples.validate = True
res = ntriples.unquote(quot)
# revert to default
ntriples.validate = False
log.debug("restype %s" % type(res))
def from_key(key, subject, predicate, object, contexts_value):
"Takes a key and subject, predicate, object; returns tuple for yield"
parts = key.split(b("^"))
if subject is None:
# TODO: i & 1: # dis assemble and/or measure to see which is faster
# subject is None or i & 1
s = from_string(parts[(3-i+0)%3+1])
else:
s = subject
if predicate is None:#i & 2:
p = from_string(parts[(3-i+1)%3+1])
else:
p = predicate
if object is None:#i & 4:
o = from_string(parts[(3-i+2)%3+1])
else:
o = object
return (s, p, o), (from_string(c) for c in contexts_value.split(b("^")) if c)
def serialize(self, stream, base=None, encoding=None, **args):
if base is not None:
warnings.warn("YTSerializer does not support base.")
if encoding is not None:
warnings.warn("YTSerializer does not use custom encoding.")
encoding = self.encoding
for triple in self.store:
stream.write(_yt_row(triple).encode(encoding, "replace"))
stream.write(b("\n"))
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 testBaseSerialize(self):
g = Graph()
g.add((URIRef('http://example.com/people/Bob'), URIRef(
'urn:knows'), URIRef('http://example.com/people/Linda')))
s = g.serialize(base='http://example.com/', format='n3')
self.assertTrue(b('<people/Bob>') in s)
g2 = ConjunctiveGraph()
g2.parse(data=s, publicID='http://example.com/', format='n3')
self.assertEqual(list(g), list(g2))
def add(self, triple, context, quoted=False, txn=None):
"""\
Add a triple to the store of triples.
"""
(subject, predicate, object) = triple
assert self.__open, "The Store must be open."
assert context != self, "Can not add triple directly to store"
Store.add(self, (subject, predicate, object), context, quoted)
_to_string = self._to_string
s = _to_string(subject, txn=txn)
p = _to_string(predicate, txn=txn)
o = _to_string(object, txn=txn)
c = _to_string(context, txn=txn)
cspo, cpos, cosp = self.__indicies
value = cspo.get(bb("%s^%s^%s^%s^" % (c, s, p, o)), txn=txn)
if value is None:
self.__contexts.put(bb(c), "", txn=txn)
contexts_value = cspo.get(bb("%s^%s^%s^%s^" % ("", s, p, o)),
txn=txn) or b("")
contexts = set(contexts_value.split(b("^")))
contexts.add(bb(c))
contexts_value = b("^").join(contexts)
assert contexts_value is not None
cspo.put(bb("%s^%s^%s^%s^" % (c, s, p, o)), "", txn=txn)
cpos.put(bb("%s^%s^%s^%s^" % (c, p, o, s)), "", txn=txn)
cosp.put(bb("%s^%s^%s^%s^" % (c, o, s, p)), "", txn=txn)
if not quoted:
cspo.put(bb("%s^%s^%s^%s^" % ("", s, p, o)),
contexts_value,
txn=txn)
cpos.put(bb("%s^%s^%s^%s^" % ("", p, o, s)),
contexts_value,
txn=txn)
cosp.put(bb("%s^%s^%s^%s^" % ("", o, s, p)),
contexts_value,
txn=txn)
self.__needs_sync = True
def test_n3(self):
g = Graph()
g.add((URIRef("http://example.com/foo"),
URIRef("http://example.com/bar"),
URIRef("http://example.com/baz")))
n3 = g.serialize(format="n3")
# Gunnar disagrees that this is right:
# self.assertTrue("<http://example.com/foo> ns1:bar <http://example.com/baz> ." in n3)
# as this is much prettier, and ns1 is already defined:
self.assertTrue(b("ns1:foo ns1:bar ns1:baz .") in n3)
def testSameSubject(self):
g=rdflib.ConjunctiveGraph()
g.get_context('urn:a').add(( rdflib.URIRef('urn:1'),
rdflib.URIRef('urn:p1'),
rdflib.URIRef('urn:o1') ))
g.get_context('urn:b').add(( rdflib.URIRef('urn:1'),
rdflib.URIRef('urn:p2'),
rdflib.URIRef('urn:o2') ))
self.assertEqual(len(g.get_context('urn:a')),1)
self.assertEqual(len(g.get_context('urn:b')),1)
s=g.serialize(format='trig')
self.assertEqual(len(re.findall(b("p1"), s)), 1)
self.assertEqual(len(re.findall(b("p2"), s)), 1)
self.assert_(b('{}') not in s) # no empty graphs!
def test_n3(self):
g = Graph()
g.add((URIRef("http://example.com/foo"),
URIRef("http://example.com/bar"),
URIRef("http://example.com/baz")))
n3 = g.serialize(format="n3")
# Gunnar disagrees that this is right:
# self.assertTrue("<http://example.com/foo> ns1:bar <http://example.com/baz> ." in n3)
# as this is much prettier, and ns1 is already defined:
self.assertTrue(b("ns1:foo ns1:bar ns1:baz .") in n3)
def serialize(self, stream, base=None, encoding=None, **args):
if base is not None:
warnings.warn("NQuadsSerializer does not support base.")
if encoding is not None:
warnings.warn("NQuadsSerializer does not use custom encoding.")
encoding = self.encoding
for context in self.store.contexts():
for triple in context:
stream.write(_nq_row(triple, context.identifier).encode(encoding, "replace"))
stream.write(b("\n"))
def testSameSubject(self):
g = rdflib.ConjunctiveGraph()
g.get_context('urn:a').add(
(rdflib.URIRef('urn:1'), rdflib.URIRef('urn:p1'),
rdflib.URIRef('urn:o1')))
g.get_context('urn:b').add(
(rdflib.URIRef('urn:1'), rdflib.URIRef('urn:p2'),
rdflib.URIRef('urn:o2')))
self.assertEqual(len(g.get_context('urn:a')), 1)
self.assertEqual(len(g.get_context('urn:b')), 1)
s = g.serialize(format='trig')
self.assertEqual(len(re.findall(b("p1"), s)), 1)
self.assertEqual(len(re.findall(b("p2"), s)), 1)
self.assertTrue(b('{}') not in s) # no empty graphs!
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, result_xml)
if frag.startswith(b('<sparql:result>')):
raise SkipTest("False negative.")
self.failUnless(frag in result_xml)
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, result_xml)
if frag.startswith(b('<sparql:result>')):
raise SkipTest("False negative.")
self.failUnless(frag in result_xml)
def test_issue_130():
raise SkipTest("Remote content change - skip for now")
g = rdflib.Graph()
try:
g.parse(location="http://linked-data.ru/example")
except:
raise SkipTest('Test data URL unparseable')
if len(g) == 0:
raise SkipTest('Test data URL empty of content')
assert b('rdf:about="http://semanticfuture.net/linked-data/example/#company"') in g.serialize(), g.serialize()
def serialize(self, stream, base=None, encoding=None, **args):
self.__serialized = {}
store = self.store
self.base = base
self.max_depth = args.get("max_depth", 3)
assert self.max_depth > 0, "max_depth must be greater than 0"
self.nm = nm = store.namespace_manager
self.writer = writer = XMLWriter(stream, nm, encoding)
namespaces = {}
possible = set(store.predicates()).union(
store.objects(None, RDF.type))
for predicate in possible:
prefix, namespace, local = nm.compute_qname(predicate)
namespaces[prefix] = namespace
namespaces["rdf"] = "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
writer.push(RDF.RDF)
if "xml_base" in args:
writer.attribute(XMLBASE, args["xml_base"])
writer.namespaces(namespaces.iteritems())
# Write out subjects that can not be inline
for subject in store.subjects():
if (None, None, subject) in store:
if (subject, None, subject) in store:
self.subject(subject, 1)
else:
self.subject(subject, 1)
# write out anything that has not yet been reached
# write out BNodes last (to ensure they can be inlined where possible)
bnodes = set()
for subject in store.subjects():
if isinstance(subject, BNode):
bnodes.add(subject)
continue
self.subject(subject, 1)
# now serialize only those BNodes that have not been serialized yet
for bnode in bnodes:
if bnode not in self.__serialized:
self.subject(subject, 1)
writer.pop(RDF.RDF)
stream.write(b("\n"))
# Set to None so that the memory can get garbage collected.
self.__serialized = None
def serialize(self, stream, base=None, encoding=None, **args):
self.__serialized = {}
store = self.store
self.base = base
self.max_depth = args.get("max_depth", 3)
assert self.max_depth > 0, "max_depth must be greater than 0"
self.nm = nm = store.namespace_manager
self.writer = writer = XMLWriter(stream, nm, encoding)
namespaces = {}
possible = set(store.predicates()).union(
store.objects(None, RDF.type))
for predicate in possible:
prefix, namespace, local = nm.compute_qname(predicate)
namespaces[prefix] = namespace
namespaces["rdf"] = "http://www.w3.org/1999/02/22-rdf-syntax-ns#"
writer.push(RDF.RDF)
if "xml_base" in args:
writer.attribute(XMLBASE, args["xml_base"])
writer.namespaces(iter(namespaces.items()))
# Write out subjects that can not be inline
for subject in store.subjects():
if (None, None, subject) in store:
if (subject, None, subject) in store:
self.subject(subject, 1)
else:
self.subject(subject, 1)
# write out anything that has not yet been reached
# write out BNodes last (to ensure they can be inlined where possible)
bnodes = set()
for subject in store.subjects():
if isinstance(subject, BNode):
bnodes.add(subject)
continue
self.subject(subject, 1)
# now serialize only those BNodes that have not been serialized yet
for bnode in bnodes:
if bnode not in self.__serialized:
self.subject(subject, 1)
writer.pop(RDF.RDF)
stream.write(b("\n"))
# Set to None so that the memory can get garbage collected.
self.__serialized = None
def testDefaultGraphSerializesWithoutName(self):
data = """
<http://example.com/thing#thing_a> <http://example.com/knows> <http://example.com/thing#thing_b> .
{ <http://example.com/thing#thing_c> <http://example.com/knows> <http://example.com/thing#thing_d> . }
"""
g = rdflib.ConjunctiveGraph()
g.parse(data=data, format='trig')
data = g.serialize(format='trig')
self.assertTrue(b('None') not in data)
def test_issue_130():
g = rdflib.Graph()
try:
g.parse(location="http://linked-data.ru/example")
except:
raise SkipTest('Test data URL unparseable')
if len(g) == 0:
raise SkipTest('Test data URL empty of content')
assert b(
'rdf:about="http://semanticfuture.net/linked-data/example/#company"'
) in g.serialize(), g.serialize()
def setUp(self):
NS = u"http://example.org/"
self.graph = Graph(store)
self.graph.parse(StringInputSource(
b("""
@prefix : <http://example.org/> .
@prefix rdf: <%s> .
@prefix rdfs: <%s> .
[ :prop :val ].
[ a rdfs:Class ].""" % (RDF, RDFS))),
format="n3")
def test_n32(self):
# this test not generating prefixes for subjects/objects
g = Graph()
g.add((URIRef("http://example1.com/foo"),
URIRef("http://example2.com/bar"),
URIRef("http://example3.com/baz")))
n3 = g.serialize(format="n3")
self.assertTrue(
b("<http://example1.com/foo> ns1:bar <http://example3.com/baz> .")
in n3)
def serialize(self, stream, base=None, encoding=None, **args):
if base is not None:
warnings.warn("NQuadsSerializer does not support base.")
if encoding is not None:
warnings.warn("NQuadsSerializer does not use custom encoding.")
encoding = self.encoding
for context in self.store.contexts():
for triple in context:
stream.write(_nq_row(
triple, context.identifier).encode(encoding, "replace"))
stream.write(b("\n"))
def testTurtleFinalDot():
"""
https://github.com/RDFLib/rdflib/issues/282
"""
g = Graph()
u = URIRef("http://ex.org/bob.")
g.bind("ns", "http://ex.org/")
g.add((u, u, u))
s = g.serialize(format='turtle')
assert b("ns:bob.") not in s
def testFinalNewline(self):
"""
http://code.google.com/p/rdflib/issues/detail?id=5
"""
import sys
import platform
if getattr(sys, 'pypy_version_info', None) or platform.system() == 'Java':
from nose import SkipTest
raise SkipTest(
'Testing under pypy and Jython2.5 fails to detect that ' + \
'IOMemory is a context_aware store')
failed = set()
for p in rdflib.plugin.plugins(None, rdflib.plugin.Serializer):
if p.name is not 'nquads':
v = self.graph.serialize(format=p.name)
lines = v.split(b("\n"))
if b("\n") not in v or (lines[-1]!=b('')):
failed.add(p.name)
self.assertEqual(len(failed), 0, "No final newline for formats: '%s'" % failed)
def testTurtleFinalDot():
"""
https://github.com/RDFLib/rdflib/issues/282
"""
g = rdflib.Graph()
u = rdflib.URIRef("http://ex.org/bob.")
g.bind("ns", "http://ex.org/")
g.add( (u, u, u) )
s=g.serialize(format='turtle')
assert b("ns:bob.") not in s
def add(self, triple, context, quoted=False, txn=None):
"""\
Add a triple to the store of triples.
"""
(subject, predicate, object) = triple
assert self.__open, "The Store must be open."
assert context != self, "Can not add triple directly to store"
Store.add(self, (subject, predicate, object), context, quoted)
_to_string = self._to_string
s = _to_string(subject, txn=txn)
p = _to_string(predicate, txn=txn)
o = _to_string(object, txn=txn)
c = _to_string(context, txn=txn)
cspo, cpos, cosp = self.__indicies
value = cspo.get(bb("%s^%s^%s^%s^" % (c, s, p, o)), txn=txn)
if value is None:
self.__contexts.put(bb(c), "", txn=txn)
contexts_value = cspo.get(
bb("%s^%s^%s^%s^" % ("", s, p, o)), txn=txn) or b("")
contexts = set(contexts_value.split(b("^")))
contexts.add(bb(c))
contexts_value = b("^").join(contexts)
assert contexts_value is not None
cspo.put(bb("%s^%s^%s^%s^" % (c, s, p, o)), "", txn=txn)
cpos.put(bb("%s^%s^%s^%s^" % (c, p, o, s)), "", txn=txn)
cosp.put(bb("%s^%s^%s^%s^" % (c, o, s, p)), "", txn=txn)
if not quoted:
cspo.put(bb(
"%s^%s^%s^%s^" % ("", s, p, o)), contexts_value, txn=txn)
cpos.put(bb(
"%s^%s^%s^%s^" % ("", p, o, s)), contexts_value, txn=txn)
cosp.put(bb(
"%s^%s^%s^%s^" % ("", o, s, p)), contexts_value, txn=txn)
self.__needs_sync = True
def testPrefixes(self):
data = """
@prefix ns1: <http://ex.org/schema#> .
<http://ex.org/docs/document1> = {
ns1:Person_A a ns1:Person ;
ns1:TextSpan "Simon" .
}
<http://ex.org/docs/document2> = {
ns1:Person_C a ns1:Person ;
ns1:TextSpan "Agnes" .
}
"""
cg = rdflib.ConjunctiveGraph()
cg.parse(data=data, format='trig')
data = cg.serialize(format='trig')
self.assert_(b('ns2: <http://ex.org/docs/') in data, data)
self.assert_(b('<ns2:document1>') not in data, data)
self.assert_(b('ns2:document1') in data, data)
def __len__(self, context=None):
assert self.__open, "The Store must be open."
if context is not None:
if context == self:
context = None
if context is None:
prefix = b("^")
else:
prefix = bb("%s^" % self._to_string(context))
return len([key for key in self.__indices[0]
if key.startswith(prefix)])
