def parse(f, on_prepare_triple=None, new_blank=None, new_literal=None):
parser = xml.parsers.expat.ParserCreate(None, "")
try:
parser.buffer_text = True
parser.specified_attributes = True
except:
pass
ontology_iri = ""
objs = []
annots = []
prefixes = {}
current_content = ""
current_attrs = None
current_blank = 0
in_declaration = False
in_prop_chain = False
before_declaration = True
last_cardinality = "0"
nb_triple = 0
if not on_prepare_triple:
def on_prepare_triple(s, p, o):
nonlocal nb_triple
nb_triple += 1
if not s.startswith("_"): s = "<%s>" % s
if not (o.startswith("_") or o.startswith('"')): o = "<%s>" % o
print("%s %s %s ." % (s, "<%s>" % p, o))
if not new_blank:
def new_blank():
nonlocal current_blank
current_blank += 1
return "_:%s" % current_blank
if not new_literal:
def new_literal(value, attrs):
value = value.replace('"', '\\"').replace("\n", "\\n")
lang = attrs.get("http://www.w3.org/XML/1998/namespacelang")
if lang: return '"%s"@%s' % (value, lang)
datatype = attrs.get("datatypeIRI")
if datatype: return '"%s"^^<%s>' % (value, datatype)
return '"%s"' % (value)
def new_list(l):
bn = bn0 = new_blank()
if l:
for i in range(len(l) - 1):
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first",
l[i])
bn_next = new_blank()
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
bn_next)
bn = bn_next
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first", l[-1])
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
else:
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
return bn0
def unabbreviate_IRI(abbreviated_iri):
prefix, name = abbreviated_iri.split(":", 1)
return prefixes[prefix] + name
def get_IRI(attrs):
if "IRI" in attrs:
iri = attrs["IRI"]
if not iri: return ontology_iri
if iri.startswith("#") or iri.startswith("/"):
iri = ontology_iri + iri
return iri
return unabbreviate_IRI(attrs["abbreviatedIRI"])
def startElement(tag, attrs):
nonlocal current_content, current_attrs, in_declaration, before_declaration, last_cardinality, in_prop_chain, ontology_iri
current_content = ""
if (tag == "http://www.w3.org/2002/07/owl#Prefix"):
prefixes[attrs["name"]] = attrs["IRI"]
elif (tag == "http://www.w3.org/2002/07/owl#Declaration"):
in_declaration = True
before_declaration = False
elif (tag in types):
iri = get_IRI(attrs)
if in_declaration: on_prepare_triple(iri, rdf_type, types[tag])
objs.append(iri)
elif (tag == "http://www.w3.org/2002/07/owl#Datatype"):
objs.append(get_IRI(attrs))
elif (tag == "http://www.w3.org/2002/07/owl#Literal"):
current_attrs = attrs
elif ((tag == "http://www.w3.org/2002/07/owl#ObjectIntersectionOf")
or (tag == "http://www.w3.org/2002/07/owl#ObjectUnionOf")
or (tag == "http://www.w3.org/2002/07/owl#ObjectOneOf")
or (tag == "http://www.w3.org/2002/07/owl#DataIntersectionOf")
or (tag == "http://www.w3.org/2002/07/owl#DataUnionOf")
or (tag == "http://www.w3.org/2002/07/owl#DisjointClasses") or
(tag == "http://www.w3.org/2002/07/owl#DisjointObjectProperties")
or
(tag == "http://www.w3.org/2002/07/owl#DisjointDataProperties")
or
(tag == "http://www.w3.org/2002/07/owl#DifferentIndividuals")):
objs.append("(")
elif ((tag == "http://www.w3.org/2002/07/owl#ObjectExactCardinality")
or (tag == "http://www.w3.org/2002/07/owl#ObjectMinCardinality")
or (tag == "http://www.w3.org/2002/07/owl#ObjectMaxCardinality")
or (tag == "http://www.w3.org/2002/07/owl#DataExactCardinality")
or (tag == "http://www.w3.org/2002/07/owl#DataMinCardinality")
or (tag == "http://www.w3.org/2002/07/owl#DataMaxCardinality")):
objs.append("(")
last_cardinality = attrs["cardinality"]
elif (tag == "http://www.w3.org/2002/07/owl#AnonymousIndividual"):
objs.append(new_blank())
elif (tag == "http://www.w3.org/2002/07/owl#SubObjectPropertyOf"):
in_prop_chain = False
elif (tag == "http://www.w3.org/2002/07/owl#ObjectInverseOf") or (
tag == "http://www.w3.org/2002/07/owl#DataInverseOf") or (
tag == "http://www.w3.org/2002/07/owl#inverseOf"):
objs.append(new_blank())
elif (tag == "http://www.w3.org/2002/07/owl#ObjectPropertyChain"):
objs.append("(")
elif (tag == "http://www.w3.org/2002/07/owl#DatatypeRestriction"):
objs.append("(")
elif (tag == "http://www.w3.org/2002/07/owl#FacetRestriction"):
objs.append(attrs["facet"])
elif (tag == "http://www.w3.org/2002/07/owl#Ontology"):
ontology_iri = attrs["ontologyIRI"]
on_prepare_triple(ontology_iri, rdf_type,
"http://www.w3.org/2002/07/owl#Ontology")
version_iri = attrs.get("versionIRI")
if version_iri:
on_prepare_triple(ontology_iri,
"http://www.w3.org/2002/07/owl#versionIRI",
version_iri)
elif (tag == "RDF") or (tag == "rdf:RDF"):
raise ValueError(
"Not an OWL/XML file! (It seems to be an OWL/RDF file)")
def endElement(tag):
nonlocal in_declaration, objs, in_prop_chain
if (tag == "http://www.w3.org/2002/07/owl#Declaration"):
in_declaration = False
objs = [] # Purge stack
elif (tag == "http://www.w3.org/2002/07/owl#Literal"):
objs.append(new_literal(current_content, current_attrs))
elif (tag == "http://www.w3.org/2002/07/owl#SubClassOf") or (
tag == "http://www.w3.org/2002/07/owl#SubObjectPropertyOf"
) or (tag == "http://www.w3.org/2002/07/owl#SubDataPropertyOf") or (
tag
== "http://www.w3.org/2002/07/owl#SubAnnotationPropertyOf"):
parent = objs.pop()
child = objs.pop()
if (tag == "http://www.w3.org/2002/07/owl#SubObjectPropertyOf"
) and in_prop_chain:
relation = "http://www.w3.org/2002/07/owl#propertyChainAxiom"
parent, child = child, parent
else:
relation = sub_ofs[tag]
on_prepare_triple(child, relation, parent)
if annots: purge_annotations((child, relation, parent))
elif (tag == "http://www.w3.org/2002/07/owl#ClassAssertion"):
child = objs.pop() # Order is reversed compared to SubClassOf!
parent = objs.pop()
on_prepare_triple(child, rdf_type, parent)
if annots: purge_annotations((child, rdf_type, parent))
elif (tag == "http://www.w3.org/2002/07/owl#EquivalentClasses") or (
tag
== "http://www.w3.org/2002/07/owl#EquivalentObjectProperties"
) or (tag == "http://www.w3.org/2002/07/owl#EquivalentDataProperties"):
o1 = objs.pop()
o2 = objs.pop()
if o1.startswith("_"):
o1, o2 = o2, o1 # Swap in order to have blank node at third position -- rapper seems to do that
on_prepare_triple(o1, equivs[tag], o2)
if annots: purge_annotations((o1, equivs[tag], o2))
elif (tag == "http://www.w3.org/2002/07/owl#ObjectPropertyDomain") or (
tag == "http://www.w3.org/2002/07/owl#DataPropertyDomain"
) or (tag == "http://www.w3.org/2002/07/owl#AnnotationPropertyDomain"):
val = objs.pop()
obj = objs.pop()
on_prepare_triple(obj,
"http://www.w3.org/2000/01/rdf-schema#domain",
val)
if annots:
purge_annotations(
(obj, "http://www.w3.org/2000/01/rdf-schema#domain", val))
elif (tag == "http://www.w3.org/2002/07/owl#ObjectPropertyRange") or (
tag == "http://www.w3.org/2002/07/owl#DataPropertyRange"
) or (tag == "http://www.w3.org/2002/07/owl#AnnotationPropertyRange"):
val = objs.pop()
obj = objs.pop()
on_prepare_triple(obj,
"http://www.w3.org/2000/01/rdf-schema#range",
val)
if annots:
purge_annotations(
(obj, "http://www.w3.org/2000/01/rdf-schema#range", val))
elif (tag in prop_types):
obj = objs.pop()
on_prepare_triple(obj, rdf_type, prop_types[tag])
elif (tag == "http://www.w3.org/2002/07/owl#InverseObjectProperties"
) or (tag
== "http://www.w3.org/2002/07/owl#InverseDataProperties"):
a, b = objs.pop(), objs.pop()
on_prepare_triple(b, "http://www.w3.org/2002/07/owl#inverseOf", a)
elif (tag == "http://www.w3.org/2002/07/owl#ObjectPropertyChain"):
start = _rindex(objs)
list_iri = new_list(objs[start + 1:])
in_prop_chain = True
objs[start:] = [list_iri]
elif (tag in disjoints):
start = _rindex(objs)
list_obj = objs[start + 1:]
tag, rel, member = disjoints[tag]
if rel and (len(list_obj) == 2):
on_prepare_triple(list_obj[0], rel, list_obj[1])
if annots: purge_annotations((list_obj[0], rel, list_obj[1]))
else:
list_iri = new_list(list_obj)
iri = new_blank()
on_prepare_triple(iri, rdf_type, tag)
on_prepare_triple(iri, member, list_iri)
if annots: purge_annotations((iri, rdf_type, tag))
del objs[start:]
elif (tag == "http://www.w3.org/2002/07/owl#ObjectPropertyAssertion"
) or (tag
== "http://www.w3.org/2002/07/owl#DataPropertyAssertion"):
p, s, o = objs[-3:]
on_prepare_triple(s, p, o)
if annots: purge_annotations((s, p, o))
del objs[-3:]
elif (tag == "http://www.w3.org/2002/07/owl#ObjectComplementOf") or (
tag == "http://www.w3.org/2002/07/owl#DataComplementOf"):
iri = new_blank()
on_prepare_triple(iri, rdf_type,
"http://www.w3.org/2002/07/owl#Class")
on_prepare_triple(iri,
"http://www.w3.org/2002/07/owl#complementOf",
objs[-1])
objs[-1] = iri
elif (tag in restrs):
iri = new_blank()
on_prepare_triple(iri, rdf_type,
"http://www.w3.org/2002/07/owl#Restriction")
on_prepare_triple(iri, "http://www.w3.org/2002/07/owl#onProperty",
objs.pop(-2))
on_prepare_triple(iri, restrs[tag], objs[-1])
objs[-1] = iri
elif (tag in card_restrs):
iri = new_blank()
on_prepare_triple(iri, rdf_type,
"http://www.w3.org/2002/07/owl#Restriction")
start = _rindex(objs)
values = objs[start + 1:]
del objs[start:]
if len(values) == 2: # Qualified
tag = qual_card_restrs[tag]
on_prepare_triple(iri,
"http://www.w3.org/2002/07/owl#onProperty",
values[-2])
if objs[-1].startswith("http://www.w3.org/2001/XMLSchema"):
on_prepare_triple(
iri, "http://www.w3.org/2002/07/owl#onDataRange",
values[-1])
else:
on_prepare_triple(iri,
"http://www.w3.org/2002/07/owl#onClass",
values[-1])
else: # Non qualified
tag = card_restrs[tag]
on_prepare_triple(iri,
"http://www.w3.org/2002/07/owl#onProperty",
values[-1])
on_prepare_triple(
iri, tag,
new_literal(
last_cardinality, {
"datatypeIRI":
"http://www.w3.org/2001/XMLSchema#nonNegativeInteger"
}))
objs.append(iri)
elif (tag == "http://www.w3.org/2002/07/owl#ObjectOneOf"):
start = _rindex(objs)
list_iri = new_list(objs[start + 1:])
iri = new_blank()
on_prepare_triple(iri, rdf_type,
"http://www.w3.org/2002/07/owl#Class")
on_prepare_triple(iri, "http://www.w3.org/2002/07/owl#oneOf",
list_iri)
objs[start:] = [iri]
elif (tag == "http://www.w3.org/2002/07/owl#ObjectIntersectionOf") or (
tag == "http://www.w3.org/2002/07/owl#ObjectUnionOf"
) or (tag == "http://www.w3.org/2002/07/owl#DataIntersectionOf") or (
tag == "http://www.w3.org/2002/07/owl#DataUnionOf"):
start = _rindex(objs)
list_iri = new_list(objs[start + 1:])
iri = new_blank()
if objs[start +
1:][0].startswith("http://www.w3.org/2001/XMLSchema"):
on_prepare_triple(
iri, rdf_type,
"http://www.w3.org/2000/01/rdf-schema#Datatype")
else:
on_prepare_triple(iri, rdf_type,
"http://www.w3.org/2002/07/owl#Class")
if (tag == "http://www.w3.org/2002/07/owl#ObjectIntersectionOf"
) or (tag
== "http://www.w3.org/2002/07/owl#DataIntersectionOf"):
on_prepare_triple(
iri, "http://www.w3.org/2002/07/owl#intersectionOf",
list_iri)
else:
on_prepare_triple(iri, "http://www.w3.org/2002/07/owl#unionOf",
list_iri)
objs[start:] = [iri]
elif (tag == "http://www.w3.org/2002/07/owl#Import"):
on_prepare_triple(ontology_iri,
"http://www.w3.org/2002/07/owl#imports",
current_content)
elif (tag == "http://www.w3.org/2002/07/owl#IRI"):
iri = current_content
if not iri: iri = ontology_iri
else:
if iri.startswith("#") or iri.startswith("/"):
iri = ontology_iri + iri
objs.append(iri)
elif (tag == "http://www.w3.org/2002/07/owl#AbbreviatedIRI"):
iri = unabbreviate_IRI(current_content)
objs.append(iri)
elif (tag == "http://www.w3.org/2002/07/owl#AnnotationAssertion"):
on_prepare_triple(objs[-2], objs[-3], objs[-1])
if annots: purge_annotations((objs[-2], objs[-3], objs[-1]))
elif (tag == "http://www.w3.org/2002/07/owl#Annotation"):
if before_declaration: # On ontology
on_prepare_triple(ontology_iri, objs[-2], objs[-1])
else:
annots.append((objs[-2], objs[-1]))
del objs[-2:]
elif (tag == "http://www.w3.org/2002/07/owl#DatatypeRestriction"):
start = _rindex(objs)
datatype, *list_bns = objs[start + 1:]
list_bns = new_list(list_bns)
bn = new_blank()
objs[start:] = [bn]
on_prepare_triple(bn, rdf_type,
"http://www.w3.org/2000/01/rdf-schema#Datatype")
on_prepare_triple(bn, "http://www.w3.org/2002/07/owl#onDatatype",
datatype)
on_prepare_triple(
bn, "http://www.w3.org/2002/07/owl#withRestrictions", list_bns)
elif (tag == "http://www.w3.org/2002/07/owl#FacetRestriction"):
facet, literal = objs[-2:]
bn = new_blank()
on_prepare_triple(bn, facet, literal)
objs[-2:] = [bn]
elif (tag == "http://www.w3.org/2002/07/owl#ObjectInverseOf") or (
tag == "http://www.w3.org/2002/07/owl#DataInverseOf") or (
tag == "http://www.w3.org/2002/07/owl#inverseOf"):
bn, prop = objs[-2:]
on_prepare_triple(bn, "http://www.w3.org/2002/07/owl#inverseOf",
prop)
objs[-2:] = [bn]
def characters(content):
nonlocal current_content
current_content += content
def purge_annotations(on_iri):
nonlocal annots
if isinstance(on_iri, tuple):
s, p, o = on_iri
on_iri = new_blank()
on_prepare_triple(on_iri, rdf_type,
"http://www.w3.org/2002/07/owl#Axiom")
on_prepare_triple(on_iri,
"http://www.w3.org/2002/07/owl#annotatedSource",
s)
on_prepare_triple(
on_iri, "http://www.w3.org/2002/07/owl#annotatedProperty", p)
on_prepare_triple(on_iri,
"http://www.w3.org/2002/07/owl#annotatedTarget",
o)
for prop_iri, value in annots:
on_prepare_triple(on_iri, prop_iri, value)
annots = []
#parser.StartNamespaceDeclHandler = startNamespace
#parser.EndNamespaceDeclHandler = endNamespace
parser.StartElementHandler = startElement
parser.EndElementHandler = endElement
parser.CharacterDataHandler = characters
try:
if isinstance(f, str):
f = open(f, "rb")
parser.ParseFile(f)
f.close()
else:
parser.ParseFile(f)
except Exception as e:
raise OwlReadyOntologyParsingError(
"OWL/XML parsing error in file %s, line %s, column %s." %
(getattr(f, "name", "???"), parser.CurrentLineNumber,
parser.CurrentColumnNumber)) from e
return nb_triple
def parse(f,
on_prepare_triple=None,
new_blank=None,
new_literal=None,
default_base=""):
parser = xml.parsers.expat.ParserCreate(None, "")
try:
parser.buffer_text = True
parser.specified_attributes = True
except:
pass
stack = [["", ""]] # List of [parse type, value] pairs
prefixes = {}
prefixess = [prefixes]
tag_is_predicate = False
current_blank = 0
current_fake_blank = 0
current_content = ""
current_attrs = None
nb_triple = 0
bns = defaultdict(set)
dont_create_unnamed_bn = False
axiom_annotation_sources = {}
axiom_annotation_props = {}
axiom_annotation_targets = {}
triples_with_unnamed_bn = defaultdict(list)
if default_base:
xml_base = default_base
if xml_base.endswith("#") or xml_base.endswith("/"):
xml_base = xml_base[:-1]
xml_dir = xml_base.rsplit("/", 1)[0] + "/"
else:
xml_base = ""
xml_dir = ""
if not on_prepare_triple:
def on_prepare_triple(s, p, o):
nonlocal nb_triple
nb_triple += 1
if not s.startswith("_"): s = "<%s>" % s
if not (o.startswith("_") or o.startswith('"')): o = "<%s>" % o
print("%s %s %s ." % (s, "<%s>" % p, o))
if not new_blank:
def new_blank():
nonlocal current_blank
current_blank += 1
return "_:%s" % current_blank
def new_fake_blank():
nonlocal current_fake_blank
current_fake_blank += 1
return "_ %s" % current_fake_blank
node_2_blanks = defaultdict(new_blank)
known_nodes = set()
if not new_literal:
def new_literal(value, attrs):
value = value.replace('"', '\\"').replace("\n", "\\n")
lang = attrs.get("http://www.w3.org/XML/1998/namespacelang")
if lang: return '"%s"@%s' % (value, lang)
datatype = attrs.get(
"http://www.w3.org/1999/02/22-rdf-syntax-ns#datatype")
if datatype: return '"%s"^^<%s>' % (value, datatype)
return '"%s"' % (value)
def new_list(l):
bn = bn0 = new_blank()
if l:
for i in range(len(l) - 1):
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first",
l[i])
bn_next = new_blank()
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
bn_next)
bn = bn_next
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first", l[-1])
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
else:
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#first",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
on_prepare_triple(
bn, "http://www.w3.org/1999/02/22-rdf-syntax-ns#rest",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nil")
return bn0
def add_to_bn(bn, type, rel, value):
if type == "COL":
value = tuple(
frozenset(bns[v]) if v.startswith("_") and (
not v in known_nodes) else v for v in value)
bns[bn].add((type, rel) + value)
else:
if value.startswith("_") and (not value in known_nodes):
value = frozenset(bns[value])
bns[bn].add((type, rel, value))
def startNamespace(prefix, uri):
nonlocal prefixes
prefixes = prefixes.copy()
prefixess.append(prefixes)
if prefix: prefixes[prefix] = uri
else: prefixes[""] = uri
def endNamespace(prefix):
nonlocal prefixes
prefixess.pop()
prefixes = prefixess[-1]
def startElement(tag, attrs):
nonlocal tag_is_predicate, current_content, current_attrs, dont_create_unnamed_bn, xml_base, xml_dir
tag_is_predicate = not tag_is_predicate
if tag_is_predicate:
if attrs.get("http://www.w3.org/1999/02/22-rdf-syntax-ns#parseType"
) == "Collection":
stack.append(["Collection", []])
elif tag == "http://www.w3.org/1999/02/22-rdf-syntax-ns#RDF":
stack.append(["RDF", ""])
namespace_base = attrs.get(
"http://www.w3.org/XML/1998/namespacebase")
if namespace_base:
xml_base = namespace_base
xml_dir = namespace_base.rsplit("/", 1)[0] + "/"
else:
iri = attrs.get(
"http://www.w3.org/1999/02/22-rdf-syntax-ns#resource")
if iri:
if not ":" in iri:
if not iri: iri = xml_base
elif iri.startswith("#"): iri = xml_base + iri
elif iri.startswith("/"): iri = xml_dir + iri[1:]
else: iri = urllib.parse.urljoin(xml_dir, iri)
stack.append(["Resource", iri])
else:
iri = attrs.get(
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nodeID")
if iri:
iri = node_2_blanks[iri]
known_nodes.add(iri)
stack.append(["Resource", iri])
else:
stack.append(["Literal", ""])
current_content = ""
current_attrs = attrs
if (tag == "http://www.w3.org/2002/07/owl#annotatedSource"
) or (tag ==
"http://www.w3.org/2002/07/owl#annotatedTarget"):
dont_create_unnamed_bn = True
else:
iri = attrs.get("http://www.w3.org/1999/02/22-rdf-syntax-ns#about",
None)
if iri is None:
iri = attrs.get(
"http://www.w3.org/1999/02/22-rdf-syntax-ns#ID", None)
if iri: iri = "#%s" % iri
if iri is None:
iri = attrs.get(
"http://www.w3.org/1999/02/22-rdf-syntax-ns#nodeID")
if iri:
iri = node_2_blanks[iri]
known_nodes.add(iri)
else:
if dont_create_unnamed_bn: iri = new_fake_blank()
else: iri = new_blank()
else:
if not ":" in iri:
if not iri: iri = xml_base
elif iri.startswith("#"): iri = xml_base + iri
elif iri.startswith("/"): iri = xml_dir + iri[1:]
else: iri = urllib.parse.urljoin(xml_dir, iri)
if tag != "http://www.w3.org/1999/02/22-rdf-syntax-ns#Description":
if not iri.startswith("_ "):
on_prepare_triple(
iri, "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
tag)
if iri.startswith("_"):
add_to_bn(
iri, "REL",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#type", tag)
if stack[-1][0] == "Collection":
stack[-1][1].append(iri)
else:
if stack[-1][0] == "Literal": stack[-1][0] = "Resource"
stack[-1][1] = iri
def endElement(tag):
nonlocal tag_is_predicate, dont_create_unnamed_bn
if tag_is_predicate:
parse_type, value = stack.pop()
if stack[-1][0] == "Collection": iri = stack[-1][1][-1]
else: iri = stack[-1][1]
if tag == "http://www.w3.org/2002/07/owl#annotatedSource":
dont_create_unnamed_bn = False
axiom_annotation_sources[iri] = value
if isinstance(value, str) and value.startswith("_ "):
triples_with_unnamed_bn[iri].insert(
0, (tag, value, parser.CurrentLineNumber,
parser.CurrentColumnNumber))
tag_is_predicate = not tag_is_predicate
return
elif tag == "http://www.w3.org/2002/07/owl#annotatedProperty":
axiom_annotation_props[iri] = value
elif tag == "http://www.w3.org/2002/07/owl#annotatedTarget":
dont_create_unnamed_bn = False
axiom_annotation_targets[iri] = value
if isinstance(value, str) and value.startswith("_ "):
triples_with_unnamed_bn[iri].append(
(tag, value, parser.CurrentLineNumber,
parser.CurrentColumnNumber))
tag_is_predicate = not tag_is_predicate
return
if parse_type == "Resource":
if not iri.startswith("_ "):
on_prepare_triple(iri, tag, value)
if iri.startswith("_"):
add_to_bn(iri, "REL", tag, value)
if value.startswith("_"):
add_to_bn(value, "INV", tag, iri)
elif parse_type == "Literal":
value = new_literal(current_content, current_attrs)
if not iri.startswith("_ "):
on_prepare_triple(iri, tag, value)
if iri.startswith("_"):
add_to_bn(iri, "REL", tag, value)
elif parse_type == "Collection":
if not iri.startswith("_ "):
on_prepare_triple(iri, tag, new_list(value))
if iri.startswith("_"):
add_to_bn(iri, "COL", tag, value)
tag_is_predicate = not tag_is_predicate
def characters(content):
nonlocal current_content
if stack[-1][0] == "Literal": current_content += content
parser.StartNamespaceDeclHandler = startNamespace
parser.EndNamespaceDeclHandler = endNamespace
parser.StartElementHandler = startElement
parser.EndElementHandler = endElement
parser.CharacterDataHandler = characters
try:
if isinstance(f, str):
f = open(f, "rb")
parser.ParseFile(f)
f.close()
else:
parser.ParseFile(f)
except Exception as e:
raise OwlReadyOntologyParsingError(
"RDF/XML parsing error in file %s, line %s, column %s." %
(getattr(f, "name", "???"), parser.CurrentLineNumber,
parser.CurrentColumnNumber)) from e
if triples_with_unnamed_bn:
content_2_bns = defaultdict(list)
for bn, content in bns.items():
if not bn.startswith("_ "):
content_2_bns[frozenset(content)].append(bn)
def rebuild_bn(content):
bn = new_blank()
content_2_bns[frozenset(content)].append(bn)
for i in content:
if i[0] == "REL":
drop, p, o = i
if not isinstance(o, str): o = rebuild_bn(o)
on_prepare_triple(bn, p, o)
elif i[0] == "INV":
drop, p, o = i
if not isinstance(o, str): o = rebuild_bn(o)
on_prepare_triple(o, p, bn)
elif i[0] == "COL":
drop, p, *l = i
l = [(isinstance(x, str) and x) or rebuild_bn(x)
for x in l]
o = new_list(l)
on_prepare_triple(bn, p, o)
else:
print(i)
raise ValueError
return bn
for axiom_iri, triples in triples_with_unnamed_bn.items():
for p, o, line, column in triples:
try:
content = bns[o]
if p == "http://www.w3.org/2002/07/owl#annotatedSource":
target = axiom_annotation_targets[axiom_iri]
if target.startswith("_"):
target = frozenset(bns[target])
candidates_bn = content_2_bns[frozenset(
content
| {("REL", axiom_annotation_props[axiom_iri],
target)})]
else:
source = axiom_annotation_sources[axiom_iri]
if source.startswith("_"):
source = frozenset(bns[source] | {(
"REL", axiom_annotation_props[axiom_iri],
target)})
candidates_bn = (content_2_bns[frozenset(
content
| {("INV", axiom_annotation_props[axiom_iri],
source)})]
or content_2_bns[frozenset(content)])
if candidates_bn: o = candidates_bn[-1]
else:
#print()
#print("rebuild", o, content)
o = rebuild_bn(content)
#print()
on_prepare_triple(axiom_iri, p, o)
except Exception as e:
raise OwlReadyOntologyParsingError(
"RDF/XML parsing error in file %s, line %s, column %s."
% (getattr(f, "name", "???"), line, column)) from e
return nb_triple