def update_rdf_for_conversion(vocabprefix, vocab_properties,
rdf_vocab_properties):
html_vocab_properties = {}
html_vocab_properties['format'] = 'text/html'
html_vocab_properties['name'] = "%s.html" % os.path.splitext(
rdf_vocab_properties['name'])[0]
html_vocab_properties['path'] = rdf_vocab_properties['path'].replace(
rdf_vocab_properties['name'], html_vocab_properties['name'])
html_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(
rdf_vocab_properties['name'], html_vocab_properties['name'])
newrdf_vocab_properties = {}
newrdf_vocab_properties['format'] = 'application/rdf+xml'
newrdf_vocab_properties['name'] = "%s_modified.rdf" % os.path.splitext(
rdf_vocab_properties['name'])[0]
newrdf_vocab_properties['path'] = rdf_vocab_properties['path'].replace(
rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
newrdf_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(
rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
graph = Graph()
graph.parse(rdf_vocab_properties['path'])
graph_ns = []
for nsprefix, nsurl in graph.namespaces():
graph_ns.append(str(nsurl))
ET._namespace_map[str(nsurl)] = str(nsprefix)
for prefix, url in namespaces.iteritems():
if not str(url) in graph_ns:
ET._namespace_map[str(url)] = str(prefix)
def_tags = [
"{http://www.w3.org/2000/01/rdf-schema#}Class".lower(),
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}Property".lower(),
"{http://www.w3.org/2002/07/owl#}ObjectProperty".lower(),
"{http://www.w3.org/2002/07/owl#}DatatypeProperty".lower(),
"{http://www.w3.org/2002/07/owl#}Class".lower(),
]
tree = ET.ElementTree(file=rdf_vocab_properties['path'])
ns_uri = vocab_properties['preferredNamespaceUri']
html_uri = html_vocab_properties['uri']
rdf_uri = rdf_vocab_properties['uri']
tree_root = tree.getroot()
#vocab= tree_root.findall("{http://www.w3.org/2002/07/owl#}Ontology")
vocab = tree_root.find("{http://www.w3.org/2002/07/owl#}Ontology")
if vocab:
#for vocab in vocabs:
if not vocab.findall("{http://purl.org/dc/elements/1.1/}identifier"):
se0 = ET.SubElement(
vocab, "{http://purl.org/dc/elements/1.1/}identifier")
se0.text = rdf_uri
if not vocab.findall("{http://purl.org/dc/terms/}isVersionOf"):
se1 = ET.SubElement(
vocab,
"{http://purl.org/dc/terms/}isVersionOf",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource":
ns_uri
})
if not vocab.findall(
"{http://purl.org/vocab/vann/}preferredNamespacePrefix"):
se2a = ET.SubElement(
vocab, "{http://purl.org/vocab/vann/}preferredNamespacePrefix")
se2a.text = vocab_properties['preferredNamespacePrefix']
if not vocab.findall(
"{http://purl.org/vocab/vann/}preferredNamespaceUri"):
se2b = ET.SubElement(
vocab, "{http://purl.org/vocab/vann/}preferredNamespaceUri")
se2b.text = vocab_properties['preferredNamespaceUri']
if not vocab.findall("{http://purl.org/dc/terms/}hasFormat"):
#Add html uri - html_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":
html_uri
})
se3c = ET.SubElement(se3b,
"{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
#ET.TreeBuilder.data('text/html')
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={"{http://www.w3.org/XML/1998/namespace}lang": "en"})
se3f.text = 'HTML'
#ET.TreeBuilder.data('HTML')
#Add rdf uri - rdf_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":
rdf_uri
})
se3c = ET.SubElement(se3b,
"{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
#ET.TreeBuilder.data('application/rdf+xml')
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={"{http://www.w3.org/XML/1998/namespace}lang": "en"})
se3f.text = 'RDF'
#ET.TreeBuilder.data('RDF')
else:
#Check the formats available and add if necessary
formats = vocab.findall("{http://purl.org/dc/terms/}hasFormat")
available_types = []
for f in formats:
type_tags = f.findall(
".//{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
for type_tag in type_tags:
if type_tag.attrib.get(
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource"
):
ftype = type_tag.attrib.get(
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource"
)
elif type_tag.text:
ftype = type_tag.text
if ftype and 'html' in ftype.lower():
available_types.append('html')
elif ftype and 'rdf' in ftype.lower():
available_types.append('rdf')
if not 'html' in available_types:
#Add html file - vocabfile_html
se3a = ET.SubElement(vocab,
"{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":
html_uri
})
se3c = ET.SubElement(
se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
#ET.TreeBuilder.data('text/html')
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={
"{http://www.w3.org/XML/1998/namespace}lang": "en"
})
se3f.text = 'HTML'
#ET.TreeBuilder.data('HTML')
if not 'rdf' in available_types:
#Add rdf file - vocabfile
se3a = ET.SubElement(vocab,
"{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":
rdf_uri
})
se3c = ET.SubElement(
se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
#ET.TreeBuilder.data('application/rdf+xml')
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={
"{http://www.w3.org/XML/1998/namespace}lang": "en"
})
se3f.text = 'RDF'
#ET.TreeBuilder.data('RDF')
else:
vocab = ET.SubElement(
tree_root,
"{http://www.w3.org/2002/07/owl#}Ontology",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about": ns_uri
})
se0 = ET.SubElement(vocab,
"{http://purl.org/dc/elements/1.1/}identifier")
se0.text = rdf_uri
se1 = ET.SubElement(
vocab,
"{http://purl.org/dc/terms/}isVersionOf",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource": ns_uri
})
se2a = ET.SubElement(
vocab, "{http://purl.org/vocab/vann/}preferredNamespacePrefix")
se2a.text = vocab_properties['preferredNamespacePrefix']
se2b = ET.SubElement(
vocab, "{http://purl.org/vocab/vann/}preferredNamespaceUri")
se2b.text = vocab_properties['preferredNamespaceUri']
#Add html uri - html_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about": html_uri
})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={"{http://www.w3.org/XML/1998/namespace}lang": "en"})
se3f.text = 'HTML'
#Add rdf uri - rdf_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(
se3a,
"{http://purl.org/dc/dcmitype/}Text",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about": rdf_uri
})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(
se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
se3f = ET.SubElement(
se3d,
"{http://www.w3.org/2000/01/rdf-schema#}label",
attrib={"{http://www.w3.org/XML/1998/namespace}lang": "en"})
se3f.text = 'RDF'
terms = tree_root.getiterator()
#terms = vocab.getiterator()
for term in terms:
if term.tag.lower().strip() in def_tags:
defby = None
defby = term.find(
"{http://www.w3.org/2000/01/rdf-schema#}isDefinedBy")
if not defby:
se4 = ET.SubElement(
term,
"{http://www.w3.org/2000/01/rdf-schema#}isDefinedBy",
attrib={
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource":
ns_uri
})
#Move ontology to the first element
tree_root.remove(vocab)
tree_root.insert(0, vocab)
tree.write(newrdf_vocab_properties['path'])
#tree_root.write(newrdf_vocab_properties['path'])
return (newrdf_vocab_properties, html_vocab_properties)
class TriplesDatabase(object):
"""A database from the defined triples"""
def __init__(self):
self._open = False
def open(self, filename, graphClass=None):
"""
Load existing database at 'filename'.
"""
if filename is None:
if graphClass is None:
self.graph = Graph()
else:
self.graph = graphClass()
else:
assert os.path.exists(filename), (
"%s must be an existing database" % (filename,))
path, filename = os.path.split(filename)
self.graph = sqliteBackedGraph(path, filename)
self._open = True
def query(self, rest, initNs=None, initBindings=None):
"""
Execute a SPARQL query and get the results as a SPARQLResult
{rest} is a string that should begin with "SELECT ", usually
"""
assert self._open
if initNs is None:
initNs = dict(self.graph.namespaces())
if initBindings is None: initBindings = {}
sel = select(self.getBase(), rest)
ret = self.graph.query(sel, initNs=initNs, initBindings=initBindings,
DEBUG=False)
return ret
def getBase(self):
d = dict(self.graph.namespaces())
return d.get('', RDFSNS)
def addTriple(self, s, v, *objects):
"""
Make a statement/arc/triple in the database.
Strings, ints and floats as s or o will automatically be coerced to
RDFLiteral(). It is an error to give a RDFLiteral as v, so no
coercion will be done in that position.
2-tuples will be coerced to bnodes.
If more than one object is given, i.e.
addTriple(a, b, c1, c2, c3)
this is equivalent to:
addTriple(a,b,c1); addTriple(a,b,c2); addTriple(a,b,c3)
"""
assert self._open
assert len(objects) >= 1, "You must provide at least one object"
if canBeLiteral(s):
s = RDFLiteral(s)
bnode = None
for o in objects:
if canBeLiteral(o):
o = RDFLiteral(o)
elif isinstance(o, tuple) and len(o) == 2:
if bnode is None:
bnode = BNode()
self.addTriple(bnode, *o)
o = bnode
assert None not in [s,v,o]
self.graph.add((s, v, o))
def dump(self):
assert self._open
io = StringIO()
try:
self.graph.serialize(destination=io, format='n3')
except Exception, e:
import sys, pdb; pdb.post_mortem(sys.exc_info()[2])
return io.getvalue()
def update_rdf_for_conversion(prefix, vocab_properties, rdf_vocab_properties):
#(id, base, prefix) = get_vocab_base(vocabfile)
html_vocab_properties = {}
html_vocab_properties['format'] = 'text/html'
html_vocab_properties['name'] = "%s.html" % os.path.splitext(
rdf_vocab_properties['name'])[0]
html_vocab_properties['path'] = rdf_vocab_properties['path'].replace(
rdf_vocab_properties['name'], html_vocab_properties['name'])
html_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(
rdf_vocab_properties['name'], html_vocab_properties['name'])
newrdf_vocab_properties = {}
newrdf_vocab_properties['format'] = 'application/rdf+xml'
newrdf_vocab_properties['name'] = "%s_modified.rdf" % os.path.splitext(
rdf_vocab_properties['name'])[0]
newrdf_vocab_properties['path'] = rdf_vocab_properties['path'].replace(
rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
newrdf_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(
rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
graph = Graph()
graph.parse(rdf_vocab_properties['path'])
subject = None
for s in graph.subjects(namespaces['rdf']['type'],
URIRef(namespaces['owl']['Ontology'])):
subject = s
#graph2 = Graph()
graph_ns = []
for nsprefix, nsurl in graph.namespaces():
graph_ns.append(str(nsurl))
for prefix, url in namespaces.iteritems():
if not str(url) in graph_ns:
graph.bind(prefix, URIRef(url))
#properties = get_vocab_properties(prefix)
#subject = None
#for s in graph.subjects(namespaces['dc']['title'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dcterms']['title'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dc']['creator'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dcterms']['creator'], None):
# subject = s
formatNode1 = BNode()
formatNode2 = BNode()
#Add vocabulary properties identifier and format
graph.add((subject, namespaces['dc']['identifier'],
URIRef(rdf_vocab_properties['uri'])))
graph.add((subject, namespaces['dcterms']['isVersionOf'],
URIRef(vocab_properties['preferredNamespaceUri'])))
graph.add((subject, namespaces['dcterms']['hasFormat'],
URIRef(rdf_vocab_properties['uri'])))
graph.add((subject, namespaces['dcterms']['hasFormat'],
URIRef(html_vocab_properties['uri'])))
graph.add((subject, namespaces['vann']['preferredNamespaceUri'],
URIRef(vocab_properties['preferredNamespaceUri'])))
graph.add((subject, namespaces['vann']['preferredNamespacePrefix'],
URIRef(vocab_properties['preferredNamespacePrefix'])))
graph.add((URIRef(html_vocab_properties['uri']), namespaces['rdf']['type'],
URIRef(namespaces['dctype']['Text'])))
graph.add((URIRef(html_vocab_properties['uri']),
namespaces['dc']['format'], formatNode1))
graph.add((formatNode1, namespaces['rdf']['value'], Literal('text/html')))
graph.add((formatNode1, namespaces['rdfs']['label'], Literal('HTML')))
graph.add((formatNode1, namespaces['rdf']['type'],
URIRef(namespaces['dcterms']['IMT'])))
graph.add((URIRef(rdf_vocab_properties['uri']), namespaces['rdf']['type'],
URIRef(namespaces['dctype']['Text'])))
graph.add((URIRef(rdf_vocab_properties['uri']), namespaces['dc']['format'],
formatNode2))
graph.add((formatNode2, namespaces['rdf']['value'],
Literal('application/rdf+xml')))
graph.add((formatNode2, namespaces['rdfs']['label'], Literal('RDF')))
graph.add((formatNode2, namespaces['rdf']['type'],
URIRef(namespaces['dcterms']['IMT'])))
#Add rdfs:isDefinedBy for each class / property / term of the vocabulary
#Find if schema is rdfs / owl. This defines the possible types (rdf:type) for each class / property / term
#testo = vocab_type_definitions_test['rdfs']
#subjects = []
#subs = graph.subjects(namespaces['rdf']['type'], URIRef(testo))
#for s in subs:
# subjects.append(s)
#if subjects:
# objects = vocab_type_definitions_rdfs
#else:
# objects = vocab_type_definitions_owl
#For all subjects that are of the type found above, add rdfs:isDefinedBy
#for o in objects:
# subs = graph.subjects(namespaces['rdf']['type'], o)
# for s in subs:
# graph.add((s, namespaces['rdfs']['isDefinedBy'], URIRef(vocab_properties['preferredNamespaceUri'])))
list_of_terms = get_terms(rdf_vocab_properties['path'])
for s in list_of_terms:
graph.add((URIRef(s), namespaces['rdfs']['isDefinedBy'],
URIRef(vocab_properties['preferredNamespaceUri'])))
rdf_str = None
rdf_str = graph.serialize(format="pretty-xml")
#f = codecs.open(newrdf_vocab_properties['path'], 'w', 'utf-8')
f = codecs.open(newrdf_vocab_properties['path'], 'w')
f.write(rdf_str)
f.close()
return (newrdf_vocab_properties, html_vocab_properties)
def update_rdf_for_conversion(vocabprefix, vocab_properties, rdf_vocab_properties):
html_vocab_properties = {}
html_vocab_properties['format'] = 'text/html'
html_vocab_properties['name'] = "%s.html"%os.path.splitext(rdf_vocab_properties['name'])[0]
html_vocab_properties['path'] = rdf_vocab_properties['path'].replace(rdf_vocab_properties['name'], html_vocab_properties['name'])
html_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(rdf_vocab_properties['name'], html_vocab_properties['name'])
newrdf_vocab_properties = {}
newrdf_vocab_properties['format'] = 'application/rdf+xml'
newrdf_vocab_properties['name'] = "%s_modified.rdf"%os.path.splitext(rdf_vocab_properties['name'])[0]
newrdf_vocab_properties['path'] = rdf_vocab_properties['path'].replace(rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
newrdf_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
graph = Graph()
graph.parse(rdf_vocab_properties['path'])
graph_ns = []
for nsprefix, nsurl in graph.namespaces():
graph_ns.append(str(nsurl))
ET._namespace_map[str(nsurl)] = str(nsprefix)
for prefix, url in namespaces.iteritems():
if not str(url) in graph_ns:
ET._namespace_map[str(url)] = str(prefix)
def_tags = [
"{http://www.w3.org/2000/01/rdf-schema#}Class".lower(),
"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}Property".lower(),
"{http://www.w3.org/2002/07/owl#}ObjectProperty".lower(),
"{http://www.w3.org/2002/07/owl#}DatatypeProperty".lower(),
"{http://www.w3.org/2002/07/owl#}Class".lower(),
]
tree = ET.ElementTree(file=rdf_vocab_properties['path'])
ns_uri = vocab_properties['preferredNamespaceUri']
html_uri = html_vocab_properties['uri']
rdf_uri = rdf_vocab_properties['uri']
tree_root = tree.getroot()
#vocab= tree_root.findall("{http://www.w3.org/2002/07/owl#}Ontology")
vocab= tree_root.find("{http://www.w3.org/2002/07/owl#}Ontology")
if vocab:
#for vocab in vocabs:
if not vocab.findall("{http://purl.org/dc/elements/1.1/}identifier"):
se0 = ET.SubElement(vocab, "{http://purl.org/dc/elements/1.1/}identifier")
se0.text = rdf_uri
if not vocab.findall("{http://purl.org/dc/terms/}isVersionOf"):
se1 = ET.SubElement(vocab, "{http://purl.org/dc/terms/}isVersionOf", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource":ns_uri})
if not vocab.findall("{http://purl.org/vocab/vann/}preferredNamespacePrefix"):
se2a = ET.SubElement(vocab, "{http://purl.org/vocab/vann/}preferredNamespacePrefix")
se2a.text = vocab_properties['preferredNamespacePrefix']
if not vocab.findall("{http://purl.org/vocab/vann/}preferredNamespaceUri"):
se2b = ET.SubElement(vocab, "{http://purl.org/vocab/vann/}preferredNamespaceUri")
se2b.text = vocab_properties['preferredNamespaceUri']
if not vocab.findall("{http://purl.org/dc/terms/}hasFormat"):
#Add html uri - html_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":html_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
#ET.TreeBuilder.data('text/html')
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'HTML'
#ET.TreeBuilder.data('HTML')
#Add rdf uri - rdf_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":rdf_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
#ET.TreeBuilder.data('application/rdf+xml')
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'RDF'
#ET.TreeBuilder.data('RDF')
else:
#Check the formats available and add if necessary
formats = vocab.findall("{http://purl.org/dc/terms/}hasFormat")
available_types = []
for f in formats:
type_tags = f.findall(".//{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
for type_tag in type_tags:
if type_tag.attrib.get("{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource"):
ftype = type_tag.attrib.get("{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource")
elif type_tag.text:
ftype = type_tag.text
if ftype and 'html' in ftype.lower():
available_types.append('html')
elif ftype and 'rdf' in ftype.lower():
available_types.append('rdf')
if not 'html' in available_types:
#Add html file - vocabfile_html
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":html_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
#ET.TreeBuilder.data('text/html')
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'HTML'
#ET.TreeBuilder.data('HTML')
if not 'rdf' in available_types:
#Add rdf file - vocabfile
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":rdf_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
#ET.TreeBuilder.data('application/rdf+xml')
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'RDF'
#ET.TreeBuilder.data('RDF')
else:
vocab = ET.SubElement(tree_root, "{http://www.w3.org/2002/07/owl#}Ontology", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":ns_uri})
se0 = ET.SubElement(vocab, "{http://purl.org/dc/elements/1.1/}identifier")
se0.text = rdf_uri
se1 = ET.SubElement(vocab, "{http://purl.org/dc/terms/}isVersionOf", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource":ns_uri})
se2a = ET.SubElement(vocab, "{http://purl.org/vocab/vann/}preferredNamespacePrefix")
se2a.text = vocab_properties['preferredNamespacePrefix']
se2b = ET.SubElement(vocab, "{http://purl.org/vocab/vann/}preferredNamespaceUri")
se2b.text = vocab_properties['preferredNamespaceUri']
#Add html uri - html_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":html_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'text/html'
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'HTML'
#Add rdf uri - rdf_vocab_properties['uri']
se3a = ET.SubElement(vocab, "{http://purl.org/dc/terms/}hasFormat")
se3b = ET.SubElement(se3a, "{http://purl.org/dc/dcmitype/}Text", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}about":rdf_uri})
se3c = ET.SubElement(se3b, "{http://purl.org/dc/elements/1.1/}format")
se3d = ET.SubElement(se3c, "{http://purl.org/dc/terms/}IMT")
se3e = ET.SubElement(se3d, "{http://www.w3.org/1999/02/22-rdf-syntax-ns#}value")
se3e.text = 'application/rdf+xml'
se3f = ET.SubElement(se3d, "{http://www.w3.org/2000/01/rdf-schema#}label", attrib={"{http://www.w3.org/XML/1998/namespace}lang":"en"})
se3f.text = 'RDF'
terms = tree_root.getiterator()
#terms = vocab.getiterator()
for term in terms:
if term.tag.lower().strip() in def_tags:
defby = None
defby = term.find("{http://www.w3.org/2000/01/rdf-schema#}isDefinedBy")
if not defby:
se4 = ET.SubElement(term, "{http://www.w3.org/2000/01/rdf-schema#}isDefinedBy", attrib={"{http://www.w3.org/1999/02/22-rdf-syntax-ns#}resource":ns_uri})
#Move ontology to the first element
tree_root.remove(vocab)
tree_root.insert(0, vocab)
tree.write(newrdf_vocab_properties['path'])
#tree_root.write(newrdf_vocab_properties['path'])
return (newrdf_vocab_properties, html_vocab_properties)
def update_rdf_for_conversion(prefix, vocab_properties, rdf_vocab_properties):
#(id, base, prefix) = get_vocab_base(vocabfile)
html_vocab_properties = {}
html_vocab_properties['format'] = 'text/html'
html_vocab_properties['name'] = "%s.html"%os.path.splitext(rdf_vocab_properties['name'])[0]
html_vocab_properties['path'] = rdf_vocab_properties['path'].replace(rdf_vocab_properties['name'], html_vocab_properties['name'])
html_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(rdf_vocab_properties['name'], html_vocab_properties['name'])
newrdf_vocab_properties = {}
newrdf_vocab_properties['format'] = 'application/rdf+xml'
newrdf_vocab_properties['name'] = "%s_modified.rdf"%os.path.splitext(rdf_vocab_properties['name'])[0]
newrdf_vocab_properties['path'] = rdf_vocab_properties['path'].replace(rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
newrdf_vocab_properties['uri'] = rdf_vocab_properties['uri'].replace(rdf_vocab_properties['name'], newrdf_vocab_properties['name'])
graph = Graph()
graph.parse(rdf_vocab_properties['path'])
subject = None
for s in graph.subjects(namespaces['rdf']['type'], URIRef(namespaces['owl']['Ontology'])):
subject = s
#graph2 = Graph()
graph_ns = []
for nsprefix, nsurl in graph.namespaces():
graph_ns.append(str(nsurl))
for prefix, url in namespaces.iteritems():
if not str(url) in graph_ns:
graph.bind(prefix, URIRef(url))
#properties = get_vocab_properties(prefix)
#subject = None
#for s in graph.subjects(namespaces['dc']['title'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dcterms']['title'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dc']['creator'], None):
# subject = s
#if not subject:
# for s in graph.subjects(namespaces['dcterms']['creator'], None):
# subject = s
formatNode1 = BNode()
formatNode2 = BNode()
#Add vocabulary properties identifier and format
graph.add((subject, namespaces['dc']['identifier'], URIRef(rdf_vocab_properties['uri'])))
graph.add((subject, namespaces['dcterms']['isVersionOf'], URIRef(vocab_properties['preferredNamespaceUri'])))
graph.add((subject, namespaces['dcterms']['hasFormat'], URIRef(rdf_vocab_properties['uri'])))
graph.add((subject, namespaces['dcterms']['hasFormat'], URIRef(html_vocab_properties['uri'])))
graph.add((subject, namespaces['vann']['preferredNamespaceUri'], URIRef(vocab_properties['preferredNamespaceUri'])))
graph.add((subject, namespaces['vann']['preferredNamespacePrefix'], URIRef(vocab_properties['preferredNamespacePrefix'])))
graph.add((URIRef(html_vocab_properties['uri']), namespaces['rdf']['type'], URIRef(namespaces['dctype']['Text'])))
graph.add((URIRef(html_vocab_properties['uri']), namespaces['dc']['format'], formatNode1))
graph.add((formatNode1, namespaces['rdf']['value'], Literal('text/html')))
graph.add((formatNode1, namespaces['rdfs']['label'], Literal('HTML')))
graph.add((formatNode1, namespaces['rdf']['type'], URIRef(namespaces['dcterms']['IMT'])))
graph.add((URIRef(rdf_vocab_properties['uri']), namespaces['rdf']['type'], URIRef(namespaces['dctype']['Text'])))
graph.add((URIRef(rdf_vocab_properties['uri']), namespaces['dc']['format'], formatNode2))
graph.add((formatNode2, namespaces['rdf']['value'], Literal('application/rdf+xml')))
graph.add((formatNode2, namespaces['rdfs']['label'], Literal('RDF')))
graph.add((formatNode2, namespaces['rdf']['type'], URIRef(namespaces['dcterms']['IMT'])))
#Add rdfs:isDefinedBy for each class / property / term of the vocabulary
#Find if schema is rdfs / owl. This defines the possible types (rdf:type) for each class / property / term
#testo = vocab_type_definitions_test['rdfs']
#subjects = []
#subs = graph.subjects(namespaces['rdf']['type'], URIRef(testo))
#for s in subs:
# subjects.append(s)
#if subjects:
# objects = vocab_type_definitions_rdfs
#else:
# objects = vocab_type_definitions_owl
#For all subjects that are of the type found above, add rdfs:isDefinedBy
#for o in objects:
# subs = graph.subjects(namespaces['rdf']['type'], o)
# for s in subs:
# graph.add((s, namespaces['rdfs']['isDefinedBy'], URIRef(vocab_properties['preferredNamespaceUri'])))
list_of_terms = get_terms(rdf_vocab_properties['path'])
for s in list_of_terms:
graph.add((URIRef(s), namespaces['rdfs']['isDefinedBy'], URIRef(vocab_properties['preferredNamespaceUri'])))
rdf_str = None
rdf_str = graph.serialize(format="pretty-xml")
#f = codecs.open(newrdf_vocab_properties['path'], 'w', 'utf-8')
f = codecs.open(newrdf_vocab_properties['path'], 'w')
f.write(rdf_str)
f.close()
return (newrdf_vocab_properties, html_vocab_properties)
