def process_journal(records, writer, mappings):
record, fields = majority_vote(records, ('Journal',), mappings)
if record.get('issn'):
uri = URIRef('urn:issn:%s' % record['issn'])
graph_uri = URIRef('/graph/issn/%s' % record['issn'])
elif record.get('x-nlm-ta'):
uri = URIRef('/id/journal/%s' % sluggify(record['x-nlm-ta']))
graph_uri = URIRef('/graph/journal/%s' % sluggify(record['x-nlm-ta']))
elif record.get('name'):
uri = URIRef('/id/journal/%s' % sluggify(record['name']))
graph_uri = URIRef('/graph/journal/%s' % sluggify(record['name']))
else:
sys.stderr.write("Unidentifiable: %s" % record)
return
for id, _ in fields['id']:
mappings['id'][id] = uri
mappings['journal'][uri] = graph_uri.split('/', 3)[-1]
writer.send((uri, RDF.type, FABIO.Journal, graph_uri))
for key, predicate in JOURNAL_DATA_PROPERTIES:
if key in record:
writer.send((uri, predicate, Literal(record[key]), graph_uri))
if isinstance(record.get('publisher'), URIRef):
writer.send((uri, DCTERMS.publisher, record['publisher'], graph_uri))
def detect_namespace(rdf):
"""Try to automatically detect the URI namespace of the vocabulary.
Return namespace as URIRef.
"""
# pick a concept
conc = rdf.value(None, RDF.type, SKOS.Concept, any=True)
if conc is None:
logging.critical(
"Namespace auto-detection failed. "
"Set namespace using the --namespace option.")
sys.exit(1)
ln = localname(conc)
ns = URIRef(conc.replace(ln, ''))
if ns.strip() == '':
logging.critical(
"Namespace auto-detection failed. "
"Set namespace using the --namespace option.")
sys.exit(1)
logging.info(
"Namespace auto-detected to '%s' "
"- you can override this with the --namespace option.", ns)
return ns
def discussion_as_graph(self, discussion_id):
self.ensure_discussion_storage(None)
from assembl.models import Discussion
d_storage_name = self.discussion_storage_name()
d_graph_iri = URIRef(self.discussion_graph_iri())
v = get_virtuoso(self.session, d_storage_name)
discussion_uri = URIRef(
Discussion.uri_generic(discussion_id, self.local_uri()))
subjects = list(v.query(
"""SELECT DISTINCT ?s WHERE {
?s assembl:in_conversation %s }""" % (discussion_uri.n3())))
subjects.append([discussion_uri])
# print len(subjects)
cg = ConjunctiveGraph(identifier=d_graph_iri)
for (s,) in subjects:
# Absurdly slow. DISTINCT speeds up a lot, but I get numbers.
for p, o in v.query(
'SELECT ?p ?o WHERE { graph %s { %s ?p ?o }}' % (
d_graph_iri.n3(), s.n3())):
cg.add((s, p, o))
for (s, o, g) in v.query(
'''SELECT ?s ?o ?g WHERE {
GRAPH ?g {?s catalyst:expressesIdea ?o } .
?o assembl:in_conversation %s }''' % (discussion_uri.n3())):
cg.add((s, CATALYST.expressesIdea, o, g))
# TODO: Add roles
return cg
def view(rtype, rid):
if '.' in rid:
rid, suffix = rid.rsplit('.', 1)
else:
suffix = None
path = rtype + '/' + rid
uri = URIRef(app.config['RESOURCE_BASE'] + path)
#if template:
services = app.config['SERVICES']
rq = render_template("queries/%s.rq" % rtype,
prefixes=RQ_PREFIXES, this=uri.n3(), services=services)
fmt = _conneg_format(suffix)
if fmt == 'rq':
return rq, 200, {'Content-Type': 'text/plain'}
res = run_query(app.config['ENDPOINT'], rq)
#else:
# url = data_base + path + '.n3'
# res = requests.get(url)
graph = to_graph(res.content)
this = graph.resource(uri)
if fmt in ('html', 'xhtml'):
return render_template(rtype + '.html',
path=path, this=this, curies=graph.qname)
else:
headers = {'Content-Type': MIMETYPES.get(fmt) or 'text/plain'}
fmt = {'rdf': 'xml', 'ttl': 'turtle'}.get(fmt) or fmt
return graph.serialize(format=fmt), 200, headers
def get_synthesis_contributors(self, id_only=True):
# author of important extracts
from .idea_content_link import Extract
from .auth import AgentProfile
from .post import Post
from sqlalchemy.sql.functions import count
local_uri = AssemblQuadStorageManager.local_uri()
discussion_storage = \
AssemblQuadStorageManager.discussion_storage_name()
idea_uri = URIRef(self.uri(local_uri))
clause = '''select distinct ?annotation where {
%s idea:includes* ?ideaP .
?annotation assembl:resourceExpressesIdea ?ideaP }'''
extract_ids = [x for (x,) in self.db.execute(
SparqlClause(clause % (
idea_uri.n3(),),
quad_storage=discussion_storage.n3()))]
r = list(self.db.query(AgentProfile.id, count(Extract.id)).join(
Post, Post.creator_id==AgentProfile.id).join(Extract).filter(
Extract.important == True, Extract.id.in_(extract_ids)))
r.sort(key=lambda x: x[1], reverse=True)
if id_only:
return [AgentProfile.uri_generic(a) for (a, ce) in r]
else:
ids = [a for (a, ce) in r]
order = {id: order for (order, id) in enumerate(ids)}
agents = self.db.query(AgentProfile).filter(AgentProfile.id.in_(ids)).all()
agents.sort(key=lambda a: order[a.id])
return agents
def discussion_as_graph(self, discussion_id):
from assembl.models import Discussion, AgentProfile
local_uri = self.local_uri()
discussion = Discussion.get(discussion_id)
d_storage_name = self.discussion_storage_name()
d_graph_iri = URIRef(self.discussion_graph_iri())
v = get_virtuoso(self.session, d_storage_name)
discussion_uri = URIRef(
Discussion.uri_generic(discussion_id, local_uri))
subjects = [s for (s,) in v.query(
"""SELECT DISTINCT ?s WHERE {
?s assembl:in_conversation %s }""" % (discussion_uri.n3()))]
subjects.append(discussion_uri)
participant_ids = list(discussion.get_participants(True))
profiles = {URIRef(AgentProfile.uri_generic(id, local_uri))
for id in participant_ids}
subjects.extend(profiles)
# add pseudo-accounts
subjects.extend((URIRef("%sAgentAccount/%d" % (local_uri, id))
for id in participant_ids))
# print len(subjects)
cg = ConjunctiveGraph(identifier=d_graph_iri)
self.add_subject_data(v, cg, subjects)
# add relationships of non-pseudo accounts
for ((account, p, profile), g) in v.triples((None, SIOC.account_of, None)):
if profile in profiles:
cg.add((account, SIOC.account_of, profile, g))
# Tempting: simplify with this.
# cg.add((profile, FOAF.account, account, g))
for (s, o, g) in v.query(
'''SELECT ?s ?o ?g WHERE {
GRAPH ?g {?s catalyst:expressesIdea ?o } .
?o assembl:in_conversation %s }''' % (discussion_uri.n3())):
cg.add((s, CATALYST.expressesIdea, o, g))
return cg
def instance_view_jsonld(request):
from assembl.semantic.virtuoso_mapping import AssemblQuadStorageManager
from rdflib import URIRef, ConjunctiveGraph
ctx = request.context
user_id = authenticated_userid(request) or Everyone
permissions = get_permissions(
user_id, ctx.get_discussion_id())
instance = ctx._instance
if not instance.user_can(user_id, CrudPermissions.READ, permissions):
return HTTPUnauthorized()
discussion = ctx.get_instance_of_class(Discussion)
if not discussion:
raise HTTPNotFound()
aqsm = AssemblQuadStorageManager()
uri = URIRef(aqsm.local_uri() + instance.uri()[6:])
d_storage_name = aqsm.discussion_storage_name(discussion.id)
v = get_virtuoso(instance.db, d_storage_name)
cg = ConjunctiveGraph(v, d_storage_name)
result = cg.triples((uri, None, None))
#result = v.query('select ?p ?o ?g where {graph ?g {<%s> ?p ?o}}' % uri)
# Something is wrong here.
triples = '\n'.join([
'%s %s %s.' % (uri.n3(), p.n3(), o.n3())
for (s, p, o) in result
if '_with_no_name_entry' not in o])
return aqsm.quads_to_jsonld(triples)
def post(self):
chan = URIRef(self.get_argument('chan'))
sub = self.settings.currentSub()
chanKey = Literal(chan.rsplit('/', 1)[1])
old = sub.get_levels().get(chanKey, 0)
sub.editLevel(chan, 0 if old else 1)
def test_creation_with_unknown_ns(self):
uri = 'http://localhost:8000/resource/aggregation/ton-smits-huis/454'
predicate = RDFPredicate(uri)
graph = Graph()
graph.add((URIRef(uri), FOAF.name, Literal("sjoerd")))
subject = list(graph.subjects())[0]
uri_ref = URIRef(uri)
assert uri_ref.n3() == "ns1:454"
assert predicate is not None
assert predicate.label is not None
def get_service():
g = new_graph()
me = URIRef(url_for('get_service', _external=True))
g.add((me, RDF.type, SERVICE_TYPE))
for db_resource in service_graph.subjects(RDF.type, PARTITION.Root):
db_resource = URIRef(
url_for('get_resource', rid=db_resource.replace(URI_PREFIX, ""), _external=True))
g.add((me, CONTAINMENT_LINK, db_resource))
response = make_response(g.serialize(format='turtle'))
response.headers['Content-Type'] = 'text/turtle'
return response
def enterNewAnnotationMenu(self):
'''
Interactive input for a new annotation
'''
self.printNewAnnotationMenu()
i = 1
for year in self.yearsAnnotations:
print '{}) {}'.format(i,year["year"])
i += 1
print
year = raw_input('Table to annotate: ')
cell = raw_input('Cell to annotate: ')
author = raw_input('Author: ')
corrected = raw_input('Corrected value (leave blank if none): ')
flag = raw_input('Flag: ')
graphURI = URIRef(self.yearsAnnotations[int(year)-1]["uri"])
d2sGraphURI = graphURI.replace("cedar-project.nl", "www.data2semantics.org")
annoURI = URIRef(d2sGraphURI + '/NOORDBRABANT/' + cell)
cellURI = annoURI.replace("annotations", "data")
# Create the new annotation
query = """
PREFIX oa: <http://www.w3.org/ns/openannotation/core/>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
INSERT INTO GRAPH <""" + graphURI + """>
{
<""" + annoURI + """> a oa:Annotation;
oa:annotated \"""" + str(datetime.datetime.now().strftime("%Y-%m-%d")) + """\"^^xsd:date;
oa:annotator \"""" + author + """\";
oa:generated \"""" + str(datetime.datetime.now().strftime("%Y-%m-%d")) + """\"^^xsd:date;
oa:generator <https://cedar-project.nl/tools/cedar-demo.py>;
oa:hasBody [ rdf:value \"""" + corrected + ' ' + flag + """\" ];
oa:hasTarget <""" + cellURI + """>;
oa:modelVersion <http://www.openannotation.org/spec/core/20120509.html> .
}
"""
# query = "INSERT INTO GRAPH <http://cedar-project.nl/annotations/VT_1859_01_H1> {<http://a> rdf:type <http:b>}"
print query
self.sparql.setQuery(query)
self.sparql.setReturnFormat(JSON)
self.results = self.sparql.query().convert()
def tableView_objectValueForTableColumn_row_(self, tableView, tableColumn, row):
id = tableColumn.identifier()
uri = self.resources[row]
if id=="uri":
base =self.context
base = base.split("#", 1)[0]
uri = URIRef(uri.replace(base, "", 1)) # relativize
return uri
elif id=="label":
return self.redfoot.label(uri, "")
elif id=="comment":
return self.redfoot.comment(uri, "")
else:
return ""
def check_valid_uri(self, uri):
""" checks to see if a uri is valid """
valid = False
if isinstance(uri, str):
uri_out = False
try:
uri_test = URIRef(uri)
uri_out = uri_test.n3()
except:
# some sort of error thrown, so not valid
valid = False
if isinstance(uri_out, str):
valid = True
return valid
def __init__(self, store, identifier=None, graph=None):
if graph is not None:
assert identifier is None
np = store.node_pickler
identifier = md5()
s = list(graph.triples((None, None, None)))
s.sort()
for t in s:
identifier.update("^".join((np.dumps(i) for i in t)))
identifier = URIRef("data:%s" % identifier.hexdigest())
super(GraphValue, self).__init__(store, identifier)
for t in graph:
store.add(t, context=self)
else:
super(GraphValue, self).__init__(store, identifier)
def term(self, term, use_prefix=True):
if isinstance(term, Namespace):
term = URIRef(term)
if term is None:
return RDF.nil
elif not hasattr(term, 'n3'):
return self.term(Literal(term))
elif use_prefix and isinstance(term, URIRef):
return self.uri(term)
elif isinstance(term, Literal):
if term.datatype in (XSD.double, XSD.integer, XSD.float, XSD.boolean):
return unicode(term).lower()
elif isinstance(term, Namespace):
return unicode(term)
return term.n3()
def bulk_update(self, named_graph, graph, size, is_add=True):
"""
Bulk adds or deletes. Triples are chunked into n size groups before
sending to API. This prevents the API endpoint from timing out.
"""
context = URIRef(named_graph)
total = len(graph)
if total > 0:
for set_size, nt in self.nt_yielder(graph, size):
if is_add is True:
logger.debug("Adding {} statements to <{}>.".format(set_size, named_graph))
self.update(u'INSERT DATA { GRAPH %s { %s } }' % (context.n3(), nt))
else:
logger.debug("Removing {} statements from <{}>.".format(set_size, named_graph))
self.update(u'DELETE DATA { GRAPH %s { %s } }' % (context.n3(), nt))
return total
def __init__(self, parent, identifier=None, meta=None):
self.parent = parent
if identifier is None:
identifier = BNode()
self.identifier = URIRef(identifier)
self.meta = MetaData(self, meta)
self.meta.generate()
def get_idea_ids_showing_post(cls, post_id):
"Given a post, give the ID of the ideas that show this message"
# This works because of a virtuoso bug...
# where DISTINCT gives IDs instead of URIs.
from .generic import Content
discussion_storage = \
AssemblQuadStorageManager.discussion_storage_name()
post_uri = URIRef(Content.uri_generic(
post_id, AssemblQuadStorageManager.local_uri()))
return [int(id) for (id,) in cls.default_db.execute(SparqlClause(
'''select distinct ?idea where {
%s sioc:reply_of* ?post .
?post assembl:postLinkedToIdea ?ideaP .
?idea idea:includes* ?ideaP }''' % (post_uri.n3(),),
quad_storage=discussion_storage.n3()))]
class DOI(DOIMetadata):
"""find and download a pdf for the doi given
>>> DOI('10.1016/0166-218X(92)00170-Q')
"""
headers = generic_headers
headers['Host'] = 'gen.lib.rus.ec'
headers['Referer'] = 'http://gen.lib.rus.ec/scimag/'
url = 'http://gen.lib.rus.ec/scimag/?s={}&journalid=&v=&i=&p=&redirect=1'
def __init__(self, doi, *args, **kwargs):
super().__init__(doi, *args, **kwargs)
self.url = URIRef(self.url.format(self.doi))
r = requests.get(self.url, headers=self.headers)
r.raise_for_status()
self.links = re.compile(r'a href="([^"]+pdf)"').findall(r.text)
link, *links = self.links
r = requests.get(link, stream=True)
self.filename = Literal('.'.join((doi.replace('/','_'), 'pdf')))
with open(self.filename, 'wb') as fd:
for chunk in r.iter_content(1024*10):
fd.write(chunk)
self.path = URIRef(urljoin('file:', pjoin(os.getcwd(), self.filename)))
self.graph.add((self.path, URIRef('http://purl.org/dc/terms/identifier'), self.doi))
self.graph.commit()
def canvas_and_images_graph(graph, canvas_uri):
canvas_uri = URIRef(canvas_uri)
canvas_graph = Graph()
canvas_graph += graph.triples((canvas_uri, None, None))
qres = graph.query("""SELECT ?image_anno ?image WHERE {
?image_anno a oa:Annotation .
?image_anno oa:hasTarget %s .
?image_anno oa:hasBody ?image .
?image a ?type .
FILTER(?type = dcmitype:Image || ?type = dms:Image || ?type = dms:ImageChoice) .
}""" % (canvas_uri.n3()), initNs=ns)
for image_anno, image in qres:
canvas_graph += graph.triples_choices(([image_anno, image], None, None))
return canvas_graph
def term(self, term, use_prefix=True):
if isinstance(term, (Namespace, ClosedNamespace)):
term = URIRef(namespace_to_uri(term))
if term is None:
return RDF.nil
elif not hasattr(term, 'n3'):
return self.term(Literal(term))
elif use_prefix and isinstance(term, URIRef):
return self.uri(term)
elif isinstance(term, Literal):
if term.datatype in (XSD.double, XSD.integer, XSD.float, XSD.boolean):
return unicode(term).lower()
elif use_prefix and term.datatype: # Abbreviate datatype if possible
datatype_term = self.uri(term.datatype)
return '"%s"^^%s' % (term, datatype_term)
elif isinstance(term, Namespace):
return unicode(term)
return term.n3()
def relativize(self, uri):
base = URIRef(self.base)
basedir = URIRef(self.base if base.endswith('/') else base.rsplit('/', 1)[0])
if base is not None:
if uri == base:
uri = URIRef('')
elif uri == basedir:
uri = URIRef('.')
elif uri.startswith(basedir + '/'):
uri = URIRef(uri.replace(basedir + '/', "", 1))
return uri
class ContentProvider(object):
#implements(IContentProvider)
def __init__(self, config):
self.config = config
self.jena = JenaHelper(config)
self.harvester = URIRef(config['harvesteruri'])
data = self.jena.graphquery("DESCRIBE %s" % self.harvester.n3())
#import pdb; pdb.set_trace()
self.originatingSource = data.value(self.harvester, ANDS.originatingSource) or Literal(config['originatingsource'])
self.groupDescription = data.value(self.harvester, ANDS.groupDescription) or Literal(config['groupdescription'])
self.item_sparql_query = resource_string(__name__, "item_sparql.sparql")
def set_logger(self, log):
"""Set the logger instance for this class
"""
self.log = log
def update(self, from_date=None):
"""Harvests new content added since from_date
returns a list of content_ids that were changed/added,
this should be called before get_contents is called
"""
query = resource_string(__name__, "items_to_harvest.sparql")
self._content = self.jena.selectquery(query % {'harvester': self.harvester})
return self._content
def count(self):
"""Returns number of content objects in the repository
returns None if number is unknown, this should not be
called before update is called
"""
return len(self._content)
def get_content_ids(self):
"""returns a list/generator of content_ids
"""
return self._content
def get_content_by_id(self, id):
"""Return content of a specific id
"""
# assume id is URIRef instance
g = Graph(identifier=URIRef(id))
#print self.item_sparql_query %{"subject":id, 'harvester': self.harvester}
data = self.jena.graphquery(self.item_sparql_query %{"subject": id,
'harvester': self.harvester}, g)
#print data
# FIXME: make tese conditional
data.add((URIRef(id), ANDS.originatingSource, self.originatingSource))
data.add((URIRef(id), ANDS.groupDescription, self.groupDescription))
return data
def fixit(current_object):
"""
Read the def data structure and replace all string URIs with URIRef entities
:param current_object: the piece of the data structure to be fixed
:return current_object: the piece repaired in place
"""
from rdflib import URIRef
if isinstance(current_object, dict):
for k in current_object.keys():
current_object[k] = fixit(current_object[k])
elif isinstance(current_object, list):
for i in range(0, len(current_object)):
current_object[i] = fixit(current_object[i])
elif isinstance(current_object, basestring):
if current_object.startswith("http://"):
current_object = URIRef(current_object)
elif current_object.startswith("xsd:"):
current_object = cast_to_rdflib(current_object)
return current_object
def __init__(self, config):
self.config = config
self.jena = JenaHelper(config)
self.harvester = URIRef(config['harvesteruri'])
data = self.jena.graphquery("DESCRIBE %s" % self.harvester.n3())
#import pdb; pdb.set_trace()
self.originatingSource = data.value(self.harvester, ANDS.originatingSource) or Literal(config['originatingsource'])
self.groupDescription = data.value(self.harvester, ANDS.groupDescription) or Literal(config['groupdescription'])
self.item_sparql_query = resource_string(__name__, "item_sparql.sparql")
def match_sparql_ask(transition, event, token, fsa):
"""
The 'sparql-ask' matcher.
With this matcher,
transition conditions are interpreted as the WHERE clause of a SPARQL Ask query,
where variable ?obs is bound to the considered obsel,
and prefix m: is bound to the source trace URI.
"""
m_ns = fsa.source.model_uri
if m_ns[-1] != '/' and m_ns[-1] != '#':
m_ns += '#'
history = token and token.get('history_events')
if history:
pred = URIRef(history[-1])
first = URIRef(history[0])
else:
pred = None
first = None
condition = transition['condition']
## this would be the correct way to do it
# initBindings = { "obs": URIRef(event), "pred": pred, "first": first }
## unfortunately, Virtuoso does not support VALUES clauses after the ASK clause,
## which is how SPARQLUpdateStore handles initBindings
## so we generate that clause in the condition instead
condition = """
BIND (%s as ?obs)
BIND (%s as ?pred)
BIND (%s as ?first)
""" % (
URIRef(event).n3(),
pred.n3() if pred else '""', # simulating NULL
first.n3() if first else '""', # simulating NULL
) + condition
## thank you for nothing Virtuoso :-(
return fsa.source_obsels_graph.query(
"ASK { %s }" % condition,
initNs={"": KTBS, "m": m_ns},
# initBindings=initBindings, # not supported by Virtuoso :-(
).askAnswer
def convert(self, name, qname, attrs):
if name[0] is None:
name = URIRef(name[1])
else:
name = URIRef("".join(name))
atts = {}
for (n, v) in attrs.items(): #attrs._attrs.iteritems(): #
if n[0] is None:
att = URIRef(n[1])
else:
att = URIRef("".join(n))
if att.startswith(XMLNS) or att[0:3].lower()=="xml":
pass
elif att in UNQUALIFIED:
#if not RDFNS[att] in atts:
atts[RDFNS[att]] = v
else:
atts[URIRef(att)] = v
return name, atts
def get_contributors(self):
# anyone who contributed to any of the idea's posts
local_uri = AssemblQuadStorageManager.local_uri()
discussion_storage = \
AssemblQuadStorageManager.discussion_storage_name()
idea_uri = URIRef(self.uri(local_uri))
clause = '''select count(distinct ?postP), count(distinct ?post), ?author where {
%s idea:includes* ?ideaP .
?postP assembl:postLinkedToIdea ?ideaP .
?post sioc:reply_of* ?postP .
?post sioc:has_creator ?author }'''
r = self.db.execute(
SparqlClause(clause % (
idea_uri.n3(),),
quad_storage=discussion_storage.n3()))
r = [(int(cpp), int(cp), 'local:AgentProfile/' + a.rsplit('/',1)[1]
) for (cpp, cp, a) in r]
r.sort(reverse=True)
return [a for (cpp, cp, a) in r]
def addDir(self, dirPath):
"""add description of this disk directory and every parent directory"""
if not dirPath.startswith(self.topDir):
raise ValueError("%s is not in %s" % (dirPath, self.topDir))
dirUri = uriOfFilename(self.rootUri, self.topDir, dirPath)
dirUri = URIRef(dirUri.rstrip('/') + '/')
stmts = [(dirUri, RDF.type, PHO['DiskDirectory']),
(dirUri, PHO['filename'], Literal(dirPath)),
(dirUri, PHO['basename'], Literal(os.path.basename(dirPath)))]
try:
parentUri = self.addDir(os.path.dirname(dirPath))
stmts.append((dirUri, PHO.inDirectory, parentUri))
except ValueError:
pass
self.graph.add(triples=stmts,
context=URIRef("http://photo.bigasterisk.com/scan/fs"))
return dirUri
def buildGraphFromJson(jsonData):
graph = Graph()
jsonDict = json.loads(jsonData)
for sub, predDict in jsonDict.items():
subToPut = None
if sub.startswith("http"):
subToPut = URIRef(sub.encode("utf-8"))
else:
subToPut = BNode(sub.encode("utf-8"))
for pred, objDictList in predDict.items():
for objDict in objDictList:
if objDict["type"] == "literal" or objDict["value"] == "*":
print("{} {} {} ({})".format(subToPut, pred, objDict["value"].encode("utf-8"), objDict["type"]))
graph.add((subToPut, URIRef(pred), Literal(objDict["value"].encode("utf-8"))))
elif objDict["type"] == "uri":
print("{} {} {}".format(subToPut.encode('utf-8'), pred, objDict["value"].encode('utf-8')))
graph.add((subToPut, URIRef(pred), URIRef(objDict["value"].encode("utf-8"))))
else:
print("{} {} {}".format(subToPut, pred, objDict["value"].encode("utf-8")))
graph.add((subToPut, URIRef(pred), BNode(objDict["value"].encode("utf-8"))))
return graph
def liftingtabs(mydb, cursor, cursorupdate):
graph = Graph() # graph for the dataset
docgraph1 = Graph() # graph for the documentation drawing
docgraph2 = Graph() # graph for the documentation drawing
# add namespaces
graph = apply_namespaces(graph)
docgraph1 = apply_namespaces(docgraph1)
docgraph2 = apply_namespaces(docgraph2)
# get the ones we need here
STCATH = get_namespace(graph, 'stcath')
CRM = get_namespace(graph, 'crm')
doci1 = 2 # msid with no liftingtabs
doci2 = 1307 # msid with liftingtabs, pagemarker id: 23
# deal with thesaurus concepts
graph.add((URIRef("http://w3id.org/lob/concept/2833"), RDF.type,
CRM["E55_Type"]))
graph.add((URIRef("http://w3id.org/lob/concept/2833"), RDFS.label,
Literal("board strap markers", lang="en")))
graph.add((URIRef("http://w3id.org/lob/concept/1658"), RDF.type,
CRM["E57_Material"]))
graph.add((URIRef("http://w3id.org/lob/concept/1658"), RDFS.label,
Literal("tanned skin", lang="en")))
graph.add((URIRef("http://w3id.org/lob/concept/1197"), RDF.type,
CRM["E57_Material"]))
graph.add((URIRef("http://w3id.org/lob/concept/1197"), RDFS.label,
Literal("tawed skin", lang="en")))
graph.add((URIRef("http://w3id.org/lob/concept/5429"), RDF.type,
CRM["E55_Type"]))
graph.add((URIRef("http://w3id.org/lob/concept/5429"), RDFS.label,
Literal("adhering", lang="en")))
graph.add((URIRef("http://w3id.org/lob/concept/4045"), RDF.type,
CRM["E55_Type"]))
graph.add((URIRef("http://w3id.org/lob/concept/4045"), RDFS.label,
Literal("nailing", lang="en")))
graph.add((URIRef("http://stcath.overturnin"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.overturnin"), RDFS.label,
Literal("over turn-in attaching", lang="en")))
graph.add((URIRef("http://stcath.underturnin"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.underturnin"), RDFS.label,
Literal("under turn-in attaching", lang="en")))
graph.add((URIRef("http://stcath.brokenoff"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.brokenoff"), RDFS.label,
Literal("broken off", lang="en")))
graph.add(
(URIRef("http://stcath.brokenandsewn"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.brokenandsewn"), RDFS.label,
Literal("broken and sewn", lang="en")))
graph.add((URIRef("http://stcath.missing"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.missing"), RDFS.label,
Literal("missing", lang="en")))
graph.add((URIRef("http://stcath.sound"), RDF.type, CRM["E55_Type"]))
graph.add(
(URIRef("http://stcath.sound"), RDFS.label, Literal("sound",
lang="en")))
graph.add((URIRef("http://stcath.worn"), RDF.type, CRM["E55_Type"]))
graph.add(
(URIRef("http://stcath.worn"), RDFS.label, Literal("worn", lang="en")))
graph.add((URIRef("http://stcath.detached"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.detached"), RDFS.label,
Literal("detached", lang="en")))
graph.add((URIRef("http://stcath.dangling"), RDF.type, CRM["E55_Type"]))
graph.add((URIRef("http://stcath.dangling"), RDFS.label,
Literal("dangling", lang="en")))
docgraph1.add((URIRef("http://w3id.org/lob/concept/2833"), RDF.type,
CRM["E55_Type"]))
docgraph1.add((URIRef("http://w3id.org/lob/concept/2833"), RDFS.label,
Literal("board strap markers", lang="en")))
docgraph2.add((URIRef("http://w3id.org/lob/concept/2833"), RDF.type,
CRM["E55_Type"]))
docgraph2.add((URIRef("http://w3id.org/lob/concept/2833"), RDFS.label,
Literal("board strap markers", lang="en")))
docgraph2.add((URIRef("http://w3id.org/lob/concept/1658"), RDF.type,
CRM["E55_Type"]))
docgraph2.add((URIRef("http://w3id.org/lob/concept/1658"), RDFS.label,
Literal("tanned skin", lang="en")))
docgraph2.add((URIRef("http://w3id.org/lob/concept/5429"), RDF.type,
CRM["E55_Type"]))
docgraph2.add((URIRef("http://w3id.org/lob/concept/5429"), RDFS.label,
Literal("adhering", lang="en")))
docgraph2.add(
(URIRef("http://stcath.underturnin"), RDF.type, CRM["E55_Type"]))
docgraph2.add((URIRef("http://stcath.underturnin"), RDFS.label,
Literal("under turn-in attaching", lang="en")))
docgraph2.add(
(URIRef("http://stcath.brokenoff"), RDF.type, CRM["E55_Type"]))
docgraph2.add((URIRef("http://stcath.brokenoff"), RDFS.label,
Literal("broken off", lang="en")))
# 1_3_LiftingTabs
cursor.execute(
"SELECT mss.msuuid, mss.cataloguename, lt.msid, lt.yesnonk FROM MSs mss INNER JOIN `1_3_LiftingTabs` lt on mss.id=lt.msid"
)
rows = cursor.fetchall()
for row in rows:
msuuid = URIRef(row["msuuid"], str(STCATH))
if row["yesnonk"] == "no":
graph.add((msuuid,
CRM["NTP46_is_not_composed_of_physical_thing_of_type"],
URIRef("http://w3id.org/lob/concept/2833")))
if row["msid"] == doci1:
docgraph1.add(
(msuuid,
CRM["NTP46_is_not_composed_of_physical_thing_of_type"],
URIRef("http://w3id.org/lob/concept/2833")))
docgraph1.add((msuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph1.add(
(msuuid, RDFS.label, Literal(row["cataloguename"], lang="en")))
if row["msid"] == doci2:
docgraph2.add((msuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph2.add(
(msuuid, RDFS.label, Literal(row["cataloguename"], lang="en")))
# LiftingTabs
cursor.execute(
"SELECT lt.id, mss.msuuid, mss.cataloguename, lt.msid, lt.partadditionuuid, lt.leftliftingtabuuid, lt.rightliftingtabuuid, lt.location, lt.material, lt.attachment, lt.turnin FROM MSs mss INNER JOIN `LiftingTabs` lt on mss.id=lt.msid"
)
rows = cursor.fetchall()
for row in rows:
shelfmark = row["cataloguename"]
msuuid = URIRef(row["msuuid"], str(STCATH))
# lifting tabs
locations = row["location"].split(",")
# TODO: deal with option "Centre, board"
for location in locations:
location = location.strip()
if location == "Left board" or location == "Both boards" or location == "Right board":
if location == "Left board" or location == "Both boards":
if row["leftliftingtabuuid"] is None:
newuuid = str(uuid.uuid4())
leftliftingtabuuid = URIRef(newuuid, str(STCATH))
# update the database
sql = "UPDATE LiftingTabs SET leftliftingtabuuid=%s WHERE id=%s"
val = (newuuid, row["id"])
cursorupdate.execute(sql, val)
mydb.commit()
else:
leftliftingtabuuid = URIRef(row["leftliftingtabuuid"],
str(STCATH))
graph.add((leftliftingtabuuid, RDF.type,
CRM["E22_Man-Made_Object"]))
graph.add((leftliftingtabuuid, CRM["P2_has_type"],
URIRef("http://w3id.org/lob/concept/2833")))
graph.add((leftliftingtabuuid, RDFS.label,
Literal("Left board marker of " + shelfmark,
lang="en")))
if row["material"] == "Tanned leather":
graph.add((leftliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1658")))
elif row["material"] == "Tawed leather":
graph.add((leftliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1197")))
if location == "Right board" or location == "Both boards":
if row["rightliftingtabuuid"] is None:
newuuid = str(uuid.uuid4())
rightliftingtabuuid = URIRef(newuuid, str(STCATH))
# update the database
sql = "UPDATE LiftingTabs SET rightliftingtabuuid=%s WHERE id=%s"
val = (newuuid, row["id"])
cursorupdate.execute(sql, val)
mydb.commit()
else:
rightliftingtabuuid = URIRef(
row["rightliftingtabuuid"], str(STCATH))
graph.add((rightliftingtabuuid, RDF.type,
CRM["E22_Man-Made_Object"]))
graph.add((rightliftingtabuuid, CRM["P2_has_type"],
URIRef("http://w3id.org/lob/concept/2833")))
graph.add((rightliftingtabuuid, RDFS.label,
Literal("Right board marker of " + shelfmark,
lang="en")))
if row["material"] == "Tanned leather":
graph.add((rightliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1658")))
elif row["material"] == "Tawed leather":
graph.add((rightliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1197")))
if row["partadditionuuid"] is None:
newuuid = str(uuid.uuid4())
partadditionuuid = URIRef(newuuid, str(STCATH))
# update the database
sql = "UPDATE LiftingTabs SET partadditionuuid=%s WHERE id=%s"
val = (newuuid, row["id"])
cursorupdate.execute(sql, val)
mydb.commit()
else:
partadditionuuid = URIRef(row["partadditionuuid"],
str(STCATH))
graph.add(
(partadditionuuid, RDF.type, CRM["E79_Part_Addition"]))
graph.add((partadditionuuid, RDFS.label,
Literal("Addition of board markers to " + shelfmark,
lang="en")))
graph.add((partadditionuuid, CRM["P110_augmented"], msuuid))
if location == "Both boards" or location == "Left board":
graph.add((partadditionuuid, CRM["P111_added"],
leftliftingtabuuid))
if location == "Both boards" or location == "Right board":
graph.add((partadditionuuid, CRM["P111_added"],
rightliftingtabuuid))
if row['attachment'] == "Glued":
graph.add(
(partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://w3id.org/lob/concept/5429")))
elif row['attachment'] == "Nailed":
graph.add(
(partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://w3id.org/lob/concept/4045")))
if row['turnin'] == "Under turn-in":
graph.add(
(partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://stcath.underturnin")))
elif row['turnin'] == "Over turn-in":
graph.add(
(partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://stcath.overturnin")))
if location == "Both boards":
# TODO: Mark the location of the lifting tabs when the board foredge place is available
#graph.add((leftliftingtabuuid, CRM["P55_has_current_location"], ...))
#graph.add((rightliftingtabuuid, CRM["P55_has_current_location"], ...))
pass
elif location == "Right board":
# TODO: Mark the location of the lifting tabs when the board foredge place is available
#graph.add((rightliftingtabuuid, CRM["P55_has_current_location"], ...))
pass
elif location == "Left board":
# TODO: Mark the location of the lifting tabs when the board foredge place is available
#graph.add((leftliftingtabuuid, CRM["P55_has_current_location"], ...))
pass
if row["msid"] == doci2:
docgraph2.add(
(leftliftingtabuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph2.add((leftliftingtabuuid, RDFS.label,
Literal("Left board marker of " + shelfmark,
lang="en")))
docgraph2.add((leftliftingtabuuid, CRM["P2_has_type"],
URIRef("http://w3id.org/lob/concept/2833")))
docgraph2.add(
(rightliftingtabuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph2.add((rightliftingtabuuid, RDFS.label,
Literal("Right board marker of " + shelfmark,
lang="en")))
docgraph2.add((rightliftingtabuuid, CRM["P2_has_type"],
URIRef("http://w3id.org/lob/concept/2833")))
docgraph2.add(
(partadditionuuid, RDF.type, CRM["E79_Part_Addition"]))
docgraph2.add((partadditionuuid, RDFS.label,
Literal("Addition of board markers to " + shelfmark,
lang="en")))
docgraph2.add(
(partadditionuuid, CRM["P111_added"], leftliftingtabuuid))
docgraph2.add(
(partadditionuuid, CRM["P111_added"], rightliftingtabuuid))
docgraph2.add((partadditionuuid, CRM["P110_augmented"], msuuid))
docgraph2.add(
(leftliftingtabuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph2.add((leftliftingtabuuid, RDFS.label,
Literal("Left board marker of " + shelfmark,
lang="en")))
docgraph2.add(
(rightliftingtabuuid, RDF.type, CRM["E22_Man-Made_Object"]))
docgraph2.add((rightliftingtabuuid, RDFS.label,
Literal("Right board marker of " + shelfmark,
lang="en")))
docgraph2.add(
(partadditionuuid, RDF.type, CRM["E79_Part_Addition"]))
docgraph2.add((partadditionuuid, RDFS.label,
Literal("Addition of board markers to " + shelfmark,
lang="en")))
docgraph2.add(
(partadditionuuid, CRM["P111_added"], leftliftingtabuuid))
docgraph2.add(
(partadditionuuid, CRM["P111_added"], rightliftingtabuuid))
docgraph2.add((partadditionuuid, CRM["P110_augmented"], msuuid))
docgraph2.add((leftliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1658")))
docgraph2.add((rightliftingtabuuid, CRM["P45_consists_of"],
URIRef("http://w3id.org/lob/concept/1658")))
docgraph2.add((partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://w3id.org/lob/concept/5429")))
docgraph2.add((partadditionuuid, CRM["P32_used_general_technique"],
URIRef("http://stcath.underturnin")))
# LiftingTabsCondition
cursor.execute(
"SELECT ltc.id, lt.msid, mss.msuuid, mss.cataloguename, lt.leftliftingtabuuid, lt.rightliftingtabuuid, ltc.condition, ltc.conditionuuid, ltc.leftboard FROM `LiftingTabsCondition` ltc LEFT JOIN `LiftingTabs` lt ON ltc.liftingtabid=lt.id INNER JOIN MSs mss ON mss.id=lt.msid"
)
rows = cursor.fetchall()
for row in rows:
shelfmark = row["cataloguename"]
msuuid = URIRef(row["msuuid"], str(STCATH))
if row['leftliftingtabuuid'] is not None:
leftliftingtabuuid = URIRef(row["leftliftingtabuuid"], str(STCATH))
# elif row['leftliftingtabuuid'] is None:
# if row["leftboard"] == 1: # we have a left lifting tab condition but no left lifting tab
# print(str(row["msid"]) + ", ")
if row['rightliftingtabuuid'] is not None:
rightliftingtabuuid = URIRef(row["rightliftingtabuuid"],
str(STCATH))
# elif row['rightliftingtabuuid'] is None:
# if row["leftboard"] == 0: # we have a right lifting tab condition but no right lifting tab
# print(str(row["msid"]) + ", ")
if row["leftboard"] == 1: # this is the left board
if row["conditionuuid"] is None:
newuuid = str(uuid.uuid4())
conditionuuid = URIRef(newuuid, str(STCATH))
# update the database
sql = "UPDATE LiftingTabsCondition SET conditionuuid=%s WHERE id=%s"
val = (newuuid, row["id"])
cursorupdate.execute(sql, val)
mydb.commit()
else:
conditionuuid = URIRef(row["conditionuuid"], str(STCATH))
graph.add((conditionuuid, RDF.type, CRM["E3_Condition_State"]))
graph.add(
(conditionuuid, RDFS.label,
Literal("Condition of left board marker of " + shelfmark,
lang="en")))
graph.add(
(leftliftingtabuuid, CRM["P44_has_condition"], conditionuuid))
if row["msid"] == doci2:
docgraph2.add(
(conditionuuid, RDF.type, CRM["E3_Condition_State"]))
docgraph2.add(
(conditionuuid, RDFS.label,
Literal("Condition of left board marker of " + shelfmark,
lang="en")))
docgraph2.add((leftliftingtabuuid, CRM["P44_has_condition"],
conditionuuid))
docgraph2.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.brokenoff")))
elif row["leftboard"] == 0: # this is the right board
if row["conditionuuid"] is None:
newuuid = str(uuid.uuid4())
conditionuuid = URIRef(newuuid, str(STCATH))
# update the database
sql = "UPDATE LiftingTabsCondition SET conditionuuid=%s WHERE id=%s"
val = (newuuid, row["id"])
cursorupdate.execute(sql, val)
mydb.commit()
else:
conditionuuid = URIRef(row["conditionuuid"], str(STCATH))
graph.add((conditionuuid, RDF.type, CRM["E3_Condition_State"]))
graph.add(
(conditionuuid, RDFS.label,
Literal("Condition of right board marker of " + shelfmark,
lang="en")))
graph.add(
(rightliftingtabuuid, CRM["P44_has_condition"], conditionuuid))
if row["msid"] == doci2:
docgraph2.add(
(conditionuuid, RDF.type, CRM["E3_Condition_State"]))
docgraph2.add(
(conditionuuid, RDFS.label,
Literal("Condition of right board marker of " + shelfmark,
lang="en")))
docgraph2.add((rightliftingtabuuid, CRM["P44_has_condition"],
conditionuuid))
docgraph2.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.brokenoff")))
if row["condition"] == "Broken off":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.brokenoff")))
elif row["condition"] == "Broken and Sewn":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.brokenandsewn")))
elif row["condition"] == "Missing":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.missing")))
elif row["condition"] == "Sound":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.sound")))
elif row["condition"] == "Worn":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.worn")))
elif row["condition"] == "Detached":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.detached")))
elif row["condition"] == "Dangling":
graph.add((conditionuuid, CRM["P2_has_type"],
URIRef("http://stcath.dangling")))
# documentation drawing
dot1 = visualise_graph(docgraph1, 'MS without board strap markers',
"forth")
dot2 = visualise_graph(docgraph2, 'MS with board strap markers', "forth")
dot1.render('liftingtabs/liftingtabs-1.gv', format='svg')
dot2.render('liftingtabs/liftingtabs-2.gv', format='svg')
# serialise the graph
graph.serialize(destination='liftingtabs/liftingtabs.ttl',
format='turtle',
encoding="utf-8")
docgraph1.serialize(destination='liftingtabs/liftingtabs-doc-1.n3',
format='n3',
encoding="utf-8")
docgraph2.serialize(destination='liftingtabs/liftingtabs-doc-2.n3',
format='n3',
encoding="utf-8")
def extract_metadata(item):
g = item.get_graph()
m = defaultdict(list)
#for s,p,o in g.triples((URIRef(item.uri), ag.NAMESPACES[dc]['identifier'], None)):
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dc']['title']):
m['title'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dc']['identifier']):
m['identifier'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dc']['description']):
m['description'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dc']['creator']):
m['creator'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dc']['subject']):
m['subject'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['abstract']):
m['abstract'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['created']):
try:
dt = formatDate(str(o))
except:
dt = o
m['created'].append(dt)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['description']):
m['description'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['hasVersion']):
m['hasVersion'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['identifier']):
m['identifier'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['isVersionOf']):
m['isVersionOf'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['license']):
m['license'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['mediator']):
m['mediator'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['modified']):
try:
dt = formatDate(str(o))
except:
dt = o
m['modified'].append(dt)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['publisher']):
m['publisher'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['rights']):
m['rights'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['subject']):
m['subject'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['dcterms']['title']):
m['title'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['oxds']['isEmbargoed']):
m['isEmbargoed'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['oxds']['embargoedUntil']):
try:
dt = formatDate(str(o))
except:
dt = o
m['embargoedUntil'].append(dt)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['oxds']['currentVersion']):
m['currentVersion'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['bibo']['doi']):
m['doi'].append(o)
for o in g.objects(URIRef(item.uri), ag.NAMESPACES['ore']['aggregates']):
m['aggregates'].append(o)
return dict(m)
from rdflib import URIRef
import io
from sdoterm import *
from localmarkdown import Markdown
VOCABURI="http://schema.org/"
CORE = "core"
DEFTRIPLESFILESGLOB = ["data/*.ttl","data/ext/*/*.ttl"]
LOADEDDEFAULT=False
TERMS={}
EXPANDEDTERMS={}
TERMSLOCK = threading.Lock()
RDFLIBLOCK = threading.Lock()
DATATYPEURI = URIRef(VOCABURI+"DataType")
ENUMERATIONURI = URIRef(VOCABURI+"Enumeration")
THINGURI = URIRef(VOCABURI+"Thing")
class SdoTermSource():
TYPE = "Class"
PROPERTY = "Property"
DATATYPE = "Datatype"
ENUMERATION = "Enumeration"
ENUMERATIONVALUE = "Enumerationvalue"
REFERENCE = "Reference"
TERMCOUNTS=None
SOURCEGRAPH=None
MARKDOWNPROCESS=True
def entity_uri(value):
return URIRef(ui_url('entities', value))
def collection_uri(value):
return URIRef(ui_url('collections', value))
s = row['s'].strip().replace(' ', '_')
p = row['p']
o = row['o'].strip().replace(' ', '_')
triples += [(s, p, o)]
# RDF generation
cso_namespace = 'https://cso.kmi.open.ac.uk/topics/'
swkg_namespace = 'https://swkg.kmi.open.ac.uk/entity/'
relation_namespace = 'https://swkg.kmi.open.ac.uk/relation/'
g = Graph()
for (s, p, o) in triples:
s_URI = ''
if s in csoTopics:
s_URI = URIRef(cso_namespace + s)
else:
s_URI = URIRef(swkg_namespace + s)
o_URI = ''
if s in csoTopics:
o_URI = URIRef(cso_namespace + o)
else:
o_URI = URIRef(swkg_namespace + o)
p_URI = URIRef(relation_namespace + p)
g.add((s_URI, p_URI, o_URI))
print(g.serialize("SemWebKG.rdf", format="xml"))
print(g.serialize("SemWebKG.nt", format="nt"))
def country_uri(value):
return URIRef('iso-3166-1:%s' % value)
def BNode():
return URIRef(uuid4().urn)
def eventos(uri, nombre, fecha, descripcion, lugar, media, image, uriTime,
mapa, duracion, linkURI):
if media != "No disponible":
g.add((URIRef(uri), EVENT.illustrate, URIRef(media)))
g.add((URIRef(uri), RDF.type, EVENT.Event))
g.add((URIRef(uri), FOAF.depiction, URIRef(image)))
g.add((URIRef(uri), RDFS.label, Literal(nombre, lang='es')))
g.add((URIRef(uri), RDFS.comment, Literal(descripcion, lang='es')))
#Fecha y duracion
g.add((URIRef(uri), EVENT.time, URIRef(uriTime)))
g.add((URIRef(uriTime), RDF.type, TIME.Interval))
g.add((URIRef(uriTime), TIME.at, Literal(fecha, datatype=XSD.dateTime)))
g.add((URIRef(uriTime), TIME.duration,
Literal(duracion, datatype=XSD.duration)))
#Lugar
g.add((URIRef(uri), EVENT.place, URIRef(mapa)))
g.add((URIRef(mapa), RDFS.label, Literal(lugar, lang='es')))
g.add((URIRef(uri), UMBEL.isAbout, URIRef(rutaAnillo['Eventos.rdf'])))
g.add((URIRef(uri), RDFS.seeAlso, URIRef(linkURI)))
g.add((URIRef(uri), VCARD.category, Literal("EVENTOS", lang='es')))
def main():
parser = argparse.ArgumentParser(description='Feature extraction')
parser.add_argument('-im',
'--im',
metavar='image',
nargs='+',
dest='im',
type=str,
required=True,
help='Images to calculate features on')
parser.add_argument('-md',
'--md',
metavar='metadata',
dest='md',
type=str,
required=False,
nargs='+',
help='Clinical data on patient (DICOM)')
parser.add_argument('-sem',
'--sem',
metavar='semantics',
dest='sem',
type=str,
required=False,
nargs='+',
help='Semantic Features')
parser.add_argument('-seg',
'--seg',
metavar='segmentation',
dest='seg',
type=str,
required=True,
nargs='+',
help='Segmentation to calculate features on')
parser.add_argument('-para',
'--para',
metavar='Parameters',
nargs='+',
dest='para',
type=str,
required=False,
help='Parameters')
parser.add_argument('-out',
'--out',
metavar='Features',
dest='out',
type=str,
required=False,
help='Patient features output (HDF)')
args = parser.parse_args()
if type(args.im) is list:
args.im = ''.join(args.im)
if type(args.seg) is list:
args.seg = ''.join(args.seg)
if type(args.out) is list:
args.out = ''.join(args.out)
featureVector = CalcFeatures(image=args.im, mask=args.seg)
if 'rdf' in args.out:
# Write output to rdf
# import rdflib and some namespace
from rdflib import Graph, URIRef, BNode, Literal, Namespace
from rdflib.namespace import RDF, FOAF
# convert python object to RDF
print "-----------------------------------------------------------"
print " RDF Output:"
print ""
Img = Graph()
lung1_image = URIRef("http://example.org/CT-Image")
Img.add((lung1_image, RDF.type, FOAF.Image))
list_key = featureVector.keys()
list_value = featureVector.values()
for i in range(len(list_key)):
tmp_value = Literal(list_value[i])
tmp_name = list_key[i]
Img.add((lung1_image, FOAF.tmp_name, tmp_value))
print Img.serialize(format='turtle')
# Create a rdf file for storing output
Img.serialize(args.out, format="pretty-xml")
elif 'hdf5' in args.out:
# Write output to hdf5
import numpy as np
import pandas as pd
# Assign features to corresponding groups
shape_labels = list()
shape_features = list()
histogram_labels = list()
histogram_features = list()
GLCM_labels = list()
GLCM_features = list()
GLRLM_labels = list()
GLRLM_features = list()
GLSZM_labels = list()
GLSZM_features = list()
for featureName in featureVector.keys():
if 'shape' in featureName:
shape_labels.append(featureName)
shape_features.append(featureVector[featureName])
if 'firstorder' in featureName:
histogram_labels.append(featureName)
histogram_features.append(featureVector[featureName])
if 'glcm' in featureName:
GLCM_labels.append(featureName)
GLCM_features.append(featureVector[featureName])
if 'glrlm' in featureName:
GLRLM_labels.append(featureName)
GLRLM_features.append(featureVector[featureName])
if 'glszm' in featureName:
GLSZM_labels.append(featureName)
GLSZM_features.append(featureVector[featureName])
# Convert feature to single dictionary containing PD series
features = dict()
pandas_dict = dict(zip(shape_labels, shape_features))
shape_dict = dict()
shape_dict['all'] = pd.Series(pandas_dict)
shape_features = pd.Series(shape_dict)
features['shape_features'] = shape_features
pandas_dict = dict(zip(histogram_labels, histogram_features))
histogram_dict = dict()
histogram_dict['all'] = pd.Series(pandas_dict)
histogram_features = pd.Series(histogram_dict)
features['histogram_features'] = histogram_features
GLCM_dict = dict(zip(GLCM_labels, GLCM_features))
GLRLM_dict = dict(zip(GLRLM_labels, GLRLM_features))
GLSZM_dict = dict(zip(GLSZM_labels, GLSZM_features))
texture_features = dict()
texture_features['GLCM'] = pd.Series(GLCM_dict)
texture_features['GLRLM'] = pd.Series(GLRLM_dict)
texture_features['GLSZM'] = pd.Series(GLSZM_dict)
texture_features = pd.Series(texture_features)
features['texture_features'] = texture_features
# We also return just the arrray
image_feature_array = list()
for _, feattype in features.iteritems():
for _, imfeatures in feattype.iteritems():
image_feature_array.extend(imfeatures.values)
image_feature_array = np.asarray(image_feature_array)
image_feature_array = image_feature_array.ravel()
panda_labels = ['image_features', 'image_features_array']
panda_data = pd.Series([features, image_feature_array],
index=panda_labels,
name='Image features')
panda_data.to_hdf(args.out, 'image_features')
g.bind("dcterms", DCTERMS)
g.bind("owl", OWL)
g.bind("pro", pro)
g.bind("proles", proles)
g.bind("prov", prov)
g.bind("ti", ti)
#############################
# #
# Persons #
# #
#############################
for person in root.findall('.//tei:person', tei):
person_id = person.get('{http://www.w3.org/XML/1998/namespace}id')
person_uri = URIRef(base_uri + '/person/' + person_id)
person_ref = '#' + person_id
# person
g.add((person_uri, RDF.type, crm.E21_Person))
# same as
same_as = person.get('sameAs')
if same_as is not None:
same_as = same_as.split()
i = 0
while i < len(same_as):
same_as_uri = URIRef(same_as[i])
g.add((person_uri, OWL.sameAs, same_as_uri))
i += 1
def test_read(self):
sx = None
for s in self.g.subjects(predicate=URIRef("https://example.org/e"), object=URIRef("https://example.org/f")):
sx = s
assert sx == URIRef("https://example.org/d")
def __init__(self, uri: URIRef, namespaces: Union[dict, NamespaceManager,
None]):
self.uri = URIRef(uri)
self.namespaces = namespaces
# -*- coding: utf-8 -*-
import json
from pyexcel_xlsx import get_data
from rdflib import Graph, URIRef, Literal, RDF, RDFS, XSD, Namespace
ns = Namespace("sw-kreusch")
g = Graph()
typePLZ = URIRef("http://dbpedia.org/ontology/zipCode")
typeCounty = URIRef("http://dbpedia.org/ontology/county")
typeiPhoneModel = URIRef("https://www.wikidata.org/wiki/Q2766")
hasVEK = URIRef("hasVEK")
isInCounty = URIRef("isInisInCounty")
inceptedOn = URIRef("inceptedOn")
g.add((hasVEK, RDF.type, RDF.Property))
g.add((hasVEK, RDFS.range, XSD.nonNegativeInteger))
g.add((inceptedOn, RDF.type, RDF.Property))
g.add((inceptedOn, RDFS.range, XSD.dateTime))
#Parsing the zipCode to county mapping
data = get_data("AuszugGV1QAktuell.xlsx")
counties = data["Onlineprodukt_Gemeinden_310317"]
for community in counties:
if len(community) > 13:
#Testing if the current row contains a string or not
try:
regCode = int(str(community[2]))
except UnicodeEncodeError:
continue
except ValueError:
continue
#concatenating the regCode/county code
def parse_template_instance(parent_template_id: int,
instance: Group,
nsm: NamespaceManager = None) -> AbstractTemplate:
"""Parse a stOTTR template instance.
Arguments:
* parent_template_id: ID of the template that owns the scope of the instance.
* instance: instance to parse
* nsm: Namespace manager used to expand prefixed URIs.
Returns:
An OTTR template.
"""
cross_variable = None
# if the instance uses a cross expansion
if instance.type == "cross":
# first, find the repeated variable
classic_args = list()
for pos in range(len(instance.content.arguments)):
arg = instance.content.arguments[pos]
if type(arg) is not str:
cross_variable = parse_term(arg[0], nsm=nsm)
classic_args.append(arg[0])
else:
classic_args.append(arg)
# if not found, raise error
# TODO improve error reporting
if cross_variable is None:
raise SyntaxError(
"Found an expansion mode 'cross' without a repeated variable.")
# then, replace the current instance by the inner instance
instance = instance.content
instance.arguments = classic_args
# TODO supports other expansion modes
# parse arguments to RDF Terms
ottr_arguments = list()
for arg in instance.arguments:
arg = parse_term(arg, nsm=nsm)
# unify variables found during the process, so they are unique to the local scope
if type(arg) is Variable:
new_arg = unify_var(arg, parent_template_id)
# replace the cross variable when it gets renammed
if arg == cross_variable:
cross_variable = new_arg
arg = new_arg
ottr_arguments.append(arg)
# build the concrete template object
template_name = parse_term(instance.name, nsm=nsm)
ottr_instance = None
# case 1: handle base templates (using a generic method)
if template_name in BASE_TEMPLATES:
TemplateConstructor, nb_arguments = BASE_TEMPLATES[template_name]
if len(ottr_arguments) != nb_arguments:
raise Exception(
f"The {template_name.n3()} template takes exactly {nb_arguments} arguments, but {len(ottr_arguments)} were provided"
)
params = parse_instance_arguments(parent_template_id,
ottr_arguments,
nsm=nsm)
ottr_instance = TemplateConstructor(*params)
else:
# case 2: a non-base template instance
ottr_instance = NonBaseInstance(
template_name,
parse_instance_arguments(parent_template_id,
ottr_arguments,
nsm=nsm))
# use a cross expansion operator if needed
if cross_variable is not None:
cross_name = URIRef(f"http://pyOTTR?cross={template_name}")
return CrossTemplate(cross_name, ottr_instance, cross_variable)
return ottr_instance
def test_mulpath_n3():
uri = "http://example.com/foo"
n3 = (URIRef(uri) * ZeroOrMore).n3()
assert n3 == "<" + uri + ">*"
def email_uri(value):
return URIRef('mailto:%s' % value)
def factual_data_graph(results):
for row in results:
if row != []:
for el in row['categoria']: #ogni el è un dizionario
factual_data.add((URIRef(categ + el['uri']), RDF.type, ecrm.E1_CRM_Entity))
factual_data.add((URIRef(categ + el['uri']), rdfs.label, Literal(el['label'], datatype=XSD.string)))
if len(el) > 2: #quindi non contiene la chiave 'superclasse', ma solo uri e label
factual_data.add((URIRef(categ + el['superclasse']), RDF.type, ecrm.E1_CRM_Entity))
factual_data.add((URIRef(categ + el['uri']), ecrm.P67_refers_to, URIRef(categ + el['superclasse'])))
if row['riscritture_letterarie'] != {}:
for el2 in row['riscritture_letterarie']: # el2 = dizionario
try:
factual_data.add((URIRef(work + el2['work_uri']), RDF.type, efrbroo.F1_Work))
factual_data.add((URIRef(work + el2['work_uri']), rdfs.label, Literal(el2['work_label'], datatype=XSD.string)))
#manca subwork_uri e subwork_label
for autore_uri in el2['autore_uri']:
factual_data.add((URIRef(work + autore_uri), RDF.type, efrbroo.F10_Person))
factual_data.add((URIRef(work + el2['work_uri']), rdfs.label, (URIRef(work + autore_uri))))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), RDF.type, efrbroo.F27_Work_Conception))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), ecrm.P14_carried_out_by, (URIRef(work + autore_uri))))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), efrbroo.R16_initiated, (URIRef(work + el2['work_uri']))))
# manca il link tra work uri e work label
# manca la data !!!!
except:
print('Mi sa che manca qualcosa in riscritture letterarie')
if row['fonti_medievali_e_moderne'] != {}:
for el2 in row['fonti_medievali_e_moderne']: # el2 = dizionario
try:
factual_data.add((URIRef(work + el2['work_uri']), RDF.type, efrbroo.F1_Work))
factual_data.add((URIRef(work + el2['work_uri']), rdfs.label,
Literal(el2['work_label'], datatype=XSD.string)))
# manca subwork_uri e subwork_label
# manca la data !!!!
for autore_uri in el2['autore_uri']:
factual_data.add((URIRef(work + autore_uri), RDF.type, efrbroo.F10_Person))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), RDF.type, efrbroo.F27_Work_Conception))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), ecrm.P14_carried_out_by,
(URIRef(work + autore_uri))))
factual_data.add((URIRef(work + el2['work_uri'] + '-wc'), efrbroo.R16_initiated,
(URIRef(work + el2['work_uri']))))
# manca il link tra autore uri e autore label
# manca la data !!!!
except:
print('Mi sa che manca qualcosa in fonti medievali e moderne')
if row['riscritture_cinematografiche'] != {}:
for el3 in row['riscritture_cinematografiche']:
try:
factual_data.add((URIRef(work + el3['titolo_film_uri']), RDF.type, efrbroo.F1_Work))
factual_data.add((URIRef(work + el3['titolo_film_uri']), rdfs.label, Literal(el3['titolo_film'], datatype=XSD.string)))
# manca la data !!!!
# manca autore_uri + autore label
# manca work conception
except:
print('Alc nol va in CINEMA')
print(factual_data.serialize(format='trig').decode('UTF-8'))
factual_data.serialize(destination='discarica/rdf_liv0_works.trig', format='trig')
return factual_data
def phone_uri(value):
return URIRef('tel:%s' % value)
def createHotelId(hotelId):
hotelId = URIRef('http://data.linkedevents.org/london/hotels/' + hotelId)
return hotelId
def document_uri(value):
return URIRef(ui_url('documents', value))
def replaced_by(uri, title):
new_uri = lookup(title)
if(new_uri != ''):
yse_skos.remove((None, None, URIRef(uri)))
yse_skos.add((URIRef(uri), dct.isReplacedBy, URIRef(new_uri)))
yse_skos.add((URIRef(uri), OWL.deprecated, Literal('true',datatype=XSD.boolean)))
def term(self, name):
from FuXi.Syntax.infixOWL import Class
return Class(URIRef(self + name))
g.parse('NewBase.ttl', format='n3')
with open('StudentRecord.csv', "r") as student_file:
student = csv.reader(student_file, delimiter=',')
counter = 0
for row in student:
if counter == 0:
counter += 1
else:
student_id = row[0]
student_firstname = str(row[1]).strip()
student_lastname = str(row[2]).strip()
student_email = str(row[3])
program = str(row[4])
student_enrolled_subject = row[5]
student_enrolled_snumber = row[6]
course_name = URIRef("http://focu.io/data#" +
str(row[7]).replace(" ", "%"))
student_grade = row[8]
student_enrolled_term = row[9]
b_node = BNode()
print("http://focu.io/data#" +
quote(str(student_lastname + student_firstname)))
student_node = URIRef(
"http://focu.io/data#" +
quote(str(student_lastname + student_firstname)))
print(student_node)
g.add((student_node, rdf.type, focu.Student))
g.add((student_node, foaf.givenName, Literal(student_firstname)))
g.add((student_node, foaf.familyName, Literal(student_lastname)))
g.add((student_node, v.email, Literal(student_email)))
g.add((student_node, focu.student_id, Literal(student_id)))
g.add((b_node, foaf.name, course_name))
def delete_triples(uri):
yse_skos.remove((URIRef(uri), None, None))
yse_skos.remove((None, None, URIRef(uri)))
def check_ro(base_path: Path, nested: bool = False) -> None:
manifest_file = base_path / "metadata" / "manifest.json"
assert manifest_file.is_file(), "Can't find {}".format(manifest_file)
arcp_root = find_arcp(base_path)
base = urllib.parse.urljoin(arcp_root, "metadata/manifest.json")
g = Graph()
# Avoid resolving JSON-LD context https://w3id.org/bundle/context
# so this test works offline
context = Path(get_data("tests/bundle-context.jsonld")).as_uri()
with open(manifest_file, "r", encoding="UTF-8") as fh:
jsonld = fh.read()
# replace with file:/// URI
jsonld = jsonld.replace("https://w3id.org/bundle/context", context)
g.parse(data=jsonld, format="json-ld", publicID=base)
if os.environ.get("DEBUG"):
print("Parsed manifest:\n\n")
g.serialize(sys.stdout, format="ttl")
ro = None
for ro in g.subjects(ORE.isDescribedBy, URIRef(base)):
break
assert ro is not None, "Can't find RO with ore:isDescribedBy"
profile = None
for dc in g.objects(ro, DCTERMS.conformsTo):
profile = dc
break
assert profile is not None, "Can't find profile with dct:conformsTo"
assert profile == URIRef(provenance.CWLPROV_VERSION), (
"Unexpected cwlprov version " + profile
)
paths = []
externals = []
for aggregate in g.objects(ro, ORE.aggregates):
if not arcp.is_arcp_uri(aggregate):
externals.append(aggregate)
# Won't check external URIs existence here
# TODO: Check they are not relative!
continue
lfile = _arcp2file(base_path, aggregate)
paths.append(os.path.relpath(lfile, base_path))
assert os.path.isfile(lfile), "Can't find aggregated {}".format(lfile)
assert paths, "Didn't find any arcp aggregates"
assert externals, "Didn't find any data URIs"
for ext in ["provn", "xml", "json", "jsonld", "nt", "ttl"]:
f = "metadata/provenance/primary.cwlprov.%s" % ext
assert f in paths, "provenance file missing " + f
for f in [
"workflow/primary-job.json",
"workflow/packed.cwl",
"workflow/primary-output.json",
]:
assert f in paths, "workflow file missing " + f
# Can't test snapshot/ files directly as their name varies
# TODO: check urn:hash::sha1 thingies
# TODO: Check OA annotations
packed = urllib.parse.urljoin(arcp_root, "/workflow/packed.cwl")
primary_job = urllib.parse.urljoin(arcp_root, "/workflow/primary-job.json")
primary_prov_nt = urllib.parse.urljoin(
arcp_root, "/metadata/provenance/primary.cwlprov.nt"
)
uuid = arcp.parse_arcp(arcp_root).uuid
highlights = set(g.subjects(OA.motivatedBy, OA.highlighting))
assert highlights, "Didn't find highlights"
for h in highlights:
assert (h, OA.hasTarget, URIRef(packed)) in g
describes = set(g.subjects(OA.motivatedBy, OA.describing))
for d in describes:
assert (d, OA.hasBody, URIRef(arcp_root)) in g
assert (d, OA.hasTarget, URIRef(uuid.urn)) in g
linked = set(g.subjects(OA.motivatedBy, OA.linking))
for l in linked:
assert (l, OA.hasBody, URIRef(packed)) in g
assert (l, OA.hasBody, URIRef(primary_job)) in g
assert (l, OA.hasTarget, URIRef(uuid.urn)) in g
has_provenance = set(g.subjects(OA.hasBody, URIRef(primary_prov_nt)))
for p in has_provenance:
assert (p, OA.hasTarget, URIRef(uuid.urn)) in g
assert (p, OA.motivatedBy, PROV.has_provenance) in g
# Check all prov elements are listed
formats = set()
for prov in g.objects(p, OA.hasBody):
assert (prov, DCTERMS.conformsTo, URIRef(provenance.CWLPROV_VERSION)) in g
# NOTE: DC.format is a Namespace method and does not resolve like other terms
formats.update(set(g.objects(prov, DC["format"])))
assert formats, "Could not find media types"
expected = set(
Literal(f)
for f in (
"application/json",
"application/ld+json",
"application/n-triples",
'text/provenance-notation; charset="UTF-8"',
'text/turtle; charset="UTF-8"',
"application/xml",
)
)
assert formats == expected, "Did not match expected PROV media types"
if nested:
# Check for additional PROVs
# Let's try to find the other wf run ID
otherRuns = set()
for p in g.subjects(OA.motivatedBy, PROV.has_provenance):
if (p, OA.hasTarget, URIRef(uuid.urn)) in g:
continue
otherRuns.update(set(g.objects(p, OA.hasTarget)))
assert otherRuns, "Could not find nested workflow run prov annotations"
from unittest import TestCase
from rdflib import URIRef, RDF, Graph
from ldp import NS as LDP
from ldp.dataset import NamedContextDataset, context as dataset
from ldp.globals import continents, capitals, aggregation
CONTINENTS = URIRef('http://www.telegraphis.net/data/continents')
CAPITALS = URIRef('http://www.telegraphis.net/data/capitals')
class TestNamedContext(TestCase):
def test_single_context_parse(self):
ds = NamedContextDataset()
ds.g['cont'] = ds.parse(source='test/continents.rdf',
publicID=CONTINENTS)
self.assertEqual(len(list(ds.g['cont'][::])), 112)
def test_multiple_context_parse(self):
ds = NamedContextDataset()
ds.g['cont'] = ds.parse(source='test/continents.rdf',
publicID=CONTINENTS)
self.assertEqual(len(list(ds.g['cont'][::])), 112)
ds.g['capitals'] = ds.parse(source='test/capitals.rdf',
publicID=CONTINENTS)
self.assertEqual(len(list(ds.g['cont'][::])), 2584)
def test_multiple_context_aggregation(self):
from rdflib.namespace import RDF, RDFS, FOAF, OWL, XSD, DC, DCTERMS
import json
import os
path = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
data = json.load(open(path + "/DatiPerElaborazione/Pub.geojson", "r"))
g = Graph()
cmo = Namespace("http://www.comune.milano.it/ontology/")
schema = Namespace("https://schema.org/")
g.bind("cmo", cmo)
g.bind("schema", schema)
for element in data:
uri = element["URI"]
g.add([URIRef(uri), RDF.type, cmo.Pub])
g.add([URIRef(uri), RDFS.label, Literal(element["nome"])])
g.add([URIRef(uri), cmo.localBusinessWebsite, Literal(element["website"])])
g.add([URIRef(uri), schema.email, Literal(element["email"])])
g.add([URIRef(uri), cmo.localBusinessPostalCode, Literal(element["cap"])])
g.add([URIRef(uri), schema.address, Literal(element["indirizzo"])])
g.add([
URIRef(uri), cmo.latitude,
Literal(element["lat"], datatype=XSD.float)
])
g.add([
URIRef(uri), cmo.longitude,
Literal(element["long"], datatype=XSD.float)
])
g.serialize(destination=path + '/Turtles/pub.ttl', format='turtle')
def read_manifest(item, manifest_file):
triples = []
namespaces = {}
seeAlsoFiles = []
oxdsClasses = ['http://vocab.ox.ac.uk/dataset/schema#Grouping', 'http://vocab.ox.ac.uk/dataset/schema#DataSet']
aggregates = item.list_rdf_objects(item.uri, "ore:aggregates")
g = ConjunctiveGraph()
gparsed = g.parse(manifest_file, format='xml')
namespaces = dict(g.namespaces())
#Get the subjects
subjects = {}
for s in gparsed.subjects():
if s in subjects:
continue
if type(s).__name__ == 'URIRef':
if str(s).startswith('file://'):
ss = str(s).replace('file://', '')
if manifest_file in ss:
subjects[s] = URIRef(item.uri)
else:
manifest_file_path, manifest_file_name = os.path.split(manifest_file)
ss = ss.replace(manifest_file_path, '').strip('/')
for file_uri in aggregates:
if ss in str(file_uri):
subjects[s] = URIRef(file_uri)
break
if not s in subjects:
subjects[s] = URIRef(item.uri)
else:
subjects[s] = URIRef(s)
elif type(s).__name__ == 'BNode':
replace_subject = True
for o in gparsed.objects():
if o == s:
replace_subject = False
if replace_subject:
subjects[s] = URIRef(item.uri)
else:
subjects[s] = s
#Get the dataset type
#set the subject uri to item uri if it is of type as defined in oxdsClasses
datasetType = False
for s,p,o in gparsed.triples((None, RDF.type, None)):
if str(o) in oxdsClasses:
if type(s).__name__ == 'URIRef' and len(s) > 0 and str(s) != str(item.uri) and str(subjects[s]) != str(item.uri):
namespaces['owl'] = URIRef("http://www.w3.org/2002/07/owl#")
triples.append((item.uri, 'owl:sameAs', s))
triples.append((item.uri, RDF.type, o))
elif type(s).__name__ == 'BNode' or len(s) == 0 or str(s) == str(item.uri) or str(subjects[s]) == str(item.uri):
gparsed.remove((s, p, o))
subjects[s] = item.uri
#Get the uri for the see also files
for s,p,o in gparsed.triples((None, URIRef('http://www.w3.org/2000/01/rdf-schema#seeAlso'), None)):
if type(o).__name__ == 'URIRef' and len(o) > 0:
obj = str(o)
if obj.startswith('file://'):
obj_path, obj_name = os.path.split(obj)
obj = obj.replace(obj_path, '').strip('/')
for file_uri in aggregates:
if obj in str(file_uri):
seeAlsoFiles.append(file_uri)
gparsed.remove((s, p, o))
#Add remaining triples
for s,p,o in gparsed.triples((None, None, None)):
triples.append((subjects[s], p, o))
return namespaces, triples, seeAlsoFiles
def browser_cerca():
"""
Permite la comunicacion con el agente via un navegador
via un formulario
"""
global product_list
if request.method == 'GET':
return render_template('cerca.html', products=None)
elif request.method == 'POST':
# Peticio de cerca
if request.form['submit'] == 'Cerca':
logger.info("Enviando peticion de busqueda")
# Content of the message
contentResult = ECSDI['Cerca_productes_' + str(get_count())]
# Graph creation
gr = Graph()
gr.add((contentResult, RDF.type, ECSDI.Cerca_productes))
# Add restriccio nom
nom = request.form['nom']
if nom:
# Subject nom
subject_nom = ECSDI['RestriccioNom' + str(get_count())]
gr.add((subject_nom, RDF.type, ECSDI.RestriccioNom))
gr.add(
(subject_nom, ECSDI.Nom, Literal(nom,
datatype=XSD.string)))
# Add restriccio to content
gr.add((contentResult, ECSDI.Restringe, URIRef(subject_nom)))
marca = request.form['marca']
if marca:
subject_marca = ECSDI['Restriccion_Marca_' + str(get_count())]
gr.add((subject_marca, RDF.type, ECSDI.Restriccion_Marca))
gr.add((subject_marca, ECSDI.Marca,
Literal(marca, datatype=XSD.string)))
gr.add((contentResult, ECSDI.Restringe, URIRef(subject_marca)))
min_price = request.form['min_price']
max_price = request.form['max_price']
if min_price or max_price:
subject_preus = ECSDI['Restriccion_Preus_' + str(get_count())]
gr.add((subject_preus, RDF.type, ECSDI.Rango_precio))
if min_price:
gr.add(
(subject_preus, ECSDI.Precio_min, Literal(min_price)))
if max_price:
gr.add(
(subject_preus, ECSDI.Precio_max, Literal(max_price)))
gr.add((contentResult, ECSDI.Restringe, URIRef(subject_preus)))
seller = get_agent_info(agn.SellerAgent, DirectoryAgent,
UserPersonalAgent, get_count())
gr2 = send_message(
build_message(gr,
perf=ACL.request,
sender=UserPersonalAgent.uri,
receiver=seller.uri,
msgcnt=get_count(),
content=contentResult), seller.address)
index = 0
subject_pos = {}
product_list = []
for s, p, o in gr2:
if s not in subject_pos:
subject_pos[s] = index
product_list.append({})
index += 1
if s in subject_pos:
subject_dict = product_list[subject_pos[s]]
if p == RDF.type:
subject_dict['url'] = s
elif p == ECSDI.Marca:
subject_dict['marca'] = o
elif p == ECSDI.Modelo:
subject_dict['modelo'] = o
elif p == ECSDI.Precio:
subject_dict['precio'] = o
elif p == ECSDI.Nombre:
subject_dict['nombre'] = o
elif p == ECSDI.Peso:
subject_dict['peso'] = o
product_list[subject_pos[s]] = subject_dict
return render_template('cerca.html', products=product_list)
# --------------------------------------------------------------------------------------------------------------
# Peticio de compra
elif request.form['submit'] == 'Comprar':
products_checked = []
for item in request.form.getlist("checkbox"):
item_checked = []
item_map = product_list[int(item)]
item_checked.append(item_map['marca'])
item_checked.append(item_map['modelo'])
item_checked.append(item_map['nombre'])
item_checked.append(item_map['precio'])
item_checked.append(item_map['url'])
item_checked.append(item_map['peso'])
products_checked.append(item_checked)
logger.info("Creando la peticion de compra")
# Content of the message
content = ECSDI['Peticion_compra_' + str(get_count())]
# Graph creation
gr = Graph()
gr.add((content, RDF.type, ECSDI.Peticion_compra))
# Asignar prioridad a la peticion (asignamos el contador de mensaje)
gr.add((content, ECSDI.Prioridad,
Literal(get_count(), datatype=XSD.integer)))
# Creacion de la ciudad (por ahora Barcelona) --------------------------------------------------------------
subject_ciudad = ECSDI['Ciudad_' +
str(random.randint(1, sys.float_info.max))]
gr.add((subject_ciudad, RDF.type, ECSDI.Ciudad))
gr.add((subject_ciudad, ECSDI.Nombre,
Literal(41.398373, datatype=XSD.float)))
gr.add((subject_ciudad, ECSDI.Latitud,
Literal(2.188247, datatype=XSD.float)))
gr.add((subject_ciudad, ECSDI.Longitud,
Literal('Barcelona', datatype=XSD.string)))
# Creacion del sobre (Compra) ------------------------------------------------------------------------------
subject_sobre = ECSDI['Compra_' +
str(random.randint(1, sys.float_info.max))]
gr.add((subject_sobre, RDF.type, ECSDI.Compra))
gr.add((subject_sobre, ECSDI.Pagat, Literal(0,
datatype=XSD.integer)))
gr.add((subject_sobre, ECSDI.Enviar_a, URIRef(subject_ciudad)))
total_price = 0.0
for item in products_checked:
total_price += float(item[3])
# Creacion del producto --------------------------------------------------------------------------------
subject_producto = item[4]
gr.add((subject_producto, RDF.type, ECSDI.Producto))
gr.add((subject_producto, ECSDI.Marca,
Literal(item[0], datatype=XSD.string)))
gr.add((subject_producto, ECSDI.Modelo,
Literal(item[1], datatype=XSD.string)))
gr.add((subject_producto, ECSDI.Nombre,
Literal(item[2], datatype=XSD.string)))
gr.add((subject_producto, ECSDI.Precio,
Literal(item[3], datatype=XSD.float)))
gr.add((subject_producto, ECSDI.Peso,
Literal(item[5], datatype=XSD.float)))
gr.add(
(subject_sobre, ECSDI.Productos, URIRef(subject_producto)))
gr.add((subject_sobre, ECSDI.Precio_total,
Literal(total_price, datatype=XSD.float)))
gr.add((content, ECSDI.Sobre, URIRef(subject_sobre)))
seller = get_agent_info(agn.SellerAgent, DirectoryAgent,
UserPersonalAgent, get_count())
answer = send_message(
build_message(gr,
perf=ACL.request,
sender=UserPersonalAgent.uri,
receiver=seller.uri,
msgcnt=get_count(),
content=content), seller.address)
products_matrix = []
for item in answer.subjects(RDF.type, ECSDI.Producto):
product = [
answer.value(subject=item, predicate=ECSDI.Marca),
answer.value(subject=item, predicate=ECSDI.Modelo),
answer.value(subject=item, predicate=ECSDI.Nombre),
answer.value(subject=item, predicate=ECSDI.Precio)
]
products_matrix.append(product)
return render_template('endSell.html', products=products_matrix)
