def search_predicates_object(self, pred, worksets):
'''
Search (predicate, object) pair in a workset
'''
sd = SparqlDao()
workset = ''
for ws in worksets:
workset += '<' + ws + '>'
if "http" in pred:
qry_string = FileOps().open('query/search_predicates_uri_ws.rq')
qry_string = qry_string.format('<'+pred+'>', workset)
else:
qry_string = FileOps().open('query/search_predicates_literal_ws.rq')
qry_string = qry_string.format('<'+pred+'>', workset)
original = sd.autocomplete_sparql(self.endpoint, qry_string)
count = 0
preds = []
for orig in original:
count += int(orig[0])
for data in original:
preds.append({ "predicate": data[1], "weight": self._calculate_weight(data[0], count), "count": int(count)})
return json.dumps(preds)
def search_predicates(self, searchterm):
'''
Find predicates associated with an object and calculates the weightings
@todo: put this into named graph with time.
@todo: filter this for workset or non-workset.
'''
original = []
sd = SparqlDao()
simil = Similarities()
if "http" in searchterm:
qry_string = FileOps().open('query/search_predicates_uri.rq')
qry_string = qry_string.format('<'+searchterm+'>')
else:
qry_string = FileOps().open('query/search_predicates_literal.rq')
qry_string = qry_string.format('"'+searchterm+'"')
original = sd.autocomplete_sparql(self.endpoint, qry_string)
count = 0
preds = []
for orig in original:
count += int(orig[0])
for data in original:
preds.append({ "predicate": data[1], "weight": self._calculate_weight(data[0], count), "count":int(count)})
return json.dumps(preds)
def join_graphs(self, graph):
merged_data = []
original = []
sd = SparqlDao()
#qry_string = FileOps().open('query/original_author.rq')
#qry_string = qry_string.format(graph)
#original = sd.run_remote_sparql(self.endpoint, qry_string)
#get the other graphs linked by VIAF
qry_string = FileOps().open('query/walk_path_query.rq')
qry_string = qry_string.format(graph)
merged_data = sd.run_remote_sparql(self.endpoint, qry_string)
link = []
unlink = []
for merge in merged_data:
for orig in original:
if orig[2] == merge[2] and orig[1] == merge[1]:
link.append(merge)
if merge[1] not in link:
unlink.append(merge)
return json.dumps({"original":original,"link":link,"difference":self.deduplicate_data(unlink,link),"similarity":self.calculate_similarity_perc(link,original)})
def search_subject(self, searchterm):
'''
Find subjects associated with a predicate
@todo: put this into named graph with time.
@todo: filter this for workset or non-workset.
'''
original = []
sd = SparqlDao()
if "http" in searchterm:
qry_string = FileOps().open('query/search_subjects_uri.rq')
qry_string = qry_string.format('<'+searchterm+'>')
else:
qry_string = FileOps().open('query/search_subjects_literal.rq')
qry_string = qry_string.format('"'+searchterm+'"')
original = sd.autocomplete_sparql(self.endpoint, qry_string)
terms = []
for data in original:
terms.append({'id': data[0], 'value': data[1]})
return json.dumps(terms)
def search_data_type(self, term, data_type):
'''
Perform substring search on data limited by a type and return JSON
'''
original = []
sd = SparqlDao()
qry_string = FileOps().open('query/search_string_type.rq')
qry_string = qry_string.format(term, 'http://eeboo.oerc.ox.ac.uk/eeboo/' + data_type)
original = sd.autocomplete_sparql(self.endpoint, qry_string)
terms = []
for data in original:
terms.append({'id': data[0], 'value': data[1]})
return json.dumps(terms)
def search_data(self, term):
'''
Perform substring search on data and return JSON
'''
original = []
sd = SparqlDao()
qry_string = FileOps().open('query/search_string.rq')
qry_string = qry_string.format(term)
original = sd.autocomplete_sparql(self.endpoint, qry_string)
terms = []
for data in original:
terms.append({'id': data[0], 'value': data[1]})
return json.dumps(terms)
def get_original(self, graphs):
'''
Gets the base graphs for the request
'''
original = []
sd = SparqlDao()
searchterm = ''
for g in graphs:
searchterm += '<' + str(g) + '>'
qry_string = FileOps().open('query/original_author.rq')
qry_string = qry_string.format(searchterm)
original = sd.run_remote_sparql(self.endpoint, qry_string)
return json.dumps(original)