def remove_actor_bad_uris(self):
self.__sparql = SPARQLWrapper(
'http://localhost:5820/BestLinkedMovieDatabase/update')
self.__sparql.setMethod(POST)
self.__sparql.setReturnFormat(JSON)
self.__sparql.setQuery(Queries.remove_bad_actor_dbp_refs())
self.__sparql.query()
def match_actors(self, uri_chunks: Tuple):
response = None
with open('Data/DBPediaActors.ttl', 'wb') as output:
for uri_chunk in uri_chunks:
uri_string = "(" + "".join(uri + ","
for uri in uri_chunk)[:-1] + ")"
self.__sparql.setQuery(
Queries.dbpedia_uris_to_resources(uri_string, 'Person'))
response = self.__sparql.query().convert()
output.write(response)
output.flush()
return response
def clean_actor_uris(self, uris) -> Graph or None:
cleaned_graph = Graph()
self.__sparql.setQuery(
Queries.lmdb_actors_to_dbpedia_movie_uri_pairs())
response = self.__sparql.query().convert()
pairs = list(
map(Stardog.__to_movie_clean_uri_pair,
response["results"]["bindings"]))
if len(pairs) == 0:
return None
for lmdb_resource, dbp_uri in pairs:
lmdb_resource = URIRef(lmdb_resource)
dbp_uri = URIRef(
dbp_uri.replace('"', "").replace("'", "").replace("`", ""))
cleaned_graph.add((lmdb_resource, namespace.OWL.sameAs, dbp_uri))
cleaned_graph.serialize('Data/CleanedLMDBActorSameAs.ttl',
format='turtle')
return cleaned_graph # It is now safe to import the cleaned data using stardog.
def get_actors(self) -> Tuple:
self.__sparql.setQuery(Queries.lmdb_actors_to_dbpedia_uris())
response = self.__sparql.query().convert()
uris = list(map(Stardog.__to_uri, response["results"]["bindings"]))
return tuple(tuple(uris[i:i + 100]) for i in range(0, len(uris), 100))