def main():
logging.info(f"Request to {TWSS_RESOURCES_URI}")
twss_graph = OwlMovieRepository.read(TWSS_RESOURCES_URI)
# Change this flag to use local data
local = False
if local:
dbpedia_graph = OwlMovieRepository.read(DBPEDIA_PERSONS_FILE)
wiki_graph = OwlMovieRepository.read(WIKIDATA_PERSONS_FILE)
else:
names = get_persons_names(twss_graph)
logging.info("Querying remote persons")
# Warining making too many requests on dbpedia server may block you
dbpedia_graph = query_dbpedia_persons(names)
wiki_graph = query_wikidata_persons(names)
remote_persons_graph = dbpedia_graph + wiki_graph
merged_graph = merge_graphs(twss_graph, remote_persons_graph)
logging.info(f"Done! Writing graph in {EXTENDED_PERSONS_FILE}")
OwlMovieRepository.write(
path_file=EXTENDED_PERSONS_FILE,
graph=merged_graph,
namespaces=NAMESPACES,
)
def main():
# Init arguments parser
parser = argparse.ArgumentParser()
parser.add_argument("-v", "--verbose", action="store_true")
parser.add_argument("-o", "--offline", action="store_true")
args = parser.parse_args()
# Init logger
logging.basicConfig(
level=logging.DEBUG if args.verbose else logging.INFO,
format="%(asctime)s - %(levelname)s - %(message)s",
)
# Evaluate arguments
if not args.offline:
logging.info("Running Oscar script")
oscars.main()
logging.info("Running person enricher script")
person_enricher.main()
# Merge enrichments
logging.info("Enriching the graph")
enriched_graph = build_enrichment_graph()
OwlMovieRepository.write(path_file=ENRICHED_GRAPH_FILE,
graph=enriched_graph,
namespaces=NAMESPACES)
logging.info(f"Done! Enriched graph saved in {ENRICHED_GRAPH_FILE}")
def query_dbpedia_persons(names):
DBPEDIA_URL = "http://dbpedia.org/sparql"
logging.info(f"Request to {DBPEDIA_URL}")
persons_regex = "(" + "|".join(names) + ")"
sparql = SPARQLWrapper(DBPEDIA_URL, returnFormat=RDFXML)
sparql.setQuery(DBPEDIA_PERSONS.format(persons_regex=persons_regex))
results = sparql.queryAndConvert()
OwlMovieRepository.write(path_file=DBPEDIA_PERSONS_FILE,
graph=results,
namespaces=NAMESPACES)
return results
def query_wikidata_persons(names):
WIKIDATA_URL = "https://query.wikidata.org/sparql"
logging.info(f"Request to {WIKIDATA_URL}")
persons_regex = "(" + "|".join(names) + ")"
sparql = SPARQLWrapper(WIKIDATA_URL, returnFormat=RDFXML)
sparql.setQuery(WIKIDATA_PERSONS.format(persons_regex=persons_regex))
results = sparql.queryAndConvert()
OwlMovieRepository.write(
path_file=WIKIDATA_PERSONS_FILE,
graph=results,
namespaces=NAMESPACES,
)
return results
def main():
logging.info("Reading graphs")
twss_resources = get_twss_resources_graph()
oscar_winners_graph = get_oscar_winners_graph(twss_resources)
logging.info("Matching remote actors with locals")
merged_graph = twss_resources + oscar_winners_graph
result = merged_graph.query(COMBINE_REMOTE_AND_LOCAL_ACTORS)
logging.info(f"Done! Writing graph in {OSCAR_WINNERS_FILE}")
OwlMovieRepository.write(
path_file=OSCAR_WINNERS_FILE, graph=result.graph, namespaces=NAMESPACES
)
def get_oscar_winners_graph(twss_resources):
logging.info("Looking for actors who were directed by Oscar winners.")
if OSCAR_WINNERS_CACHE_FILE.exists():
logging.info("Reading Oscar winners from cache")
oscar_winners_graph = OwlMovieRepository.read(source=OSCAR_WINNERS_CACHE_FILE)
else:
logging.info("Reading Oscar winners from web")
oscar_winners_graph = build_oscar_winners_graph(twss_resources)
logging.info(
f"Writing the cache file of Oscar winners "
f"in {OSCAR_WINNERS_CACHE_FILE}."
)
OwlMovieRepository.write(
path_file=OSCAR_WINNERS_CACHE_FILE,
graph=oscar_winners_graph,
namespaces=NAMESPACES,
)
return oscar_winners_graph
def write_links():
twss_graph = OwlMovieRepository.read(ORIGINAL_DATASET_FILE)
links_graph = Graph()
twss_actors_uris = get_actors_uris(twss_graph)
dbpedia_actors = get_dbpedia_actors(twss_actors_uris)
for dbpedia_actor, twss_actor_uri in zip(dbpedia_actors, twss_actors_uris):
dbpedia_actor_name = to_dbpedia_actor_name(twss_actor_uri)
if len(dbpedia_actor) == 0:
logging.error(f"Not found owl:sameAs for {dbpedia_actor_name}")
else:
logging.debug(f"Found owl:sameAs for dbpedia_{dbpedia_actor_name}")
links_graph.add(
(
twss_actor_uri,
OWL.sameAs,
get_dbpedia_actor_uri(dbpedia_actor, dbpedia_actor_name),
)
)
OwlMovieRepository.write(LINKS_FILE, links_graph, namespaces=NAMESPACES)