How to delete the specified triples on the isql command line of virtuoso? These triples are stored in the file
remove.nt, and the number of triples to be deleted is 400000.
I know that SPARQL DELETE can be used to delete specified triples, but it requires manual copying of triples to the command line, which does not meet my requirements.
Virtuoso version: 07.20.3235
What are your requirements ?
I would presume you have graph(s) in the a Virtuoso database, from which you are seeking to delete 400K triples in the
remove.nt file ?
SPARQL DELETE would be the recommended method of deleting triples , using physical triples or triples pattern that selectivity delete a number of triples. How was the
remove.nt file generated and does it have selective patterns of triples that are to be deleted ie with a given attribute (subject, predicate, object) value or might some of the triples all be in a single graph that could be deleted with the SPARQL DROP GRAPH command ?
Thanks @hwilliams .
My requirements are as follows:
I have lots of triples in the a Virtuoso database, and want to delete 400K triples in the remove.nt file, 400K triples is a part of the database and cannot be queried through SPARQL statements.
That is to say, if I want to delete some triples in Virtuoso, I must use SPARQL or specify a small number of triples. It is not possible to delete all triples contained in the specified triples file?
There are no SPARQL specifications/methods for deleting triples based on the content of a RDF dataset file.
In Virtuoso this could possibly be achieved using
remove.nt file containing the triples to be removed into a temporary graph in Virtuoso using the RDF Bulk Loader
log_enable(3,1); as recommended when deleting large number of triples in Virtuoso.
Run the following SQL query to delete the triples directly from the
delete from rdf_quad a where exists (select 1 from rdf_quad b where a.s = b.s and a.p = b.p and a.o = b.o and b.g = iri_to_id('temp_graph_name') );
temp_graph_name is the name of the temp graph the
remove.nt triples have been loaded into.
Note, as always, before embarking on such an operation, ensure a Full Backup of the Virtuoso database is in place, should you need to revert back to the previous state.