Hi Patrick, Sorry, it was easy after all :)
now, something extra, we have about 3-4M triples that we update occasionally by clearing the gragh and inserting the new triples. 1) does the index affect the clear/update procedure? and if it does, are there any commands to call before / after to speed it up 2) (a little out of topic), but i read somewhere that by disabling the http host the update is much faster. Is there a function to call from a script? e.g. disable http (maybe disable index) clear / update (re-enable index) enable http Regards, Dimitris On Thu, Mar 24, 2011 at 12:30 PM, Patrick van Kleef <[email protected]>wrote: > Hi Dimitris, > > > i have a problem with the bif:contains function. >> It does not work and i could find if there is any configuration options to >> enable it. >> >> Server/Virtuoso: version 06.01.3127, on Linux (x86_64-pc-linux-gnu), >> Single Edition >> SPARQL Endpoint: http://el.dbpedia.org/sparql >> >> For the same query, the bif:contains function does not return any results, >> while with Filter regex does >> e.g. >> >> PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> >> A) >> select ?o ?l where { ?o rdfs:label ?l. >> ?l bif:contains "Linux". } >> B) >> select ?o ?l where { ?o rdfs:label ?l. >> FILTER(regex(?l,"Linux")). } >> >> I tested the (A) option in the dbpedia.org/sparql and it works fine >> >> Execute urls >> A) >> http://el.dbpedia.org/sparql?default-graph-uri=http%3A%2F%2Fel.dbpedia.org&should-sponge=&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E%0D%0A%0D%0Aselect+%3Fo+%3Fl+where+{%0D%0A%3Fo+rdfs%3Alabel+%3Fl.%0D%0A%3Fl+bif%3Acontains+%22Linux%22.%0D%0A}&format=text%2Fhtml&debug=on&timeout=<http://el.dbpedia.org/sparql?default-graph-uri=http%3A%2F%2Fel.dbpedia.org&should-sponge=&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E%0D%0A%0D%0Aselect+%3Fo+%3Fl+where+%7B%0D%0A%3Fo+rdfs%3Alabel+%3Fl.%0D%0A%3Fl+bif%3Acontains+%22Linux%22.%0D%0A%7D&format=text%2Fhtml&debug=on&timeout=> >> >> B) >> http://el.dbpedia.org/sparql?default-graph-uri=http%3A%2F%2Fel.dbpedia.org&should-sponge=&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E%0D%0A%0D%0Aselect+%3Fo+%3Fl%0D%0Awhere+{%0D%0A%3Fo+rdfs%3Alabel+%3Fl.%0D%0AFILTER%28regex%28%3Fl%2C%22Linux%22%29%29.%0D%0A}&format=text%2Fhtml&debug=on&timeout=<http://el.dbpedia.org/sparql?default-graph-uri=http%3A%2F%2Fel.dbpedia.org&should-sponge=&query=PREFIX+rdfs%3A+%3Chttp%3A%2F%2Fwww.w3.org%2F2000%2F01%2Frdf-schema%23%3E%0D%0A%0D%0Aselect+%3Fo+%3Fl%0D%0Awhere+%7B%0D%0A%3Fo+rdfs%3Alabel+%3Fl.%0D%0AFILTER%28regex%28%3Fl%2C%22Linux%22%29%29.%0D%0A%7D&format=text%2Fhtml&debug=on&timeout=> >> >> > It looks like you did not initialize the RDF freetext index. > > To create index on all triples just perform the following command using > isql: > DB.DBA.RDF_OBJ_FT_RULE_ADD (null, null, 'All'); > Then force a synchronization: > > DB.DBA.VT_INC_INDEX_DB_DBA_RDF_OBJ (); > > > Please read: > > http://docs.openlinksw.com/virtuoso/sparqlextensions.html > > > Best regards, > > Patrick > -- Kontokostas Dimitris
