Hi, After further testing I see that tdbstats appears to only summarise the contents of the default graph. If all of the data exists in named graphs then it generates output that indicates that the indexes are empty (i.e. zero triple counts, no predicate counts).
Is that the expected behaviour? Cheers, L. On Tue, Nov 6, 2012 at 4:21 PM, Leigh Dodds <[email protected]> wrote: > Hi, > > I'm having problems generating TDB stats for some data that I've > indexed using tdbloader2 in Apache 2.7.4. > > The data is a copy of the Ordnance Survey data available from here: > > http://archive.org/download/kasabi/ordnance-survey-linked-data.gz > > From the Apache 2.7.4 directory I run tdbloader2 asking it to index > the unzipped nquads file. > > The indexer runs fine and I can query the data in fuseki without > problem. However if I run > > bin/tdbstats --loc /path/to/indexes > > I get: > > (stats > (meta > (timestamp > "2012-11-06T16:16:05.798+00:00"^^<http://www.w3.org/2001/XMLSchema#dateTime>) > (run@ "2012/11/06 16:16:05 GMT") > (count 0)) > (other 0)) > > I also extracted the first 1m quads (head -n1000000) into a separate > file and indexed that. The result is the same. Indexes appear to be > created OK but tdbstats still produces an empty count. > > I'm using Ubuntu on a 32bit system. > > Any ideas? > > Cheers, > > L. > > -- > Leigh Dodds > Freelance Technologist > Open Data, Linked Data Geek > t: @ldodds > w: ldodds.com > e: [email protected] -- Leigh Dodds Freelance Technologist Open Data, Linked Data Geek t: @ldodds w: ldodds.com e: [email protected]
