Hi Joachim!
This should now be fixed in jena master. You will need to reindex using
the new jena.textindexer. If you don't want to build Jena yourself, you
can wait for the daily snapshot build to pick this up.
https://repository.apache.org/content/groups/snapshots/org/apache/jena/
-Osma
On 08/02/16 10:11, Osma Suominen wrote:
Hi Joachim!
Thanks for reporting. I've confirmed this and opened an issue in JIRA:
https://issues.apache.org/jira/browse/JENA-1133
-Osma
05.02.2016, 17:12, Neubert, Joachim kirjoitti:
I have a graphField defined for a Fuseki 2.3.0 text index EntityMap.
When I now load data via tdbloader and text.indexer
java -cp $FUSEKI_HOME/fuseki-server.jar tdb.tdbloader --loc=$TDB_DIR
java -cp $FUSEKI_HOME/fuseki-server.jar jena.textindexer
according to an inspection of the index with luke, index entries
"urn:x-arq:DefaultGraph" for the graphField are created.
However, when I use HTTP to load additional data like that:
curl -X POST --silent -H "Content-Type: application/x-turtle" -d
@$SCFILE $SWDSKOS_ENDPOINT/data?default
index entries "urn:x-arq:DefaultGraphNode" are created.
The text query seems to aim at the latter, ('(text:test) AND
graph:urn\:x\-arq\:DefaultGraphNode'), which seems to be the reason
for my searches creating empty results.
When I comment out the graphField statement and recreate the store,
searches bring up the expected results.
Cheers, Joachim
--
Osma Suominen
D.Sc. (Tech), Information Systems Specialist
National Library of Finland
P.O. Box 26 (Kaikukatu 4)
00014 HELSINGIN YLIOPISTO
Tel. +358 50 3199529
[email protected]
http://www.nationallibrary.fi
