On 12/06/12 02:15, Eric Scott wrote:
I'm encountering a problem inserting into named graphs when doing so
programmatically.
Pardon my clojure, but I think its intent is pretty clear, let me know
and I'll translate to java-ish pseudocode:
(let [model (TDBFactory/createModel (ensure-directories-exist
"/home/escott/test1TDB")),
update-request (UpdateFactory/create)
]
(.add update-request "Create Graph <http://myDB.com/ng>")
(.add update-request "INSERT DATA {graph <http://myDB.com/ng>
{<http://myDB.com/a><http://myDB.com/b><http://myDB.com/c>}}")
(UpdateAction/execute update-request model)
(.close model)))
This runs without error, and is similar to code I've called in the past
which works fine, but does not insert into a named graph.
However, when I make (say) this call:
(let [model (TDBFactory/createModel "/home/escott/test1TDB")
]
(sparql-query model "select ?g where {graph ?g {}}")))
I get no errors, but an empty response.
I have also entered exactly these update requests in fuseki and gotten
the expected behavior (a listing of named graphs, including the one I
just created).
Here are my maven spex:
[org.apache.jena/jena-core "2.7.0-incubating"]
[org.apache.jena/jena-arq "2.9.0-incubating"]
[org.apache.jena/jena-tdb "0.9.0-incubating"]
Is this a bug, or am I doing it wrong?
It is because you are creating a model not a dataset.
You then try to update the model by adding another model (the named graph).
SPARQL Update acts on a dataset/graphstore. If you pass a model to the
update engine, it wraps it in a general purpose, but in-memory, dataset
with the passed-in model as the default graph.
Your code adds a model "INSERT DATA {graph..." but that is not in the
TDB dataset.
Better is to create a TDB dataset and update that.
TDBFactory/createDataset
By the way - this route is now blocked - you'll get an exception saying
"no such graph" because it now uses an datasetwrapper that only has one
graph.
Andy
Thanks,
