Thank you, Claude and Andy, for your answers, which are very helpful.
However, I am still in the dark about how to do link Jena and MySQL.
I have found information about Jena/MySQL:
http://jena.sourceforge.net/DB/mysql-howto.html
This uses a classclass com.hp.hpl.jena.db.DBConnection
which was last seen in jena-2.5.4.jar, and since deprecated and gone.
If I understand it correctly, it was replaced by SDB, but here
http://jena.apache.org/documentation/sdb/
I read that
/Use of SDB for new applications is not recommended./
/This component is "maintenance only"./
/TDB <http://jena.apache.org/documentation/tdb/index.html>is faster,
more scalable and better supported than SDB.
/
/
/
But TDB is not MySQL.
So what is now the way to use Jena with MySQL?
Thank you, and with kind regards,
Michel
/
/
On 24/09/2013 07:01, Andy Seaborne wrote:
On 23/09/13 11:14, Claude Warren wrote:
Seems to me it would be faster to load the data at the jena interface
rather than at the fuseki interface. Since duplicates are ignored you
could just open a connection to the model and start adding triples as
you
read them from the json file.
if you don't want to take your system offline to do this you could do it
against a new database and then use mysql replication or
export/import to
move the data.
I don't know if the jena/mysql implementation allows for parallel
access to
the mysql database without conflicts.
It does - it all uses JDBC transactions.
In the case of SDB, you can modify the SQL database with sdbload while
Fuseki is running.
This is not true of TDB where Fuseki&HTTP are the equivalent of JDBC
because TBD is inside Fuseki.
Andy
Claude
On Mon, Sep 23, 2013 at 10:39 AM, Michel de Lange <
[email protected]> wrote:
dear all,
My aim is to take an existing library catalogue (which I have a
large json
file), and put this into a triple store. It is about books and authors,
andI have about 50,000 records. My naive strategy is to go through my
input, and every time I come to a new book, I do a sparql query to
look up
whether this already exists inthe triple store. If it doesnot, I
will do
somekind of update query to put this book, and its authors, intothe
triple
store. Apart from the fact that I don't know how to do such an
update (I
posted a separate question about that), I wonder if this is a good
strategy
at all. Should I be doing update queries, or should the output of my
program bean rdf file, which I then ingest separately? Or should I
be doing
something else completely?
The ontology is VIVO, the triple store is mysql, which I access through
fuseki, with a java program and the jena libraries.
My question is: what would be a good way, at a high level, to
accomplish
this?
Many thanks for your attention, any help is greatly appreciated.
Michel
