armchar

There is a client to go with this project. Note that this is but a rudimentary prototype, particularly the client.

Quick start

All the commands should be run from the project root. First, we need to build the ontology files:

( cd Ontology ; make install )

The second step is to initialise the TDB database. You need to have the Jena library available and use its tdbloader2 script. Run the following commands from the project root.

JDIR=$HOME/prg/apache-jena-3.16.0/bin/
rm -rf tdb
mkdir -p tdb
$JDIR/tdbloader2 --loc tdb Ontology/cieran.ttl Ontology/contested.ttl 

Finally, we build and start the web server. This requires docker. Depending on the docker installation, you may or may not need to change the permissions and ownership of the tdb directory.

mvn install || exit
mkdir -p payara
cp target/armchar-1.0-SNAPSHOT.war payara/armchar.war
docker build -t armchar .
docker run --name armchar -p 8080:8080 -v `pwd`/payara:/opt/payara/deployments -v `pwd`/serverdata:/opt/payara/serverdata  -v `pwd`/tdb:/opt/payara/tdb  armchar

The server takes a while to start. You can test it by downloading character data, for instance,

curl http://localhost:8080/armchar/Character/cieran/1217/Summer > summer1217.json

Note that it takes a couple of minutes to initialise. This is a one-time cost, however, as the inference graph is built.