TPF Streaming Query Executor
The TPF endpoint must have time-annotations for all dynamic triples using some annotation method. The client can accept a regular SPARQL query, and it will detect any dynamic triple patterns and will stream its results.
Paper with more details about this concept: TODO
Install the latest version from GitHub:
$ git clone firstname.lastname@example.org:rubensworks/TPFStreamingQueryExecutor $ cd TPFStreamingQueryExecutor $ npm install
A docker container can be started like this:
docker build -t tpfqs-client . && docker run \ -e "QUERY=<SPARQL query>" \ -e "TARGET=http://<server>:3000/<datasource>" \ -e "CACHING=true" \ -e "INTERVAL=true" \ -e "DEBUG=true" \ --rm tpfqs-client graphs
To start the train departure information demo, run:
cd bin && ./startLiveTrainServer.sh
To start the live music demo from Q-Music, run:
cd bin && ./startLiveMusicServer.sh
Execute custom queries
To be able to run custom queries, the TPF server will require a dataset that contains time annotations for the dynamic triples using some type of RDF annotation.
It is advised to read through the
live-ldf-server script for an example on how to annotate triples on-the-fly and storing them at some TPF endpoint.
querytrain [annotationtype] script shows how to call the actual