This jQuery widget is a browser-based user interface to the Comunica SPARQL client. It allows users to execute SPARQL queries over one or multiple heterogeneous interfaces, such as Triple Pattern Fragments interfaces. The widget is used here in order to primarily work with the data from the Fast and Slow project.
- Run
yarn install
to fetch dependencies and build the browser version of the client code. - Run
yarn run dev
to run a local Web server (yarn run dev-prod
for minified production output). - Edit datasources in
settings.json
and queries in thequeries
folder, and runqueries-to-json
to compile both of them in a single JSON file. - Run
yarn run build
to generate a production version in thebuild
folder.
The original Comunica SPARQL engine is written for the Node.js environment. The Webpack library makes it compatible with browsers.
The query engine itself runs in a background thread using Web Workers. The user interface (ldf-client-ui.js
) instructs the worker (ldf-client-worker.js
) to evaluate queries by sending messages, and the worker sends results back.
Configure your widget by editing the settings.json file.
Next, edit the queries directory in which you should insert the queries that will be present by default in the widget.
Build the Docker container as follows:
docker build -t comunica-sparql-widget .
After that, you can run your newly created container by mounting your current folder to the Docker container:
docker run -p 3000:80 -it --rm comunica-sparql-widget
Settings and queries can be passed at runtime by mounting your custom queries.json
to the Docker container:
# Compile queries.json from settings.json and the files in the queries folder
./queries-to-json
# Provide the compiled queries.json at runtime
docker run -v $(pwd)/queries.json:/usr/share/nginx/html/queries.json -p 3000:80 -it --rm comunica-sparql-widget
Access on http://localhost:3000
The Linked Data Fragments jQuery Widget was originally written by Ruben Verborgh and ported for Comunica SPARQL by Ruben Taelman.
The adaptation for the Fast and Slow project was made by the Fast and Slow team.
This code is copyrighted by Ghent University – imec and released under the MIT license.