The Data Shape Server serves a stored knowledge graph schema information (e.g., the classes, the properties and their connections) to a client that can use it, e.g., in autocompletion of SPARQL queries over the data set with the corresponding schema.
The Data Shape Server is used in the context of ViziQuer tool (https://github.com/LUMII-Syslab/viziquer) to support auto-completion of visual queries over RDF databases.
node.js
installation Link, version ≥ 18- access to the
PostgreSQL
database containing the data schema information (the meta information about the endpoint to be queried).
Steps to start the data shape server locally, using an existing DSS database:
cd server
- run
npm ci
once to install the DSS dependencies - create
.env
file fromsample.env
and enter there the DB connection string and the port number the DSS app will be listening to - run
npm run dev
to start DSS in development mode ornpm start
to start DSS in production mode
To set up a new DSS database, please follow the steps from here.
When the database is created, import at least one schema, following the steps from here.
Then you can proceed as in the section Getting started
.
The Data Shape Server has been developed at Institute of Mathematics and Computer Science, University of Latvia, https://lumii.lv, with partial support from Latvian Science Council project lzp-2021/1-0389 "Visual Queries in Distributed Knowledge Graphs" (since 2022).