# Database and Ontop Setup
In this tutorial, we run a DuckDB database of museums that includes information about the museums themselves, their employees and their exhibits.
# Database setup
Procedure to set up the datebase for the following exercises:
- Unzip the archive of DuckDB (duckdb.zip).
- That's it! The data directory contains the already prepared databases, compatible with DuckDB version 0.7.1.
- If you want to have a first look at the database, you can open it with the DuckDB client application that can be downloaded from the official site (opens new window).
# Ontop SPARQL endpoint setup
For this tutorial, we use a SPARQL endpoint provided by ontop to test our mappings and lenses. Here, we walk through the preparation of the endpoint via Ontop CLI, but a docker-image-based endpoint can be created in a similar fashion, as explained in the endpoint tutorial.
- Download Ontop CLI (opens new window) and unzip it to a directory, which is denoted as
- Copy the DuckDB jdbc driver to
$ONTOP_CLI_DIR/jdbcfrom the duckdb.zip archive.
- Copy the directory
datacontaining the database file
- Paste your mapping and lenses files, as well as the ontology and properties files into a new directory called
- You will create the mapping and lenses files throughout the course of this tutorial. Alternatively, the duckdb.zip archive contains one directory for each of the lens tutorials, with sample mappings and lenses already pre-made.
- The ontology and properties files will be the same throughout all tutorials. They are already provided in the duckdb.zip archive.
- Start the Ontop endpoint. On Mac/Linux:
$ONTOP_CLI_DIR/ontop endpoint \ --ontology=input/ontology.ttl \ --mapping=input/mapping.obda \ --lenses=input/lenses.json \ --properties=input/duckdb.properties \ --cors-allowed-origins=http://yasgui.org # if needed
ontop endpoint ^ --ontology=input/ontology.ttl ^ --mapping=input/mapping.obda ^ --lenses=input/lenses.json ^ --properties=input/duckdb.properties ^ --cors-allowed-origins=http://yasgui.org
- Open the web interface of the SPARQL endpoint to try some queries: http://localhost:8080/ (opens new window)