sparql endpoint and sampo-ui interface for NuMAD coin finds data.
load initial data from csv into postgres database. Cleans up data to be imported in a more usable way.
Source of truth for the data
Maps the data in real time from the database to a sparql endpoint using the nomisma ontology for the most part.
The nomisma ontology is extremely loosely defined without any domains or ranges for any class. Because of that some liberty was taken in how to assign literal properties, while still trying to follow what nomisma does as closely as possible.
If they ever release a new ontology version that is better defined it would be a good idea to migrate to that.
The sampo client sends requests to the sampo server to get results and facets etc.
The sampo server takes requests from the client and sends sparql queries to the ontop endpoint to get data from the VKG.
Install just to be able to run commands from the justfile.
Download postgres jdbc driver for ontop:
curl -o ontop/jdbc/postgresql-42.7.8.jar https://jdbc.postgresql.org/download/postgresql-42.7.8.jarYou also need the original csv file named numad-data.csv in the data directory.
This config uses local pmtiles files for the base layers of maps. For this to work you either need
to download the files belgium.pmtiles and europe.pmtiles into sampoConfigs/sampo/assets/maps or edit
the configs of the coins.json search perspective to point to where your base layer files are hosted.
The pmtiles files were downloaded using scripts from this repository.
Rebuild, run import script and rerun everything:
just reinitRerun import script:
just importRun everything
docker compose up -dStop running
docker compose downDevelopment by Ghent Centre for Digital Humanities - Ghent University. Funded by the GhentCDH research projects.
