A small (prototype) single page app which matches a list of names against a Solr index of entities (places, people, corporate bodies) and lets the user refine and export structured (CSV) data.
The Solr schema (on coredir) is not yet optimised and is quite slow to index due to the
use of a Beider Morse filter analyser for the optional phonetic matching.
The datasets used are the Geonames allCountries.zip
and this set of entities from the EHRI portal database.
You can set up a dev instance by running sudo docker-compose up from the project directory, which will start an
instance of Solr 6.6.6 using the supplied core config on port 8984. You can then run the ./bin/ingest script to
download the datasets and import them into Solr using, respectively, the CSV and JSON update handlers. Note: this will
take quite some time since Solr has to ingest ~1.5GB of Geonames data.
Running the app on port 9000 can be done by starting the SBT shell and running the run command.
