See documentation
Provide a reusable web service to search for concepts from an existing SKOS taxonomy.
- Provide a suggestion/auto-completion from a list of concepts indexed
- Search for concepts from their names
- Get a concept by its identifier
Requires Java (version 6+) and Maven (version 2+).
Run mvn clean package
at the root of the project. This will build a "fat" JAR (shaded JAR containing all its dependencies).
You can start the application by running the following command at the root of the project:
java -jar target/skos-suggester-1.0-SNAPSHOT.jar server src/main/resources/configuration.yml
(This assumes that you have set up your configuration file at src/main/resources/configuration.yml
, see src/main/resources/configuration.yml.example
for a sample configuration.)
After having packaged the application, run:
java -jar target/skos-suggester-1.0-SNAPSHOT.jar skosimport -f [path to your file] [path to your configuration file]
-f
is the location of the SKOS RDF file to import- Use
--format
to override the default format (N-TRIPLES).
Alternatively, if you have a Jena TDB store instead of a file:
java -jar target/skos-suggester-1.0-SNAPSHOT.jar tdbimport -d [path to the store] [path to your configuration file]
See fabfile.py
(requires Fabric, see requirements.txt
, install it using pip install -r requirements.txt
).
To automatically import the latest FAST topics, use the following command:
fab ENV update_topics
The script assumes that unzip
and tdbloader2
are installed.
Run at the root of the project:
mvn test
Basic load testing has been done using Locust.
To run the load test, first install it using pip
and then run:
locust --hostname "http://127.0.0.1:8080"
See locustfile.py
at the root of the project for more information.
Build the image by running at the root of the repository (assuming you have docker.io installed):
docker build -t skos-solr .
Start the container by running:
docker run -d -p 8983:8983 skos-solr
Or alternatively, if you have fig
installed, just run fig up
.