Journal Observatory backend

This repository contains the source code of the backend developed in the Journal Observatory project. The backend consists of a GraphDB triplestore, a SPARQL endpoint, and a REST API.

GraphDB

GraphDB is the triplestore backend for the Journal Observatory project. It can be run on the free-tier. It provides a triplestore, SPARQL endpoint, and a graphical user interface to manage the server.

Deployment

GraphDB can be deployed using Docker:

docker run \
    --name graphdb \
    --env GDB_HEAP_SIZE=12G \
    --publish 7200:7200 \
    --volume ./graphdb/graphdb.properties:/opt/graphdb/conf/graphdb.properties \
    --volume ./graphdb_data:/opt/graphdb/data/repositories \
    khaller/graphdb-free

Or using Docker Compose and the docker-compose.yml file:

docker-compose up -d graphdb

PAD API

The PAD API is a Flask application that serves PADs over a REST API. It is meant to provide simple access to the PADs stored in the triplestore, as a limited alternative to the SPARQL endpoint. The API has two modes:

• /api/pads: Get a list of PADs based on an optional search query
• /pad/<id>: View a single pad in a JSON-LD, Trig or Graphical format

The documentation of the API can be found at /apidocs

Deployment

The PAD API is configurable with the following environment variables:

• APP_ENVIRONMENT
• APP_HOST
• APP_PORT
• APP_SPARQL_HOST
• APP_SPARQL_QUERY_PATH
• APP_GLOBAL_LIMIT

The PAD API can be run in a python development environment by installing the requirements and serving the Flask application.

pip install -r requirements.txt
PYTHONPATH="./pad_api/src" flask -A api.application:api --debug run --host "${APP_HOST}" --port "${APP_PORT}"

It can also be run in a Docker container:

cd pad_api
docker build -t pad_api
docker run \
    --name pad_api
    --env APP_ENVIRONMENT=develop
    --env APP_HOST=0.0.0.0
    --env APP_PORT=5001
    --env APP_SPARQL_HOST=http://graphdb:7200
    --env APP_SPARQL_QUERY_PATH=/repositories/pad
    --env APP_GLOBAL_LIMIT=50
    --publish "${APP_PORT}:${APP_PORT}"
    pad_api

Or using Docker Compose and the docker-compose.yml file:

docker-compose up -d pad_api