The PPO data server is a machine level interface to the database back-end storing all indexed results from the ppo-data-pipeline. Front end interfaces which call the ppo-data-server are the global PPO data portal and the rppo R package.
Query functions that return instance data run through a node.js reverse proxy to the elasticsearch database service, which is run on a different server, secured by an opening through a firewall via a dedicated port. Query functions that return trait data elements from the Plant Phenology Ontology utilize the rdflib.js library for parsing the ontology, and cache elements on the server to be returned for requests. A cronjob is utilized to update cache results once per night so any changes in the PPO itself will take up to 24 hours to register. The purpose of the cacheing is to improve speed of responses.
The URL endpoint prefix for all services is at https://www.plantphenology.org/api/. Following are the services that live below this endpoint. Please note the versions of the service endpoints below which contain different current versions, indicated by v1 or v2. It is important to reference the correct version for each service to return the documented responses:
- v1/query Query the PPO data store
- v2/download Downlaod results as a package from data store
- v2/ppo Lookup terms from the ontology proxy