The current ontology is sourced from ThothEntities, built into tables with jQuery, and then iframed into docs.diffbot.com.
This presents a few issues:
Algolia search does not work for the ontology. Making it confusing to users.
The multi-layered complexity presents issues when it comes to keeping documentation fresh and updated.
Front-end load times are very slow (while processing the JSON into the DOM)
This solution aims to solve the above by:
Building semantic markdown files for the ontology.
Decoupling front-end and back-end for the ontology. Programmatically generated markdown files now host the front-end, while the backend (JSON ontology) continues to be updated by the respective maintainers
Since the ontology is rendered as static markdown files, there is no performance hit
There are instructions to build in ./website/generateOntology.js and appended to the README.
I don't know what our CD pipeline looks like right now, but it would be ideal if we could just add node generateOntology in there somewhere so we don't have to worry about fresh ontology again :)
The current ontology is sourced from ThothEntities, built into tables with jQuery, and then iframed into docs.diffbot.com.
This presents a few issues:
This solution aims to solve the above by:
There are instructions to build in
./website/generateOntology.js
and appended to the README.