These are the services that back our Content API. The Content API exposes our non-catalogue content: at the moment, that's just articles, but in the future it might include endpoints for events, exhibitions, and functional content.
/pipeline: runs on a schedule, queries for updated content, transforms it and puts it into an index in the content ES cluster./api: queries the Elasticsearch index/webhook: a Lambda handler exposed via a function URL for handling Prismic webhook events and publishing them to a EventBridge bus/unpublisher: receives events published by the webhook, extracts document IDs from them and removes them from the ES index.
- Deployment is all done in CI, see
.buildkite/pipeline.yml huskywill set up a pre-commit hook for autoformatting when you first install dependencies