The catalogue pipeline populates the search index for our online catalogue search.
This includes:
- fetching records from source catalogues and keeping them up-to-date
- transforming records into a single, common model
- combining records from multiple sources, where appropriate
- creating an Elasticsearch index which can be queried by the catalogue API
This GitBook space is meant to provide a high-level overview of the catalogue pipeline and its design. These docs are meant for Wellcome Collection developers who want to learn about the project, or for colleagues at other institutions who want to build something similar.
It does not contain specific operational details, e.g. how to deploy specific services. Those are kept inside the code repository.
The catalogue pipeline code is in https://github.com/wellcomecollection/catalogue-pipeline