Skip to content

Latest commit

 

History

History
46 lines (26 loc) · 2.31 KB

004-Named-Queries.md

File metadata and controls

46 lines (26 loc) · 2.31 KB
Raw

Named Queries

Current

You can ask the DLCS to select from images where one of the metadata values of the image matches some value, then order by one of the other values. The DLCS can return this query as a IIIF Manifest, or a PDF.

Wellcome use the PDF feature, and we use the manifest named query extensively on other projects to generate skeleton IIIF for further enrichment, starting with just the images and a little bit of metadata.

Proposed

  • Extract and refactor to new service
  • Projections: IIIF 2, 3; PDF
  • Revisit the named query syntax, and the tools to build named queries in the DLCS portal
  • Pass in link to IIIF Manifest that uses the Image API endpoints; PDF projection reads the text content of canvases and includes it in the PDF (Fireball enhancement) => proper text in PDF
  • Research need for other projections like ePUB, MOBI, etc.

Note that there is a class of named queries that doesn't require that the DLCS hosts the images, but that might not be a named query any more.

Suggested named query URL form (public view)

In the current orchestrator, a named query conforms to

/(pdf|iiif-resource)/customer/named-query-name/{*params}

where:

  • first slot is identified by the orchestrator as a named query - we could add more top level query types later, for other projection types
  • last slot is as many /p1/p2/p3... as you want to feed it

This could be better.

/q/customer/named-query-name/{*params}

...where q indicates that this is a named query of any type, always routed to Orchestrator Orchestrator then works out what the handler is by looking up the name in the third segment (e.g., /my-pdf-query/)

Orchestrator maintains a map of customer, queries, handler, so it can forward it on to (potentially) separate services, e.g., a PDF generator, a Manifest generator.

It would make for more useful URLs. It would also mean you weren't allowed to have two different types of query with the same name in the same customer, but that's OK. That's good.

The problem with a current URL like https://dlcs.digirati.io/pdf/2/eden/space-test/1 is "what do I call it?"

/pdf/.../pdf/

Instead, you can just have /q/2/pdf/... (or whatever we choose as the "named query" signifier in the first path segment, doesn't have to be /q/). At the moment we have to keep reserving words in the first path segment.