From the outset, the Wagtail search module has sought to tackle a problem that's largely independent of Wagtail as a CMS: providing full text search for Django models that is easy to configure and works seamlessly across a variety of backends, both in-database and through external services such as Elasticsearch. It's a goal we're still working towards, and in this talk I'll introduce some of the challenges we face:

• Building on Django's familiar APIs for filtering and querysets, even when queries aren't using the database
• Keeping external data stores in sync with the database, to handle details such as user permissions
• Managing backends with widely differing capabilities, that have a habit of being moving targets between releases, and effectively testing across many different configurations
• Getting the subtle UX details right to ensure that search works in the way that users expect

Finally, I will show how the major redevelopment of the search module happening as part of Wagtail 7.2 aims to tackle these challenges more effectively.

Attendees will gain insight into how search works under the hood, and learn how to work with the updated search APIs coming in 7.2.