This plugin is an attempt to make the often difficult search indexing process easy and code first in Umbraco 8 projects. It requires a property value converter to be available, to get strongly typed content, but once that is enabled, like Models Builder, the setup is straight forward.
The plugin is based on NEST, but uses the lowlevel client to do most of the processing, giving a fast and responsive alternative to Lucene and Examine. Furthermore a lot of default settings are already set, but the app component exposes all configurations and adds a few helper methods, like adding and keeping track on indices.
Connection Pools, Search and indexing is all supported and the plugin uses a handler strategy to inject custom functionality for your document types. If you use nested components the indexer can also read recursively through the elements and fetch search text from them, as long as they have a handler attached.
To the left is an example of how to add a complex index and of a custom document handler. In this example, the document has a MetaTitle that is computed and the client requested a category filtering on the 3rd level, so all content in this index has a category field.
The Search Text is the content that the search function looks through and the properties are available data to retrieve in the result or as a filter during the search operation. Along with the search service is a dashboard for umbraco backoffice where the backoffice users can reindex the different indices or access the kibana entity on the elastic search server.