Skip to content

Search

Archivy supports two search engines:

  1. Elasticsearch - an incredibly powerful solution that is however harder to install.
  2. ripgrep, much more lightweight but also less powerful.

These allow archivy to index and provide full-text search on their knowledge bases.

You can select these in the archivy init script, and the the config docs also has more information on this.

Elasticsearch

Elasticsearch, is a complex and extendable search engine that returns high-quality results.

Instructions to install and run the service which needs to be running when you use Elasticsearch can be found here.

Append these two lines to your elasticsearch.yml config file:

http.cors.enabled: true
http.cors.allow-origin: "http://localhost:5000"

Then, when you run archivy init simply specify you have enabled ES to integrate it with archivy.

You will now have full-text search on your knowledge base!

If you're adding ES to an existing knowledge base, use archivy index to sync any changes.

Elasticsearch can be a hefty dependency, so if you have any ideas for something more light-weight that could be used as an alternative, please share on this thread.

Ripgrep

Ripgrep, on the other hand, is much more lightweight and small, but is also a bit limited in its functionality.

Follow these instructions to install ripgrep.

Then simply specify you want to use it during the archivy init script (or edit the config to add it).