Airsonic: enhancement : Integrate Elastic search

[dorel14]

hello
I think it would be a nice option having elastic search integrated so that it might be easier to search along big library

https://github.com/biconou/SubsonicES.

A poc in french here
https://www.elastic.co/fr/blog/subsonic-utilise-elasticsearch-comme-base-de-donnees-orientee-documents#

[biconou]

Definitely yes !
So you red my article. Nice.
Note that in that work I completely replaced the SQL database by Elasticsearch. I don't think it will be a goal for the Airsonic community.
But I've planned to write some plugin to use Elasticsearch as an additional search data store (if any other people interested in that).
In my article, I show the benefit of using Kibana to browse datas.
I also had the Idea that an Elasticsearch cluster could be used to index music content over the whole Airsonic community (dream !!).

[dorel14]

Yes i red it
i agree in the fact to make airsonic use it as a search engine .best thing would be indexing metadata and have some 'facettes' for the different fields.Also perhaps having Kibana for stats (Number of artist , Number of play .......)

[biconou]

Note that it will require to deploy two more services beside Airsonic : Elasticsearch and Kibana.
This can be solve with docker-compose for instance.

[jvoisin]

Airsonic is already using Lucene internally, what would be the point to have an Elastic/Kibana instance on top of it?

[dorel14]

Hello
Elastic search make research very easy with facets , it makes also very easy to filters the results and mix a lot of criteria for example to prepare a playslist
If lucene can do the same , it 's ok for me

[jvoisin]

From the first lines of Elasticsearch's wikipedia article:

Elasticsearch is a search engine based on the Lucene library. It provides a distributed, multitenant-capable full-text search engine with an HTTP web interface and schema-free JSON documents.

So it seems that we can close this issue.