I am looking for a way to integrate an additional way to search the Evergreen database. PostgreSQL is great and versatile but I could imagine a scenario where we would allow the local server administrator to implement an alternative in some circumstances. For example:
opac_search_engine_def = [
'keyword => 'elasticsearch',
'title' => 'postgres',
'series' => 'postgres'
.....
]
staff_search_engine_def = [
'keyword => 'postgres',
'title' => 'postgres',
'series' => 'postgres'
.....
]
This could start small and slowly grow. The Elasticsearch team maintains a Perl API: https://www.elastic.co/guide/en/elasticsearch/client/perl-api/current/index.html
Therefore, I believe this could be used inside of Evergreen with the results coming back in the same compatible methods interpreted by the rest of the Evergreen code.
I understand that search is complicated and feature-rich. I realize this would create yet another piece of the stack but I think it could be worth it! There are a ton of built-in Elasticsearch features not the least of which is "Phrase suggester" https://www.elastic.co/guide/en/elasticsearch/reference/current/search-suggesters.html
Thoughts on this are welcome. Trying to get a conversation about it.
I know the koha folks have an elasticsearch option too!
I haven't worked much with elasticsearch, but I've really enjoyed working with solr. It's really easy to tune queries. It also has a result grouping feature that might help out with implementing metarecord search (https:/