Search proxy

[juokaz]

Summary

A layer that forwards any requests to an existing help desk server implementation, but indexes any updates and provides full-text search capabilities to clients.

Requirements

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in IETF RFC 2119.

Specification

The service MUST forward all requests to the upstream server unmodified.

The service SHOULD augment responses sent from the upstream server to the client with information on how to perform searches (e.g. through embedded XForms, XHTML Forms or URI Templates).

[juokaz]

I propose using Elasticsearch as that is very easy to get up working and also supports data synchronization using Rivers. We are definitely not going to implement Rivers as that requires writing a Java plugin, which is quite a lot of work.

So we store documents in elasticsearch containing a list of searchable fields and a reference to actual ticket in the ticket system(s). So once the results are retrieved from the search index, results are looked up from the servers and returned. OR tickets are stored as cached copies as part of the document and returned immediately with a timestamp to identify the age.

[juokaz]

You can follow it here https://github.com/juokaz/restfest-search-proxy