Ingesting node

[sharon.c]

I used to use Logstash to ingest data to Elasticsearch, now I plan to change it to logstash, but there is one functionality I have to rely on, I need to generate a hash to uniquely identify a document

This is the logstash code I have, I how can I have the same functionality in ingesting processor as the following code in logstash

        ruby {
           code => "require 'digest/md5';
                    event['@metadata']['computed_id'] = Digest::MD5.hexdigest(event['filed1'].to_s + event['field2'].to_s + event['record_time'].to_s  );
                    end "
        }

I have two questions

1. Is there any way to have external md5 library to generate hash code based on some fields values?
2. In ingesting node, how do I set ['@metadata']['computed_id'] for elasticsearch to pick up the document id for each document

[shanec (Shane Connelly)]

Oddly, I was fiddling around this past weekend with the ingest node and made a processor to do this. It does MD5, SHA1, and SHA256. https://github.com/eskibars/elasticsearch-ingest-hashfields . It only works on string data currently, but should be easy enough to extend if you need others.

[sharon.c]

Thank you so much for your answer.
I also need to make this hash as _id field. Can you provide an example how to do it?
_id field is important for duplication in our elasticsearch index, it makes sure we do not have repeated documents.