Convert a metadata String field into array


(Panda) #1

Hi

I have an HTTP JSON Input as follows :
{
"mapping": {
"index": "indexname",
"activity": {
"action": "script_update",
"fieldList": ["field1","field2"]
},
"document_type": "_doc",
"document_id": "1",
"routing": "1"
}
}

I have converted the above data in @metadata fields as it was not needed in final elasticsearch output.
Now I need to run a script in which I need to iterate on the "fieldList" array of @metadata and access each element as seperate String.

Currently I am unable to convert the String to array of comma separated string (from @metadata).

I have tried using mutate + split filter but it didn't work.

    input{
	http {
		host => "127.0.0.1"
		port => 8080
	}
}
filter{
	mutate{
		add_field => { 
			"[@metadata][index]" => "%{[mapping][index]}" 
		}
	}	
	mutate{
		add_field => { 
			"[@metadata][action]" => "%{[mapping][activity][action]}" 
		}
	}
	mutate{
		add_field => { 
			"[@metadata][fieldlist]" => "%{[mapping][activity][fieldlist]}" 
		}
	}

	mutate{
		add_field => { 
			"[@metadata][userId]" => "%{[mapping][activity][userId]}" 
		}
	}
	mutate{
		add_field => {	
			"[@metadata][document_id]" => "%{[mapping][document_id]}" 
		}
	}
	mutate{
		add_field => { 
			"[@metadata][document_type]" => "%{[mapping][document_type]}" 
		}
	}	
	mutate{
		add_field => {	
			"[@metadata][routing]" => "%{[mapping][routing]}" 
		}
	}
}
output{

	  if [@metadata][action] == "script_update" {
			stdout { codec => rubydebug { metadata => true }} 
			
			elasticsearch{
				hosts => ["localhost:9200"]
				index => "%{[@metadata][index]}"
				document_type => "%{[@metadata][document_type]}"
				document_id => "%{[@metadata][document_id]}"
				routing => "%{[@metadata][routing]}"
				action => "update"
				script_lang => "painless"
				script_type => "inline" 
				script => ' ' 
                           /* Add script to iterate on field list and update the corresponding field in elasticsearch with value 2 */
				}
			}	
}	 

Any help would be appreciated.


(Panda) #2

Any update on this ?