You can simply have another relationship from GetKafka to MergeContent essentially allowing MergeContent to have the original message so it could be merged. See attached image for example:
Oleg [cid:1F6F4F5A-85C9-4809-AECD-645E4F5054D3] On Oct 13, 2015, at 5:22 AM, yejug <[email protected]<mailto:[email protected]>> wrote: Hi I'm new in NiFi and trying to implement simple workflow: 1) */GetKafka /*> *original encoded message* (AVRO/MSGPACK/SOME ANOTHER FORMAT...) 2) */ConvertFromXXXToJson /*> JSON message 3) */EvaluateJsonPath /*> read some fields *from decoded Json* and put into FlowFile attribute (in my case I simple read /timestamp / field) 4) */MergeContent/* > merge *original messages* used /timestamp / attribute as "correlation" attribute 5) */PutHDFS /*> save *batch of original messages* into some directory on HDFS The problem is: Yet after step 2) I loss original messages readed from kafka, and only parsed JSONs are available. It is possible via standard NiFI processors\services set access original messages at step 4), without developing custom ConvertXXXtoJson processors (in simple case copy-pasting) and transmit input flowFiles to new custom relation ? -- View this message in context: http://apache-nifi-developer-list.39713.n7.nabble.com/Ingest-Original-data-from-External-system-by-data-s-dependent-condition-tp3093.html Sent from the Apache NiFi Developer List mailing list archive at Nabble.com<http://Nabble.com>.
