Hello,
I am one of the committers for Apache NiFi (incubating). I am looking to integrate NiFi with Spark streaming. I have created a custom Receiver to receive data from NiFi. I’ve tested it locally, and things seem to work well. I feel it would make more sense to have the NiFi Receiver in the Spark codebase along side the code for Flume, Kafka, etc., as this is where people are more likely to look to see what integrations are available. Looking there, though, it seems that all of those are “fully integrated” into Spark, rather than being simple Receivers. Is Spark interested in housing the code for Receivers to interact with other services, or should this just reside in the NiFi codebase? Thanks for any pointers -Mark
