HeartSaVioR commented on issue #22282: [SPARK-23539][SS] Add support for Kafka headers in Structured Streaming URL: https://github.com/apache/spark/pull/22282#issuecomment-524063520 @dongjinleekr Raised PR to your fork (https://github.com/dongjinleekr/spark/pull/1) for addressing review comments from @zsxwing . What I suggested was reflecting either one from here: > You may either make them ThreadLocal (though it may grow incrementally so don't recommend), or just follow the previous approach of KafkaRecordToUnsafeRowConverter and initialize per caller (would be each task). instead of rolling back. The main point of review comments from @zsxwing is, all `Record => xxx` function values are meant to be "critical path" as it would be executed per record. New initialization (especially initializing heavy instance which may require some calculation) should be avoided if possible.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
