vinothchandar commented on issue #18298: URL: https://github.com/apache/hudi/issues/18298#issuecomment-4084600730
> However, after HoodieStreamer converts that transformed dataset into JavaRDD / GenericRecord and writes through the Hudi write path, that plan-level lineage is no longer directly visible to lineage systems We could move more of that into DataFrame code (it's a worthy pursuit). But I also wonder this should be a standard ask from Spark lineage tracking tools. e.g. https://docs.datahub.com/docs/metadata-integration/java/acryl-spark-lineage uses Spark listeners to do this generically at the Spark layer. All in all - I think this is not a Hudi problem. You probably want to track something across all Spark jobs. and we can do this cleanly with some listeners? I believe that is the standard approach across OpenLineage, DataHub etc. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
