tweise commented on PR #23777: URL: https://github.com/apache/flink/pull/23777#issuecomment-1837588962
LGTM from my side, based on a quick review after @mxm had already covered the details. Regarding complexity: FLIP-27 sources are without doubt more complex for the source implementation/author. That was the price to push past the limitations of the (eventually too simplistic) old SourceFunction and accommodate the rather diverse and complex requirements of feature rich source connectors. Part of FLIP-27 was to reduce implementation complexity by providing building blocks like abstractions for readers and fetchers. Changes proposed here serve a limited/specific scenario for testing. The goal is not to also modify the consuming tests, which would be risky from a regression angle and perhaps not feasible. While it takes a bit of effort to understand the implementation, I also see no obvious winner from an abstraction perspective, at least not one that would be helpful for general connector needs. Please also consider that this effort was delayed till now, lacking criticality beyond finally sunsetting SourceFunction. Given that @afedulov has already put significant effort into this, I would personally focus further efforts elsewhere. -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
