yanghua commented on pull request #2868: URL: https://github.com/apache/hudi/pull/2868#issuecomment-828903309
> > I still insist that we need to include kafka-related dependencies. If you look back at the HoodieFlinkStreamerV2 class. What is it in essence? It is just a program written using Flink DataStream API, which is specific (Kafka -> Hudi) > > No, on one says that they don't know how to add a connector jar or actually few people use the `HoodieFlinkStreamerV2` tool. "one says that they don't know how to add a connector jar" -> I recommend we package it into the bundle for users. It's not that users won't, but users should not or may not need to perceive these things. This is a question of user experience. According to your logic, what reason do you think users will not use FlinkWriteClient directly? Why should we guide users to use Flink SQL? Can't users write the FlinkStreamer class by themselves? All of this is to shield users from details as much as possible, let the framework provide out-of-the-box capabilities as much as possible, and provide a good experience as much as possible? Is not it? "actually few people use the `HoodieFlinkStreamerV2` tool" -> Actually, there still few users use the flink write client, because it is still not production-ready for 0.8, you know. IMO, we do not get enough samples about your result. I have never understood why we cannot include the kafka connector to provide convenience to some users who do not use SQL. And it should provide a consistent experience with DeltaStreamer based on Spark. Otherwise, don't call "FlinkStreamerXXX". -- 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]
