Hi Matthias, That makes things so much clearer. Thanks for your help understanding the codebase better.
Speaking of the code base of Kafka Streams, I've got typo fixes and docs improvements locally here and there scattered throughout the code of Kafka Streams, what are the proper steps to get them merged? Just a pull request? Or should I file an JIRA issue and...please guide. Appreciated. Pozdrawiam, Jacek Laskowski ---- https://about.me/JacekLaskowski Mastering Spark SQL https://bit.ly/mastering-spark-sql Spark Structured Streaming https://bit.ly/spark-structured-streaming Mastering Kafka Streams https://bit.ly/mastering-kafka-streams Follow me at https://twitter.com/jaceklaskowski On Mon, Feb 26, 2018 at 6:37 PM, Matthias J. Sax <matth...@confluent.io> wrote: > Thanks for the feedback. > > You are of course right, that we cannot remove toString() -- we used the > @deprecated annotation to imply that the overwrite will be removed and > toString() will fall back to Object.toString() in the future and thus > not provide any useful information (at least this was the plan). > > Using localThreadMetadata() within toString() and print whatever it > return might be an alternative. However, I am not sure if we gain much. > Also, KafkaConsumer/Producer/AdminClient don't overwrite toString() -- > IMHO, not overwriting toString() aligns better with other parts of the > code. > > About @Evolving: that is correct, too. We just try to be nice and > provide backward compatibility even if we never promised :) > > > -Matthias > > On 2/26/18 1:38 AM, Jacek Laskowski wrote: > > Hi, > > > > I've just found that KafkaStreams.toString [1] is deprecated, but I think > > that it does not make sense. > > > > The parameterless toString is simply part of the java.lang.Object > contract > > and I don't think it could ever get deprecated (unless it is by Object). > > > > I think it'd be much better if the method used whatever it's recommended > > for a toString-like functionality (which seems that > > KafkaStreams.localThreadsMetadata [2] is or something based on that). > > > > Thoughts? > > > > p.s. I also think that since KafkaStreams is marked > > as @InterfaceStability.Evolving using @Deprecated markers does not add > much > > if anything. I thought that Evolving was to say that literally everything > > could change without warning at any time, couldn't it? > > > > [1] > > https://github.com/apache/kafka/blob/trunk/streams/src/ > main/java/org/apache/kafka/streams/KafkaStreams.java?utf8=%E2%9C%93#L900 > > > > [2] > > https://github.com/apache/kafka/blob/trunk/streams/src/ > main/java/org/apache/kafka/streams/KafkaStreams.java?utf8=%E2%9C%93#L1061 > > > > Pozdrawiam, > > Jacek Laskowski > > ---- > > https://about.me/JacekLaskowski > > Mastering Spark SQL https://bit.ly/mastering-spark-sql > > Spark Structured Streaming https://bit.ly/spark-structured-streaming > > Mastering Kafka Streams https://bit.ly/mastering-kafka-streams > > Follow me at https://twitter.com/jaceklaskowski > > > >
