Status

Current state: Under Discussion

Discussion thread: here [Change the link from the KIP proposal email archive to your own email thread]

JIRA: KAFKA-4217

Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).

Motivation

Currently, KStream.transform transforms each record of the input stream into zero or more records in the output stream. A Transformer object is applied on each input record. The transformation is stateful, i.e., the transformer accesses state stores to compute output records. The signature of KStream.transform specifies a single key-value pair as output record. However, the user can emit multiple records for each input record if she calls for each output record method forward on the ProcessorContext object that is given to the transformer. By calling forward on the context, the user looses strong typing of the output records, i.e., if a user emits an output record that is not of the type specified in the transformer, the error is only thrown at run time and not at compile time. Furthermore, the public API misses a stateful transformation that transforms the value of each record of the input stream into zero or more records with same key but potentially different values in the output stream. Such a transformation would be the stateful version of KStream.flatMapValues.

Stateful transform operations that ensure strong typing that are more consistent with the stateless map operations, would make the public API safer to use, better comprehensible, and more complete.

Public Interfaces

<K1, V1> KStream<K1, V1> flatTransform(final TransformerSupplier<? super K, ? super V, ? extends Iterable<? extends KeyValue<? extends K1, ? extends V1>>> transformerSupplier,

                                       final String... stateStoreNames);

<VR> KStream<K, VR> flatTransformValues(final ValueTransformerSupplier<? super V, ? extends Iterable<? extends VR>> valueTransformerSupplier,

                                        final String... stateStoreNames);

<VR> KStream<K, VR> flatTransformValues(final ValueTransformerWithKeySupplier<? super K, ? super V, ? extends Iterable<? extends VR>> valueTransformerSupplier,

                                        final String... stateStoreNames);

Briefly list any new interfaces that will be introduced as part of this proposal or any existing interfaces that will be removed or changed. The purpose of this section is to concisely call out the public contract that will come along with this feature.

A public interface is any change to the following:

• Binary log format

• The network protocol and api behavior

• Any class in the public packages under clientsConfiguration, especially client configuration

  • org/apache/kafka/common/serialization

  • org/apache/kafka/common

  • org/apache/kafka/common/errors

  • org/apache/kafka/clients/producer

  • org/apache/kafka/clients/consumer (eventually, once stable)

• Monitoring

• Command line tools and arguments

• Anything else that will likely break existing users in some way when they upgrade

Proposed Changes

Describe the new thing you want to do in appropriate detail. This may be fairly extensive and have large subsections of its own. Or it may be a few sentences. Use judgement based on the scope of the change.

Compatibility, Deprecation, and Migration Plan

• What impact (if any) will there be on existing users?
• If we are changing behavior how will we phase out the older behavior?
• If we need special migration tools, describe them here.
• When will we remove the existing behavior?

Rejected Alternatives

If there are alternative ways of accomplishing the same thing, what were they? The purpose of this section is to motivate why the design is the way it is and not some other way.