hi Normally, this kind of issue needs a "JIRA" and a "KIP". That is a rule of kafka community.
The jira is used to describe your idea briefly, display the fix version, and other details. Also, we can keep tech discussion over there. The KIP is used to discuss the public APIs or big changes. Hence, please file jira for your PR first. Best, Chia-Ping Maxim Fortun <m...@maxf.net> 於 2024年8月15日 週四 下午9:55寫道： > I have described it in a pull request mentioned in the original email. > Pasting below: > [image: 16812.png] > > Feature/add custom produce request parsers by maxfortun · Pull Request > #16812 · apache/kafka <https://github.com/apache/kafka/pull/16812> > github.com <https://github.com/apache/kafka/pull/16812> > <https://github.com/apache/kafka/pull/16812> > > Add ability to specify a custom produce request parser. A custom produce > request parser would allow to intercept all incoming messages before they > get into the broker and apply broker wide logic to the messages. This could > be a trace , a filter, or a transform(such as lineage). > The inline new ProduceRequest(new ProduceRequestData(new > ByteBufferAccessor(buffer), version), version) was moved out of > ProduceRequest.java into its own class ProduceRequestParser.java, and > ProduceRequest class was augmented with the dynamic loading of a specified > parser class. When no class is specified, the default one is loaded > maintaining full backward compatibilty. > > > On Aug 14, 2024, at 2:13 PM, Chia-Ping Tsai <chia7...@gmail.com> wrote: > > I submitted a request(https://issues.apache.org/jira/browse/INFRA-25451) > for > > it last week, but received no response. > > that is a known issue :( > > Maybe you can file a jira first and then create a google doc (or other > online doc) to describe your idea. > > Best, > Chia-Ping > > >
