Hi Mykhailo, I've been mulling over your KIP today. I think that what you're proposing makes sense. I'm having a little trouble wrapping my head around the exact problem with the current API, though...
It sounds like you're saying that you want to have different key and value serdes for the same type, which is not possible in conjunction with implicitly substituted serdes. (Because there's no way to express at the type level whether the serde is for keys or not). I guess what I'm struggling with is why you actually want to have different key and serdes for the same type. Maybe this is not relevant to the proposal itself, but I think it would help with the cost/benefit analysis, especially since the proposal would change approx. every method in the streams-scala DSL. I do have some concrete questions, but I'd rather hold onto them until I feel like I understand the situation. Thanks again! -John On Wed, Nov 6, 2019 at 11:25 AM John Roesler <j...@confluent.io> wrote: > > Hey Mykhailo, > > I just wanted to let you know that I'm looking at your proposal, but > it'll take me a little while to re-activate the Scala part of my > brain. > > For everyone's benefit, here's the link to the KIP: > https://cwiki.apache.org/confluence/display/KAFKA/KIP-513%3A+Distinguish+between+Key+and+Value+serdes+in+scala+wrapper+library+for+kafka+streams > > Thanks for starting this discussion! > -John > > On Sun, Oct 20, 2019 at 7:11 AM Михаил Ерёменко <yeryomen...@gmail.com> wrote: > > > > Hi! > > > > I would like to start discussion. > > > > Regards, > > Mykhailo Yeromenko
