The KIP says: > In WindowBytesStoreSupplier, we will add:
Should it be `SessionBytesStoreSupplier` ? What do you mean by > Users customizing the SessionBytesStoreSupplier should not implement this function. From my understanding, this is a breaking change for all users implementing a custom `session window store`. I would expect that there are very few users but the KIP should state this clearly as a breaking API change. Nit: the JIRA link seems to be wrong. -Matthias On 6/28/18 10:12 AM, John Roesler wrote: > +1 > > On Thu, Jun 28, 2018 at 4:39 AM Damian Guy <damian....@gmail.com> wrote: > >> +1 >> >> On Thu, 28 Jun 2018 at 02:16 Ted Yu <yuzhih...@gmail.com> wrote: >> >>> +1 >>> >>> On Wed, Jun 27, 2018 at 4:40 PM, Bill Bejeck <bbej...@gmail.com> wrote: >>> >>>> +1 >>>> >>>> -Bill >>>> >>>> On Wed, Jun 27, 2018 at 7:39 PM Guozhang Wang <wangg...@gmail.com> >>> wrote: >>>> >>>>> Hello folks, >>>>> >>>>> I'd like to start a voting thread on KIP-330. I've intentionally >>> skipped >>>>> the discuss phase since it is a pretty straight-forward public API >>> change >>>>> and should actually be added since day one. The bug fix of KAFKA-7071 >>>>> helped us to discover this overlook. >>>>> >>>>> >>>>> -- >>>>> -- Guozhang >>>>> >>>> >>> >> >
signature.asc
Description: OpenPGP digital signature
