while HTTP-style (string, string) are the most common and most familiar, there is a very significant impact on msg size, especially given that some payloads are literally a few integers (think stock quotes) and would be dwarfed by an http-like header segment.
I think we're ok with not allowing for repeatable headers because: 1. either theyre set by multiple parties that are clashing by mistake - in this case downstream parsers will fail no matter what we do as the payloads in those headers are presumably different 2. or theyre set by the same "component"/system/plugin in which case the owner of the code can read-modify-write the same single header. On Tue, Jan 10, 2017 at 10:38 PM, Ewen Cheslack-Postava <e...@confluent.io> wrote: > not necessarily advocating for supporting this, just suggesting that we > think about it. As you add these features and get closer to mapping to > other systems, people will inevitably try to map them. Headers are an area > where, if we're going to add them, it's worth considering compatibility as > someone will inevitably come and complain that system X does Y with headers > and we should also support Y because any decent system that provides > headers will do so. >
