Hi Anastasia,
Is the concern that deserialization would have no guarantee as to the order
> it appears?
>
No, it's simply that when reading the logs screeds of unformatted JSON can
be hard to read, especially when more important fields come later in an
object (or after an array). This is also a p
Hi Tom,
I'm glad it makes more sense. I modified the KIP just to make it a little
more clear as well.
Is the concern that deserialization would have no guarantee as to the order
it appears? Since ObjectNodes hold the tree structure in a LinkedHashMap,
the iterating order to deserialize would be d
Hi Anastasia,
Thanks for those changes. I eventually figured out that the
XYZJsonDataConverter are the new classes from the factoring-out of the
Jackson dependency (KAFKA-10384), after which the proposal made much more
sense to me. Apologies for being slow on the uptake there.
I can see now that
Hi Tom,
Thanks for your input!
1. I'll add more details for the RequestConvertToJson and
XYZJsonDataConverter classes. Hopefully it will be more clear, but just to
answer your question, RequestConvertToJson does not return a
XYZJsonDataConverter, but rather it returns a JsonNode which will be
ser
Hi Anastasia,
Thanks for the KIP, I can certainly see the benefit of this. I have a few
questions:
1. I think it would be helpful to readers to explicitly illustrate the
RequestConvertToJson and XYZJsonDataConverter classes (e.g. with method
signatures for one or two methods), because currently i
