Yup, Francis got it right. There are POJOs in Avatica which Jackson
(un)marshals the JSON in-to/out-of and logic which constructs the POJOs
from Protobuf and vice versa.
In some hot-code paths, there are implementations in the server which
can use protobuf objects directly (to avoid extra deserialization).
F21 wrote:
I am not sure what you mean here. The phoenix query server (which is
based on avatica, which is a subproject in the Apache Calcite project)
accepts both Protobufs and JSON depending on the value of
"phoenix.queryserver.serialization".
The server implements readers that will convert the Protobuf or JSON
request into its internal data structures to perform further processing.
On 4/09/2016 7:19 PM, Cheyenne Forbes wrote:
It would be great if this question could be answered.
Thank you,
Cheyenne
