I'm working on an asynchronous Clojure protocol buffer client.
Most of Riak's protocol buffer message codes are sequential and alternate between request and response. But some codes are missing. I first noticed this issue in the context of the Java client, but now I am pretty sure this is a Riak issue.
To make my client logic make more sense, I've defined the missing codes as shown below. I starred the problematic message codes.
I consider the code mix-up (above) to be a mistake. Using unique codes for messages is a cleaner way to go, and that is how a bulk of the API works. I admit that only people writing clients would probably notice it.
It would be nice to get it cleaned up if possible.
/cc @rzezeski since the codes are in relation to Yokozuna. I'm setting the milestone to be 2.1 for now since we are too late in the 2.0 cycle to change the API.
I'm working on an asynchronous Clojure protocol buffer client.
Most of Riak's protocol buffer message codes are sequential and alternate between request and response. But some codes are missing. I first noticed this issue in the context of the Java client, but now I am pretty sure this is a Riak issue.
To make my client logic make more sense, I've defined the missing codes as shown below. I starred the problematic message codes.
I consider the code mix-up (above) to be a mistake. Using unique codes for messages is a cleaner way to go, and that is how a bulk of the API works. I admit that only people writing clients would probably notice it.
It would be nice to get it cleaned up if possible.