Thanks for drafting this. Made some minor edits and added some comments.
Thanks,
Jun
On Thu, Nov 29, 2012 at 3:17 PM, Jay Kreps wrote:
> I started trying to document the 0.8 protocol from the code and write a
> guide to client implementation. This is meant to be a more user-friendly
> and up-t
Not quite. I noticed a few bugs that I pointed out in the previous email. I
would like to correct these today before we officially freeze it. So maybe
give me through Monday? At that point it would be absolutely fantastic to
have someone try to work off the document rather than off our code, and
im
Is the 0.8 protocol pretty much frozen at this point? If so, I'd like to start
updating my python client.
Also, I'm guessing 0.8 will be backwards compatible with 0.7 (due to the API
versioning)
-David
On Nov 29, 2012, at 7:47 PM, Jay Kreps wrote:
> Okay here is a proposal to address the issu
Okay here is a proposal to address the issues I raised.
1, 2. Correlation id. This is not strictly speaking needed, but it is maybe
useful for debugging to be able to trace a particular request from client
to server. So we will extend this across all the requests.
3. For metadata response I will t
I started trying to document the 0.8 protocol from the code and write a
guide to client implementation. This is meant to be a more user-friendly
and up-to-date version of the proposal wiki we had on the protocol changes.
Here is what I wrote up so far:
https://cwiki.apache.org/confluence/display/K
