On 24 Mar 2011, at 23:35, Robbie Gemmell wrote:
Whilst I don't particularly disagree with including either of the
changes
from a risk point of view, I don't really see this is something all
that
necessary for 0.10 either.
It has been this way for the past 2 releases and is thus the only
way anyone
who has ever used it must have done so. I'd just update the
documentation
(which needs to be done anyway due to the previous point), update
it for
0.12 and then note it can be used both ways going forward.
Robbie
-----Original Message-----
From: Justin Ross [mailto:[email protected]]
Sent: 24 March 2011 23:27
To: [email protected]
Subject: Re: [jira] [Updated] (QPID-3156) Java client
implementation of the
Address syntax for Destinations throws a ClassCastException when
used with
the documented 'True' or 'False' values for node durability
Hi, Andrew. (Pounce!) Do you think this change represents high
utility and
low risk, and should therefore go to 0.10?
Justin,
If the tradeoff is utility against risk, then there should be no real
issues with the changes, since the documentation is clearly
incorrect. However, changing the *documentation* for 0.10 instead, to
reflect the reality of the (broken) code would be simpler, and we
could then leave the fix on trunk. This is basically what Robbie said
above, as well.
Andrew.
--
-- andrew d kennedy ? do not fold, bend, spindle, or mutilate ;
-- http://grkvlt.blogspot.com/ ? edinburgh : +44 7582 293 255 ;
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:[email protected]
