On 05/21/2012 09:09 AM, Kevin Wolf wrote:
Am 21.05.2012 15:59, schrieb Luiz Capitulino:
On Fri, 18 May 2012 19:08:42 +0200
Paolo Bonzini<pbonz...@redhat.com> wrote:
Modified QMP commands
=====================
As we have discussed on the ML, we're not going to extend QMP commands.
I understand your reasoning, and since the beginning I thought this was
something useful to do, but we've already settled for not doing this.
I also think that we shouldn't have exceptions, as in practice this means
we're extending commands anyway. So either, we do it or we don't.
What's the reason for disallowing command extensions? I'd find it rather
silly to maintain ten different functions that all do basically the
same, just that some of them have more optional parameters than others...
How does a client figure out if the command supports the new options? And start
writing the Python code for it before answering... Even if we exposed the
schema via QMP, checking via schema interpretation would be a nightmare.
Regards,
Anthony Liguori
Kevin
