On 04.02.2016 18:33, Daniel P. Berrange wrote:
On Mon, Feb 01, 2016 at 03:40:20PM +0300, Nikolay Shirokovskiy
wrote:
> remoteConnectUnregisterCloseCallback is not quite good.
> if it is given a callback function different from that
> was registered before then local part will fail silently. On
> the other hand we can not gracefully handle this fail
> as the remote part is already unregistered.
We could sanity check the callback before unregistering the
remote part. Or you could do the local unregister first since
if the remote part then fails, it is harmless - we'll see the
close event frm the server still, but we won't dispatch it.
I'd prefer sanity check. The second option makes reregistering
impossible (however i doubt this is really a usecase). Will resend
soon.
> There are a lot of options to fix it. I think of totally
> removing the callback argument from unregistering. What's
> the use of it?
We can't remove it since that would change ABI.
Yes, I forgot about this.