From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53149) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dXiEP-0004YV-DR for qemu-devel@nongnu.org; Wed, 19 Jul 2017 02:13:10 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dXiEO-0000Jj-5z for qemu-devel@nongnu.org; Wed, 19 Jul 2017 02:13:09 -0400 Date: Wed, 19 Jul 2017 08:12:59 +0200 From: Kevin Wolf Message-ID: <20170719061259.GB4617@noname.redhat.com> References: <1500385286-21142-1-git-send-email-armbru@redhat.com> <20170718160825.GW11927@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170718160825.GW11927@redhat.com> Subject: Re: [Qemu-devel] [PATCH for-2.10 00/10] Correct two minor QMP interface design flaws List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Daniel P. Berrange" Cc: Markus Armbruster , qemu-devel@nongnu.org, eblake@redhat.com, mreitz@redhat.com, qemu-block@nongnu.org, quintela@redhat.com, dgilbert@redhat.com Am 18.07.2017 um 18:08 hat Daniel P. Berrange geschrieben: > On Tue, Jul 18, 2017 at 03:41:16PM +0200, Markus Armbruster wrote: > > blockdev-add and migrate-set-parameters overload empty strings to mean > > something entirely different. See my memo "qapi: Stop abusing > > "special" values for something entirely different" for details. > > > > Message-ID: <87379zhrhn.fsf@dusky.pond.sub.org> > > https://lists.gnu.org/archive/html/qemu-devel/2017-07/msg04526.html > > > > This series deprecates these usages of "" in favour of JSON null. > > Because we're so close to the 2.10 freeze, the implementation is > > intentionally stupid: rewrite null to "" at first opportunity. The > > proper way to do it would be rewriting "" to null, but that requires > > fixing up code to work with null. There are TODO comments for that. > > I'm willing to take care of them in the next development cycle. > > I tested the migrate-set-parameters QMP command with this applied and > I can still pass "" as the tls-hostname. > > When using query-migrate-parameters we still see "" too. > > So from POV of avoiding regressions for existing usage, we look OK > here. Opportunity for a follow-up patch: Automate these tests. Kevin