From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43359) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d8Rvl-0004d8-VW for qemu-devel@nongnu.org; Wed, 10 May 2017 09:45:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d8Rvi-0003UV-T7 for qemu-devel@nongnu.org; Wed, 10 May 2017 09:45:30 -0400 Received: from mx1.redhat.com ([209.132.183.28]:50350) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1d8Rvi-0003UP-Mt for qemu-devel@nongnu.org; Wed, 10 May 2017 09:45:26 -0400 From: Markus Armbruster References: <20170505201128.12099-1-ehabkost@redhat.com> <20170505201128.12099-4-ehabkost@redhat.com> <20170505210923.GP3482@thinpad.lan.raisama.net> Date: Wed, 10 May 2017 15:45:23 +0200 In-Reply-To: <20170505210923.GP3482@thinpad.lan.raisama.net> (Eduardo Habkost's message of "Fri, 5 May 2017 18:09:23 -0300") Message-ID: <87ziekzuz0.fsf@dusky.pond.sub.org> MIME-Version: 1.0 Content-Type: text/plain Subject: Re: [Qemu-devel] [PATCH v2 3/3] string-input-visitor: Support alternate types List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eduardo Habkost Cc: Eric Blake , Michael Roth , qemu-devel@nongnu.org Eduardo Habkost writes: > On Fri, May 05, 2017 at 03:53:40PM -0500, Eric Blake wrote: [...] >> Slightly simpler for ignoring int for now, while still something we >> could add in later. I've been wanting to have an alternate for >> 'int'/'str' for InetAddress port, since we want to support named ports >> but most often use just integers. On the command line, port=1 is fine, >> but in QMP, we currently have to spell it port="1". That's a case where >> we'd allow a pairing of any string with an integer, rather than just an >> enum. > > Does that mean we already have an use case where we will have to > relax the restrictions on ambiguous enums? :) A rather weak one. Certainly too weak by itself to make me open another can of string-input-visitor worms. > I won't mind at all if we remove the runtime detection of > ambiguous enums. It will make the code much simpler. That's the idea :)