On Thu, Dec 04, 2008 at 01:15:23PM +0100, Guido Günther wrote:
On Thu, Dec 04, 2008 at 10:56:25AM +0000, Daniel P. Berrange wrote:
On Wed, Dec 03, 2008 at 06:20:12PM +0100, Guido G?nther wrote:
On Sun, Nov 30, 2008 at 12:43:48PM +0100, Guido Günther wrote:
From 87db4a698ed9b49294c0f94137fc6beef13bd4e8 Mon Sep 17 00:00:00 2001 From: =?utf-8?q?Guido=20G=C3=BCnther?= <agx@sigxcpu.org> Date: Tue, 25 Nov 2008 13:02:43 +0100 Subject: [PATCH] differentiate between active and inactive configs
by honoring the VIR_DOMAIN_XML_INACTIVE flag. O.k. to commit this part as a start so you can readily use it vor lxc?
ACK, assuming 'make check' still passes. It does here, sure. Patch Applied now. I've added the flag to the functions currently needed for qemu but if nobody objects I'd like to add them to:
virDomainDeviceDefParse virDomainDefParseString virDomainHostdevSubsysUsbDefParseXML virDomainDiskDefParseXML virDomainFSDefParseXML virDomainNetDefParseXML virDomainInputDefParseXML virDomainSoundDefParseXML virDomainHostdevDefParseXML virDomainChrDefParseXML
I'm about to release an intermediate 0.5.1 release with the bug and small improvement fixes since 0.5.0, I assume the more generic patch can wait after that, but that sounds right to me in principle. Attached patch adds a flag parameter to the above functions and passes VIR_DOMAIN_XML_INACTIVE from the upper levels. I see two advantages in
On Thu, Dec 04, 2008 at 02:16:06PM +0100, Daniel Veillard wrote: this: * symmetric interfaces to the XML parsing functions (why has virDomainDefParseFile a flags argument while virDomainDefParseString hasn't) * other drivers can use this very easily (IIRC lxc has some use for it) otherwise there's not too much gain so I'm a bit undecided if this is actually necessary. -- Guido