On Tue, Dec 21, 2010 at 08:50:04PM +0100, Matthias Bolte wrote:
2010/12/21 Eric Blake <eblake@redhat.com>:
On 12/21/2010 11:34 AM, Chris Lalancette wrote:
All, I'll preface this by saying that I'm not 100% sure I'm correct. But I still think there may be an API break that was introduced with the VMware player driver. In include/libvirt/virterror.h, VIR_FROM_VMWARE was added to the *middle* of the enum for virErrorDomain. If any clients of libvirt happen to be using hardcoded numbers, then they will now have the wrong number for everything after VIR_FROM_VMWARE.
virterror.h is public, we should fix it now before the next release,
Absolutely ! BTW I will send a mail about next release suggestion later today, now that my move is over and I'm back online.
because these values are passed over the wire in RPC calls, and if the server on one machine uses different values than the client on another machine, then the client can misinterpret the error values.
We might explicit write out all values of the enum members. That'll make it more obvious that adding in the middle is a bad idea.
agreed too,
Can we add a make syntax-check rule to avoid future breakage, like we have for the XDR protocol?
PS: Sorry, for not noticing this during review of the VMware driver.
Well it happens, but yes an automated extra check for all enums in the public APIs would be good, I think this could be also managed with libvirt-api.xml extracted values, assuming we put the numbers in. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel@veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/