On Tue, Jun 30, 2015 at 01:48:41PM +0100, Daniel P. Berrange wrote:
On Tue, Jun 30, 2015 at 02:38:24PM +0200, Christophe Fergeau wrote:
> On Tue, Jun 30, 2015 at 01:28:19PM +0100, Daniel P. Berrange wrote:
> > On Tue, Jun 30, 2015 at 01:25:28PM +0100, Zeeshan Ali (Khattak) wrote:
> > > On Tue, Jun 30, 2015 at 12:41 PM, Christophe Fergeau
> > > <cfergeau(a)redhat.com> wrote:
> > > > Hey,
> > > >
> > > > Looks good to me, I'd name the type GVirNetworkDhcpLease rather
than
> > > > GVirNetworkDHCPLease, this is consistent with
> > > > GVirConfigCapabilitiesCpuModel (and not CPUModel).
> > >
> > > Hm.. I was trying to keep it consistent with the underlying
> > > 'virNetworkDHCPLease'.
> >
> > I agree - IMHO acronyms should always be capitalized in our
> > APIs - use of Cpu is a bug (which we can't fix), but we should
> > not add more of such bugs. Likewise MAC, rather than Mac
>
> Fine with me then. Do you mean even in method names, or just in type
> names?
Just types - for method names we use lowercase + underscores everywhere
which is fine as its in keeping with GLib common practice.
We could have some typedef xxxCpuyyy xxxCPUyyy; if we want to 'fix'
these wrong capitalizations.
Christophe