On Tue, Jun 26, 2007 at 12:00:01PM +0100, Richard W.M. Jones wrote:
> Daniel Veillard wrote:
> >>I haven't updated the Python bindings, but will do so next.
> >
> > it's probable that those 2 calles will be handled automatically
> >since the arg is a known structure and the return value is a string.
> >It's a matter of rebuilding the API xml with the new headers once applied
> >and let the generator do its job.
>
> I'll check. I'm really ill today ...
no hurry :-)
> >Can we avoid using asprintf, especially in new code, I don't think the
> >convenience wins over the portability problem, thanks
> >Or it's just a conveninent way to indicate where a new path creation
> >routine should be added...
>
> Even better, can we add asprintf to the code and compile it into libvirt
> if configure.in detects that it is absent from libc? Asprintf is really
> useful...
And then I would not call it asprintf, and would rather have it used
on all arches. We need a routine for building paths dynamically, that's where
we use asprintf in teh code, and we could do very easilly with a vararg function
using a MAX_PATH buffer. Using a specialized function is more in line with
the actual code and allows for a simpler implementation.
> >>--- src/xen_unified.h 30 Apr 2007 16:57:15 -0000 1.3
> >>+++ src/xen_unified.h 26 Jun 2007 09:51:05 -0000
> >>@@ -50,6 +50,9 @@ struct _xenUnifiedPrivate {
> >> * xen_unified.c.
> >> */
> >> int opened[XEN_UNIFIED_NR_DRIVERS];
> >>+
> >>+ /* Canonical URI. */
> >>+ char *name;
> >> };
> >
> > is that the canonical URI or the FQDN ? Maybe the field need to be
> > renamed
> >(and possibly the comment).
>
> It's the canonical URI. For example NULL || "xen" -> "xen:///":
so let's rename the field to canonical_uri !
Daniel
--
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard | virtualization library http://libvirt.org/
veillard@xxxxxxxxxx | libxml GNOME XML XSLT toolkit http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/