[Libguestfs] [PATCH] New API: is_whole_device

[Matthew Booth]

On Fri, 2013-02-08 at 09:42 +0000, Richard W.M. Jones wrote:
> On Thu, Feb 07, 2013 at 02:24:10PM +0000, Matthew Booth wrote:
> > ---
> >  daemon/devsparts.c   | 27 +++++++++++++++++++++++++++
> >  generator/actions.ml | 17 +++++++++++++++++
> >  src/MAX_PROC_NR      |  2 +-
> >  3 files changed, 45 insertions(+), 1 deletion(-)
> > 
> > diff --git a/daemon/devsparts.c b/daemon/devsparts.c
> > index 9a2ce9a..1939f2a 100644
> > --- a/daemon/devsparts.c
> > +++ b/daemon/devsparts.c
> > @@ -245,6 +245,33 @@ do_part_to_partnum (const char *part)
> >  }
> >  
> >  int
> > +do_is_whole_device (const char *device)
> > +{
> > +  #define PREFIX "/dev/"
> > +
> > +  /* Shouldn't be possible because we already sanity checked our input */
> > +  if (!STRPREFIX (device, PREFIX)) return 0;
> 
> There's no need for this test.
> 
> > +  CLEANUP_FREE char *devpath = NULL;
> > +  if (asprintf (&devpath, "/sys/block/%s/device",
> > +                device + strlen (PREFIX)) == -1)
> > +  {
> > +    reply_with_perror ("asprintf");
> > +    return -1;
> > +  }
> > +
> > +  struct stat statbuf;
> > +  if (stat (devpath, &statbuf) == -1) {
> > +    if (errno == ENOENT || errno == ENOTDIR) return 0;
> > +
> > +    reply_with_perror ("stat");
> > +    return -1;
> > +  }
> > +
> > +  return 1;
> > +}
> 
> This test seems fine.  Is it worth adding S_ISDIR (statbuf.st_mode)?

It's actually a symlink. I'll add that and also a comment describing
what it's looking for, then push it.

Matt