Re: implement asnprintf, take 2

[Jeff Johnston <jjohnstn at redhat dot com>]

Christopher Faylor wrote:
> On Thu, May 03, 2007 at 03:12:24PM -0400, Jeff Johnston wrote:
> > Eric Blake wrote:
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > >
> > > According to Jeff Johnston on 4/30/2007 3:28 PM:
> > > > Hi Eric,
> > > >
> > > >  I'm not seeing an explanation for the following code change you have
> > > > made in vfprintf.c:
> > > >
> > > > @@ -1247,13 +1201,13 @@ number:                 if ((dprec = prec) >= 0)
> > > >                                if (_fpvalue == 0) {
> > > >                                        /* kludge for __dtoa
> > > > irregularity */
> > > >                                        PRINT ("0", 1);
> > > > -                                       if (expt < ndig || (flags & ALT)
> > > > != 0) {
> > > > +                                       if (ndig || flags & ALT) {
> > > >                                                PRINT (decimal_point, 1);
> > > >                                                PAD (ndig - 1, zeroes);
> > > >                                        }
> > > >
> > > > Note the check expt < ndig has been replaced with ndig.
> > > Any other holdups from checking the rest of the patch in once this hunk is
> > > reverted?
> > No, please go ahead.  I was looking at a more clever way to do the 
> > ELIX_LEVEL checking, but it doesn't work.
>
> Do we still need to worry about ELIX anymore, Jeff?

We will continue to maintain it.

-- Jeff J.