Re: humanize_number(3) for dd(1) summary?

[Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>]

> >>>> Can't dd_snprintf() be also optional if it occupies ~2.5KiB?
> >>>
> >>> I can; there's one trade off though:
> >>>
> >>> - the SMALLPROG dd(1) becomes smaller (by about 2.2kiB), which is still
> >>> approx. 200 bytes bigger than the original SMALLPROG dd(1)
> >>>
> >>> - the normal dd(1) becomes bigger (approx. 600B), as we don't factor
> >>> posix/human/quiet summary() anymore.
> >>
> >> Forgot to ask -- which way you prefer? I don't mind, personally.
> >
> > I'd like smaller SMALLPROG one, for people who want dd(1)
> > on rescue media to zap leftover labels on their disks etc.
> 
> okay, last call, this time with an updated dd(1).

Looks fine. Thanks.

Some late comments for minor nits:

> --- args.c    16 Sep 2011 16:06:23 -0000      1.35
> +++ args.c    5 Nov 2011 01:42:13 -0000
> @@ -64,11 +64,13 @@ static void       f_cbs(char *);
>  #ifdef       NO_CONV
>  __dead
>  #endif
> +
>  static void  f_conv(char *);

Probably these lines should be
>> #ifdef       NO_CONV
>> __dead static void   f_conv(char *);
>> #else
>> static void  f_conv(char *);
>> #endif
or leave no newline?

> --- dd.1      22 Dec 2010 09:42:53 -0000      1.23
> +++ dd.1      5 Nov 2011 01:42:14 -0000
> @@ -32,7 +32,7 @@
>  .\"
>  .\"  @(#)dd.1        8.2 (Berkeley) 1/13/94
>  .\"
> -.Dd December 22, 2010
> +.Dd November 05, 2011

wizd(8) would say November 5, 2011 ?

---
Izumi Tsutsui