On Sun, Apr 20, 2008 at 3:43 PM, Jim Meyering <[EMAIL PROTECTED]> wrote: > > "Olaf van der Spek" <[EMAIL PROTECTED]> wrote: > > On Sat, Apr 19, 2008 at 6:23 PM, Jim Meyering <[EMAIL PROTECTED]> wrote: > >> "Olaf van der Spek" <[EMAIL PROTECTED]> wrote: > >> >> > Could [df] use --human-readable by default? > >> > >> >> > >> >> No, that would break many scripts that get the value from the third > or > >> >> fourth column and compare it to some constant, for example. > >> > > >> > What about doing it only if the output is a terminal? > >> > >> Sorry, but no. > >> coreutils has a policy of not making behavior depend on that. > >> There are a few exceptions, but those are mandated. > > > > Hmm, that makes sense. Still, it'd be nice if humans wouldn't have to > > suffer due to this. > > Isn't there a common option, like --raw or --script, that scripts > > could use to get machine readable output? > > I guess df isn't the only app with this problem. > > Humans can use an alias or shell function to add > the options of their choice.
> Tastes/needs differ. That's certainly true. How many users would prefer the current default over -h and why? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
