svn commit: r345365 - head/bin/date

[Rodney W. Grimes]

> On Thu, Mar 21, 2019 at 9:24 AM Ian Lepore <ian at freebsd.org> wrote:
> 
> > On Thu, 2019-03-21 at 06:47 +0000, Warner Losh wrote:
> > > Author: imp
> > > Date: Thu Mar 21 06:47:23 2019
> > > New Revision: 345365
> > > URL: https://svnweb.freebsd.org/changeset/base/345365
> > >
> > > Log:
> > >   Remove -n flag, fix setting date / time
> > >
> >
> > IMO, the -n flag should still be accepted, but ignored, to provide
> > compatibility with anybody's old scripts or other automation that uses
> > date -n, since the new default behavior is exactly what the old -n flag
> > provided.
> >
> > On the other hand, the fact that date(1) has failed to set the date for
> > about 3 months shows that it isn't a heavily-used command.
> >
> 
> I read this a few hours ago and thought "nope, well, maybe" and set it
> aside. I'm still ambivalent. I've never used -n when setting the date in my
> entire career. I didn't even know what it was when this bug was brought up.

When timed was kicked out of base some people rasied there hands about
the fact they used it, perhaps go back to that and ask them about this
use of -n and if just defaulting to always doing that is or isnt
an issue for them.

> so on the one hand it's easy enough to ignore, on the other we have enough
> ignore for compatibility junk in the tree...

Another item to add to the long list of things to consider in a deprication
policy.

> Warner
-- 
Rod Grimes                                                 rgrimes at freebsd.org