bug#23110: seq apparent bug

[Pádraig Brady]

On 07/04/16 15:40, Bernhard Voelker wrote:
> reopen 23110
> thanks
>
> On 04/07/2016 11:46 AM, Ruediger Meier wrote:
> > I understand that this issue is not a bug. But it wouldn't be also not a
> > bug if coreutils would behave like BSD:
> >
> > $ seq 1 0 10 ; echo $?
> > seq: zero increment
> > 1
>
> Ah, okay: as there's prior art in the BSDs [0], the attached
> proposes the same for GNU coreutils.
>
> [0] http://ftp.netbsd.org/pub/NetBSD/NetBSD-release-7/src/usr.bin/seq/seq.c

<Resend since I don't see this on the lists after a day>

I see FreeBSD has seq since 9.0.

I agree that we should avoid repeating output with "0" STEP.

The patch is fine as is and good to push.

Do we want to deal with these cases spinning the cpu,
in further patches?

   seq 1 nan 1
   seq 1 .0000000000000000000000000000001 1

As an aside, I see FreeBSD requires the STEP to be in the right direction
when FIRST != LAST, or it will also exit with error.
GNU will just output nothing in that case.

cheers,
Pádraig.