GNU bug report logs - #16061
Error in the message for src/shuf.c:73 in 8.22-pre3

Previous Next

Full log

Message #23 received at 16061 <at> debbugs.gnu.org (full text, mbox):

From: Pádraig Brady <P <at> draigBrady.com>
To: Bernhard Voelker <mail <at> bernhard-voelker.de>
Cc: Philipp Thomas <pth <at> suse.de>, Assaf Gordon <assafgordon <at> gmail.com>,
 16061 <at> debbugs.gnu.org
Subject: Re: bug#16061: Error in the message for src/shuf.c:73 in 8.22-pre3
Date: Fri, 06 Dec 2013 01:45:23 +0000

On 12/06/2013 01:36 AM, Bernhard Voelker wrote:
> On 12/06/2013 02:26 AM, Bernhard Voelker wrote:
>> On 12/06/2013 02:13 AM, Pádraig Brady wrote:
>>> I suppose we could make -r require that -n is specified,
>>> but I'm not sure.
>>
>> Regarding this in the above context - no, because then you would
>> have to specify -n1 when "picking 1 random item". Saving the user
>> from typing -n1 seems to be _the_ reason for assuming n=1.
> 
> OTOH there may be cases where you don't know the number of items
> beforehand, but you want shuf to output that number of random items
> from the input regardless of repetitions:
> 
>   $ gen_data | shuf -r
> 
> Shuf will know the number of elements and may easily pick that
> number incl. repetitions.
> 
> If you want to pick just one, you know it and it's easy to say -n1,
> but it's not that easy in the above case.

But with random selection would there ever be a link
between input and output cardinality?
I can't think of a case that there would be.

Pádraig.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.