GNU bug report logs - #10016
ls -lk is wrong

Previous Next

Package: coreutils;

Reported by: "Alan Curry" <pacman-cu <at> kosh.dhis.org>

Date: Fri, 11 Nov 2011 00:04:01 UTC

Severity: normal

Done: Paul Eggert <eggert <at> cs.ucla.edu>

Bug is archived. No further changes may be made.

Full log


View this message in rfc822 format

From: Pádraig Brady <P <at> draigBrady.com>
To: Paul Eggert <eggert <at> cs.ucla.edu>
Cc: 10016 <at> debbugs.gnu.org, Eric Blake <eblake <at> redhat.com>, Alan Curry <pacman-cu <at> kosh.dhis.org>, Jim Meyering <jim <at> meyering.net>
Subject: bug#10016: ls -lk is wrong
Date: Sat, 12 Nov 2011 00:56:29 +0000
On 11/11/2011 09:06 PM, Paul Eggert wrote:
> The long option shouldn't be --kilobyte, since ls -k means
> 1024 not 1000.  So I suppose it should be --kibibyte.
> 
> It's a little tricky, since -k means --block-size=1K
> for df and du as well, and I assume this won't change,
> since df -k and du -k conform to POSIX. (Surely there's
> no need to add --kibibyte to du and df -- why should
> we make df and du more confusing merely because
> ls must be more confusing? :-).
> 
> So does the following sound plausible?
> 
> Add --kibibyte to 'ls', make it equivalent to -k, change
> -k so that it conforms to POSIX, and have --block-size
> override -k.  But leave df and du alone

I'm reluctant to add a new option which no one will really use.
But I concur, given the hits from:

http://codesearch.google.com/#search/&q=ls\%20.*--block%20lang:^shell$&p=1&type=cs

-k really isn't used in that context, and replacing --block with -k
in the above query returns no hits.

cheers,
Pádraig.




This bug report was last modified 13 years and 253 days ago.

Previous Next


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