Possible changes to the output of "find -ls" - reactions?

[James Youngman]

Per the bug described at http://savannah.gnu.org/bugs/?18227

Package: findutils
Version: 4.2.28-1
Severity: normal

The docs say that find -ls behaves as ls -dils. This is not true:

$ find /dev/hda -ls
 1629    0 brw-rw----   1 root     disk              Nov  7 12:23 /dev/hda
$ ls -dils /dev/hda
1629 0 brw-rw---- 1 root disk 3, 0 Nov  7 12:23 /dev/hda
$

Find should print the device numbers for device nodes.

Greetings
Marc


... it looks like the findutils implementation does not quite match
its documentation when the file being examined is a device node.  This
looks to me like a case where it is reasonable to change find's
behaviour.  But is anybody relying on the current format?    There are
other differences between find and "ls -dils" even in the POSIX locale
:-

$ LANG=C; LC_ALL=C; LC_TIME=C; export LC_ALL LC_TIME LANG; f="/dev/hda
/ /home/james/.bashrc"; /bin/ls -dils $f; echo =========; for i in $f;
do find $i -maxdepth 0 -ls; done
    2 1 drwxr-xr-x 28 root  root  1024 Feb 24 08:57 /
 5664 0 brw-rw----  1 root  disk  3, 0 Feb 24 09:34 /dev/hda
672499 4 -rwxr-xr-x  1 james users   52 Sep 10  2000 /home/james/.bashrc
=========
 5664    0 brw-rw----   1 root     disk              Feb 24 09:34 /dev/hda
    2    1 drwxr-xr-x  28 root     root         1024 Feb 24 08:57 /
672499    4 -rwxr-xr-x   1 james    users          52 Sep 10  2000
/home/james/.bashrc



So are there any objections to changes to the format of the output of
"-ls"?   I'm specifically considering adding device numbers and
perhaps changing the alignment.

James.