Follow-up Comment #1, bug #16214 (project coreutils):
Hmm. It seems -- and this is thanks to Michael Polyemakos and Stephen Harris
-- what I found isn't a bug, but an undocumented (I think) "feature":
a->i k->m == +1 -> +12 hours
n->y == -1 -> -12 hours
ie the result of TZ=UTC date -d "12:00 $v" where v changes is:
a --> Wed Mar 29 13:00:00 UTC 2006
b --> Wed Mar 29 14:00:00 UTC 2006
c --> Wed Mar 29 15:00:00 UTC 2006
d --> Wed Mar 29 16:00:00 UTC 2006
e --> Wed Mar 29 17:00:00 UTC 2006
f --> Wed Mar 29 18:00:00 UTC 2006
g --> Wed Mar 29 19:00:00 UTC 2006
h --> Wed Mar 29 20:00:00 UTC 2006
i --> Wed Mar 29 21:00:00 UTC 2006
k --> Wed Mar 29 22:00:00 UTC 2006
l --> Wed Mar 29 23:00:00 UTC 2006
m --> Thu Mar 30 00:00:00 UTC 2006
n --> Wed Mar 29 11:00:00 UTC 2006
o --> Wed Mar 29 10:00:00 UTC 2006
p --> Wed Mar 29 09:00:00 UTC 2006
q --> Wed Mar 29 08:00:00 UTC 2006
r --> Wed Mar 29 07:00:00 UTC 2006
s --> Wed Mar 29 06:00:00 UTC 2006
t --> Wed Mar 29 05:00:00 UTC 2006
u --> Wed Mar 29 04:00:00 UTC 2006
v --> Wed Mar 29 03:00:00 UTC 2006
w --> Wed Mar 29 02:00:00 UTC 2006
x --> Wed Mar 29 01:00:00 UTC 2006
y --> Wed Mar 29 00:00:00 UTC 2006
My use of "a year ago" is not really valid syntax, then. I was fooled by a
lookalike! :-)
_______________________________________________________
Reply to this item at:
<http://savannah.gnu.org/bugs/?func=detailitem&item_id=16214>
_______________________________________________
Message sent via/by Savannah
http://savannah.gnu.org/
_______________________________________________
Bug-coreutils mailing list
Bug-coreutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-coreutils
