bin/62134: strange date -v results

[Danijel Tasov]

The following reply was made to PR bin/62134; it has been noted by GNATS.

From: Danijel Tasov <dt at vosat.de>
To: Edwin Groothuis <edwin at mavetju.org>
Cc: FreeBSD-gnats-submit at FreeBSD.org
Subject: Re: bin/62134: strange date -v results
Date: Sat, 31 Jan 2004 14:07:53 +0100

 On Sat, Jan 31, 2004 at 01:30:43PM +1100, Edwin Groothuis wrote:
 > 	When the date is adjusted to a specific value that doesn't
 > 	actually exist (for example March 26, 1:30 BST 2000 in the
 > 	Europe/London timezone), the date will be silently adjusted
 > 	forwards in units of one hour until it reaches a valid time.
 > 	When the date is adjusted to a specific value that occurs
 > 	twice (for example October 29, 1:30 2000), the resulting
 > 	timezone will be set so that the date matches the earlier
 > 	of the two times.
 > 
 > Loop i=2:
 >     step 1. current date is 2004-01-31
 >     step 2. -v2004y: set current year to 2004, date will become 2004-01-31
 >     step 3. -v2m: set current month to 2, date will become 2004-02-31
 > 	    Unfortunatly this date doesn't exist, so it takes the next
 > 	    best value and date will become 2004-03-01.
 >     step 4. -v17d: set current day to 17, date will become 2004-03-17
 
 Thank you for your answer. My mistake was, that I assumed that
 the execution of all -v flags is atomic. I've read the quoted
 paragraph from date(1), but I would never come to the idea, that
 the -v flags are executed one after another.
 
 Can somebody close this PR please?
 
                                                -DaTa-
 
 -- 
 Jabber: data at jabber.pilgerer.org ICQ: 84074839 e-Mail: dt at vosat.de
 http://wiki.pilgerer.org/ - Die Kommunikations- und Kooperations-   | () ASCII
 plattform fuer Unix-Fans und alle die es werden wollen.  Macht mit! | /\ RIBBON
 # "zieh deinen xterm groesser und fertig..." - Spaceelk