From: Vladimir V. Zolotych
Subject: parse-time
Date: Tue, 08 May 2001 16:54:06 +0000
Message-ID: <3AF824AE.F44833AC@eurocom.od.ua>

Help me please with time functions
(CMUCL 18c).

* (format-universal-time t (get-universal-time))
Tuesday, 5/8/01 07:45:30 pm [-2]
NIL
Ok, no questions.

* (format-universal-time t (encode-universal-time 30 46 19 8 5 2001))
Tuesday, 5/8/01 07:46:30 pm [-2]
NIL
Also good.

* (format-universal-time t (parse-time "May 8, 2001 19:49:01"))
Tuesday, 5/8/01 08:49:01 pm [-2]
                ^^ why an hour later here ?
NIL
* 

What is the matter ? Something with time zone or DST I didn't
understand ?

* (decode-universal-time (parse-time "May 8, 2001 19:52:01"))
1
52
20 <<< ??
8
5
2001
1
T
-2
* 

-- 
Vladimir Zolotych                         ······@eurocom.od.ua

• Re: parse-time Marc Spitzer
• Re: parse-time Pierre R. Mai

From: Marc Spitzer
Subject: Re: parse-time
Date: Tue, 08 May 2001 19:43:46 +0000
Message-ID: <slrn9fginh.15bb.marc@oscar.eng.cv.net>

In article <·················@eurocom.od.ua>, Vladimir V. Zolotych wrote:
> Help me please with time functions
> (CMUCL 18c).
> 
> * (format-universal-time t (get-universal-time))
> Tuesday, 5/8/01 07:45:30 pm [-2]
> NIL
> Ok, no questions.
> 
> * (format-universal-time t (encode-universal-time 30 46 19 8 5 2001))
> Tuesday, 5/8/01 07:46:30 pm [-2]
> NIL
> Also good.
> 
> * (format-universal-time t (parse-time "May 8, 2001 19:49:01"))
> Tuesday, 5/8/01 08:49:01 pm [-2]
>                 ^^ why an hour later here ?
> NIL
> * 
> 
> What is the matter ? Something with time zone or DST I didn't
> understand ?
> 
> * (decode-universal-time (parse-time "May 8, 2001 19:52:01"))
> 1
> 52
> 20 <<< ??
> 8
> 5
> 2001
> 1
> T
> -2
> * 
> 
> -- 
> Vladimir Zolotych                         ······@eurocom.od.ua


Could it be daylight savings time has something to do with it?

marc

From: Pierre R. Mai
Subject: Re: parse-time
Date: Thu, 10 May 2001 09:04:25 +0000
Message-ID: <87snidr2w6.fsf@orion.bln.pmsf.de>

"Vladimir V. Zolotych" <······@eurocom.od.ua> writes:

> * (format-universal-time t (get-universal-time))
> Tuesday, 5/8/01 07:45:30 pm [-2]
> NIL
> Ok, no questions.
> 
> * (format-universal-time t (encode-universal-time 30 46 19 8 5 2001))
> Tuesday, 5/8/01 07:46:30 pm [-2]
> NIL
> Also good.
> 
> * (format-universal-time t (parse-time "May 8, 2001 19:49:01"))
> Tuesday, 5/8/01 08:49:01 pm [-2]
>                 ^^ why an hour later here ?
> NIL
> * 

The problem is that CMUCL's parse-time doesn't take into account
daylight-savings time for defaulted formats.  Hence it gets the
universal time it calculates wrong by an hour during DST, for all
formats which don't specify enough time-zone information that
defaulting isn't necessary.

I'll have a look if this can be fixed easily, but parse-time is a
non-trivial code-base, and some rewriting will be necessary, I fear.

Regs, Pierre.

-- 
Pierre R. Mai <····@acm.org>                    http://www.pmsf.de/pmai/
 The most likely way for the world to be destroyed, most experts agree,
 is by accident. That's where we come in; we're computer professionals.
 We cause accidents.                           -- Nathaniel Borenstein