I have two Lisp systems running on a Win2k box. I'm in the UK, our
daylight saving time started last week. They both still decode the
time as GMT and no daylight saving. The Win2k box kind of knows
about BST (it describes the time as `GMT daylight saving' or something).
I think the Win2k box knows it's in the UK.
Is it Windows that's fouled up or is this a bug in both the Lisps
which think they are in the US (and am I right that daylight saving
time starts this coming weekend in the US?)
Our Unix boxes get it right (though an old solaris box, running 5.6,
is confused).
--tim
> I have two Lisp systems running on a Win2k box. I'm in the UK, our
> daylight saving time started last week. They both still decode the
> time as GMT and no daylight saving. The Win2k box kind of knows
> about BST (it describes the time as `GMT daylight saving' or something).
It may not be connected, but this sounds suspiciously similar to a known
Windows bug:
http://msdn.microsoft.com/visualc/headlines/2001.asp
The known bug is about April 1 being a Sunday, and daylight savings starting
on "the first Sunday in April".
Perhaps your problem is somehow similar?
_______________________________________________________________________________
Don Geddis http://shop.goto.com ······@goto.com
VP of Technology, GoTo Shopping Phone 650-403-2220
1820 Gateway Drive, Suite 300, San Mateo, CA 94404 Fax 650-403-2201
In article <···············@tfeb.org>, Tim Bradshaw <···@tfeb.org> wrote:
>I have two Lisp systems running on a Win2k box. I'm in the UK, our
>daylight saving time started last week. They both still decode the
>time as GMT and no daylight saving. The Win2k box kind of knows
>about BST (it describes the time as `GMT daylight saving' or something).
>I think the Win2k box knows it's in the UK.
>
>Is it Windows that's fouled up or is this a bug in both the Lisps
>which think they are in the US (and am I right that daylight saving
>time starts this coming weekend in the US?)
>
>Our Unix boxes get it right (though an old solaris box, running 5.6,
>is confused).
You haven't said what Lisp implementation you're running. But I think most
Lisps expect the OS to supply information about time zones (there's no
standard way to tell a Common Lisp implementation what zone you're in or
when DST changes).
Have the Lisp systems been running since before the change to summer time?
Maybe it only asks the OS for the time zone when it first starts up. There
might be an implementation-dependent function to reinitialize the internal
state of the time decoder, or you might need to restart Lisp.
--
Barry Margolin, ······@genuity.net
Genuity, Burlington, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.
Barry Margolin <······@genuity.net> writes:
>
> You haven't said what Lisp implementation you're running. But I think most
> Lisps expect the OS to supply information about time zones (there's no
> standard way to tell a Common Lisp implementation what zone you're in or
> when DST changes).
Sorry. LispWorks and Allegro, both currentish versions.
>
> Have the Lisp systems been running since before the change to summer time?
> Maybe it only asks the OS for the time zone when it first starts up. There
> might be an implementation-dependent function to reinitialize the internal
> state of the time decoder, or you might need to restart Lisp.
>
No, they've both restarted - I don't trust the machine to stay up for
that long (:-).
My intuition is that it's Windows, but that's just because it's
usually safe to blame Windows for all bad things...
--tim
Tim Bradshaw <···@tfeb.org> writes:
> Barry Margolin <······@genuity.net> writes:
>
> > You haven't said what Lisp implementation you're running. But I think most
> > Lisps expect the OS to supply information about time zones (there's no
> > standard way to tell a Common Lisp implementation what zone you're in or
> > when DST changes).
>
> Sorry. LispWorks and Allegro, both currentish versions.
Shame on you, Tim. How 'bout if I paste this back to you in a few years
and ask you what the referent versions were...?
(not (typep (get-universal-time) 'implementation-version))
> [...] My intuition is that it's Windows, but that's just because it's
> usually safe to blame Windows for all bad things...
Would that we could teach this practice to the unwashed masses.
I think they use the same idiom, but with "Lisp" substituted strategically
in a few places.
Kent M Pitman <······@world.std.com> writes:
>
> Shame on you, Tim. How 'bout if I paste this back to you in a few years
> and ask you what the referent versions were...?
>
> (not (typep (get-universal-time) 'implementation-version))
>
Ah, but I hope that I won't be using Windows in a few years (:-).
ACL 6.0 and LW 4.1.20 professional.
> Would that we could teach this practice to the unwashed masses.
> I think they use the same idiom, but with "Lisp" substituted strategically
> in a few places.
My guess is that since they *both* give exactly the same wrong answer
it's likely to be a common cause. Of course the standard answer its
proibably that all the brackets cause the date to be wrong...
--tim
[ Tim Bradshaw ]
> I have two Lisp systems running on a Win2k box. I'm in the UK, our
> daylight saving time started last week. They both still decode the
> time as GMT and no daylight saving. The Win2k box kind of knows
> about BST (it describes the time as `GMT daylight saving' or something).
> I think the Win2k box knows it's in the UK.
I heard at work that Allegro uses US daylight savings. This finally
led me to hand the time control stuff over to cron.
> Is it Windows that's fouled up or is this a bug in both the Lisps
> which think they are in the US (and am I right that daylight saving
> time starts this coming weekend in the US?)
http://webexhibits.org/daylightsaving/g.html
Martin
--
"An ideal world is left as an exercise to the reader."
-Paul Graham, On Lisp