On Sun, 4 Nov 2001, Rich Bowen wrote:
>On Sat, 3 Nov 2001, Nick Tonkin wrote:
>> So does this mean that you would not support extending Date::ISO to
>> provide a method to output a date in YYY-MM-DD format by default? It
>> sounds like it does. In that case I would have to think that
>> creating Date::MySQL would be appropriate.
>
>No, it does not mean that at all. Did you see my next note?
>
>Allow me to repeat.
>
>If it seems clear that nobody uses the YYYY-WW-D format, which does
>indeed seem to be the case, it makes a lot of sense to modify Date::ISO
>to output the YYYY-MM-DD format instead.
>
>My only concern is that changing the API breaks actual production code.
>
>However, since there was a *ghastly* bug in Date::ISO for about 3 months
>before anyone mentioned it to me, it seems likely to assume that the
>module is not exactly getting thousands of users.
>
>I am almost done writing my book (God be praised) and will put this on
>my list of things to work on when that is out of the way. It's on my
>todo list in my Palm.
>
>> On Fri, 2 Nov 2001, Rich Bowen wrote:
>> > On Sat, 3 Nov 2001, Ilmari Karonen wrote:
>> > > On Fri, 2 Nov 2001, Nick Tonkin wrote:
>> > > > Nevertheless, in order to smooth the ruffled feathers of (my fellow) Brits
>> > > > and other Euros, I shall change my module so that it is required to
>> > > > provide a format specification. I only have 'us' and 'eu' at this
>> > > > point: I suppose 'iso' would be redundant since that's how MySQL handles
>> > > > dates.
>> > >
>> > > May I suggest 'epoch' and 'ical' as additional formats?
>> > >
>> > > Mind you, I think it might be better if you could work with the author
>> > > of Date::ISO to combine the features of the two modules. It seems that,
>> > > for some reason, Date::ISO has no method to directly produce an ISO date
>> > > string, even though it can parse them. With the addition of such a
>> > > method, Date::ISO should be able to do most of what your module does. I
>> > > have no idea to about the internal error checks in the two modules, but
>> > > I see no reason why those could not be combined.
>> >
>> > I was sure that there was an iso method, which output the iso formatted
>> > date.
>> >
>> > rbowen@rhiannon:~% perl -MDate::ISO -le 'my $d=Date::ISO->new( epoch =>
>> > time ); print $d->iso;'
>> > 2001-W44-6
>> >
>> > Unfortunately, the "default" ISO date format is this year-week-day
>> > format. At least that's what I gathered from all the web sites that I
>> > read about this format.
I disagree.
In ISO 8601:1986, the basic format for the complete representation of a
date is 19991231 and the extended format for a date is 1999-12-31 (see
section 5.2.1 Calendar Date and in particular 5.2.1.1 Complete
Representation).
There is discussion about week numbers, but it occurs much later in the
standard (section 5.2.3 Date Identified by calendar week and day numbers).
[I know there is also ISO 8601:2000, but I don't have a copy of it to
quote from, and I understand that the changes are minor.]
As to web sites that document this, the best I know is:
http://www.cl.cam.ac.uk/~mgk25/iso-time.html
The opening section launches straight into YYYY-MM-DD format.
--
Jonathan Leffler #include <disclaimer.h>
STSM, IBM Data Management Solutions. Phone: +1 650-926-6921
Email: [EMAIL PROTECTED], [EMAIL PROTECTED]
Guardian of DBD::Informix v1.00.PC1 -- http://dbi.perl.org
"I don't suffer from insanity; I enjoy every minute of it!"
