Re: Semantics of date vs. date/time EXDATEs

["Graham Gilmore" <grahamg@xxxxxxxxxxx>]

----- Original Message -----
From: "John Stracke" <francis@xxxxxxxx>
To: <ietf-calendar@xxxxxxx>
Sent: 28 February, 2001 14:02
Subject: Re: Semantics of date vs. date/time EXDATEs


> Federico Mena Quintero wrote:
>
> > Let's say I have a recurring event.  If I add EXDATE properties,
what
> > is the difference between making them date or date/time values?
>
> Uh...<checks RFC-2445, 4.8.5.1>...it doesn't say.  Good catch.
>
> > Does a date/time value only affect occurrences whose start time is
> > precisely that date/time?  And does a date value affect all
> > occurrences that happen during that day
>
> That's what I would guess would be the expected behavior.  Anybody
else?
>

    I would interpret it that way as well.

> The problem is that EXDATE says:
>
>      excluding any start date and times which fall within the union
>      of start date and times generated by any specified "EXRULE"
>      and "EXDATE" properties.
>
> ...but doesn't define how that union is generated.

    There's a couple of related issues already listed at
http://www.softwarestudio.org/iCal/2445Issues.html :

        - http://www.imc.org/ietf-calendar/mail-archive/msg02600.html ,
where Frank Dawson suggests that mixed DATE & DATE-TIME value types be
prohibited in an iCalendar object.

        - http://www.imc.org/ietf-calendar/mail-archive/msg04949.html ,
where Damon Chaplin explores some interpretations of DATE values for
different properties.

    Both of these are marked as 'open'.  Does that mean there hasn't
been a consensus on them, or just that they haven't been formally added
to the spec yet?

    Graham