Phillip.Guerra at nkch.org
Wed Feb 22 19:45:47 UTC 2006
I hesitate to reply, as I'm a relative newbie to this list, and I may be
way off base. However, would it not be easier to implement a coding
scheme for this field that maps an occurrence of a day within a month
W-D where W is the Day of the Week, 1 - Sun, 2 - Mon thru 7 - Sat
O is the occurrence of that day within the month 1-1st,
2-2nd, 3-3rd, 4-4th, 5 - Last
1-1 translates to 'the first Sunday'
1-5 translates to 'the last Sunday'
7-3 translates to 'the 3rd Saturday' and so on...
It would seem to remove any misinterpretation. Just a newbie thought...
I, also think that since the historical references are not going to
change with respect to the day, so that it might as well be represented
with the actual day and date.
Again, just a newbie thought...
From: Dave Cantor [mailto:Dave at Cantor.mv.com]
Sent: Wednesday, February 22, 2006 1:27 PM
To: tz at lecserver.nci.nih.gov
Subject: RE: First Sunday
On 22 Feb 2006 at 13:45, Andy Lipscomb wrote:
> And while we're on that subject, there are several lastSuns and a
lastSat in Zone entries (which inherently refer to just one occurrence).
Would it be better to replace those with the appropriate actual days?
Maybe, but that won't work for a rule indicating lastSun in February
which may be in effect both during leap years and during non-leap years.
(I didn't check -- I'm guessing that there are none.)
Groton, CT 06340-3731 Dave at Cantor.mv.com
More information about the tz