[ciapug] calendars

[Scott Phillips]

Well, mktime() produces a 32-bit int containing the number of seconds since 
midnight of Jan 1, 1970, right?  I'm thinking that the presence (or lack) 
of day-time wouldn't be immediately apparent.  That's where the flag 
thought came from.


At 02:34 PM 5/5/2005 -0500, you wrote:
>If no day-time is given, I think it is safe to assume it is all day. 
>However, a flag meaning as much would work fine.
>
>--Tony
>
>Scott Phillips wrote:
>>Good points.  I'm going to complicate matters slightly.
>>I'd like the date to be required but start/end times optional.  How 
>>should I keep track of whether this is an all-day event (no specific 
>>time) or meeting with time specified.  Perhaps an all-day-flag field?
>>Trivial questions?  Probably.  But I like to hear how others approach 
>>things and this list has been real quiet lately!
>>
>>At 01:42 PM 5/5/2005 -0500, you wrote:
>>
>>><snip>
>>>I would say start & duration, since you can always calculate end time 
>>>from the other 2 and it's easier to do offsets than diffs between times.
>>></snip>
>>>
>>>Depends on how you store you date/time fields. I think using native DBMS 
>>>date/time data types are worthless and, instead, we opt to use unix 
>>>timestamps, when possible (since the timestamp can't work well before 
>>>1970 or whatever it is)
>>>
>>>--Tony
>>>_______________________________________________
>>>ciapug mailing list
>>>ciapug@cialug.org
>>>http://cialug.org/mailman/listinfo/ciapug
>>
>>_______________________________________________
>>ciapug mailing list
>>ciapug@cialug.org
>>http://cialug.org/mailman/listinfo/ciapug
>_______________________________________________
>ciapug mailing list
>ciapug@cialug.org
>http://cialug.org/mailman/listinfo/ciapug