On Mar 10, 2019, at 14:14, al davis <[email protected]> wrote:
> 
> Not a surprise ... For that instant, the time is 2 AM, but one second
> later it's 3:00:01.  The exact time of 3 AM does not exist.  It's just
> like 2:30 AM.

Not quite. In the United States, the forward transition looks like this:

[…]
01:59:58
01:59:59
03:00:00
03:00:01
[…]

So 03:00:00 actually does exist. The entire 2 AM hour however is skipped.

(Please note that this applies to forward transitions in the *United States*; 
other jurisdictions apply DST on different dates and at different times, where 
they apply it at all) (not all US jurisdictions apply it either).


> What would you expect it to do with an event scheduled at 2:30 AM, in
> the middle of the hour being skipped?  I am not asking what it does.  I
> am asking what it should do.

That’s an easy one: skip it! 02:30:00 doesn’t exist in that case.


> What should it do for an event scheduled at exactly 2 AM?

Ditto.

The really “interesting” cases are those that arise in the Autumn (‘fall back’) 
transition, where the clock does this:

[…]
01:58:59
01:59:59
01:00:00
01:00:01
[…]

So here, you get *two* 1 AM hours, occurring back-to-back. So, what ought to 
happen to an event that is scheduled to execute at (say) 01:15:00? Should it 
run just the first time? The second? *Both*?

But to answer Rob’s original question: the behavior of RDCatch at a DST 
transition is currently “undefined” (which is a weaselly way of saying “we have 
no idea what it does”).

Cheers!


|---------------------------------------------------------------------|
| Frederick F. Gleason, Jr. |             Chief Developer             |
|                           |             Paravel Systems             |
|---------------------------------------------------------------------|
|         A room without books is like a body without a soul.         |
|                                                                     |
|                                                         -- Cicero   |
|---------------------------------------------------------------------|


_______________________________________________
Rivendell-dev mailing list
[email protected]
http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev

Reply via email to