Re: [calsify] SKIP was Re: AD review of draft-ietf-calext-rscale-03

[Barry Leiba <[e-mail]>]

I think those descriptions are clear, and they look correct and complete to
me.  I'd be happy with them if the working group agrees.

But let me present an alternative formulation:

1. If the date is invalid because the month is not a valid month in the
currect year, the SKIP is to the next (forward) or previous (backward)
valid month.

2. Otherwise, the SKIP is to the next (forward) or previous (backward)
valid day.

I think that covers the situation.  If you think you need to explain the
different "day" situations further, which is never a bad thing, that can be
added as explanatory text.  Yes/no?

Barry

On Monday, February 2, 2015, Cyrus Daboo <[e-mail]> wrote:

> Hi,
> (Dealing with Barry's AD review issues #2 and #8 in a separate thread)
>
> OK, so I think you have identified a legitimate issue with SKIP. I would
> like to clarify the skip behavior as described below. I will propose
> concrete changes to the draft once we have agreed on the basic concepts.
>
> When a recurrence generates an invalid date, the following applies when
> SKIP is set to FORWARD or BACKWARD:
>
> 1) If the invalid date was generated from a rule starting on a leap day or
> a leap month: the SKIP value for forward/backward is then a day or a month,
> respectively.
>
> 2) If the invalid date was generated from a rule starting on a
> day-of-month not valid in the current month: the SKIP value for
> forward/backward is to the nearest valid day.
>
> 3) If the invalid date was generated from a rule with a BYMONTHDAY that
> results in a day-of-month not valid in the current month: the SKIP value
> for forward/backward is to the nearest valid day.
>
> Note that the above "rules" for SKIP are independent of the actual RSCALE
> value, which I believe is the correct approach (because otherwise we would
> need some kind of registry or an indicator from ICU as to what the behavior
> should be for each calendar scale).
>
> Can other RRULE experts please review the above and make sure those three
> descriptions are valid?
>
> --
> Cyrus Daboo
>
>