Re: [calsify] JSCalendar: fractional seconds

[Ken Murchison <[e-mail]>]

Would it make sense to change the parameter name to SUBSECOND as its a 
known term?

This also would allow extending recurrence rules to have a matching 
SUBSECONDLY frequency if needed.  As stated early in the thread, a 
libical user wants to extend RRULE to do exactly this.


On 6/6/19 9:07 AM, Robert Stepanek wrote:
> This is a draft specification, happy about any input:
>
> 2.1.  FRACSEC parameter
>
>    Parameter name:  FRACSEC
>
>    Purpose:  This parameter is used to define fractional seconds for
>       time values and durations.  It is meant to preserve time precision
>       on time values and duration with sub-second precision, without
>       increasing the time value range within iCalendar.
>
>    Description:  This parameter MAY be specified on properties of type
>       DATE-TIME or DURATION.  The integral part of the float value MUST
>       be zero.  The value MUST NOT be negative.  iCalendar
>       implementations MAY ignore this parameter in date time arithmetic.
>       Implementations MUST ignore presence of the FRACSEC parameter on
>       RECURRENCE-ID properties when determining recurrence overrides.
>       If present on a RECURRENCE-ID property, its value MUST match the
>       FRACSEC parameter value on the DTSTART property.
>
>    Format Definition:
>
>    This parameter is defined by the following notation:
>
>                                        fracsec-param = float
>
>    Example:  DTSTART;FRACSEC=0..03:20190605T133015
>
>
>
> On Thu, Jun 6, 2019, at 1:56 PM, Ken Murchison wrote:
>> As a related topic, a libical user/developer submitted a PR to add 
>> millisecond precision to time values. I haven’t committed it because 
>> it simply changes the DATETIME values which certainly isn’t backwards 
>> compatible. I suggested a new VALUE type but even that isn’t 
>> compatible with unaware clients.
>>
>> Regardless, there does appear to be *some* desire to have fractional 
>> seconds support in iCalendar.
>>
>> --
>> Kenneth Murchison
>> Cyrus Development Team
>> FastMail US LLC
>> [e-mail]
>>
>>
>>
>> On Thu, Jun 6, 2019, at 8:04 AM, Marten Gajda wrote:
>>>
>>> Is this parameter is meant to preserve the fractional seconds for 
>>> round-trips or are iCalendar clients encouraged to use the value?
>>>
>>> I mean, would a client that's aware of X-FRACSEC have to match 
>>> RECURRENCE-IDs ,RDATES and EXDATES taking the fractional seconds 
>>> into account?
>>>
>>> I think we should advise against that and make clear that this is 
>>> not meant to increase the precision of times in iCalendar, but 
>>> merely to preserve it.
>>>
>>> Btw, how do we cope with unaware clients doing partial updates? I.e. 
>>> a client which is not aware of X-FRACSEC adds a new 
>>> recurrence-override (not adding X-FRACSEC to RECURRENCE-ID). When 
>>> being converted back to JSCalendar, this would result in an 
>>> additional instance instead of an override.
>>>
>>> Marten
>>>
>>> Am 06.06.19 um 13:49 schrieb Robert Stepanek:
>>>> It just occurred to me that the current JSCalendar RFC draft 
>>>> ambiguously defines if fractional seconds are allowed in the 
>>>> LocalDate time type.
>>>>
>>>> To be clear: we intend to allow fractional seconds in the start 
>>>> property (and any other LocalDate property). The next version of 
>>>> the RFC draft will define this more clearly.
>>>>
>>>> Unfortunately, fractional second date-times and durations can not 
>>>> be round-tripped out of the box with iCalendar. We will define an 
>>>> iCalendar extension parameter in the informational guide for 
>>>> mapping iCalendar and JSCalendar 
>>>> (draft-ietf-calext-jscalendar-icalendar):
>>>>
>>>>   * X-FRACSEC of value type INTEGER
>>>>   * This parameter MAY be set on iCalendar properties of type
>>>>     DATE-TIME or DURATION. It MUST NOT be set more than once per
>>>>     property.
>>>>   * The value of this parameter defines the fractional seconds of
>>>>     the DATE-TIME or DURATION type.
>>>>   * This will allow iCalendar implementations  not being aware of
>>>>     this extension parameter to display the time value "good
>>>>     enough" with second precision.
>>>>
>>>>
>>>> Please let me know if you prefer another approach.
>>>>
>>>> Cheers,
>>>> Robert
>>>>
>>>> _______________________________________________
>>>> calsify mailing list
>>>> [e-mail]  <mailto:[e-mail]>
>>>> https://www.ietf.org/mailman/listinfo/calsify
>>>>
>>> -- 
>>> Marten Gajda
>>> CEO
>>>
>>> dmfs GmbH
>>> Schandauer Straße 34
>>> 01309 Dresden
>>> GERMANY
>>>
>>> phone: [phone]
>>> email:[e-mail]  <mailto:[e-mail]>
>>>
>>> Managing Director: Marten Gajda
>>> Registered address: Dresden
>>> Registered No.: AG Dresden HRB 34881
>>> VAT Reg. No.: DE303248743
>>> _______________________________________________
>>> calsify mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/calsify
>>>
>>
>> _______________________________________________
>> calsify mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/calsify
>>
>
>
> _______________________________________________
> calsify mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/calsify

-- 
Ken Murchison
Cyrus Development Team
FastMail US LLC