Re: [VCARDDAV] Fwd: Last Call: <draft-daboo-et-al-icalendar-in-xml-08.txt> (xCal: The XML format for iCalendar) to Proposed Standard

On 04/14/2011 06:21 PM, Filip Navara wrote:
> On Thu, Apr 14, 2011 at 5:57 PM, Peter Saint-Andre<stpeter@stpeter.im>  wrote:
>> Just FYI. This document is intended to be in sync with vcardxml
>> regarding XML structure, so folks here might want to review both
>> documents with that in mind...
> That's surely not the case, vcardxml uses different encoding for
> parameters as far as I can see. This was apparently deliberate change
> in vcardxml:
>
> "o  All parameters now use value elements.  This affects type, and pref."
>
> Aside from that there's another problem with the ENCODING parameter.
> iCalendar (RFC 5545) is very vague with the definition and usage of
> the ENCODING parameter and as such it is possible, albeit pretty
> useless, to create the following property:
>    "SEQUENCE;ENCODING=BASE64:MQ=="
We have

  seq = "SEQUENCE" seqparam ":" integer CRLF
        ; Default is "0"

        seqparam   = *(";" other-param)

where

other-param   = (iana-param / x-param)

      iana-param  = iana-token "=" param-value *("," param-value)
      ; Some other IANA-registered iCalendar parameter.

      x-param     = x-name "=" param-value *("," param-value)
      ; A non-standard, experimental parameter.

so I think the above is just invalid.

The encoding param is really just allowed on ATTACH as far as I can see.

Arnaud
> At least one of the leading iCalendar parsing libraries actually has
> such content in the .ics files in test suite.
>
> draft-daboo-et-al-icalendar-in-xml states the following:
>
> "As required by [RFC5545], the ENCODING paramater MUST be present and
> set to "BASE64", and the VALUE parameter must be present and set to
> "BINARY"."
>
> and
>
> "iCalendar uses a base64 encoding for binary data.  The base64
> encoding MUST remain when converted to XML."
>
> The problem is that this is in conflict with the vague definition in
> RFC5545, where ENCODING=BASE64 can be used for non-BINARY values.
>
> Best regards,
> Filip Navara
> _______________________________________________
> VCARDDAV mailing list
> VCARDDAV@ietf.org
> https://www.ietf.org/mailman/listinfo/vcarddav