Re: [VCARDDAV] Default charset

[Peter Saint-Andre <[e-mail]>]

On 2/22/10 7:58 AM, Cyrus Daboo wrote:
> Hi folks,
> I just came across the following link:
> <http://microformats.org/wiki/vcard-errata>. This claims that RFC2426
> was wrong to remove the CHARSET parameter on properties (which I
> personally do not believe - and no formal IETF errata has ever been
> posted for that). Apparently some clients require CHARSET parameters on
> properties for VERSION:3.0 too:
> <http://help.wugnet.com/office/Outlook-vCard-UTF-ftopict1163209.html>.
> 
> Clearly we need to clean up this mess. Now, the current 4.0 draft does
> state clearly that for a MIME object with a Content-Type header, the
> default charset is UTF-8 (Section 10.1) for the newly registered media
> type text/vcard. What it does not do is explain what charset to use for
> a "standalone" file (dealing with the microformats issue). iCalendar is
> explicit in stating that the default charset for an iCalendar object is
> UTF-8 (<http://tools.ietf.org/html/rfc5545> Section 3.1.4). I believe
> the 4.0 specification should be just as explicit about this.

+1

> Proposed changes:
> 
> - Add a new Section 3.4 with text matching that in RFC5545 Section 3.1.4:
> 
>    3.4.  Character Set
> 
>       There is not a property parameter to declare the charset used in a
>       property value.  The default charset for vCard data is UTF-8
>       as defined in [RFC3629].

The term "default" might be taken to imply that one can override the
default, but that seems not to be the case.

>       The "charset" Content-Type parameter MUST be used in MIME transports
>       to specify the charset being used.

Which is always UTF-8?

> - First paragraph of Section 4.1 needs to be adjusted to remove
> reference to charset given the addition of Section 3.4 as above.
> 
> - Add an "Internationalization Considerations" with text matching that
> in RFC5545 (note this may be stronger than we want in that it mandates
> use of utf-8 only):
> 
>    XX.  Internationalization Considerations
> 
>       Applications MUST generate vCard data in the UTF-8 charset and
>       MUST accept vCard data in the UTF-8 or US-ASCII charset.

Is this indeed stronger than we want? That is, do we ever want any
charset other than UTF-8? I don't think so.

Peter

-- 
Peter Saint-Andre
https://stpeter.im/