[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

ENCODING/CHARSET



Currently I'm implementing vCard export of address information from an 
application and read carefully more than one time over the specification.

But I'm still confused about the topics "encoding" and "character set". 

If I understand the spec right, the default encoding of a vCard object is 
7-Bit and can be overiden by using the "ENCODING" property parameter.

Q: If my vCard objects contain 8-Bit characters (think of all the 
   German-Umlaut-characters) do I *have* to use this property?

Q: If I use QUOTED-PRINTABLE as encoding, do I have to specify the used
   character set using the CHARSET property (which would be ISO-8859-1
   in my case)?

I've done all this and think my implementation is correct but I observe
a different behaviour of many other applications with vCard support.
E.g. if I use Netscape Messenger and include a vCard with 8-Bit characters
I cannot see any ENCODING or CHARSET property! The vCard is displayed 
correct, though.

Further I noticed incomplete support of the vCard SDK concerning the 
ENCODING property. Against my posting two days ago there *is* support for 
reading vCard objects. 
But if e.g. the N property is QUOTED-PRINTABLE encoded (e.g. like

N:ENCODING=QUOTED-PRINTABLE;CHARSET=ISO-8859-1:Familyame=E4;Surname)

this property gets parsed as F:"Familynameš;Surname" and not as
F:"Familynameš" G:"Surname".

If I omit the ENCODING and CHARSET property from the above example
and use 8-Bit characters instead, the resulting vCard object gets parsed
correct! But that's IMHO not according to the spec :(

Can anyone shed some light on these issues?

Michael
-- 
Michael Wiedmann   office:  <michael.wiedmann@xxxxxxxxx>
                   private: <mw@xxxxxxxxxxxxxxxxxx>