Re: valueencoding production (Was: draft-ietf-ldapbis-filter-03.txt)

[Hallvard B Furuseth <h.b.furuseth@usit.uio.no>]

Kurt D. Zeilenga writes:
> Should <valueencoding> be something like:
> 
>       valueencoding  = 0*(normal / escaped)
>       normal         = UTFN / UTFMB
>       UTFN           = %x01-27 / %x2b-5b / %x5d-7f
>       escaped        = "\" hex hex     
>       hex            = %x30-39 / %x41-46 / %x61-66
> 
> where UTFM is as defined in [Models]:

Hm?  Wasn't the recenlty discussed change done in order to allow
non-UTF8 data (octet strings) as value encodings?  This change would
disallow them again.


BTW, I've just noticed that the requirement to quote %x80-%xff dates
to the -01 version of the draft, not to the octetstring change.
That draft restricts unquoted characters to
      normal  = %x01-27 / %x2b-5b / %x5d-7f
rfc2254 does not do that, it just says that
      value   = AttributeValue from Section 4.1.6 of [rfc2251].

-- 
Hallvard