RE: AttributeTypeValue and binary

["Ramsay, Ron" <Ron.Ramsay@ca.com>]

Kurt,

Jim did write 'unless an option is present'. He didn't cover the case of
binary syntax, either. Anyway, I like it.

Ron.

-----Original Message-----
From: Kurt D. Zeilenga [mailto:Kurt@OpenLDAP.org]
Sent: Thursday, 25 January 2001 12:12
To: Jim Sermersheim
Cc: mcs@netscape.com; ietf-ldapbis@OpenLDAP.org
Subject: Re: AttributeTypeValue and binary



At 05:52 PM 1/24/01 -0700, Jim Sermersheim wrote:
>My suggested change is the result of trying to keep the existing wording
intact as much as possible, and also not address any issues other than the
wrapper confusion. Another wordy possibility is:
>
>A field of type AttributeValue is an OCTET STRING containing an encoding of
the value syntax of the companion attribute type. When the syntax of an
attribute type defines a string encoding, that encoding is used unless an
option is present in the companion AttributeDescription which overrides the
string-based encoding. For example, if the "binary" option is present, the
value will be the binary encoding of the value.
>
>This addresses the "wrapper" issue, the "default encoding" issue, and the
"future encodings" issue.

This has the problem in that the default encoding for certificate
is binary.  That is, when all attributes are requested and the
entry contains a userCertificate attribute, the attribute should
be returned as userCertificate;binary.