RE: The 'any' attribute type

[Ron Ramsay <Ron.Ramsay@OpenDirectory.com.au>]

Yes, I have a problem.

The advantage of the use of 1.1 is that it isn't a valid OID. It
therefore requires no special processing as it will never be found in
the directory. Maybe you could try something like unix wildcard matching
where *;ja will match <anything>;ja.

The problem with 1.1.1 is it can only be used once. You really need a
taxonomy:

iso(1)
    invalid-attr(1)
    kludge(2)
        a-silly-idea(1)
        another-silly-idea(2)
            being-silly-this-way(1)
            being-silly-that-way(2)
        placeholder(3)
    kludge(3)

(You can't use 1.1.1 having already terminated at 1.1.)

(Got DAP?)

> -----Original Message-----
> From:	Jim Sermersheim [SMTP:JIMSE@novell.com]
> Sent:	Friday, August 27, 1999 10:17 AM
> To:	ietf-ldapext@netscape.com
> Subject:	The 'any' attribute type
> 
> Last year, Bruce Greenblatt brought up the notion of using a
> nomenclature like "*;lang-ja" in order to specify that 'all'
> attributes which are stored using the Japanese lang tag be returned
> from a search operation. There was a problem with that, since the * is
> a special character which represents an attribute description string
> (not just an attribute type).
> 
> We're at a point where we need to define a new attribute type option,
> and it would be really nice if this functionality existed. Currently,
> the special (retired) oid 1.1 is used to specify 'no' attributes
> should be returned. Does anyone see a problem with introducing a new
> special oid (maybe 1.1.1) which represents 'all' attributes should be
> returned?
> 
> Bruce's example would look like: 1.1.1;lang-ja
> 
> Jim