Re: object class 'alias'

[David Chadwick <d.w.chadwick@salford.ac.uk>]

Date forwarded: 	Mon, 12 Jul 1999 00:05:03 -0700 (PDT)
Date sent:      	Mon, 12 Jul 1999 00:04:51 -0700
To:             	"Kurt D. Zeilenga" <Kurt@OpenLDAP.Org>
From:           	Bruce Greenblatt <bgreenblatt@dtasi.com>
Subject:        	Re: object class 'alias'
Copies to:      	ietf-ldapext@netscape.com
Forwarded by:   	ietf-ldapext@netscape.com

> Kur
> 
> You've misunderstood the spec.  The structural object class of an entry is
> the bottommost one in the object class hierarchy when it is created.

**N.B. This is NOT in fact a correct interpretation. N.B.** The text 
does not state this, although I agree that it is not crystal clear. It 
states

This object class is the most subordinate object
class of the entry's structural object class superclass chain.

It does not state

An entry's structural object class is the most subordinate object 
class in a superclass chain.


e.g. in a hierarchical object class chain say A>B>C>D where A is 
top, C may be the structural object class of one entry and D may be 
the structural object class of another entry. The entry says what its 
structural object class is via the operational attribute  
"structuralObjectClass". Thus the first entry it is of object class A, B 
and C, whereas the second entry is of object class A, B, C and D.

Hope this clarifies things.

> Clause 4.4 is attempting to say that the structural object class of an
> entry ultimately descends from either top or alias.  This confusing
> terminology is taken from the X.500 series.  It is my understanding that
> this particular clause was written in a bar in Rhodes... 

Wish it had been, but unfortunately we never managed to have a 
meeting in Rhodes. (might have been a bar in Pheonix though :-)

>E.g. 8.1.19 of
> x.501 says:
> 
> With respect to a particular entry, the single structural object class
> used to determine the DIT Content Rule and DIT Structure Rule applying to
> the entry. This object class is indicated by the structuralObjectClass
> operational attribute. This object class is the most subordinate object
> class of the entry's structural object class superclass chain.
> 
> Similarly, 8.3.2 says:
> 
> An object or alias entry is characterised by precisely one structural
> object class superclass chain which has a single structural object class
> as the most subordinate object class. This structural object class is
> referred to as the structural object class of the entry.
> 
> To the best of my knowledge, X.500 never gives a term for the other object
> classes in the chain other than abstract.  

In fact they will probably be structural object classes (for different 
entries), as in my example above.
I agree it can be a bit confusing
David

> 
> Bruce
> 
> >PS:
> >
> >In my review I noticed the an inconsistency.
> >RFC2252 implies the "alias" object class is ABSTRACT
> >while RFC2256 declares it as STRUCTURAL.
> >
> >  RFC2252, 4.4, para 1:
> >    In general every entry will contain an abstract class ("top" or
> >    "alias")....
> >
> >  RFC2256, 7.2:
> >    ( 2.5.6.1 NAME 'alias' SUP top STRUCTURAL MUST aliasedObjectName )
> >
> >
> >
> 
> 


***************************************************

David Chadwick
IS Institute, University of Salford, Salford M5 4WT
Tel +44 161 295 5351  Fax +44 161 745 8169
Mobile +44 790 167 0359
*NEW* Email D.W.Chadwick@salford.ac.uk *NEW*
Home Page  http://www.salford.ac.uk/its024/chadwick.htm
Understanding X.500  http://www.salford.ac.uk/its024/X500.htm
X.500/LDAP Seminars http://www.salford.ac.uk/its024/seminars.htm
Entrust key validation string MLJ9-DU5T-HV8J

***************************************************