[Date Prev][Date Next]
Re: How to :
On Friday 18 January 2002 19:24, you wrote:
> At 03:42 AM 2002-01-17, De Leeuw Guy wrote:
> >I would like to carry on this discussion with you.
> >I have been developing LDAP software client since 1999 and the problems
> >encountered are always the same :
> >- missing an attribute specifying the number of "children entries??" of
> > its DN (in order to improve the performance in a tree).
> A future version of OpenLDAP may provide a "has subordinates" boolean
> attribute. For a number of technical reasons, I don't expect
> OpenLDAP to provide a "number of subordinates" attribute.
Maybe a first step in this direction could be a schema flag indicating
that an object mght containother objects. Something like:
(Of course with checking in the server to inhibit the creation of
objects below an object with X-OPENLDAP-CONTAINER == '0')
> >- missing a function able to return the attributes of a class.
> A client can discover the attributes of an object class using
> subschema discover and an then request only these attributes to
> be returned. An extension allowing the client to directly request
> the return of attributes allowed by an object class would be nice.
> In fact, such an extension is in the works (see HEAD sources).
> >- if some attributes are empty, to enable them to inherit the values of
> > the above (parent) class (cases mentioned hereafter)
> > Regarding what we debated recently, don't you think that, in an "object"
> > logic and in certain situations, this functionality speaks for oneself ?
> While not having the semantics you describe, there are X.500
> collective attributes. An extension is being considered
> by the IETF which specifies their use in LDAP. I'm working on
> an implementation for OpenLDAP (but don't expect it to be released
> anytime soon).
Peter Marschall | eMail: firstname.lastname@example.org
Scheffelstraße 15 | email@example.com
97072 Würzburg | Tel: 0931/14721
PGP: D7 FF 20 FE E6 6B 31 74 D1 10 88 E0 3C FE 28 35