[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Schema define?

--Le mercredi 22 mai 2002 15:52 +0800 bwu@gwcom.com écrivait:

Dear all,

I am new to LDAP and trying to define my own schema for my needs
and I have the following questions:
For example,
        attributeType ( NAME 'name'
                EQUALITY caseIgnoreMatch
                SUBSTR caseIgnoreSubstringsMatch
                SYNTAX{32768} )

This attribute have unique oid but my company haven't get a
Unique oid  from IANA. What oids should I assign so I will not affect the
system.  Do I really need to get one oid for my cmopany?
Will there be some virtual ids that won't use??

It seems to work just as well if you just omit the OID!

Actually that's not legal, since in the RFC2252 syntax the "numericoid" is not optional. I suppose it's better to have one, since everyone says so. The admin guide (section 8.2.1) even admonishes:

Under no circumstances should you use a fictious OID!

So we have requested one for our university (<http://www.iana.org/cgi-bin/enterprise.pl>).

But while waiting to receive our OID, I just leave that space blank in our local.schema and it works perfectly.

I read related rfcs but I still confused about EQUALITY, SUBSTR and
ORDERING  Can someone explain a little bit between EQUALITY and SUBSTR
as well as ORDERING(does it for numerics)?

The "name" should not be crashed in attribute and objectclass, any
suggested or standard  methods to make it easy to read and keep its
meaning but remains unique easily?

On this last point see the admin guide, section 8.2 (<http://www.openldap.org/doc/admin/schema.html#Extending Schema>).

David Olivier
Klebs gardien Alpages CRI courrier brebis Lyon 2 Lumière