Grouper Users - Open Discussion List

["Michael R. Gettes" <>]

Yup!  :-)  I am VERY happy to see this getting resolved.  THANK YOU!

/mrg

On Jun 10, 2008, at 13:20, Kathryn Huxtable wrote:

> I'd actually talked with Michael Gettes about doing something like  
> this last fall and then forgot. -K
>
> On Jun 10, 2008, at 11:48 AM, Kathryn Huxtable wrote:
>
> > Well, then, I'll add a regex for the next version. -K
> >
> > On Jun 10, 2008, at 8:57 AM, Tom Zeller wrote:
> >
> > > On Mon, Jun 9, 2008 at 4:01 PM, Tom Barton <>  
> > > wrote:
> > > Kathryn Huxtable wrote:
> > >
> > > When I was at KU, we used a different provisioning mechanism,  
> > > which allowed us to specify a string prefix for provisioning. Our  
> > > provisioning would only manage values in the attribute beginning  
> > > with that prefix, leaving other values alone. This enabled us to  
> > > have multiple provisioning mechanisms provisioning into  
> > > ePentitlement.
> > >
> > > Maybe something like this would be useful in ldappc...
> > >
> > > Yes, or a regex identifying the value space that ldappc "owns".
> > >
> > > Tom
> > >
> > > I've been looking at using ldappc and (coincidentally) just  
> > > yesterday came across this same exception regarding adding an  
> > > already existing objectclass to a Fedora/Red Hat directory server  
> > > entry. In our home-grown provisioning software, we used 'managed  
> > > values' which were evaluated as a regex, similar to TomB's  
> > > suggestion.
> > >
> > > <ProvisionedAttribute id="objectClass" ...
> > >  <ManagedValue>top</ManagedValue>
> > >  <ManagedValue>person</ManagedValue>
> > >  <ManagedValue>organizationalPerson</ManagedValue>
> > >  <ManagedValue>user</ManagedValue>
> > >  ...
> > >
> > > TomZ