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

Re: Thinking about operational attribute numSubordinates...



> At 10:19 AM 2001-12-21, Kurt D. Zeilenga wrote:
> >The question to be answered is how to pass the information
> >about.  A hack would be to put a varible in the op structure
> >which was 0 (not present) or 1 (false) or 2 (true) (set by
> >the and have backend_operational() generate the attribute
> >needed.  A better approach would be to think out how such
> >information, in general, should be passed about.
> 
> I should also note that I'm fine with a quick hack for now.
> We likely need to do a few quick hacks in this area before
> we'll have enough operational experience to come up with an
> approach to meet our present and future needs.

I think we could easily leave the decision on what operational
attributes are ahndled to the backend itself, by using 
backend_operational() (as suggested by a remark 
in servers/slapd/result.c, if I remember correctly) and adding
an optional backend routine (be_operational() ?).
The call to backend_operational already passes both the backend 
and the entry; for back-ldbm and back-bdb "hasSubordinates" can
be easily computed on the fly.  This allows the backend to
add any supported operational attributes, if available.  I guess
we'd better avoid this overhead for the attributes that are not 
explicitly required...

Pierangelo.