durusmail: durus-users: Usage pattern of "gen_oid_record"
Usage pattern of "gen_oid_record"
2006-04-17
2006-04-17
2006-04-18
2006-04-18
Durus documentation (Re: [Durus-users] Usage pattern of "gen_oid_record")
2006-04-19
Re: Durus documentation (Re: [Durus-users] Usage pattern of "gen_oid_record")
2006-04-19
Usage pattern of "gen_oid_record"
David Binger
2006-04-17
On Apr 17, 2006, at 6:37 AM, Jesus Cea wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Durus standard backend defines a "gen_oid_record()" method that
> gives a
> generator that iterates over all registers in the storage. This is
> used,
> I suppose, to upgrade object classes when necessary. Or to do backups.
>
> Problem is, if during the "gen_oid_record()" the client does an object
> commit, my BerkeleyDB backend storage can't garantee that the iterator
> doesn't skip an object or gives an already visited object.
>
> So my question is: How is currently the usage pattern of
> "gen_oid_record()"?. In particular, does the client object commit
> to the
> storage while iterating over the storage content?

Here, when we use gen_oid_record() it is in an update script, and
we do not commit in the middle of the iteration.  I think changing the
index while iterating over the index is not supported.


reply