I am trying to use the following approach to get a unique sequence value
without actually persisting an object into the table associated with
the sequence. (I really only want the unique sequence of values. I.e. if
the DB provided an Oracle style sequence, I'd probably just do a SQL
nextval() call on it.). I guess I'm attempting to construct a
psuedo-sequence of sorts with a one-column table that I never insert
into, but has an OJB HL sequence associated with it.
I'm using the approach as given in the OJB Documentation here:
======================================================================
http://db.apache.org/ojb/docu/guides/sequencemanager.html#Force+computation+of+unique+values
Identity oid = broker.serviceIdentity().buildIdentity(Object
newPersistentObject);
=======================================================================
(The persistent object that I'm providing is an instance of
OccHeadingGroupSequenceVO which is given in the repository
class-descriptor below.)
For a single junit test run, I am seeing a sequence of unique values.
However, between runs, I'm seeing the same exact sequence - i.e. it has
the same starting value! I'm not seeing the OBJ_HL_SEQ table being
updated as expected either.
I'm using the following sequence manager:
org.apache.ojb.broker.util.sequence.SequenceManagerHighLowImpl
I am seeing the exact same behavour with both MySQL and HSQL.
Is this correct behavour and I'm just misusing the buildIdentity() call?
Or am I missing something else?
Here is my repository class-descriptor:
===========================================================
<class-descriptor
class="OccHeadingGroupSequenceVO"
table="occ_heading_group_sequence"
>
<!-- Primary Key
$DEFAULT(pk,SEQ:occ_heading_group_sequence_seq) -->
<field-descriptor
id="1"
name="pk"
column="pk"
jdbc-type="INTEGER"
nullable="false"
primarykey="true"
autoincrement="true"
sequence-name="occ_heading_group_sequence_seq"
/>
</class-descriptor>
============================================================
Note that in a second test in which I actually do inserts into the
occ_heading_group_sequence table, everything behaves as expected.
At this point I'm assuming that I may be missing something on a
conceptual level and someone can help me out with the information I've
given. If however, the devil is in the details, let me know and I'll go
there.
Or, if someone has a better approach to what I'm wanting, I'm listening...
Thanks!
John
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
