We have an existing database that uses GUID's as it's primary key for across
all tables. The PK's are defined as uniqueidentifier types, and their
default value is 'newid()' -- which is a call to generate a new GUID for
that ID.

My goal is to continue to use this pattern for generation of sequence
numbers for our implementation of OJB. And if I read it right, that means I
want to use the 'database generated sequences' concept in OJB, with the
SequenceManagerNextValImpl Sequence Manager.

Now, on this page here
(http://db.apache.org/ojb/howto-use-db-sequences.html), it talks about
needing to name your database sequences when you create your table (in the
example it is 'UniqueIdentifier') and that name is then used as the
attribute 'sequence-name' for the field-descriptor. But in our situation,
there is really no Sequence defined, with a name associated it, we are
really just using the default value that the table has associated with it.

So, how would I go about merging the concepts together? Can I use the
SequenceManagerNextValImpl and not give the  sequence-name attribute for my
PK field-descriptor -- will it just default then to the value genrated by
the database?

Thanks in advance!
-Andrew


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to