In Oracle, where sequences are used for PKs, I never set a table for
HI abstract entities. The adaptor generates a sequence with the
entity name, rather than the table name.
On Nov 4, 2008, at 2:49 PM, Mike Schrag wrote:
So Chuck says for Frontbase "It does for FrontBase as the PK
generation sequences as associated with a physical table in
FrontBase. ". That's puzzling to me. i.e., AMPerson is a physical
table. Why does the parent need to map to a physical table at all
for HI?
So for FrontBase, sequences are attached to tables, so you must have
a table to have a sequence. In PG, that's not true. So it's
technically possible that the PG plugin doesn't generate this empty
table and instead just generates the sequence for the parent
entity. But I SUSPECT it's implemented at a lower level ... That's
a total guess, though. It's certainly possible the PG plugin is
smart enough to NOT do this for abstract entity horizontal
superclasses. It's sort of tricky, though, since I suppose you
could have a mix of horizontal and single table inheritance (eek),
at which point it doesn't know that in the future you might not need
that abstract superclass table.
ms
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list ([email protected])
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/webobjects-dev/kenlists%40anderhome.com
This email sent to [EMAIL PROTECTED]
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list ([email protected])
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
