I just can't help but adding my 0.01 EUR here again ... What's the use of
that feature (apart from feature-completeness)?
>This will only complicate things on the adaptor layer of cayenne. If it
>were to save round-trips to the database I'd be happy but this isn't the
>case here. So what's the use?
>
>Maybe I'm too puristic here? :-)
For my own needs, it is because Cayenne may not be the only interface into
the DB, so you can't assume that all PK are generated by it. The assumption
y'all make pretty commonly is that Cayenne is the only insertor into the
DB, which is not the case in my project. If it isn't and there are tables
with autoincrement keys, then the DB needs to generate those natively and
have Cayenne access them.
This is a philosophical question that I can't answer. Which is better for
Cayenne as a whole?
1) Requiring that if Cayenne inserts into a DB with autoincrement tables,
no non-Cayenne systems insert (which is how it is now)
2) Complicating the adaptors and doing the bookkeeping of whether keys are
generated internally or externally.
d
This archive was generated by hypermail 2.0.0 : Mon Jan 06 2003 - 16:08:27 EST