I'm not saying it does work -- I'm saying it should be made to work. :)
On 9/6/05, Gentry, Michael (Contractor) <michael_gentr..anniemae.com> wrote:
> Ah. I've always ignored that since I don't have Oracle.
>
> So, if it is checked, have PostgreSQL adapter use sequences, otherwise
> use PK table? I'd think the GUI would need to be tweaked, too (an
> Oracle part and a PostgreSQL part)?
>
> Thanks,
>
> /dev/mrg
>
>
> -----Original Message-----
> From: Mike Kienenberger [mailto:mkienen..mail.com]
> Sent: Tuesday, September 06, 2005 11:09 AM
> To: cayenne-deve..bjectstyle.org
> Subject: Re: Cayenne 1.2 + PostgreSQL
>
>
> Second half of the DbEntity "Entity tab"
> "Customize PK generation"
>
> On 9/6/05, Gentry, Michael (Contractor) <michael_gentr..anniemae.com>
> wrote:
> > I don't know ... Is that "to sequence or not to sequence" in the
> > modeler? I don't see it in mine. I don't have Oracle, either.
> >
> > I'm not opposed to the sequences, I just think there should be
> > relatively transparent support for the auto_pk_support table and/or
> > sequences, especially for existing data.
> >
> > /dev/mrg
> >
> > -----Original Message-----
> > From: Mike Kienenberger [mailto:mkienen..mail.com]
> > Sent: Tuesday, September 06, 2005 10:59 AM
> > To: cayenne-deve..bjectstyle.org
> > Subject: Re: Cayenne 1.2 + PostgreSQL
> >
> >
> > Why can't PSQL use the same "to sequence or not to sequence" option
> > preferences that are available for Oracle?
> >
> > On 9/6/05, Gentry, Michael (Contractor) <michael_gentr..anniemae.com>
> > wrote:
> > > Or just have an option in the modeler to choose which scheme to use
> > and
> > > let the adapter use the original or new scheme based on the setting?
> > >
> > > /dev/mrg
> > >
> > >
> > > -----Original Message-----
> > > From: Andrus Adamchik [mailto:andru..bjectstyle.org]
> > > Sent: Tuesday, September 06, 2005 10:52 AM
> > > To: cayenne-deve..bjectstyle.org
> > > Subject: Re: Cayenne 1.2 + PostgreSQL
> > >
> > >
> > > Right, if you go from 1.1 to 1.2 on PostgreSQL, you'd have to create
> > > PK sequences. I guess what we can do to make it smoother is
> extending
> > > sequence creation syntax to pick current max value per CAY-314:
> > >
> > > select setval('pk_my_table', coalesce(max(id) + 1, 200)) from
> my_table
> > >
> > > Andrus
> > >
> > >
> > > On Sep 6, 2005, at 10:39 AM, Gentry, Michael ((Contractor)) wrote:
> > >
> > > > I've been flopping between 1.1 and 1.2 and when I tried running
> one
> > of
> > > > my demo apps, written using 1.1, under 1.2 with PostgreSQL it
> > > > completely
> > > > bombed. I couldn't do INSERTs at all because it kept doing a
> SELECT
> > > > nextval('pk_table') to try to get primary keys. Well, being a 1.1
> > > > application, I didn't have sequences in my database, I had the
> > > > auto_pk_support table for primary keys.
> > > >
> > > > Does this mean if someone has something written in 1.1/PostgreSQL
> > and
> > > > then tries to use 1.2, they have migration problems?
> > > >
> > > > Thanks,
> > > >
> > > > /dev/mrg
> > > >
> > > >
> > >
> > >
> >
>
This archive was generated by hypermail 2.0.0 : Tue Sep 06 2005 - 11:25:50 EDT