Andrew Dunstan wrote:
>
>
> Bruce Momjian wrote:
> > Tom Lane wrote:
> >
> >> Perhaps an appropriate thing to do is separate out the representation
> >> change from the other new features, and apply just the latter for now.
> >> Or maybe we should think about having two versions of hstore. This
> >> is all tied up in the problem of having a decent module infrastructure
> >> (which I hope somebody is working on for 8.5). I don't know where
> >> we're going to end up for 8.5, but I'm disinclined to let a fairly
> >> minor contrib feature improvement break upgrade-compatibility before
> >> we've even really started the cycle.
> >>
> >
> > I can just have pg_migrator detect hstore and require it be removed
> > before upgrading; we did that already for 8.3 to 8.4 and I am assuming
> > we will continue to have cases there pg_migrator just will not work.
> >
> >
>
> The more things you exclude the less useful the tool will be. I'm
> already fairly sure it will be unusable for all or almost all my clients
> who use 8.3.
Sorry to hear that. You have studied the existing limitations in the
README, right?
I think it is important to report cases where pg_migrator doesn't work,
but I don't think we will ever avoid such cases. We can't stop Postgres
from moving forward, so my bet is we are always going to have such cases
where pg_migrator doesn't work.
I can't imagine losing a huge percentage of pg_migrator users via hstore
changes, especially since you can migrate hstore manually and still use
pg_migrator.
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com
+ If your life is a hard drive, Christ can be your backup. +