Re: PostgreSQL Gotchas

On Thu, Oct 06, 2005 at 05:29:14PM -0500, Jim C. Nasby wrote:
> On Thu, Oct 06, 2005 at 12:29:51PM -0700, David Fetter wrote:
> > * Problems that will be fixed in the next version of PostgreSQL.
> > This means that problems get on developers' radar and get fixed.
> > I suppose by some extremely un-generous method of assessment, this
> > could be a gotcha.
> >
> > * Incomplete information, e.g. not mentioning Slony-I as an
> > upgrade path.
>
> From the few times I've emailed the author he seems happy to accept
> better text. I'm sure he'd be happy to mention slony as an upgrade
> option (though it would be best if there was a good doc describing
> this we could point to). Likewise I believe he'd note things that
> will be fixed in a given version (such as 8.1 or 8.2).

I'll draft something up :)

> > * One deviation from the standard that won't be changed: fold-to-lower.
>
> Does the standard even specify which case you fold to?

It specifies fold-to-upper, and that's just wrong. :P

> I agree, this is only a gotcha if you're used to the very
> non-standard behavior in MySQL.
>
> > * Nits so minor as not to be worth addressing (non-optional AS in
> > FROM clauses vs. SQL standard's making that AS optional there)
>
> *shrug* I think it's useful to be able to see why something you
> might expect to work doesn't.

I suppose it's good to have those. There are other things I think of
as bugs, and would like to have addressed internally before I send
them off to gotchas land. PG's incomplete support for DOMAINs, for
example, or the not-quite-it handling of composite types, or things
that are missing (imho) important features...

Cheers,
D
--
David Fetter david(at)fetter(dot)org http://fetter.org/
phone: +1 510 893 6100 mobile: +1 415 235 3778

Remember to vote!