On Mon, Apr 28, 2008 at 2:24 PM, Zeugswetter Andreas OSB SD
<Andreas.Zeugswetter@[EMAIL PROTECTED]
> wrote:
> I think you are not considering existing btree indexes here
> (for the reordering case) ?
You're quite right, I had not considered existing indexes. There's no
easy way to deal with that other than rebuilding them. :(
I *still* think someone with a big table would prefer to drop/create
their indexes rather than go through a nasty ALTER COLUMN which would
have the seemingly much worse outcome of rebuilding their whole table
AND any indexes. But whatever - I'll implement option 1 and submit it,
as a marked improvement over the status quo. If I can make option 2
work fast enough I'll consider submitting it as a feature improvement
thereafter, but given the general consensus for option 1 I'm not
pu****ng option 2 much any more.
Cheers
Tom
--
Sent via pgsql-hackers mailing list (pgsql-hackers@[EMAIL PROTECTED]
)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
