Tom Lane wrote:
> worse --- in the first place there are severe performance issues
> associated with unindexed foreign-key checks, and in the second place
> there is the foot-gun problem that you might forget to re-add the
> indexes at all.
MySQL has a syntax in ALTER TABLE similar to PGSQL's DISABLE/ENABLE
TRIGGER to do the same for KEYS. That way you will not run the risk of
forgetting the INDEX (though you could still forget to enable them).
Potentially there could be a switch in pg_restore to create all KEYS as
disabled initially and automatically enable them after the import is
complete. Note that it could be smart enough to give special treatment
to foreign keys.
In response to
pgsql-hackers by date
|Next:||From: Simon Riggs||Date: 2006-02-24 08:48:31|
|Subject: Re: memory context for tuplesort return values|
|Previous:||From: Michael Glaesemann||Date: 2006-02-24 07:24:15|
|Subject: Re: PostgreSQL unit tests|