Re: Doc patch: Document names of automatically created constraints and indexes

Peter Eisentraut <peter_e@gmx.net>

From: Peter Eisentraut <peter_e@gmx.net>
To: "Karl O. Pinc" <kop@meme.com>
Cc: Robert Haas <robertmhaas@gmail.com>, Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@postgresql.org
Date: 2012-11-24T23:02:41Z
Lists: pgsql-hackers
On Wed, 2012-11-21 at 23:56 -0600, Karl O. Pinc wrote:
> No, wait.   If constraint "name_of_primary_key" is an internal
> and is to change over time, how do you deal with dropping,
> now, a primary key constraint that was created, then, before
> some change to the internal name.  And you wouldn't want
> to accidentally remove a user-created constraint
> that just happened to have the same name as the internal
> primary key constraint name, especially in the case
> where there's a real primary key constraint created
> under an old naming convention.

Internally, dependencies are tracked by OID, not by name, so this isn't
a problem.