Re: CATUPDATE confusion? - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: CATUPDATE confusion?
Date
Msg-id 54FB0708.7070200@gmx.net
Whole thread Raw
In response to Re: CATUPDATE confusion?  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: CATUPDATE confusion?  (Stephen Frost <sfrost@snowman.net>)
List pgsql-hackers
On 3/7/15 12:31 AM, Tom Lane wrote:
> Peter Eisentraut <peter_e@gmx.net> writes:
>> On 12/29/14 7:16 PM, Adam Brightwell wrote:
>>> Given this discussion, I have attached a patch that removes CATUPDATE
>>> for review/discussion.
> 
>> committed this version
> 
> Hmm .. I'm not sure that summarily removing usecatupd from those three
> system views was well thought out.  pg_shadow, especially, has no reason
> to live at all except for backwards compatibility, and clients might well
> expect that column to still be there.  I wonder if we'd not be better off
> to keep the column in the views but have it read from rolsuper.

I doubt anyone is reading the column.  And if they are, they should stop.

pg_shadow and pg_user have been kept around because it is plausible that
a lot of tools want to have a list of users, and requiring all of them
to change to pg_authid at once was deemed too onerous at the time.  I
don't think this requires us to keep all the details the same forever.




pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: Improving test coverage of extensions with pg_dump
Next
From: Stephen Frost
Date:
Subject: Re: CATUPDATE confusion?