Re: BUG #13442: ISBN doesn't always roundtrip with text

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr>
Cc: Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, bz(at)mailinator(dot)com, PostgreSQL Bugs List <pgsql-bugs(at)postgresql(dot)org>
Subject: Re: BUG #13442: ISBN doesn't always roundtrip with text
Date: 2015-07-27 21:37:01
Message-ID: 17091.1438033021@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Fabien COELHO <coelho(at)cri(dot)ensmp(dot)fr> writes:
> The usual exit strategy is:
> DROP TYPE isbn CASCADE;
> Which works fine... but as a side effect silently remove ISBN columns that
> may exist in user table. I do not think it is a good idea for an extension
> update script to do that to a librarian:-)

Certainly.

> Bar any idea around this issue, I think that an extension upgrade script
> is really unadvisable, and the user should do a dump data, drop extension,
> create extension, restore data.

AFAICS, that isn't exactly an improvement. You're telling the user
"if you have an ISBN column, you're screwed, and we are going to make it
as painful as it can possibly be to get out of that situation".

I think we'd be better off trying to migrate to a situation where these
type names all still exist but they all act like ISBN13. I'm not sure
what the stages on that journey are.

regards, tom lane

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Michael Paquier 2015-07-27 23:54:31 Re: Memory leak in XLOG reader facility when decoding records (caused by WAL refactoring)
Previous Message Fabien COELHO 2015-07-27 21:04:33 Re: BUG #13442: ISBN doesn't always roundtrip with text