Thank you all for the explanation. I'll work around the issue. It's nice to understand the thought process even though I might disagree with it.
Sent from my iPhone
On Feb 25, 2012, at 13:10, Guillaume Lelarge <guillaume(at)lelarge(dot)info> wrote:
> On Sat, 2012-02-25 at 09:23 -0700, Scott Ribe wrote:
>> On Feb 25, 2012, at 9:18 AM, Brian Weaver wrote:
>>> Thanks for the pointer. Is it just me that finds it the behavior of pg_restore odd? If the default installation since 9.0 has PL/PgSQL installed then why does pg_restore still emit statements to create the language? As a developer by trade it smells like a bug.
>> It's pg_dump that's emitting the command to create the language. If you ran pg_dump from 9.0+, it would not do so.
> Not quite true. pg_dump from 9.0 does save the language definition, but
> it uses the new CREATE OR REPLACE statement for languages, so that, when
> you restore it in a 9.0+ database that already has the same language, it
> won't complain with an error message.
> BTW, it isn't odd that pg_dump 9.0 save the language definition. Having
> by default the plpgsql language when you create a database doesn't mean
> you can't drop it.
>> This is an example of why the standard advice for upgrading is to use the newer pg_dump against the older database
In response to
pgsql-admin by date
|Next:||From: Wissem||Date: 2012-02-27 10:17:11|
|Subject: Re: Automatic Failover Hot Standby / Streaming replication|
|Previous:||From: Guillaume Lelarge||Date: 2012-02-25 18:10:09|
|Subject: Re: Problems with pg_restore (plpgsql already exists)|