Re: ALTER TABLE INHERIT vs collations - Mailing list pgsql-hackers

From Tom Lane
Subject Re: ALTER TABLE INHERIT vs collations
Date
Msg-id 11689.1303070959@sss.pgh.pa.us
Whole thread Raw
In response to Re: ALTER TABLE INHERIT vs collations  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
Peter Eisentraut <peter_e@gmx.net> writes:
> On Sat, 2011-04-16 at 18:23 -0400, Tom Lane wrote:
>> Right at the moment, ALTER INHERIT doesn't verify that collations
>> match in a proposed inheritance child.

> It should be prevented, I think.

BTW, on looking through the source to find the cause, I see that the
reason it's not prevented is that MergeAttributesIntoExisting() failed
to check for collation match.  But the other three places in tablecmds.c
that check for child column type matches do check collation too.  So it
seems clear that this was just an oversight in one case.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: MMAP Buffers
Next
From: Radosław Smogura
Date:
Subject: Re: MMAP Buffers