Re: collation problem on 9.1-beta1

From: Marc Cousin <cousinmarc(at)gmail(dot)com>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: collation problem on 9.1-beta1
Date: 2011-06-02 15:28:49
Message-ID: 4DE7AC31.40805@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On 02/06/2011 14:09, Peter Eisentraut wrote:
> On ons, 2011-05-11 at 14:58 -0400, Tom Lane wrote:
>> Marc Cousin<cousinmarc(at)gmail(dot)com> writes:
>>> I've been starting to work on a 'what's new in 9.1' like i did last
>>> year, and am faced with what I feel is a bug, while building a demo case
>>> for collation.
>>
>>> Here it is:
>>
>>> SELECT * from (values ('llegar'),('llorer'),('lugar')) as tmp
>>> order by 1 collate "es_ES.utf8";
>>> ERROR: collations are not supported by type integer at character 74
>>
>> This isn't a bug, or at least not one we're going to fix. ORDER BY
>> column-number is a legacy syntax that doesn't support many options, and
>> COLLATE is one of the ones that it doesn't support.
>>
>> (The actual technical reason for this is that COLLATE turns the argument
>> into a general expression, not something we can special-case. You would
>> get the same error from writing "1 COLLATE something" anyplace else.)
>
> Well, I'm just counting how many more people are going to complain about
> this before fixing it. Possibly in 9.2.
I wasn't complaining at all, just wondering if this was intended or not :)

But I'm sure that I won't be the only one to be caught by this, as it
took me about ten minutes to fall in this trap. That was the main reason
I reported this problem in the first place :)

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Robert Haas 2011-06-02 17:31:33 Re: Unlogged tables cannot be truncated twice
Previous Message Artiom Makarov 2011-06-02 15:21:03 Re: BUG #6048: TRUNCATE vs TRUNCATE CASCADE: misleading message