Re: Case-Insensitve Text Comparison - Mailing list pgsql-hackers

From Andrew Sullivan
Subject Re: Case-Insensitve Text Comparison
Date
Msg-id 20080602182412.GD67993@commandprompt.com
Whole thread Raw
In response to Re: Case-Insensitve Text Comparison  (Jeff Davis <pgsql@j-davis.com>)
Responses Re: Case-Insensitve Text Comparison  (Tino Wildenhain <tino@wildenhain.de>)
List pgsql-hackers
On Mon, Jun 02, 2008 at 10:29:30AM -0700, Jeff Davis wrote:

> What if you had a CHECK constraint that was locale-sensitive? Would the
> constraint only be non-false (true or null) for records inserted under
> the same locale? That's not very useful.

It would seem that this is one of the important cases that needs to be
worked out.  I wasn't suggesting that per-session locale (or whatever
we want to call it) is _easy_ or, for that matter, even possible; just
that it would solve a large number of the problems that people
complain about.

In fact, I suspect that what we really need is something a little more
like "in-database locale" or something.
> I think if you want some special treatment of text for some users, it
> should be explicit. 

Yes.  Also, not just text.  Think of currency, numeric separators, &c.

A

-- 
Andrew Sullivan
ajs@commandprompt.com
+1 503 667 4564 x104
http://www.commandprompt.com/


pgsql-hackers by date:

Previous
From: "Pavel Stehule"
Date:
Subject: Re: Proposal: new function array_init
Next
From: Stefan Kaltenbrunner
Date:
Subject: Re: Where can I find the doxyfile?