Date: Wed, 23 Jun 2021 10:21:38 +0000
From: Telford Tendys <psql@lnx-bsp.net>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>
Subject: Re: Unicode FFFF Special Codepoint should always collate high.
Message-ID: <20210623102138.GA18446@mail>
References: <20210622083918.GA12063@mail>
 <CA+hUKGKnYyS01bVmY7p6ozBgB0ZA=ReSkyWXG5bbHUt6fh-aig@mail.gmail.com>
 <20210623032418.GB12063@mail>
 <CA+hUKGJF0C0_rch1emO7id0QxDNa2s0pPtQCVjuTPM3mqLs+hg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CA+hUKGJF0C0_rch1emO7id0QxDNa2s0pPtQCVjuTPM3mqLs+hg@mail.gmail.com>
OpenPGP: id=0A153029; url=http://lnx-bsp.net/pubkey.text;
 preference=signencrypt
User-Agent: Mutt/1.5.20 (2009-12-10)
Archived-At: 
 <https://www.postgresql.org/message-id/20210623102138.GA18446%40mail>
Precedence: bulk

On 21-06-23 18:18, Thomas Munro wrote:
> Just in case you didn't know, PostgreSQL knows how to convert a prefix
> search for 'aar%' into a range search with a related trick in some
> limited circumstances:

Thanks for the suggestion, I probably should learn those tricks but
for the time being, would prefer to live with a slower search and get it
eventually done the way I want.


> Seems a little light on references and justifications for the
> expectiation.  ICU's results could be useful in the discussion.

I trust those guys, they will figure it out. I strongly predict that
they will keep the behaviour consistent with RHEL 7.

Is there an easy way to make normal Linux glibc utilities (e.g. sort)
use a locale from the ICU library? There's a package availble one Centos-8
here:

    libicu-60.3-2.el8_1.x86_64

Trouble is that only a few applications use it, and I can't find any way
to plug-in / plug-out this functionality. Introducing postgresql details to
the bugzilla ticket will muddy the water and create a aura of diffuse
responsibility. What I've found is generally where there's a lot of words,
people don't read them.