BUG #2448: Incorrect UTF-8 ILIKE behaviour (similar to #2261)

The following bug has been logged online:

Bug reference:      2448
Logged by:          Ryszard Lach
Email address:      rla@debian.org
PostgreSQL version: 8.1.3
Operating system:   Debian
Description:        Incorrect UTF-8 ILIKE behaviour (similar to #2261)
Details:

I'm sorry if it is a duplicate of Bug #2261, I couldn't find a postgresql
bug tracking system, so could not check its status.

My environment:

initdb launched in LANG=pl_PL.UTF-8
database in UTF-8

I don't know how to quote polish letters here, so let's say I have a letter
'x' which is oacute and has polish upper case counterpart 'X' (upper-case
oacute).

Testing queries are:

select 'x' like lower('X');

returns true, and

select 'x' ilike 'X';

returns false.

First query proves that there is correct mapping between 'x' and 'X'
available, so why the second one cannot use it?

R.