From hlj@posix.com Thu Oct 15 10:22:55 1992
Received: from netcomsv.netcom.com by dkuug.dk with SMTP id AA06732
  (5.65c8/IDA-1.4.4j for  /usr/spool/uucppublic/uumap/mlist SC22WG15 dkuug.dk /usr/spool/archive/JTC1/SC22/WG15); Fri, 16 Oct 1992 01:32:12 +0100
Received: from posix.COM by netcomsv.netcom.com with UUCP (4.1/SMI-4.1)
	id AA19385; Thu, 15 Oct 92 17:31:28 PDT
Received: by posix.COM (5.64/A/UX-AMR-1.0)
	id AA16809; Thu, 15 Oct 92 17:22:56 PDT
Message-Id: <9210160022.AA16809@posix.COM>
Subject: Re: (SC22WG15RIN.246) substitute
To: sc22wg15@dkuug.dk
Date: Thu, 15 Oct 92 17:22:55 PDT
From: Hal Jespersen <hlj@posix.com>
In-Reply-To: <199210152259.AA01799@dkuug.dk>; from "Keld J|rn Simonsen" at Oct 15, 92 11:59 pm
Organization: POSIX Software Group, 447 Lakeview Way, Redwood City, CA 94062
Phone: +1 (415) 364-3410	FAX: +1 (415) 364-4498
X-Mailer: ELM [version 2.2 PL0]
X-Charset: ASCII
X-Char-Esc: 38

> 
> Substitute specification in the LC_COLLATE section of localedef
> 
> DS proposes to use the wording contained in ISO/IEC 9945-2 DIS
> annex G.

Keld, I can find no such wording in Annex G.  In any event, even if you
go back to an earlier draft, we need to have compelling rationale for
why this is a required feature.  We also need a solution to the
previous situation of how do you define the LC_COLLATE substitute
feature in terms of regular expressions, when REs themselves are
dependent on LC_COLLATE--a recursive definition.

Hal