From owner-freebsd-hackers Mon Oct 16 08:56:03 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id IAA12323 for hackers-outgoing; Mon, 16 Oct 1995 08:56:03 -0700 Received: from expo.x.org (expo.x.org [198.112.45.11]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id IAA12318 for ; Mon, 16 Oct 1995 08:55:58 -0700 Received: from exalt.x.org by expo.x.org id AA21966; Mon, 16 Oct 95 11:55:26 -0400 Received: from localhost by exalt.x.org id LAA07471; Mon, 16 Oct 1995 11:55:25 -0400 Message-Id: <199510161555.LAA07471@exalt.x.org> To: ache@astral.msk.su Cc: hackers@freefall.FreeBSD.org Subject: Re: A couple problems in FreeBSD 2.1.0-950922-SNAP In-Reply-To: Your message of Mon, 16 Oct 1995 04:44:58 EST. Organization: X Consortium Date: Mon, 16 Oct 1995 11:55:24 EST From: "Kaleb S. KEITHLEY" Sender: owner-hackers@FreeBSD.org Precedence: bulk >>>Bruce, I agree with you and POSIX agree with us. >>>Default table must be strict ASCII per POSIX. >>>So, we can close this table propogating subject. >>Does POSIX mention ASCII? ANSI allows other encodings >>of course, and is only strict for letters and digits, >>but only letters are very important. >Well 1003.1 says something about minimal subset to run C pgms. >Comparing ASCII and 8859-1 is clear that ASCII is more minimal. ANSI/POSIX/ISO C require that *at least* (my emphasis) A..Z, a..z, 0..9, !"#$&'()*+,-./:;<=>?[/]^_{~ , the space character, and control characters for horizontal tab, vertical tab, and form feed. For source files you must have a newline. For the execution environment the character set must also have control characters for alert, backspace, carriage return and new line. The only other restriction are on the decimal digits, 0..9, which must have a value one greater than the value of the previous. ASCII meets this criteria. ISO8859-1 also meets this criteria. There is not requirement that you use the most minimally compliant character set. I do not understand your insistence that the C locale be strictly ASCII and no more. I claim that it would be useful in the C locale if the default char- type table were to be nominally populated for 8859-1, i.e. excluding the upper and lower designations. You claim that doing so will break Russian and Japanese users. I claim that these users won't be using the C locale anyway. The fact that your locale support is broken and the utility programs are broken is merely an argument that they should be fixed. :-) -- Kaleb