From owner-freebsd-hackers Mon Oct 16 18:44:08 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id SAA03043 for hackers-outgoing; Mon, 16 Oct 1995 18:44:08 -0700 Received: from sovcom.kiae.su (sovcom.kiae.su [144.206.136.1]) by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id SAA03038 for ; Mon, 16 Oct 1995 18:44:04 -0700 Received: by sovcom.kiae.su id AA06088 (5.65.kiae-1 ); Tue, 17 Oct 1995 04:31:11 +0300 Received: by sovcom.KIAE.su (UUMAIL/2.0); Tue, 17 Oct 95 04:31:10 +0300 Received: (from ache@localhost) by ache.dialup.demos.ru (8.6.11/8.6.9) id EAA00837 for kaleb@x.org; Tue, 17 Oct 1995 04:25:24 +0300 To: "Kaleb S. KEITHLEY" Cc: hackers@freebsd.org References: <199510170105.VAA25559@exalt.x.org> In-Reply-To: <199510170105.VAA25559@exalt.x.org>; from "Kaleb S. KEITHLEY" at Mon, 16 Oct 1995 21:05:24 EST Message-Id: Organization: Olahm Ha-Yetzirah Date: Tue, 17 Oct 1995 04:25:24 +0300 (MSK) X-Mailer: Mail/@ [v2.40 FreeBSD] From: =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= (aka Andrey A. Chernov, Black Mage) X-Class: Fast Subject: Re: A couple problems in FreeBSD 2.1.0-950922-SNAP Lines: 49 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 1986 Sender: owner-hackers@freebsd.org Precedence: bulk In message <199510170105.VAA25559@exalt.x.org> Kaleb S. KEITHLEY writes: >> I know. And it is so when you don't have LANG or your char width > 8bits. >> When you have LANG and your char width <= 8bits and you have >> ENABLE_STARTUP_LOCALE variable set, lets call this case as >> "ANSI extention". As you can see, this extention is well-controlled >> by ENABLE_STARTUP_LOCALE. >And this marvelous "extension" breaks programs, as you've found with >the XFree86 xterm. Well, shure, it seems that you not expect that propogating code table to 8859-1 BREAKS XTERM in the same way! Because when LANG set to 8859-1 our cases are equal! >> Most of my agrument isn't premise but different behaviour of all >> is*() macros for different charsets. What 8859-1 program treats >> as isalpha isn't isalpha in other charset, for both input/output. >> When program bound its input/output by is*() macros, it leads >> into big trouble into your variant. >I claim this is a non-issue. Terry claims it's expected behavior. Please >provide an example or other proof that this is not the case. Pretty simple, here some example of code: do { *s++ = getchar(); } while (!ispunct(*s)); When 8859-1 is set it stops on D7 and it will be right. When 8859-5 is set is stops on D7 too, but D7 is letter. >Because your way of fixing is in violation of ANSI/POSIX/ISO. Well, it isn't only "my" way, I saw the same behavior in Xenix internationalization f.e. :-) I agree now that it can't be default case, so I withdraw my proposal to change ENABLE to DISABLE. >When it's active it breaks correct programs. When it's inactive it isn't It breaks no more than table propopgating, just right in the same way. -- Andrey A. Chernov : And I rest so composedly, /Now, in my bed, ache@astral.msk.su : That any beholder /Might fancy me dead - FidoNet: 2:5020/230.3 : Might start at beholding me, /Thinking me dead. RELCOM Team,FreeBSD Team : E.A.Poe From "For Annie" 1849