Date: Tue, 17 Oct 1995 14:26:18 +0300 (MSK) From: =?KOI8-R?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= (aka Andrey A. Chernov, Black Mage) <ache@astral.msk.su> To: core@freebsd.org, hackers@freebsd.org, Poul-Henning Kamp <phk@critter.tfs.com> Subject: Re: Locale stuff: call for conclusion. Message-ID: <omQBvWmCL1@ache.dialup.demos.ru> In-Reply-To: <3031.813918926@critter.tfs.com>; from Poul-Henning Kamp at Tue, 17 Oct 1995 09:35:26 %2B0100 References: <3031.813918926@critter.tfs.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <3031.813918926@critter.tfs.com> Poul-Henning Kamp writes: >(Core-members: Please send a vote to me on this) Here is some non-clarified things left. >I belive that if not everything, then at least at lot, have been said, >(sometimes even over and over again) but nothing has yet been done. >I think we can also agree to the following things: >1. The crt0.s thing is a kludge, which should go away as soon as > something more sane is implemented. >2. The crt0.s thing is actually a violation of some stds. >3. There is no harm to make the default case (unsetenv LANG & all that) > be ISO-8859-1, and this will make a lot of people a lot more happy. > It may not help Andrey much, but it is a large part of our userbase. I am happy with my hack right now and was against because it cause some problems for our userbase (i.e. when my hack is inactive), namely is*() macros do wrong things, but Terry point it as a 'feature', not a 'bug'. >4. The right solution is to go through the programs, one by one and DTRT > to them, if it makes sense. A lot of programs shouldn't care at all, > cp(1) for instance, whereas other programs it does indeed matter, > sort(1) for instance. >Based on this I suggest that we: >A) execute 3. I agree. >B) execute 1. in the other sense of the word, and get crt0.s cleaned up. > This will happen January 1st 1996 even if A) and/or C) hasn't happened. As you sayd > as something more sane is implemented. I am completely against of removing this hack when nothing sane implemented, so date doesn't play role here. >C) execute 4, as soon as anybody volounteer to do it. Well, as we voting, here additional issue (D), if you care of what sane can be implemented instead. By Terry idea we can switch from XPG/4 to XPG/3, it allows as safely use setlocale() in 8bit clean programs (not care of multibyte runic chars which isn't really used by Terry words). Currently all ports setlocale() soft and vi is very broken for runic chars, i.e. all such stuff assumes XPG/3 instead XPG/4 we have. Switching to XPG/3 allowing as to fix almost every 8bit clean program by simple adding setlocale(LC_ALL, "") into main(). -- 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
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?omQBvWmCL1>