Date: Fri, 8 Jan 2010 09:11:23 +0300 From: Andrey Chernov <ache@nagual.pp.ru> To: Bruce Evans <brde@optusnet.com.au> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org, Konstantin Belousov <kib@freebsd.org> Subject: Re: svn commit: r201604 - head/lib/libc/gen Message-ID: <20100108061123.GA13136@nagual.pp.ru> In-Reply-To: <20100108030937.Q56559@delplex.bde.org> References: <201001052020.o05KKVEp002337@svn.freebsd.org> <20100107050409.GA85573@nagual.pp.ru> <20100108030937.Q56559@delplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Jan 08, 2010 at 03:29:07AM +1100, Bruce Evans wrote: > Nothing has turned here yet. We just didn't want to risk changing > alphasort() yet. I mean the comment itself. > POSIX adds an EINVAL error and a requirement that errno no be modified > if there is no error, so that errors from strcoll() can be detected > (still messily -- you have to set errno to != EINVAL before the call). > Maybe falling back to strcmp() is good enough for the errnor handling. Currently strcoll() never use EILSEQ and always falback to ASCII when ordering is equal. In the future, when EILSEQ will be used, it can fallback to ASCII too, producing stable ordering. -- http://ache.pp.ru/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100108061123.GA13136>