Date: Wed, 28 May 2008 18:16:54 +0200 From: Joerg Sonnenberger <joerg@britannica.bec.de> To: cvs-src@freebsd.org Subject: Re: cvs commit: src/usr.bin/cpio Makefile bsdcpio.1 cmdline.c config_freebsd.h cpio.c cpio.h cpio_platform.h err.c matching.c matching.h pathmatch.c pathmatch.h src/usr.bin/cpio/test Makefile main.c test.h test_0.c test_basic.c test_format_newc.c ... Message-ID: <20080528161654.GB13294@britannica.bec.de> In-Reply-To: <20080528135832.GA14844@dragon.NUXI.org> References: <200805261715.m4QHFZUK070554@repoman.freebsd.org> <20080526172717.GA93432@freebsd.org> <483AFE87.6020103@freebsd.org> <20080528013528.GA97270@dragon.NUXI.org> <1211981232.1637.5.camel@localhost> <20080528135832.GA14844@dragon.NUXI.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--tThc/1wpZn/ma/RB Content-Type: text/plain; charset=us-ascii Content-Disposition: inline On Wed, May 28, 2008 at 06:58:32AM -0700, David O'Brien wrote: > Please read this as: "Please don't touch pax". It works fine as-is, is > small, is only used by folks that know it exists (many BSD users don't), > is BSDL'ed, pretty much cause no harm or extra work for us, etc... It doesn't work as-is. Consider the attached pax archive. I strongly believe it is a valid "pax Interchange Format" archive as defined by the Single Unix Specification. FreeBSD's pax can't handle this. NetBSD's pax is somewhat better as it at least detects that an extended header is used, but can't deal with it correctly. Now you might consider long symlinks an obscure edge case, but they happen in real world situations. Mono f.e. has a number of such long symlinks. Joerg --tThc/1wpZn/ma/RB--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080528161654.GB13294>