Date: Wed, 5 Sep 2001 09:34:04 -0400 (EDT) From: Mikhail Teterin <mi@aldan.algebra.com> To: sobomax@FreeBSD.org Cc: obrien@FreeBSD.org, ache@nagual.pp.ru, ru@FreeBSD.org, ports@FreeBSD.org, kris@obsecurity.org Subject: Re: cvs commit: src/lib/libc/stdlib strtol.c strtoll.c strtoq.c strtoul.c strtoull.c strtouq.c Message-ID: <200109051334.f85DY7o63331@aldan.algebra.com> In-Reply-To: <3B9612A4.7F70AE5B@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 5 Sep, Maxim Sobolev wrote: >> BTW, most of the patches in the ports-tree don't have such IDs, but >> some do. Since those IDs are discarded as early as at the "make >> patch" stage, I think they are pretty useless -- they don't make it >> to the compiled binaries anyway. But what's the general opinion? > No, they are pretty useful, IMO. When the user has a problem you can > verify version of patches he has and after the problem is resolved > tell him "update patxh-xx to the rev.Y.X and you'll be fine". Can't we use the version strings of the ports' Makefiles? They are supposed to be in sync with the patches. Anyway, you don't say "update X to rev Y". You say: "update to the latest version of the port" -- nothing but the whole port, and no other version, but the latest one. Besides, you can always simply look at the timestamp of the patch. The usefullness of the rcs (and other) ids in the sources, is that they let the ids make it into the binaries, where they can be read by ident(1) and similar tools. Ids in patch files don't make into the binaries... > NetBSD for example has $NetBSD$ in all their patches, this is also > helpful when you need to steal patch from the NetBSD ports collection > and add it into FreeBSD one. I don't see, how the version string helps you steal a patch... -mi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200109051334.f85DY7o63331>