From owner-freebsd-git@freebsd.org Sun Feb 7 19:48:42 2021 Return-Path: Delivered-To: freebsd-git@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 229A45355BD for ; Sun, 7 Feb 2021 19:48:42 +0000 (UTC) (envelope-from uqs@freebsd.org) Received: from mailman.nyi.freebsd.org (unknown [127.0.1.3]) by mx1.freebsd.org (Postfix) with ESMTP id 4DYfpx73H6z3HPh for ; Sun, 7 Feb 2021 19:48:41 +0000 (UTC) (envelope-from uqs@freebsd.org) Received: by mailman.nyi.freebsd.org (Postfix) id D44415354FA; Sun, 7 Feb 2021 19:48:41 +0000 (UTC) Delivered-To: git@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id 9130C5354F8 for ; Sun, 7 Feb 2021 19:48:41 +0000 (UTC) (envelope-from uqs@freebsd.org) Received: from acme.spoerlein.net (acme.spoerlein.net [IPv6:2a05:fc87:1:5::15]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "www.spoerlein.net", Issuer "R3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4DYfpx1N5Wz3H7k for ; Sun, 7 Feb 2021 19:48:41 +0000 (UTC) (envelope-from uqs@freebsd.org) Received: from localhost (acme.spoerlein.net [IPv6:2a05:fc87:1:5:0:0:0:15]) by acme.spoerlein.net (8.16.1/8.15.2) with ESMTPS id 117JmdOF027181 (version=TLSv1.3 cipher=TLS_AES_256_GCM_SHA384 bits=256 verify=NO); Sun, 7 Feb 2021 20:48:39 +0100 (CET) (envelope-from uqs@freebsd.org) Date: Sun, 7 Feb 2021 20:48:38 +0100 From: Ulrich =?utf-8?B?U3DDtnJsZWlu?= To: Shawn Webb Cc: git@freebsd.org Subject: Re: HEADS UP: hashes changing for the freebsd-ports repo on Sunday Message-ID: References: <20210207163437.ffta4ebkndofh4ke@mutt-hbsd> <20210207174648.bzpfkw326agk2s2a@mutt-hbsd> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20210207174648.bzpfkw326agk2s2a@mutt-hbsd> User-Agent: Mutt/2.0.3 (2020-12-04) X-Rspamd-Queue-Id: 4DYfpx1N5Wz3H7k X-Spamd-Bar: ---- Authentication-Results: mx1.freebsd.org; none X-Spamd-Result: default: False [-4.00 / 15.00]; REPLY(-4.00)[] X-BeenThere: freebsd-git@freebsd.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: Discussion of git use in the FreeBSD project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 07 Feb 2021 19:48:42 -0000 On Sun, 2021-02-07 at 12:46:48 -0500, Shawn Webb wrote: >On Sun, Feb 07, 2021 at 06:21:10PM +0100, Ulrich Spörlein wrote: >> On Sun, 2021-02-07 at 11:34:37 -0500, Shawn Webb wrote: >> > On Sun, Feb 07, 2021 at 05:24:33PM +0100, Ulrich Spörlein wrote: >> > > On Wed, 2021-02-03 at 13:54:56 +0100, Ulrich Spörlein wrote: >> > > > Hi folks, >> > > > >> > > > It's another one of those instances where the hashes will change >> > > > (starting with all commits from Jan 5th 2021 forward). >> > > > >> > > > Our latest ports committers were not added to the authormap file in time >> > > > _and_ this stopped being a fatal issue in the converter a while ago :/ >> > > > >> > > > Please bear with us a while longer. >> > > > Thanks >> > > > Uli >> > > > >> > > > PS: This also affects the src and doc repos, but they will obviuously >> > > > not be changed again, see the errata notes at >> > > > https://github.com/freebsd/git_conv/tree/next >> > > >> > > This has now been pushed and the new hashes (and missing authornames!) are >> > > live. >> > >> > Sorry for what may seem like dumb questions, but I just wanted to make >> > sure: >> > >> > 1. For downstream projects that have already switched to >> > git.freebsd.org for src, do we need to deal with changed hashes >> > again? (ie, git merge --allow-unrelated-histories) >> >> Nothing changed for src. >> >> > 2. Do I now need to merge unrelated histories on ports if I use ports >> > from github? >> >> Depends a bit. There is shared history, but it forkes off way in the past, >> so if you have work on custom branch, it might indeed be easiest to merge >> with --allow-unrelated-histories. >> >> (I personally would probably write out the merge commit manually myself, >> it's much easier than figuring out what sort of flags one needs on the `git >> merge`.) >> >> Of course, if you're using rebase-based workflows, this all will be rather >> trivial to rebase. > >Thanks for the info! HardenedBSD uses a merge-based workflow since we >plan to be around a while. Switching to rebased-based would have >negative consequences to our downstream vendors. > >One more question: when ports is officially switched over to git, are >we likely to see hash more hash changes? No, of course not. Once the conversion has been finalized, none of the history or metadata will be touched retroactively. We'll then have to live with errors basically forever. Cheers Uli