Date: Sun, 29 May 2016 14:59:11 -0700 From: Cy Schubert <Cy.Schubert@komquats.com> To: Pedro Giffuni <pfg@FreeBSD.org> Cc: Cy Schubert <Cy.Schubert@komquats.com>, Cy Schubert <Cy.Schubert@cschubert.com>, src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-vendor@freebsd.org Subject: Re: svn commit: r300961 - vendor/one-true-awk/dist Message-ID: <201605292159.u4TLxBDf005850@slippy.cwsent.com> In-Reply-To: Message from Pedro Giffuni <pfg@FreeBSD.org> of "Sun, 29 May 2016 14:21:29 -0500." <1b8c118d-0743-ba8f-5796-65b165bc8efd@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <1b8c118d-0743-ba8f-5796-65b165bc8efd@FreeBSD.org>, Pedro Giffuni wr ites: > > > > On 05/29/16 14:06, Cy Schubert wrote: > > In message <201605291817.u4TIHnN7040344@slippy.cwsent.com>, Cy Schubert > > writes: > >> In message <574B2EAC.3010908@FreeBSD.org>, Pedro Giffuni writes: > >>> > >>> > >>> > >>> On 29/05/2016 12:37, Cy Schubert wrote: > >>>> In message <201605291618.u4TGItNJ024583@repo.freebsd.org>, "Pedro F. > >>>> Giffuni" w > >>>> rites: > >>>>> Author: pfg > >>>>> Date: Sun May 29 16:18:55 2016 > >>>>> New Revision: 300961 > >>>>> URL: https://svnweb.freebsd.org/changeset/base/300961 > >>>>> > >>>>> Log: > >>>>> one-true-awk: replace 0 with NULL for pointers > >>>>> > >>>>> Also remove a redundant semicolon. > >>>>> Submitted upstream already. > >>>>> > >>>>> Modified: > >>>>> vendor/one-true-awk/dist/b.c > >>>>> vendor/one-true-awk/dist/lex.c > >>>>> vendor/one-true-awk/dist/maketab.c > >>>>> vendor/one-true-awk/dist/parse.c > >>>>> vendor/one-true-awk/dist/run.c > >>>>> vendor/one-true-awk/dist/tran.c > >>>>> > >>>> Was this commit and r300962 obtained from the upline or vendor or were > >>>> these commits local to FreeBSD only? > >>>> > >>>> > >>> There is no public awk public repository AFAICT, but bwk acknowledged > >>> the submission. > >>> > >>> The change to openresolv was merged to the public repository. > >> > >> As they've acknowledged the submissions, can you please tag the new > >> versions of awk and openresolve with the correct upstream version numbers, > >> please? > > > > Additionally, if there are no new version numbers, what reason is there for > > polluting the vendor branch with local patches to it? Is not the vendor > > branch for virgin code retrieved (or received) from the vendor (or upline)? > > > > > > Heck! Sorry but this particular issue of committing local patches to the vendor branch is something I've meant to raise for a long time. This is not meant against you. If you felt that then I apologize. > > Can't you simply trust the committer knows what he is doing? It's not a matter of trust. it's a matter of history. Someone may see a certain commit a couple of years from now and wonder from where it came from, if it was from the vendor and how it was obtained from the vendor. In regard to patches submitted upstream, IMHO I don't think they belong in the vendor branch. Local patches submitted upstream and either not yet accepted or accepted but not incorporated into the vendor's code base IMO should only be imported into the vendor branch when the authorized code is either committed to or released by the vendor. Admittedly this is a gray area and open to interpretation and thus different folks on The Project may have different opinions. IMO vendor branch is only for fully accepted and committed code by the vendor. If it's a gray area then is it really vendor or is it ours? IMO it would be ours. Why? Two reasons: One. History. Secondly, should we discover some anomaly (not that your commits would cause that), being able to look at the virgin code in the vendor branch and compare it with what's in HEAD might help to understand why the anomaly. (And I talked myself out of a third reason.) > > http://roy.marples.name/projects/openresolv/info/12cb1c1fb10df107 > > For nawk there is not public repository but bwk's acknowlegement said: > > "Thanks -- that's something that I should have done long long ago." > > So I think both changes are pretty much vendor code now. If they don't have a public repository then that's cool. I suppose "obtained from:" or "discussed with:" would provide good documentation. While on this topic. I have correspondence with folks upline for software in src/ and some ports/ that provide some interesting history that cannot be captured in commit log messages. I'm at the age that I'll retire from this one day and go back to growing potatoes on the farm (well, maybe not) or more likely not be around any more. It would be nice to archive some of this correspondence one day so that those who follow can still have this history. I cc my former mentor on all correspondence with our ipfilter upline so that someone else on the project will have a copy of the correspondence going forward, just so someone on The Project has a copy should I become incapacitated or unable. I think history is important. I think how we got here is important.The fact that freebsd.org archives this email is important because if the group makes a decision because of what is in this email (not that this particular email is all that important, but you know what I mean), people can go back and see how we got here. I think that's important. Once again, if I appeared hard on you, I'm sorry. That was not my intent. -- Cheers, Cy Schubert <Cy.Schubert@komquats.com> or <Cy.Schubert@cschubert.com> FreeBSD UNIX: <cy@FreeBSD.org> Web: http://www.FreeBSD.org The need of the many outweighs the greed of the few.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201605292159.u4TLxBDf005850>