Patch will be attached after obtaining PR#.
Created attachment 268095 [details] v0 Patch has been created using 'git format-patch' and must be applied using 'git am' [0]. This is extremely important to avoid mistakes in my name and also preserve 'Author:' field. Poudriere testport ok 13.5R/14.3R amd64. I deliberately omitted changelog link as it's always the same, and the same link in every commit looks just silly. [0]: https://git-scm.com/docs/git-am
Thank you for the update. Mixing functional and style changes makes it harder to review the diff. Could you please regenerate upload the patch without style changes?
(In reply to Max Brazhnikov from comment #2) Unfortunately, no. You won't convince me to not run portclippy and portfmt [0] on each Makefile I edit as there's no reason to not do that unless this breaks port which did not happen in this case. [0]: https://docs.freebsd.org/en/books/porters-handbook/book/#testing-portclippy
Anton, sorry, it does not work this way. Ports are driven by a community of volunteers. If you do not want to cooperate with the others…, well, I wonder what kind of response do you expect to your demand. It is just easier for me to update and test the port myself, rather then dealing with your "help".
A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=4d9c9b7d7dc5cd38fda13575909de7ae092ebaa8 commit 4d9c9b7d7dc5cd38fda13575909de7ae092ebaa8 Author: Max Brazhnikov <makc@FreeBSD.org> AuthorDate: 2026-02-17 09:51:36 +0000 Commit: Max Brazhnikov <makc@FreeBSD.org> CommitDate: 2026-02-17 09:52:26 +0000 archivers/7-zip: Update to 26.00 Release notes: https://sourceforge.net/p/sevenzip/discussion/45797/thread/a1f7e08417/ PR: 293202 Reported by: Anton Saietskii <vsasjason@gmail.com> archivers/7-zip/Makefile | 2 +- archivers/7-zip/distinfo | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-)
Thank you for your cooperation.
(In reply to Max Brazhnikov from comment #4) I don't believe 10 reordered lines are more complex change than entire KDE Framework update which you did recently. Dear portmgr@ team, please step in here. makc@ asked to WORSEN my patch which was initially perfect as it didn't just update the port, but also fixed Makefile variable ordering [0], as well as style. Asking to NOT run portlint/portclippy/portfmt on the port doesn't make any sense. The only fair ask is to IMPROVE the changes, not WORSEN them. Also, makc@ didn't provide any strong rationale for their ask, and made a commit which does NOT fix the Makefile order and style issues. [0]: https://docs.freebsd.org/en/books/porters-handbook/book/#porting-order
(In reply to commit-hook from comment #5) Current status after this commit: > git status; git log -1 --oneline .; portclippy Makefile On branch main Your branch is ahead of 'origin/main' by 9 commits. (use "git push" to publish your local commits) nothing to commit, working tree clean 4d9c9b7d7dc5 archivers/7-zip: Update to 26.00 CATEGORIES MASTER_SITES DISTNAME # Maintainer block MAINTAINER COMMENT WWW # License block LICENSE LICENSE_COMB LICENSE_NAME_UNRAR LICENSE_FILE_BSD3CLAUSE LICENSE_FILE_LGPL21+ LICENSE_FILE_UNRAR LICENSE_PERMS_UNRAR # Dependencies LIB_DEPENDS # USES block USES CPE_VENDOR +DOS2UNIX_FILES # Make block MAKEFILE MAKE_ARGS # CFLAGS/CXXFLAGS/LDFLAGS block CFLAGS_aarch64 # WRKSRC block NO_WRKSUBDIR BUILD_WRKSRC -DOS2UNIX_FILES # Packaging list block PLIST_FILES PORTDOCS # Options definitions OPTIONS_DEFINE And of course, portfmt didn't run over Makefile as well. Both of the issues were originally fixed in my patch.
Resumbit the patch and I'm sure Max will commit it.
(In reply to Gleb Popov from comment #9) It is already here, as 'v0'. The part of portclippy/portfmt changes have been deliberately ignored (again, asking to improve patch is ok, asking to worsen patch literally doesn't make any sense and nobody proved otherwise). Handbook basically says 'use portclippy/portfmt', and if makc@ would -- they would get the same, bit-exact result even without me attaching any patches. This is not patch issue, but a process handling one.
(In reply to Anton Saietskii from comment #10) I mean, you reopened the bug, but did not update the patch. I see no point in doing that. The patch does not apply in its current form. What do you expect from a committer?
(In reply to Gleb Popov from comment #11) 1. To admit that attached patch isn't really difficult to review at all and that the ask to exclude style fixed was not made in a good faith. 2. To read Porter's Handbook (e.g., "use portclippy!) and Committer's guide meticulously (e.g., "Reported by:" should be used when there's NO PR). 3. To apologize for intentionally preventing the application of a completely correct patch without reason. 4. Correct their own mistakes (if it's too hard to use 'git cherry-pick', incorrect patch which doesn't fix portclippy complaints may be reverted following my patch application using the mentioned way which will also apply all metadata, including 'Author:').
(In reply to Anton Saietskii from comment #12) > To admit that attached patch isn't really difficult to review at all You can decide for other people what's difficult for them. That's simply rude. > the ask to exclude style fixed was not made in a good faith You can't know that for sure unless you can read thoughts. It might be as well that you're blaming a honest person for nothing. > o read Porter's Handbook (e.g., "use portclippy!) and Committer's guide meticulously (e.g., "Reported by:" should be used when there's NO PR). To me it feels like picking on Max here. Seriously, the "Reported by" line breaks something? To me, Max tried to give at least some attribution to you, despite you attacking him. > To apologize for intentionally preventing the application of a completely correct patch without reason. There is a reason - he did not want to review functional and stylistic changes mixed in a single patch. Again, we committers are not paid for our work, you can't demand anything from us. Not to mention the fact that Max is a maintainer of this very port. > Correct their own mistakes I believe Max does not think he made a mistake, so he has nothing to correct. Instead, ask yourself what is more important to you. If you want the port to get into an ideal shape, then just rebase your changes and refresh the patch. But if it is actually about your ego and not useful contributions - well, I doubt anyone will be interested in helping you with that.
(In reply to Gleb Popov from comment #13) > You can decide for other people what's difficult for them. That's simply rude. Of course not. In fact, this is opposite -- the appreciation of them recently doing KDE 6.23 update which is much more complex. > Seriously, the "Reported by" line breaks something? It's just proof of committer and maintainer haven't been reading respective guides which is at least strongly recommended in not obligatory. > There is a reason - he did not want to review functional and stylistic changes mixed in a single patch. Let's make it crystal clear. There is a specific order in which Makefile variables should be. For example, COMMENT should follow, not precede MAINTAINER, and if I attach the patch which put them in correct order, you cannot ask (using good faith and common sense) to exclude order change because "it's hard to review". Because it's not. > I believe Max does not think he made a mistake, so he has nothing to correct. I may be wrong, but I only see the sabotage (not the first time by the way, but ngl last time there was no patch from me) of accepting the, again, perfectly correct, not complex, and good faith patch. So please let the portmgr@ decide. Also, I would like to kindly ask you to stop trolling. Thanks in advance.
(In reply to Anton Saietskii from comment #14) So now I'm accused too for being a troll. Very well, good bye sir.
This is NOT fixed. portclippy still has complaints about variable ordering.
Sorry, didn't intend to close this.
A commit in branch 2026Q1 references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=eed392c9cba6e0930e323390a0ca9cf3111353a8 commit eed392c9cba6e0930e323390a0ca9cf3111353a8 Author: Max Brazhnikov <makc@FreeBSD.org> AuthorDate: 2026-02-17 09:51:36 +0000 Commit: Max Brazhnikov <makc@FreeBSD.org> CommitDate: 2026-03-01 16:16:24 +0000 archivers/7-zip: Update to 26.00 Release notes: https://sourceforge.net/p/sevenzip/discussion/45797/thread/a1f7e08417/ PR: 293202 Reported by: Anton Saietskii <vsasjason@gmail.com> (cherry picked from commit 4d9c9b7d7dc5cd38fda13575909de7ae092ebaa8) archivers/7-zip/Makefile | 2 +- archivers/7-zip/distinfo | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-)
(In reply to Anton Saietskii from comment #16) portclippy/portfmt are tools, they're not the definite way of ordering Ports Makefile but they can make some good suggestions but certainly less than ideal ones too.