Bug 290040 - net-p2p/tremotesf: Update to 2.9.0, de-flavorize
Summary: net-p2p/tremotesf: Update to 2.9.0, de-flavorize
Status: Closed FIXED
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: --- Affects Only Me
Assignee: Hiroki Tagato
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2025-10-06 11:34 UTC by Anton Saietskii
Modified: 2025-10-12 09:12 UTC (History)
1 user (show)

See Also:


Attachments
v0 (apply via 'git am') (2.95 KB, patch)
2025-10-06 11:35 UTC, Anton Saietskii
vsasjason: maintainer-approval+
Details | Diff
v1 (apply via 'git am') (3.61 KB, patch)
2025-10-08 12:26 UTC, Anton Saietskii
vsasjason: maintainer-approval+
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Anton Saietskii 2025-10-06 11:34:13 UTC
Patch will be attached after getting PR#.
Comment 1 Anton Saietskii 2025-10-06 11:35:51 UTC
Created attachment 264345 [details]
v0 (apply via 'git am')

Please make sure to apply this via 'git am', not 'patch <' or 'git apply'.
Comment 2 Hiroki Tagato freebsd_committer freebsd_triage 2025-10-08 01:33:44 UTC
Hi, thanks for an update to the port.

I would suggest utilizing MOVED file for informing the removal of qt5 flavor and unflavorization:

- Remove PKGNAMESUFFIX=-qt6 from Makefile
- Add entries something like the following to MOVED
net-p2p/tremotesf@qt5|net-p2p/tremotesf|2025-10-08|Upstream removed qt5 support and it is now qt6 only, unflavorize port
net-p2p/tremotesf@qt6|net-p2p/tremotesf|2025-10-08|Port is now qt6 only, unflavorize port
Comment 3 Anton Saietskii 2025-10-08 12:26:48 UTC
Created attachment 264405 [details]
v1 (apply via 'git am')

(In reply to Hiroki Tagato from comment #2)

> - Add entries something like the following to MOVED
Section 11.2.2 of Porter's Handbook [0] and the file itself are both confusing -- neither mention flavors, they indicate port only. But okay, change added.

> - Remove PKGNAMESUFFIX=-qt6 from Makefile
While appreciating your feedback, I have to respectfully disagree. As indicated in commit message, this would be POLA violation as those who have 'tremotesf-qt6' package installed (which is produced from default flavor) will not receive update in this case. If we keep suffix -- they will just run 'pkg upgrade' and voila, get 2.9.0.
Thus, I would strongly prefer to keep suffix unless there's VERY good reason to not.

[0]: https://docs.freebsd.org/en/books/porters-handbook/book/#moved-and-updating-moved
Comment 4 Hiroki Tagato freebsd_committer freebsd_triage 2025-10-10 08:42:45 UTC
(In reply to Anton Saietskii from comment #3)

I thought 'pkg upgrade' would automatically recognize the package name change if we had an appropriate MOVED entry.

I did a small experiment in my private package repository what would happen if PKGNAMESUFFIX was removed and an entry was added to MOVED. What I found was 'pkg upgrade' did nothing.

So you are absolutely right. Let's keep PKGNAMESUFFIX=-qt6 as it is now.

One small comment about a MOVED entry:
- In the following entry:
  net-p2p/tremotesf@qt6|net-p2p/tremotesf|2025-10-08|Qt 6 is the only supported version, de-flavorize
  The second field should read "net-p2p/tremotesf-qt6" instead of "net-p2p/tremotesf" (need -qt6 suffix).

With that change, the patch I think is ready to commit.
Comment 5 Anton Saietskii 2025-10-10 09:18:36 UTC
(In reply to Hiroki Tagato from comment #4)

> The second field should read "net-p2p/tremotesf-qt6" instead of "net-p2p/tremotesf"

I don't believe so:
$ grep tremotesf MOVED ; Tools/scripts/MOVEDlint.awk
net-p2p/tremotesf@qt5||2025-10-08|Upstream removed Qt 5 support
net-p2p/tremotesf@qt6|net-p2p/tremotesf-qt6|2025-10-08|Qt 6 is the only supported version, de-flavorize
 1650: audio/kid3 does not have the kf5 flavor
 4800: net-p2p/tremotesf-qt6 not found

The _port_ (and the directory) is 'net-p2p/tremotesf'.
The _package_ is 'tremotesf-qt6-2.9.0'.

MOVED lists ports, not packages (and that's why it actually doesn't work for 'pkg' I believe).
In theory, 'poudriere bulk <blah> net-p2p/tremotesf@qt6' after the patch applied should say to you 'port moved to <blah>', but this didn't ever worked for me.
Comment 6 Hiroki Tagato freebsd_committer freebsd_triage 2025-10-10 19:12:49 UTC
(In reply to Anton Saietskii from comment #5)

Ah... I forgot to check with MOVEDlint. So net-p2p/tremotesf is the correct value for "moved to" field. Now the patch is ready to go and I will commit the changes after double-checking build with poudriere. Thanks for checking.
Comment 7 commit-hook freebsd_committer freebsd_triage 2025-10-10 20:56:30 UTC
A commit in branch main references this bug:

URL: https://cgit.FreeBSD.org/ports/commit/?id=78a4d1ccdc8982114e9dcd0fb1bda4d000c28d86

commit 78a4d1ccdc8982114e9dcd0fb1bda4d000c28d86
Author:     Anton Saietskii <vsasjason@gmail.com>
AuthorDate: 2025-10-06 09:49:54 +0000
Commit:     Hiroki Tagato <tagattie@FreeBSD.org>
CommitDate: 2025-10-10 20:55:21 +0000

    net-p2p/tremotesf: update to 2.9.0 and de-flavorize

    Upstream removed Qt 5 support with this release. Despite removing
    flavors, "-qt6" suffix remains to comply with POLA.

    Additionally, adjust description to avoid confusion as other modern
    GUIs may exist too.

    Changelog: https://github.com/equeim/tremotesf2/blob/2.9.0/CHANGELOG.md

    PR:             290040
    Reported by:    Anton Saietskii <vsasjason@gmail.com> (maintainer)

 MOVED                       |  2 ++
 net-p2p/tremotesf/Makefile  | 26 +++++---------------------
 net-p2p/tremotesf/distinfo  |  6 +++---
 net-p2p/tremotesf/pkg-descr |  4 ++--
 4 files changed, 12 insertions(+), 26 deletions(-)
Comment 8 Hiroki Tagato freebsd_committer freebsd_triage 2025-10-10 20:58:37 UTC
Committed, thanks!
Comment 9 Anton Saietskii 2025-10-11 12:41:01 UTC
(In reply to commit-hook from comment #7)

I'd like to kindly ask you to elaborate on the following commit changes which I believe were completely unnecessary, even erroneous:

1. According to section 9.5 of Committer's Guide [0], first letter of subject must be capitalized which it the case for my original, perfectly compliant patch. Also, changing comma to ' and ' makes commit message longer which contradicts section 9.3 [1] and just feels like the change for the sake of change, at least I don't see any real improvement here.
2. According to section 9.14 of the same [2], 'Reported by:' should be added when there's no PR.

It would be greatly appreciated if you could explain the motivation of doing these. Thanks in advance.

[0]: https://docs.freebsd.org/en/articles/committers-guide/#_capitalize_the_first_letter_of_the_subject
[1]: https://docs.freebsd.org/en/articles/committers-guide/#_keep_subject_lines_short
[2]: https://docs.freebsd.org/en/articles/committers-guide/#_include_appropriate_metadata_in_a_footer
Comment 10 Hiroki Tagato freebsd_committer freebsd_triage 2025-10-12 09:12:33 UTC
(In reply to Anton Saietskii from comment #9)

> According to section 9.5 of Committer's Guide [0], first letter of subject must be capitalized which it the case for my original, perfectly compliant patch.

I must admit that I remembered it otherwise, which was wrong. So that is my fault. My apologies.

> Also, changing comma to ' and ' makes commit message longer which contradicts section 9.3 [1] and just feels like the change for the sake of change, at least I don't see any real improvement here.

Here I preferred glammatical correctness instead of saving a few letters. In general, commas are used to separate items in a list of three or more[1]. On the other hand, "and" is used to join two words or phrases[2]. Using "and" here does not make the subject line unneccesarily long and I think using correct English matters as well as adhering to the Committer's Guide.

[1] https://www.grammarly.com/blog/punctuation-capitalization/comma/
[2] https://dictionary.cambridge.org/dictionary/english/and

> According to section 9.14 of the same [2], 'Reported by:' should be added when there's no PR.

By "Reported by:" field, I wanted to show that the change is made by the maintainer of the port. In other words, I would like to express the port is well maintained by its maintainer. (The maintainer is doing a good work.) Although "Author:" field shows the authorship of the change, it does not show the maintainership of the port. So I used:

Reported by: Name <Email> (maintainer)
                          ^^^^^^^^^^^^

I hope those explanations make sense to you.

Thanks for reading the commit message carefully. This conversation has been a good education for me.