Created attachment 247940 [details]
git diff for audio/aqualung

updated patch with better handling of lib depends.

Comment 2 Koichiro Iwao 2024-01-25 02:52:12 UTC

Are there any other ports that you lost maintainership because of domain unavailability? If so, submit a diff to reclaim maintainership all ports you maintained.

(In reply to Koichiro Iwao from comment #2)
Yes. I agree that's the most efficient approach, and
I'm already working on that. But this pr(1) goes beyond
plain MAINTAINER reclamation; see the second patch in
comment #1.

Thanks. :)

--Chris

Comment 4 Daniel Engberg 2024-01-25 21:12:08 UTC

At least submit a port which uses non legacy libraries. GTK 2 have planned deprecation this year and FFmpeg is likely going to be sooner.

(In reply to Koichiro Iwao from comment #2)
See also bug #276620

(In reply to Daniel Engberg from comment #4)
As already stated in the original  submission of this pr(1);
I'll be upgrading this port to 1.2. Which answers your grievance
toward ffmpeg4. If GTK2 has not already been addressed in that
version. I'll work with the Author to move GTK forward.
Right now I'm forced to get all my ports reconciled. Instead
of simply working on them. :/

Thanks for all your time and consideration.

--Chris

Comment 7 Daniel Engberg 2024-01-25 23:00:39 UTC

Start by fixing compatibilty with FFmpeg first, given previous history of this port and lack of response (PR 270194) you can submit that all in one patch.

(In reply to Daniel Engberg from comment #7)
I don't understand. As mentioned twice, version 1.2 already
works with ffmpeg6. So, as I just indicated, when I get 1.2
into the tree. I can then move to moving this to support
a newer version of GTK. What am I missing here? I simply
want to get this back into the tree with my name on it. So
I can move this forward.

Comment 9 Gleb Popov 2024-03-02 13:25:40 UTC

When compiling with all options ON I get:

ld: error: duplicate symbol: timeout_tag
>>> defined at gui_main.c
>>>            aqualung-gui_main.o:(timeout_tag)
>>> defined at ports.c
>>>            aqualung-ports.o:(.bss+0x450)
cc: error: linker command failed with exit code 1 (use -v to see invocation)

Also, += assignments are useless here.

Comment 10 Rene Ladan 2024-03-03 16:35:37 UTC

Why not just upstream Github directly instead of mirroring it on your own website?

(In reply to Rene Ladan from comment #10)
> Why not just upstream Github directly instead of mirroring it on your own website?

I don't understand? The source for this port is hosted on
SorceForge (MASTERSITES= SF).
Can you please better clarify your question?

Thanks,

--Chris

Comment 12 Rene Ladan 2024-03-04 19:52:09 UTC

(In reply to Chris Hutchinson from comment #11)

So that must be an old release then.

SF: version 1.0 from 9 years ago
GH: version 1.2 from last June: https://github.com/jeremyevans/aqualung , references the SF site as "for historical purposes"

So why keep on to SF?

(In reply to Rene Ladan from comment #12)
Thanks for the clarification, Rene.
I have no intention of keeping it on SF (see several
comments on this earlier). It was my hope to get the
change here quickly, so as to accommodate the global
man/ to share/man change. In my haste to do so, I
overlooked an issue Gleb was kind enough to point
out. I'll address that today. Then start the migration
changes to GH && ffmpeg6. Which will likely be a bit
of an adventure all on it's own. :)

Thanks again.

--Chris

Created attachment 248940 [details]
final patch for audio/aqualung

> ld: error: duplicate symbol: timeout_tag
>>> defined at gui_main.c
>>>            aqualung-gui_main.o:(timeout_tag)
>>> defined at ports.c
>>>            aqualung-ports.o:(.bss+0x450)
> cc: error: linker command failed with exit code 1 (use -v to see invocation)
This was caused by audio/jack(it). Which is fairly broken in it's own right.
I've disabled it, and will be presenting a pr(1) for it.

> Also, += assignments are useless here.
Blindly followed portlint(1) w/o thinking. The + has been removed in this patch.

This should do it.
I'm moving to version 1.2 && ffmpeg6 for the next round.

Thanks for all your time ans efforts!

--Chris

Created attachment 248961 [details]
patch audio/aqualung to version 1.2

This patch updates audio/aqualung to version 1.2
uses (FreeBSD) default version of ffmpeg (6), removes
now unnecessary files/patch-configure and files/patch-ffmpeg3.
port tested against default options && *all* options.
builds and works as intended on releng/13,14 and current.

This *should* satisfy all issues raised against this port. :)

Thanks!

--Chris

Comment 16 commit-hook 2024-03-06 19:20:33 UTC

A commit in branch main references this bug:

URL: https://cgit.FreeBSD.org/ports/commit/?id=77482fd613f088409e3c2ea8b112c6b4581e89c1

commit 77482fd613f088409e3c2ea8b112c6b4581e89c1
Author:     Chris Hutchinson <portmaster@bsdforge.com>
AuthorDate: 2024-03-06 19:18:37 +0000
Commit:     Gleb Popov <arrowd@FreeBSD.org>
CommitDate: 2024-03-06 19:19:09 +0000

    audio/aqualung: Update to 1.2

    PR:             276601

 audio/aqualung/Makefile                            | 32 +++++++--------
 audio/aqualung/distinfo                            |  5 ++-
 audio/aqualung/files/patch-configure (gone)        | 46 ----------------------
 audio/aqualung/files/patch-ffmpeg3 (gone)          | 24 -----------
 .../aqualung/files/patch-src_decoder_dec__mac.cpp  | 12 +++---
 5 files changed, 23 insertions(+), 96 deletions(-)

Comment 17 Daniel Engberg 2024-03-06 21:54:19 UTC

Is there a reason why we're not using upstream release archive as recommended by Porters Handbook?
https://docs.freebsd.org/en/books/porters-handbook/book/#makefile-master_sites-github
https://github.com/jeremyevans/aqualung/releases/tag/1.2
This would also very likely get rid of USES= autoreconf

Any plans to upstream
https://cgit.freebsd.org/ports/tree/audio/aqualung/files/patch-src_decoder_dec__mac.cpp ?

(In reply to Daniel Engberg from comment #17)
I have no objections to pushing our changes upstream
and including them should they become available (assuming
they're accepted). I felt *this* work was a good "first step"
(keeping it in the tree). :)

--Chris

For the record. I've been using this port for years.
So it's functionality (and availability) is important to me.