Re: git: 519c89efe3a9 - main - audio/libmad: Update to 0.16.0

From: Stefan Esser <se_at_FreeBSD.org>
Date: Sat, 07 May 2022 10:01:22 UTC
Am 06.05.22 um 11:59 schrieb Daniel Engberg:
> The branch main has been updated by diizzy:
> 
> URL: https://cgit.FreeBSD.org/ports/commit/?id=519c89efe3a9ddc8f3a57d1bb4940ff355086576
> 
> commit 519c89efe3a9ddc8f3a57d1bb4940ff355086576
> Author:     Daniel Engberg <diizzy@FreeBSD.org>
> AuthorDate: 2022-05-06 09:47:30 +0000
> Commit:     Daniel Engberg <diizzy@FreeBSD.org>
> CommitDate: 2022-05-06 09:59:33 +0000
> 
>     audio/libmad: Update to 0.16.0
>     
>     Switch to Tenacity's fork which contains patched code for all known CVEs
>     and various build fixes. This also makes it easier to maintain as we're
>     using an active fork. Upstream has also switched to CMake.
>     
>     PR:             262874
>     Approved by:    sunpoet (maintainer timeout 4+ weeks)
[...]
> --- a/audio/libmad/files/mad.pc.in
> +++ /dev/null
> @@ -1,11 +0,0 @@
> -prefix=%%PREFIX%%
> -exec_prefix=${prefix}
> -libdir=${exec_prefix}/lib
> -includedir=${prefix}/include
> -
> -Name: mad
> -Description: MPEG Audio Decoder
> -Requires:
> -Version: %%VERSION%%
> -Libs: -L${libdir} -lmad -lm
> -Cflags: -I${includedir}

This update broke all ports that depend on libmad and use pkg-config to check
for its existence and the build details!

I see that 2 ports (audio/audacity, sysutils/mtpfs)have already been patched
to adapt to this unexpected and unannounced change.

There are tens of ports that depend on audio/libmad, and I think that it
would be better to restore the old name and contents of the pkgconfig file
instead of hunting down and fixing all dependent and now broken ports.

One of my ports is affected (multimedia/dvbcut), and I'm holding back the
commit that patches the port's project file to adapt to this change, since I
hope for a reversal of that renaming from "mad.pc" to "libmad.pc".

There are a lot of ports that expect and rely on the previous name, and we
divert from practice on all kinds of other distributions for apparently no
good reason!

Regards, STefan