Package Details: mlt-git 7.22.0.r27.30115615-1

Git Clone URL: https://aur.archlinux.org/mlt-git.git (read-only, click to copy)
Package Base: mlt-git
Description: Multimedia Framework
Upstream URL: https://github.com/mltframework/mlt
Licenses: LGPL2.1
Conflicts: mlt
Provides: mlt
Submitter: agapito
Maintainer: nfnty (evorster)
Last Packager: evorster
Votes: 43
Popularity: 0.000000
First Submitted: 2009-08-26 09:32 (UTC)
Last Updated: 2024-03-07 13:24 (UTC)

Pinned Comments

nfnty commented on 2015-12-27 08:48 (UTC)

Issues and pull requests at https://github.com/nfnty/pkgbuilds

Latest Comments

1 2 3 4 5 Next › Last »

eclairevoyant commented on 2023-05-21 02:01 (UTC) (edited on 2023-05-21 02:03 (UTC) by eclairevoyant)

Upstream does versioned releases, so the pkgver should reflect that:

pkgver() {
    git -C $_gitname describe --long --tags | sed 's/^v//;s/\([^-]*-\)g/r\1/;s/-/./g'
}

This produces the more accurate version 7.16.0.r16.56b1f62f. (A similar example is available in /usr/share/pacman/PKGBUILD-vcs.proto, but that has a leading v and assumes all tags are annotated, which they aren't here.)

epoch does not need to be bumped btw, as per pacman's vercmp, numbers are strictly greater than letters, so the new version will be considered "increasing".

Also, license should be corrected to LGPL2.1.

Riedler commented on 2023-04-02 17:38 (UTC)

paru always updates the source. I removed it anyway so I could be sure. didn't help. Although this time it got to 92%…?

I'm simply going to try again. At this rate, the next compilation has to go to 100% :I

evorster commented on 2023-04-02 17:22 (UTC)

One crucial difference is that I use Yay locally. That has a --cleanbuild option which also removes the source and tries again with a clean slate, and a failed build might very well cause a later build to fail if some of the previous build files are still hanging around.

On Yay you can also manually for this by removing the ./src directory where the PKGBUILD resides, which will force the PKGBUILD to check out a new copy of the source from it's local git repo.

Riedler commented on 2023-04-02 17:07 (UTC)

I mean, your package does that, no? Anyway, I compiled it manually while looking at your PKGBUILD and it …worked? what?

I even deleted all of the stuff paru caches for AUR compilation and retried, but the PKGBUILD produces the same error as always and manual compilation with the same commands doesn't.

huh????

I commented the custom CFLAGS and LDFLAGS again because what the hell and that doesn't fix it either, although the compilation at least got to 78% instead of the ~40% the previous runs had. But also the compilation was way slower. Maybe a job ordering thing? idk

evorster commented on 2023-04-02 16:13 (UTC)

Just chatted with the creator of MLT. He believes this error is already fixed. Can you please pull the latest version of the source from git and try again, please?

Riedler commented on 2023-04-02 15:57 (UTC)

normally yes, but I disabled them because I thought they might be the problem. Looks like they aren't.

evorster commented on 2023-04-02 15:55 (UTC)

Interesting. Do you have custom LDFLAGS of compiler flags set?

Riedler commented on 2023-04-02 15:40 (UTC)

nope, still doesn't compile on my machine

evorster commented on 2023-04-02 15:37 (UTC)

@Riedler this may be a transient thing, git repos are not always guaranteed to compile. I have just tried to compile it on my machine, and it worked flawlessly. If the problem persists, please make another comment and I'll pass it along to the upstream repo

Riedler commented on 2023-04-01 13:42 (UTC)

compilation failed. compiler says:

/data/makepkg/mlt-git/src/mlt/src/modules/avformat/factory.c:35:17: error: type of ‘link_swresample_init’ does not match original declaration [-Werror=lto-type-mismatch]
   35 | extern mlt_link link_swresample_init(mlt_profile profile, char *arg);
      |                 ^
/data/makepkg/mlt-git/src/mlt/src/modules/avformat/link_swresample.c:340:10: note: type mismatch in parameter 2
  340 | mlt_link link_swresample_init(mlt_profile profile, mlt_service_type type, const char *id, char *arg)
      |          ^
/data/makepkg/mlt-git/src/mlt/src/modules/avformat/link_swresample.c:340:10: note: ‘link_swresample_init’ was previously declared here
/data/makepkg/mlt-git/src/mlt/src/modules/avformat/link_swresample.c:340:10: note: code may be misoptimized unless ‘-fno-strict-aliasing’ is used
lto1: all warnings being treated as errors
lto-wrapper: fatal error: /usr/bin/cc returned 1 exit status
compilation terminated.
/usr/bin/ld: error: lto-wrapper failed
collect2: error: ld returned 1 exit status
make[2]: *** [src/modules/avformat/CMakeFiles/mltavformat.dir/build.make:281: out/lib/mlt/libmltavformat.so] Error 1
make[1]: *** [CMakeFiles/Makefile2:841: src/modules/avformat/CMakeFiles/mltavformat.dir/all] Error 2
/data/makepkg/mlt-git/src/mlt/src/modules/plus/factory.c:108:21: error: type of ‘producer_count_init’ does not match original declaration [-Werror=lto-type-mismatch]
  108 | extern mlt_producer producer_count_init(const char *arg);
      |                     ^
/data/makepkg/mlt-git/src/mlt/src/modules/plus/producer_count.c:632:14: note: type mismatch in parameter 2
  632 | mlt_producer producer_count_init(mlt_profile profile,
      |              ^
/data/makepkg/mlt-git/src/mlt/src/modules/plus/producer_count.c:632:14: note: type ‘mlt_service_type’ should match type ‘void’
/data/makepkg/mlt-git/src/mlt/src/modules/plus/producer_count.c:632:14: note: ‘producer_count_init’ was previously declared here
lto1: all warnings being treated as errors
lto-wrapper: fatal error: /usr/bin/cc returned 1 exit status
compilation terminated.
/usr/bin/ld: error: lto-wrapper failed
collect2: error: ld returned 1 exit status
make[2]: *** [src/modules/plus/CMakeFiles/mltplus.dir/build.make:516: out/lib/mlt/libmltplus.so] Error 1
make[1]: *** [CMakeFiles/Makefile2:1390: src/modules/plus/CMakeFiles/mltplus.dir/all] Error 2
make: *** [Makefile:136: all] Error 2

I left out everything that's not an error or warning for conciceness.

I have zero experience with C so I can't fix it myself, sorry.

1 2 3 4 5 Next › Last »