Arch Linux User Repository

If you experience the problem below when updating packages, first remove mumble-git, then update all packages including protobuf and finally compile mumble-git again.

# pacman -Syu
:: Starting full system upgrade...
resolving dependencies...
looking for conflicting packages...
error: failed to prepare transaction (could not satisfy dependencies)
:: installing protobuf (3.12.0-1) breaks dependency 'libprotobuf.so=22-64' required by mumble-git

@dvzrv: Alright, I have updated the PKGBUILD accordingly.

@Radioactiveman: No, that is the next stable release version, that is referred to from within that application.

The latest git tag is:

git ls-remote -t https://github.com/mumble-voip/mumble |cut -d '/' -f3 |sort -r |head -n1

The version this package needs to track is the latest version tag + how many commits since that tag, not some arbitrary next release version number in a file.

We have the VCS package guidelines for a reason, please stick to them.

@dvzrv: But the Git version is 1.4.0, not 1.3.0-RC2 - see mumble.pri. However appending "git version" to the description would help maybe.

@Radioactiveman: Please introduce an epoch and remove the custom _version in your PKGBUILD.

We're getting out-of-date notifications from unknowing users, thinking that 1.4.0 is out (which it is not).

The right way to get the current version string would be to use the following (as stated in the VCS package guidelines):

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

Thanks for the quick update. I can confirm building in a clean CHROOT works now.

@IncredibleLaser Fixed, thank you.

This package currently doesn't build without python installed. You'll get "Project ERROR: Unable to find the system's Python binary. Some scripts invoked during the Mumble build use Python. You can manually specify it via the MUMBLE_PYTHON environment variable (either 2 or 3)." Please add python as a make dependency. Thanks.

@C0rn3j: Sorry for the late response as well. But I have found a solution. :) Install espeak-ng instead of espeak. See also https://git.archlinux.org/svntogit/packages.git/commit/trunk?h=packages/speech-dispatcher&id=7a7cc7c07e54bd8f5682fec95eaf7e7c47a74cfa.

Sorry for the late response - yes, speech-dispatcher is installed

EDIT: I tried building it with both speech-dispatcher and espeak and then again with only either one of them installed, no TTS in all cases.