The README, the publish script and the workflow named the tarball host, its
address and filesystem paths on the web host and the runner. Hosts and paths
now come only from repository variables, and the workflow shares the work
tree and the build cache through the named Docker volumes jasp-build and
jasp-cache instead of host paths. (The runner does not expand ${{ vars }} in
container.volumes, so a variable could not have replaced the host path there.)
Also drop the unused scripts/gen-tarball-index.py, and give each module build
its own process under a time limit: a module once spun for eight hours in
jaspModuleTools' bundling step and stalled the whole build.
Claude-Session: https://claude.ai/code/session_01GGe7kUhooMFuzY6GaDb2w9
Upstream JASP ships no Linux binary, so the existing AUR package builds
everything from source including several hundred R packages. This builds
that once in CI and publishes the result as jasp-desktop-bin.
JASP links against the system R and Qt6 rather than bundling them, so a
rebuild is triggered by an Arch r/qt6-base bump as well as by a new JASP
release, and the generated PKGBUILD pins depends to the versions actually
built against.
agw