Screen share 30: Packaging and CI for screen share (#177) #248
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "screenshare-30-packaging-ci"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Stacked on: screenshare-21-app-screenshare-runtime, screenshare-18-hw-encode-linux (18 is already contained in 21, so the branch is 21 + one commit).
What I built
packaging/linux/AppRun: checks every declared load-time library (libasound.so.2,libpipewire-0.3.so.0,libglib-2.0.so.0,libgobject-2.0.so.0,libgio-2.0.so.0) withldd, names all the missing ones in one message and the packages to install on Debian/Ubuntu 22.04, Ubuntu 24.04+ (t64), Fedora and Arch (packages deduplicated), exit 127. Withoutlddit starts the client as before. This covers #243's AppRun item.release.yaml:linux-cleaninstalls exactly the declared packages (libasound2 libpipewire-0.3-0 libglib2.0-0) in the bare 22.04 and checks that--debug-infoprintsscreen capture:,stream audio:,encode:anddecode:. The release notes list ALSA/PipeWire/GLib per distro, the ScreenCast portal (Wayland) and libva as optional. A comment says whylibclangis present for thepipewirecrate's bindgen.ci.yaml: on Windows,Report disk userecords whether the run was a cold build (no target dir at the start, i.e. after the cleanup). It publishes its figures asdisk-windows-cold.txt/disk-windows-warm.txton the prereleaseci-disk-windows, replacing the previous file of the same kind. Publishing is best effort and never fails the job. This exists because this Forgejo has no job-log API, which is what kept #127 open.MIN_FREE_GBgoes from 12 to 18. The Linux libclang comment is also updated.--debug-infonow mentions the screen-share section.xtask/README.mdand thepackage.rs/libs.rsdocs are updated.crates/fluxer-voice/README.mdhas a target-dir size table (#127).libs.rsunit tests check thatLINUX_DECLAREDstays in step with AppRun, withrelease.yaml'slinux-cleaninstall line and with the README Install section.tests/package.rshas AppRun behaviour tests (all screen-share libraries missing; ALSA and PipeWire missing together). The AppImage test also asserts that it loadslibpipewire/GLib, and--debug-info(Linux and Windows) asserts the screen-share section.How I tested it
FLUXER_SCREENSHOT_TESTS=1 cargo xtask ciis green locally (base1b64a8bwas green before any change).dashinubuntu:22.04with a fakeldd, and the real extracted AppImage in a bareubuntu:22.04: it lists all five libraries and the three apt packages, exit 127.pw_/spa_symbols are exported by 22.04'slibpipewire-0.3.so.0(0.3.48-1ubuntu3.2), and all 42g_symbols by its GLib. My local AppImage was built on Arch, so it cannot fully start on 22.04 (glibc 2.39+). The full start on a clean machine is whatlinux-cleandoes on the release runner, which builds on 22.04.cargo test --workspace --no-runwith CI's settings (line-tables-only, no incremental) is 13.2 GB, plus 3.0 GB for the release build, so 16.2 GB. A plain localcargo xtask ciis about 43 GB.Deviations / for the reviewer
MIN_FREE_GB= 18 is the Linux cold figure plus headroom. That is tight against about 18 to 23 GB free after a cleanup, so worth watching on the first Windows runs.--debug-infoin CI: this was already the case throughpackage_produces_portable_exe_on_windowson Server Core. It now also asserts the screen-share section. Nothing new was needed for Windows packaging (d3d11, dxgi, dwmapi and shcore are system DLLs, checked bycheck_windows).deny.toml: no change was needed (cargo denyis green).ci-disk-windowspublish step is PowerShell that I could only parse-check locally with pwsh, not run. It is wrapped in try/catch and exits 0.Closes #177
🤖 Generated with Claude Code
Independent review: Screen share 30 (packaging-ci)
Reviewed
1b64a8b..31723bdcold (9 files: workflows, AppRun, xtask package/libs + tests, READMEs).Checks
libpipewire-0.3-0(+ GLib) inLINUX_DECLARED, AppRun, releaselinux-clean, README; the portal and libva are documented as optional. Windows--debug-infoalready runs in CI throughpackage_produces_portable_exe_on_windows(Server Core) and now also asserts the screen-share section. Those labels (screen capture:,stream audio:,encode:,decode:) matchapps/fluxer/src/debug_info.rs::render, which is not cfg-gated.libva-devand libclang (viaclang-21->libclang1-21) are present.deny.tomlis unchanged and needed no change.LINUX_DECLAREDkeeps its shape.secrets.GITHUB_TOKENonly in a curl header and never echoes it. Its catch prints only the exception message.FLUXER_SCREENSHOT_TESTS=1 cargo xtask ci: green. The 3declared_everywhereunit tests, the 4 AppRun tests andpackage_produces_appimage_on_linuxall pass.Also checked: AppRun's soname matching has no substring collisions. The
addeval only takes fixed package names. The script works underset -eu. The package names are correct for Debian/22.04, 24.04 t64, Fedora and Arch.Findings
crates/fluxer-voice/README.md, and the table says "see below". #127 stays open, which is right; the first cold run'sdisk-windows-cold.txtmust be copied into the table.MIN_FREE_GB=18is tight against the 17.9-22.7 GB free that earlier Windows cleanups recorded. The cleanup does not throw when it stays below the minimum, so this cannot turn CI red by itself, but a cold build may still run out of disk. Retune once the cold figure is published.ci-disk-windowspublish PowerShell was only parse-checked, never run. It is best effort (try/catch, exit 0), so it cannot fail the job. It creates a tagci-disk-windowson the default branch, like the existingci-log-*prereleases do.Verdict: approve. No blocking findings; local CI is green. Forgejo runner results were not checked by this review.