Files

T

enricobuehler 5c05246098 feat: M10 — FFmpeg is gone from the client

cargo tree -p punktfunk-client-session finds no ffmpeg. The host still does,
which is the whole point: pf-encode keeps libavcodec unconditionally and no
host workflow, packaging script or licence file was touched.

Deleted: crates/pf-ffvk, video_vulkan.rs, video_vaapi.rs, video_libav.rs, the
libavcodec half of video_d3d11.rs, the av_log machinery, ffmpeg::codec::Id as
the decoder's vocabulary (the quic CODEC_* wire constants now serve, which is
why the evidence table was keyed on them), DecodedImage::VkFrame and ::Dmabuf,
the presenter's AVVkFrame lane, and the ffmpeg-fallback feature with
everything behind it. DrmFrameGuard collapses from an enum to a newtype, which
removes an unsafe impl Send. Roughly 25,000 lines.

Then the CI, packaging, licensing and docs work the plan's §6 lists: the
Windows workflows lose FFMPEG_DIR, PF_FFVK_VULKAN_INCLUDE and their PATH
prepend; the MSIX loses its DLL wildcard; the client .deb stops emitting libav
sonames on its own because depends come from dpkg-shlibdeps; arch, flatpak and
nix drop the dependency; and the README's "FFmpeg 7 or 8" contract narrows to
the host.

Three defects reached users' machines in the first cut, and none was in the
deletion itself.

All three desktop Settings UIs offer vulkan, vaapi and d3d11va as stored
decoder values, so those strings sit in shipped settings files today. Refusing
them by name — which is the correct rule for a stale pin — would have bricked
every upgraded client whose owner ever touched that dropdown. They now migrate
onto the native rung for the same hardware family, at decoder construction AND
at each dialog's lookup, because a legacy value that matches no preset
displays as "Automatic" and silently rewrites the user's preference on the
next save.

M9's evidence filter was deleted on the argument that with no libavcodec twin
below, barring an unproven rung removes hardware decode rather than moving
down one rung. That is true on Windows and false on Linux for Intel and every
unknown vendor id, where prefer_vulkan_first is false and the order is
native-vaapi → native-vk: a rung that has decoded nothing anywhere sitting
above one that is 250/250 on three drivers. Every Intel Linux desktop would
have moved from libavcodec VAAPI, shipping for years, onto pf-vaadec by
default — and a rung that constructs and then produces wrong pixels leaves
only by the error-streak demotion, which this codebase already documents as
not tripping on the B580's strobing. The filter is restored as a narrow, pure,
testable rule: an unproven rung yields to a proven one, and to nothing else.
Windows deliberately passes no rung below, because that vendor family is the
one with a measured wrong-pixel report against Vulkan decode, and trading no
evidence for evidence of corruption is the wrong direction.

And the notices still said FFmpeg was bundled. The root file is what both
desktop clients include_str! and what the MSIX ships, three lines under the
new card saying no FFmpeg is bundled; Apple's Acknowledgements said it too, on
iOS, tvOS and macOS. The generator now emits four per-client files scoped by
transitive closure — 0 FFmpeg mentions in each, verified — while the root file
keeps it for the host. That also ends the standing false attribution of
ffmpeg-next, GTK4, windows-rs and the NVENC SDK to an iPhone.

Windows has no reachable box, so it was compiled instead: a cross clippy at
-D warnings on x86_64 and aarch64-pc-windows-msvc with the C toolchain stubbed
so build scripts run without linking. That gate immediately caught an
include_str! path one directory too deep, which nothing else could have.

Gates: container clippy -D warnings, 160 tests, workspace check, both Windows
targets clean, client ffmpeg count 0 and host 2. The four decode crates are
untouched, so the hardware rungs' 250/250 stands.

⚠ Owed and unrun: no GPU has executed any of this milestone. M8's on-glass
software check, M7's D3D11 and VAAPI AV1 hardware legs, and M9's field bake
all still want hardware, and the bake window and criteria remain the user's.

2026-08-07 08:58:47 +02:00

4.8 KiB

Raw Blame History

Contributing to Punktfunk

Thanks for your interest in contributing!

Licensing of contributions (inbound = outbound)

Punktfunk is dual-licensed under MIT OR Apache-2.0.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as MIT OR Apache-2.0, without any additional terms or conditions.

By opening a pull request you agree to license your contribution under these terms. This is the standard Rust-ecosystem "inbound = outbound" model; it keeps the project's licensing unambiguous (including the Apache-2.0 §5 contributor patent grant) and any future relicensing clean. You retain the copyright to your contributions.

Do not paste copyleft (or otherwise incompatibly-licensed) code

The single thing that could poison the permissive license is copied source from a copyleft project. Several adjacent projects (Sunshine, Apollo, Moonlight) are GPL-3.0. You may study them and reimplement a technique, protocol, or wire format — those are not copyrightable — but never paste their code, and do not translate a GPL implementation line-by-line. When a comment credits prior art, make clear it is an independent reimplementation, not a copy. The same applies to any third party's code under a license incompatible with MIT/Apache.

If you add a new third-party dependency, it must be permissive (MIT / Apache-2.0 / BSD / ISC / Zlib / Unicode-3.0 / etc.). about.toml holds the accepted-license allow-list; regenerate the attribution file with scripts/gen-third-party-notices.sh when the dependency tree changes.

Prerequisites

The Rust toolchain is pinned exactly in rust-toolchain.toml; rustup installs it for you the first time you build, so don't override it — a different rustc reformats files nobody touched.

The workspace links real system libraries, so a bare cargo build --workspace fails on a stock machine. The authoritative list is what CI installs, in ci/rust-ci.Dockerfile — on Ubuntu 26.04, which is what gets you FFmpeg 8:

sudo apt install build-essential clang libclang-dev pkg-config cmake \
  libavcodec-dev libavformat-dev libavutil-dev libswscale-dev libavfilter-dev libavdevice-dev \
  libpipewire-0.3-dev libopus-dev libwayland-dev libxkbcommon-dev \
  libgl-dev libegl-dev libgbm-dev \
  libgtk-4-dev libadwaita-1-dev libsdl3-dev \
  libvulkan-dev

(The last two groups are the Linux client shell and the Vulkan session presenter; skip them only if you never build those crates. libvulkan-dev is for the LOADER's pkg-config/soname — ash dlopens it, and the client links no FFmpeg at all, so no libav*-dev appears here. scripts/bootstrap-ubuntu.sh sets up an Ubuntu capture-test host — NVIDIA, Sway, PipeWire — and is not a substitute for the list above.)

Before you push

Enable the repo git hooks once per clone — they run the exact rustfmt gates CI runs (main workspace + the UMDF driver workspace) on every commit and push, so a push can never fail CI on formatting alone:

git config core.hooksPath scripts/git-hooks

Then the usual full pass. Use --locked as CI does — otherwise a silent Cargo.lock update can pass locally and fail CI:

cargo fmt --all --check
cargo clippy --workspace --all-targets --locked -- -D warnings
cargo test  --workspace --locked

Two more gates that only apply to some changes:

Touched web/ or docs-site/? CI builds and typechecks both. Run, in that directory:
```
bun install && bun run build && bun run lint
```
Build first — it generates the API client / MDX typegen that the typecheck imports.
Touched Windows- or Linux-gated code from another OS? scripts/xcheck.sh windows (or linux) type-checks and lints that platform's #[cfg(target_os = …)] code in about a second, instead of waiting for the CI job that compiles it.

Generated artifacts are checked in. include/punktfunk_core.h (cbindgen) is regenerated by the build and CI fails if the committed copy drifts. api/openapi.json is not gated — nothing in CI regenerates or diffs it, so regenerate and commit it yourself whenever you touch the management API, and copy the snapshot the docs site serves:

cargo run -p punktfunk-host -- openapi > api/openapi.json
cp api/openapi.json docs-site/public/openapi.json

Match the surrounding code's comment density and naming. Commit messages end with the Co-Authored-By trailer (see git log).

See the README's Build & test section for the extra dev commands (the FEC loss harness, the standalone C-ABI proof) and Design invariants for the rules a change is expected to hold to, and the docs site for architecture and per-platform guides.