Commit Graph
219 Commits
Author SHA1 Message Date
3AvisD 75da66c1ea docs: add scope and terminology guidance
Add docs/COMPLIANCE.md with the project's compatibility scope: no DRM/anti-cheat circumvention, no third-party game assets, and neutral terminology for issues and PRs. Link it from README and NOTICE.
2026-10-19 23:06:47 -06:00
3AvisD 54edf6fca1 patches, tools, docs: use neutral integrity-check wording and state the compatibility scope
The patch comments and the Mach-O strings filter named an anti-cheat product. Use neutral wording; the patches still differ only in comments. The platform-identity document now says the overrides are for compatibility, not a bypass of anti-cheat, telemetry, licensing or platform controls.
2026-10-19 22:47:59 -06:00
3AvisD 57660f168d flake.nix: mark the desktop entry and package as unofficial
Keep the executable names and StartupWMClass; only the user-facing name, comment and descriptions change.
2026-10-19 22:47:59 -06:00
3AvisD 5cf12f4f06 README, NOTICE: add trademark notice and remove the gameplay clip
Consolidate the non-affiliation note into a dedicated Trademarks section, replace the embedded gameplay video with a pointer to the status document, and reword the takedown note as a continuity statement. Describe the compatibility tools as referring to public API names and constants.
2026-10-19 22:47:59 -06:00
3AvisD 7947164c2c darling-lol: use an original project mark instead of the game logo
The bundled darling-lol.svg/png are the League of Legends logo, a Riot Games trademark, used as the project icon. The README already states that the project is not affiliated with or endorsed by Riot Games, so the icon should not be someone else's mark.

Replace both files with an original D monogram. Same file names, so flake.nix and the desktop entry keep working. The mark is provisional and easy to swap for a final one. No functional changes.
2026-10-19 22:36:55 -06:00
Kalah d3ecc7bea5 Merge pull request #101 from PolarSeeker77/docs/nvidia-reports
Document NVIDIA reports instead of "NVIDIA is not validated"
2026-10-19 19:08:32 +02:00
Kalah d4d5d8adc9 Merge branch 'main' into docs/nvidia-reports 2026-10-19 19:08:01 +02:00
Kalah 6865e1eb6e Merge pull request #100 from lucannez64/fix/host-dns
Use the host DNS resolver in Darling
2026-10-19 19:06:55 +02:00
lucannez64 5f71c82dc5 Bypass proxies when checking guest DNS 2026-10-19 18:22:17 +02:00
Polar bf02f7abee Document NVIDIA reports instead of "NVIDIA is not validated" 2026-10-19 16:21:26 +00:00
lucannez64 e47f05f424 Use the host DNS resolver in Darling 2026-10-19 18:11:49 +02:00
Kalah ca821262e2 Merge pull request #88 from lucannez64/fix/x11-vrr
Fix Metal VSync and VRR presentation together
2026-10-19 13:00:57 +02:00
lucannez64 791d4ad74e Keep the merged X11 patch in canonical diff format 2026-10-19 13:00:24 +02:00
lucannez64 ff8a985057 Merge current main into the tested VSync and VRR fixes 2026-10-19 12:59:58 +02:00
lucannez64 6bfdfbcac7 Merge VSync and VRR fixes for combined testing 2026-10-19 12:15:09 +02:00
Kalah 8bc63ed741 Merge pull request #96 from lucannez64/fix/nsopenpanel-close
Fix NSOpenPanel Open/Cancel not closing the panel (fixes #85)
2026-10-19 09:35:43 +02:00
lucannez64andClaude Sonnet 5.5 d3adda86fa Openpanel probe: expand the tree to the subfolder before selecting, print selection diagnostics
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
2026-10-19 09:33:30 +02:00
lucannez64andClaude Sonnet 5.5 66474a3c76 Fix openpanel probe: keep trigger note and panel alive across modal pools
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
2026-10-19 09:28:34 +02:00
lucannez64andClaude Sonnet 5.5 70a2feec42 Add standalone NSOpenPanel Open/Cancel probe for issue #85
Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
2026-10-19 09:19:42 +02:00
lucannez64andClaude Sonnet 5.5 ee322b43e1 Make NSOpenPanel Open/Cancel always end the modal session
Open with no row selected passed nil to -_setFilename:, and NSOpenPanel boxed it in an
array literal, so the action threw before the modal session was stopped and the panel
stayed up. Both buttons now go through one helper that stops the session only when
this panel is the modal window and always orders the panel out; Open accepts the
selection (filtered by canChooseFiles/canChooseDirectories and multiple-selection),
or the shown directory for directory choosers. Escape cancels.

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
2026-10-19 08:49:04 +02:00
Kalah c754315072 Merge pull request #89 from lucannez64/fix/splash-stays-mapped
Keep hidden X11 windows unmapped
2026-10-19 08:25:12 +02:00
lucannez64andClaude Sonnet 5.5 84e39e3b48 Keep hidden X11 windows unmapped
Restack and makeKey no longer map a window AppKit considers hidden, and
setIsVisible: NO always unmaps, so the client's splash stays hidden.

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
2026-10-19 08:00:41 +02:00
Kalah 5136911656 Merge pull request #87 from ekisu/fix/cocoa-window-decorations
Fix client-side window decorations and LeagueClient header dragging
2026-10-19 07:43:05 +02:00
Kalah 8fc7d4dff3 Merge branch 'main' into fix/cocoa-window-decorations 2026-10-19 07:40:37 +02:00
lucannez64 041591352e Present full-window Metal layers on the managed X11 surface for VRR 2026-10-19 07:35:16 +02:00
Kalah 388d34df19 Merge pull request #77 from lucannez64/fix/overview-stuck-super
X11: release modifiers swallowed by the GNOME overview (#74)
2026-10-19 07:17:06 +02:00
Kalah 6c9abce695 Merge branch 'main' into fix/overview-stuck-super 2026-10-19 07:13:28 +02:00
ekisu b6c70354b3 Recompile NSEvent in incremental AppKit runtime builds 2026-10-19 01:15:32 -03:00
ekisu 8f97990d03 Implement Cocoa local event monitors for native client dragging 2026-10-19 01:15:03 -03:00
ekisu 41da6d9e0a Fix X11 decorations for client-drawn Cocoa windows
Synchronize Motif decoration hints on style and titlebar-property changes, including native-window recreation and restoring ordinary WM decorations. Add a ten-case standalone Cocoa/X11 regression probe and document validation on KWin/Xwayland.
2026-10-18 21:06:22 -03:00
lucannez64 27553d8874 Fix Metal VSync propagation to window swap intervals 2026-10-19 01:10:03 +02:00
Kalah 337aa79c27 Merge pull request #80 from lucannez64/fix/rootless-update-test
Fix rootless update test after the shared-identity step
2026-10-19 01:00:21 +02:00
lucannez64andClaude Opus 5.5 cb9fe9babf tests: expect the shared-identity step at the end of rootless update
#79 made update.sh stop Darling again after the overlays and run
configure-darling-platform.py --replace-known-defaults --write. The
rootless update test still expected the old call list and failed. Stub the
identity tool, expect the stop + identity calls last, and check a failed
install never reaches them.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:58:45 +02:00
Kalah 7f544d063b Merge pull request #76 from lucannez64/fix/mic-input
Pass microphone input through for voice chat
2026-10-19 00:56:58 +02:00
lucannez64 01fdb0f28f Merge remote-tracking branch 'origin/main' into fix/mic-input 2026-10-19 00:55:17 +02:00
lucannez64andClaude Opus 5.5 d2ec718c48 AUHAL: accept Vivox's playback format and trace recorded levels
The issue #75 trace shows recording now runs, but Vivox's playback unit
(mic test loopback, other players' voices) sets interleaved Float32 stereo
at 32 kHz on input scope element 0 and got -10868. Apply the relaxed rule
to both client sides (input scope element 0, output scope element 1); the
playback path already interleaves for the device and PulseAudio resamples.

AudioUnitRender(bus 1) now traces frames and peak level (no samples) so a
log shows whether the client pulls sound or silence. The input check also
sets Vivox's playback format and expects unsigned 16-bit to be refused.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:41:10 +02:00
lucannez64andClaude Opus 5.5 2462b2b755 AUHAL: only accept float or signed integer recording formats
RenderBus pads missing frames with zeros, which is loud for unsigned PCM
(silence is 0x80 for 8-bit, 0x8000 for 16-bit). Reject unsigned and 8-bit
formats on element 1; voice engines use float or signed 16-bit.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:25:23 +02:00
Kalah 240cbc462d Merge pull request #79 from lucannez64/fix/replace-shared-identity
Provision a platform identity on setup and replace the shared serial/UUID on update
2026-10-19 00:22:48 +02:00
lucannez64andClaude Opus 5.5 c8cbb1ffe7 install-runtime.sh: provision a platform identity for new prefixes
When install-runtime.sh creates the prefix (first-time setup, regular or
Nix), it now runs configure-darling-platform.py --provision --write once
the runtime is stopped, so each new install gets its own synthetic serial,
MLB, UUID and ROM. A prefix counts as new the same way Darling decides it:
the directory did not exist before the first `darling shell`. Existing
prefixes keep their identity on reinstall and update.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:22:07 +02:00
lucannez64andClaude Opus 5.5 d443e631c2 update.sh: replace the shared serial and UUID automatically
Older setups shipped serial C02ZL0GVMD6N and UUID
6F3B2A1E-8C4D-4E5F-9A7B-1D2C3E4F5A6B to every install. After installing,
update.sh now runs configure-darling-platform.py --replace-known-defaults,
which provisions the prefix only when its saved identity still has one of
those values. Other custom values (imported serials/UUIDs, model, board,
manufacturer, ROM, MLB) are kept; prefixes with no saved identity or only
custom values are not touched.

provision() also drops a standalone system-id or provisioning_UDID derived
from the shared UUID instead of keeping it.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:17:22 +02:00
lucannez64andClaude Opus 5.5 f7be4a2f93 AUHAL: accept interleaved recording formats on element 1
Vivox sets interleaved Float32 stereo on output scope element 1 and
AUBase's canonical-only rule rejected it with -10868, so voice chat never
started recording (issue #75 trace: 20 attempts, all failing there).
A real AUHAL converts any linear PCM; element 1 now accepts native-endian
float32 or 8/16/32-bit integers in either layout, which PulseAudio records
and RenderBus copies. Playback (element 0) keeps the canonical rule.

The input check gains a Vivox-style run: interleaved stereo 44.1 kHz with
the device set on element 1.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-19 00:13:16 +02:00
Kalah 003daa7fc3 Merge pull request #78 from lucannez64/fix/close-client-cooldown-boot
close-league-client-in-game: don't block closes right after boot
2026-10-18 23:20:24 +02:00
lucannez64andClaude Opus 5.5 87484f0385 close-league-client-in-game: don't block closes right after boot
last_close started at 0.0 and the cooldown compares it with
time.monotonic(), which counts from boot. For the first 30 s of uptime
every close was refused as "in cooldown". That made the tests flaky on
freshly booted CI runners (seen on PR #76). Use None for "never closed".

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 23:16:18 +02:00
lucannez64andClaude Opus 5.5 662b3d414f X11: take modifier masks from the server's modifier map
Right Alt (keycode 108) was hard-coded as Mod5/AltGr. On layouts where it
is Mod1, the swallowed-release reconcile accepted Mod1 as held but rebuilt
the state with Mod5, so the guest could see Option and Function together.
Read the per-key masks from XGetModifierMapping (refreshed on
MappingNotify) and use them everywhere; the old table is the fallback for
unmapped keys. The probe now checks Right Alt against whichever mask the
backend reports.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 23:13:07 +02:00
lucannez64andClaude Opus 5.5 92e1b5d5ab riot-audio-input-check: detect a synchronous access callback by thread
The handler read a flag the caller wrote without synchronization, and
inferred "called inside the request" from whether the call had returned,
which misfires when the handler runs on another thread first. Compare the
handler's thread with the caller's instead and keep the shared results
atomic.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 23:08:48 +02:00
lucannez64andClaude Opus 5.5 97c0e596bc X11: release modifiers whose release the compositor swallowed
GNOME's overview delivers the Super press to the game window, eats the
release and hands the keyboard back without a FocusOut. The backend kept
Command held, so every later key and click reached the game as a Command
shortcut until a real focus change reconciled it (issue #74).

Each key, button, motion and crossing event carries the server's modifier
state; release any modifier key we track as down whose modifier is missing
from it. Synthetic events from other clients are ignored.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 23:03:23 +02:00
lucannez64andClaude Opus 5.5 a65818a7f5 Pass microphone input through CoreAudio and AVFoundation
Voice chat (Vivox) saw no capture device and Chromium was told mic
access was denied. Fixes for issue #75:

- AVCaptureDevice: audio is Authorized when CoreAudio has a default
  input device; requestAccess answers asynchronously. Video stays Denied.
- PulseAudio input device reports 2 channels on input scope, 0 on output
  (was copied from the output device).
- PulseAudio input stream: real sample/host timestamps, one device
  buffer per IOProc call, holes as silence, IOProc errors no longer cork
  the stream, record buffer attrs sized to the device buffer.
- AUHAL: two elements per scope like macOS, non-blocking
  AudioUnitRender on bus 1 with planar support, device records in the
  bus 1 format, input-only units route CurrentDevice/BufferFrameSize to
  the input device, input start errors handled.
- New offline check tools/riot-audio-input-check.cpp.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-18 22:55:11 +02:00
Kalah f4f01dae12 Merge pull request #73 from PolarSeeker77/fix/x11-transparent-windows
X11: give transparent windows an alpha visual, float translucent splash windows
2026-10-18 20:57:01 +02:00
Kalah 4ff2f11c26 Merge branch 'main' into fix/x11-transparent-windows 2026-10-18 20:56:29 +02:00
Kalah 0b02889968 Merge pull request #72 from 3AvisD/fix/close-league-client-in-game
fix(launcher): close the League client UI while a match runs
2026-10-18 20:56:18 +02:00