* feat(ubisoft): keep Denuvo game tokens per account and product until the game invalidates them
* feat(ubisoft): sync cloud saves through Ubisoft's cloud save service
* fix(xbox): turning cloud saves off passes the offline mode the game save runtime reads
* fix(ubisoft): Ubisoft games list their shared prefix, so they get a saves and cloud tab
* feat(ea): sync cloud saves through EA's cloud sync service
* fix(ea): sync at EA launch and exit, keep one rule snapshot, and check the cloud copy under the write lock
* fix(ubisoft): exact slot versions, scoped save export, and a Denuvo cache bound to its account
* fix(ea): fresh rules per sync, checksums required, backup before restore, and sign-in shown on the saves tab
* fix(ubisoft): launch-time sync on the shared prefix, saves scoped to the game, and Denuvo bound to each launch's account
* fix(ea): back up only the saves a restore replaces, refuse truncated manifests, and need a master title id
* fix(ubisoft): treat a slot whose bytes changed since the last sync as a conflict before uploading over it
* fix(ea): one resolved save folder per sync, refuse case collisions, and walk each folder once
* fix(ubisoft): sync only identified slots, keep live saves and backups safe, and tie a pending eviction to its account
* fix(cloud): a forced download fetches every saved conflict and keeps the ones it could not write
* test(ea): keep the cloud save cases in EaTest
* fix(ubisoft): report only written slots, keep drift conflicts, read whole headers, and never rename a named eviction
* fix(ea): keep real save folders in the inventory, back up any restored file, refuse duplicate or malformed cloud listings
* fix(cloud): commit Ubisoft slots one at a time and stop on a newer cloud write, name skipped and blocked slots, keep unsent conflicts
* fix(ubisoft): acknowledge an invalidation only once its marker is published and never serve a kept token after it
* fix(cloud): a cloud request the launcher cannot write still starts the game and keeps the exit upload
* fix(cloud): scope Ubisoft sync state to the account and product, compare alike slots, keep committed progress, and name partial slots
* fix(ea): ask for sign-in after a rejected refresh, upload saves reached through folder aliases
* fix(cloud): a launch that stops waiting for the cloud download claims it back, so no save is written once the game starts
* fix(ubisoft): name the account in an invalidation marker and never hold the token lock across a pipe write
* fix(ea): only invalid_grant asks for sign-in again, and case collisions count only among synced saves
* test(epic): let the launcher's ask doubles take the claim flag
* fix(ea): apply save rules in their order, match bare patterns inside the save folder, and resolve LocalLow and install folders
* fix(ubisoft): stamp a fresh identifier into a new save that has none, so it can sync
* fix(ubisoft): answer a burn ticket request as unavailable instead of leaving the game waiting
* fix(cloud): read EA manifests with the XML parser, clear conflicts whose sides match, and bound the claim and stall tests
* feat(xbox): install MSIXVC2 packages from their encrypted content boxes
* feat(xbox): run a package's custom install actions once per prefix
* fix(xbox): launch the program a launcher menu names instead of the menu
* fix(xbox): check free space before an MSIXVC2 install starts
* fix(xbox): keep a forwarded launch path relative when the install folder is reached through a link
* fix(xbox): rebuild the MSIXVC2 patch on the final legacy install
* fix(xbox): keep a backup whose release fails, and test an interrupted restore
* fix(xbox): verify resumed msixvc2 files, contain writes, decode install action arguments
* fix(xbox): keep one completion record per msixvc2 file, never follow a linked install state
* fix(xbox): drop a torn msixvc2 completion record before appending to it
* fix(xbox): repair a torn msixvc2 record from its tail instead of rereading it
* fix(agent): show the side menu's agent page whenever the tab is shown
* feat(agent): permission modes for every harness, asking first by default
* test(agent): give the drawer page tests their agent page explicitly
* fix(agent): fail turns on lost stdin writes, wait for Codex to initialize, keep approvals visible
* fix(agent): redo setup on a kept process after it failed, keep approvals in view
* fix(agent): keep full-page panels closed while an approval waits
* fix(session): show the recorded reason when a session fails to start
The ended screen now shows any recorded start-up failure message except a plain guest exit, not only PROOT_MISSING. The graphics runtime staging and game environment publish failures now record GRAPHICS_RUNTIME_FAILED and GAME_ENVIRONMENT_FAILED with their own messages instead of looking like a guest exit.
* i18n(session): translate the start failure messages
---------
Co-authored-by: Tapuuk <177422350+Tapuuk@users.noreply.github.com>
Co-authored-by: maxjivi05 <maxjividen@gmail.com>
* feat(xbox): install legacy EAppx titles through Windows Update
A legacy Store title does not ship as MSIXVC. It ships as an encrypted
EAppx bundle, which streaming rejects, so the store could list the game
and never install it.
Accept EAPPX and EAPPXBUNDLE beside MSIXVC, carry the format through the
catalogue, the install record and the cached library, and dispatch
legacy-install for those formats.
Two backend patches do the work:
- droiddeck-legacy-packages.patch lets the plan, license and extract
verbs handle a legacy package, and expands a bundle in place (nesting
bounded at three), decrypting each inner EAppx with the same keys.
- droiddeck-legacy-install.patch adds legacy-install: resolve,
license, download, extract, hoist AppxManifest.xml to the game root.
Resolution asks the Xbox content service first and Windows Update
second, because some legacy products return no files from the content
service at all. The WU endpoint needs no Xbox credential but chains to
Microsoft's ECC Product Root CA 2018, which Android's CA bundle does
not carry, so that one public root is pinned for those requests.
The manifest reader accepts AppxManifest.xml as well as
MicrosoftGame.config, since a legacy package carries only the former.
Running it against a real title exposed four things that only show up at
full size:
- A licence names a key by its 16-byte content id, while a bundle header
names it as a (content id, package id) pair. Reading the 16-byte id
yields a different pair, so a complete key file reported every key
missing and no bundle could be opened. Keys are now aligned against
the header's own ids before the header is judged.
- Windows Update answers with several updates whose PackageContentId
differs from the bundle's own. Licensing used to take the first and
fail with content-http 400; every candidate is now tried in order,
preferring the one the container is named after.
- A container already in .xodus-container is reused rather than fetched
again, so a failed run does not cost another download of tens of
gigabytes. The space preflight no longer counts a container twice for
the same reason.
- A bundle that carries add-on packages unpacks each into its own
folder, leaving the manifest one level down. Hoisting lifts the
largest folder that holds a manifest, so the game still lands at the
root beside its add-ons.
Unpacking now reports bytes written, so the install bubble fills while a
package is decrypted instead of holding a single unchanging word.
* fix(xbox): unpack a legacy package across cores and report the bytes
Unpacking one title ran as a single thread on a saturated core at about
40 MB/s. Files sit at independent offsets, so each worker now opens its own
handle and takes every Nth file, with a 1 MiB read buffer in place of the
8 KiB default that turned one 64 KiB block into eight reads. Measured on the
Forza Horizon 4 bundle, writes went from 40 MB/s to about 200 MB/s.
Two things the first full-size run exposed:
- The container was deleted only after the whole nested tree was unpacked,
so a title that holds its package as one member held the archive and the
decrypted copy at once. On this device that runs out of space partway
through the second stage. The container is now deleted once its members
are on disk, which frees the archive before they are decrypted.
- A bundle member is written as a single file, so a counter that advances
when a file finishes stays at zero for the whole of the longest stage.
The reported figure now also follows the growth of the destination while
that is the case, and never moves backwards when the container is freed.
Progress is exact during the per-file stage: each file's declared length is
added as it lands, instead of walking the whole destination every tick.
* docs(xbox): record the completed legacy install
The evidence section said a finished legacy install and its decrypted
contents were unverified, because the first run was still unpacking when it
was written. That install has now completed on the same device: the
85,219,173,025-byte container produced 79.3 GB at the game root with
AppxManifest.xml and the package DLLs directly beneath it, and the launcher
lists the title as installed.
It also records what unpacking costs, since that decided the change: one
thread held a core at 100% for about 40 MB/s and four reach about 200 MB/s,
and releasing each container once its members are on disk is what keeps the
peak footprint at one package rather than the archive plus its contents.
The launch is not part of this evidence, and the entry says so.
* fix(xbox): place the 64-bit framework beside a legacy title
A Windows Update framework category lists one package per architecture under
identifier-only names. The first package was unpacked, which on this device was
the x86 build, so a 64-bit title got 32-bit DLLs and every load failed with
STATUS_INVALID_IMAGE_FORMAT. Candidates are unpacked now and only the one whose
manifest declares ProcessorArchitecture="x64" is kept.
The framework folder was also never created, so a finished download failed at the
move with "no such file or directory".
unzip.rs belongs to this patch: legacy-install calls it and the recipe left it out,
so a rebuild from the pinned source could not compile.
* feat(xbox): report a UWP entry point at launch
An unpacked package can name a COM class as its entry point instead of
Windows.FullTrustApplication. That class is a UWP application, which expects the
app model to activate a view; Wine has no host for it, so the launch died inside
the loader with no explanation. The launch now reads the entry point and reports
uwp-entry-point, and full-trust packages are unaffected.
Forza Horizon 4 is such a package: its manifest names Forza.App. Its frameworks
are installed and the loader is clean before that point.
* fix(xbox): report why a launch was refused
A failed prepare replaced the backend message with a fixed sign-in hint, so a title
that was not installed or a runtime that refused all read as a bad account. The
reason is kept, with paths and URLs dropped, and written to the store log.
* fix(adb): dismiss the keyboard on Search and Go
Search and Go were terminal IME actions but were not handled, so focus stayed in
the field and the keyboard covered the results it had just fetched.
* fix(xbox): keep a cache row without a legacy format plain
Decoding added a format key with an empty value to every cached title, so a
library restored from storage no longer matched the one that was saved.
* test(ui): match the manifest message a legacy package reports
The message covers both a MicrosoftGame.config and an AppxManifest, so the
assertion follows the text the content dialog now prints.
* fix(xbox): refuse to start a title that needs the Windows app model
A legacy package that names an entry point other than
Windows.FullTrustApplication is a UWP application. Wine has no app-model
host, so starting its executable crashes inside the loader and the launch
looks like nothing happened. Read the entry point from the manifest the
installer already parses and stop the Play action with a message instead.
Forza Horizon 4 names Forza.App, so it installs and reports as installed
but cannot be started here yet.
* fix(xbox): dispatch on the container format of the chosen package
The catalogue records the format of the first installable desktop package, and
the picker only replaced the content id. An offer that lists MSIXVC and a legacy
container would then hand the legacy id to streaming, or the streamed id to
legacy-install, and the install failed after resolution had succeeded.
The choice now carries its own format and the catalogue value only fills a gap.
* fix(xbox): keep an extraction reason that the sanitizer rewrote
The extraction sanitizer replaces anything outside letters, digits, space, dot,
underscore and dash with a question mark, so a reason such as "(os error 28)"
arrived as "?os error 28?" and the allowlist dropped the line. The failure then
showed only the stage code, with no cause.
Add that replacement character to the class the allowlist accepts.
* fix(xbox): print the reason a launch was refused
The account channel sends a reason with a refusal, and the guest replaced it
with a fixed sign-in message, so a title that was never installed still read as
a bad account. The guest now shows the reason it was given, and a token holding
a path is dropped whole instead of having its slashes removed and its directory
names kept.
* build(xbox): leave RUST_TEST_THREADS unset when the host has no value
The recipe always passed the variable, empty when the host had none, and the
pinned test harness rejects an empty count. On an ARM64 host that runs cargo
test, so the default local build failed before compiling.
* fix(xbox): rebuild the backend patches from the tree that ran
The two patches in this branch were not the ones the verified install ran, and
one of them did not compile. `unzip.rs` inflates deflated entries but the
dependency that provides the inflater was never added, so `prepare.py` produced
a tree that failed at `flate2::read::DeflateDecoder`. Anything else missing fell
into the same gap: the install stage reported no bytes, so the bar held one
value for the whole unpack, and a container a stopped run had finished was
downloaded again.
Rebuilt from the patched source, and fixed the defects a review of the patch
found:
- Unpacking reports bytes again: the workers count each file's declared length,
and for the single-file bundle member the folder's growth stands in, which is
what the device run showed.
- A finished container is reused, so a retry after a failure past the download
does not spend hours fetching it again. The space preflight counts it as
reclaimable, since the retry truncates it first.
- An install over a folder that already has a manifest replaces the old files
instead of leaving the new package nested and reporting success. A first
install still refuses to merge into an existing target.
- The manifest search looks at the depth a nested bundle writes, not only one
level down.
- `withdraw_libraries` removes only the DLLs `copy_libraries` recorded copying,
so a framework replacement cannot delete a library the title owns.
- ZIP64 sizes and offsets stay `u64`; truncating them to `u32` read the wrong
local header past 4 GiB.
- The extraction checks the blockmap length against the footer's before writing,
which the crate only does when asked.
- The container being unpacked is never treated as a nested member, so
`extract-eappx title.eappx .` no longer re-enters and deletes its own input.
- `flate2` is declared and the lock is updated.
`prepare.py` reproduces this from a pristine checkout and `cargo test` passes on
it.
* fix(xbox): apply the legacy install patch over the Nexus 0.2.15 backend
* fix(xbox): install only the newest x64 build of a legacy title and size the space check at its real peak
* fix(xbox): install a store package that is a plain archive instead of an XVD
* fix(xbox): refuse a package entry that ends before its recorded size
* fix(xbox): credit only complete staged containers toward the free space a legacy install needs
* fix(xbox): bound each package entry to its data and check ZIP64 decoded sizes
* fix(xbox): refuse duplicate package names and clear an unfinished download before measuring space
* fix(xbox): bound package entries by the next header and stage a plain package until it validates
* fix(xbox): hold the install folder lock through a legacy install
* fix(xbox): fall back to an older build when Microsoft refuses the newest one a license
* fix(xbox): keep the registered device across sign-out
* feat(xbox): explain device-limit and license refusals, with a link to manage devices
* fix(xbox): locate each build only once licensed, roll back a failed package swap, report the final license answer
* fix(xbox): skip trial and preorder packages, read runtime behavior, keep licensing timeouts
* fix(xbox): keep a failed package swap recoverable, size bundles up front, report key-write and framework errors cleanly
* fix(xbox): drop a pending manifest check on any newer launch, keep committed backups out of restores, match the app-model order
* fix(xbox): cancel pending launch checks on a new intent, remember entries a rollback could not remove
* fix(xbox): clear the removal list before a restore moves old files back
---------
Co-authored-by: TimmyCraft <46533729+TimmyTurner51@users.noreply.github.com>
* fix(ui): lay the color picker out side by side on wide screens
On a landscape screen the single column was taller than the screen. The wheel and swatches now sit beside the sliders and the hex field, so the whole picker is visible without scrolling.
* fix(ui): give the landscape color picker a larger wheel and one closing row
On a wide screen the picker still reached both screen edges: three sliders, the hex field and the buttons were stacked beside a small wheel. The wheel now fills the height of the sliders, and the comparison chip, the hex field, Cancel and OK share one row under them, so the dialog keeps a margin. OK is centered in both layouts.
* fix(ui): keep the column color picker in landscape windows narrower than 600 dp
The side by side layout needs 382 dp for its closing row before the hex field gets any width, so a small landscape window could squeeze the field to nothing.
* fix(ui): fit the side by side color picker on short landscape screens
On a 740 x 360 dp phone the closing row sat below the screen and had to be scrolled to. The wheel now takes its size from the screen height (up to the old 216 dp), and on a short screen the sliders get a shorter track and tighter spacing, so the whole dialog fits down to about 360 dp. Screens 468 dp tall and up, such as the Odin 3, look as before.
ColorPickerLayoutTest now also checks that OK is on screen without scrolling at 740 x 360, 800 x 384 and in portrait.
* chore: drop the added comments
---------
Co-authored-by: maxjivi05 <maxjividen@gmail.com>
* fix(input): the layout editor shows the same placement as the session
Editing a layout stored for one orientation while the editor is open in
the other drew raw stored coordinates, so elements turned into flat,
overlapping bars. Route the editor through OnScreenLayout like the
on-screen controls: matched layouts are regenerated for the current
screen, everything else follows the same fit, and the dirty check
compares against what was shown when the editor opened.
* fix(input): reset and scale in the editor frame the session saves
The editor draft now carries the frame it was rebased to: the gamepad
scaler is rebuilt with the editor surface's real dimensions when edits
begin, so reset regenerates legacy controls for the display the user is
editing instead of the primary display. Stock default positions keep
their own frame aspect and are refitted into the draft frame before
being applied, so reset on a differently shaped display lands where the
editor shows and no longer marks an untouched layout as changed.
* fix(input): refit only the default positions on reset and update the test callers
---------
Co-authored-by: maxjivi05 <maxjividen@gmail.com>
* fix(audio): hold host restarts until an in-flight sink guard command finishes
* fix(audio): judge sink guard commands by exit status and reap timed-out pactl
* test(audio): hold the sink guard command until the test has checked the queued restart
* test(audio): bound the stop check so a blocking stop fails instead of hanging
* test(audio): bound the exit check while the sink guard command is paused
The provider rooted at dataDir exposed the wireless ADB pairing identity, shared_prefs, databases, store logins and Steam's saved login to any app the user picked the DroidDeck root for. Those folders are now hidden from listings and search, refused on open, and can't be created, renamed into, or carried out by renaming, moving or deleting a parent.
Co-authored-by: Tapuuk <177422350+Tapuuk@users.noreply.github.com>
The Favorites collection in the music tab was always listed, even when
no tracks had been favorited.
- Favorites is now only listed in the Collections source once it has at
least one track
- With no favorites and no other collections, the Collections tab no
longer shows
- Adding a track to Favorites brings it back
- The collection detail page still resolves Favorites by id, so an
already-open page does not close when its last track is removed
Adds a test for the empty, populated and other-collections cases.
* fix: route unsupported Xclipse generations upstream
* fix: let Auto load upstream Xclipse pairs
* fix: clear stale auto-owned Xclipse RADV pair
* fix(drivers): clear independently owned fallback selections
* fix(drivers): keep independently chosen drivers when session start reconciles an unsupported Xclipse
* fix(drivers): name the Xclipse from Vulkan before session start picks which drivers Auto clears
* fix(drivers): keep the Vulkan query off the UI thread when session start reconciles Xclipse drivers
* test(drivers): control Xclipse GPU family in auto-driver reconciliation test
* fix(gpu): guard Xclipse auto-driver cleanup when running and handle open-driver bundles
* fix(gpu): keep active RADV-Xclipse-upstream bundle on unsupported Xclipse
rawAndStorageMountViewsOverlapButDistinctFoldersDoNot canonicalizes
/mnt/media_rw/... with the host JVM. On SteamOS and Fedora Atomic /mnt is a
symlink to var/mnt, so the path resolves to /var/mnt/media_rw/..., which no
Android device produces, and the unit suite fails there. Skip the case when
the host's /mnt is a symlink.
* fix(input): keep held keyboard keys when pointer capture ends
When a game gives up its pointer lock, Android pointer capture is
released and the session dropped every held input, including keyboard
keys. In a windowed session with Relative mouse on Auto, holding W to
run and releasing the mouse button after steering stopped the
character, because the key press is never sent again.
Losing pointer capture now releases only the mouse buttons, the pen and
the mouse back-button input. Keys are still released when the window
loses focus.
* fix(input): release held keys when pointer capture ends for the session menu
Opening the session menu releases pointer capture, and the menu then
consumes key events, including the key-up. A key held at that moment
would stay pressed in the guest after the menu closed. Capture loss with
the menu open now releases keys as before; other capture loss keeps them.
* fix(input): release held keys whenever another surface takes the keyboard
Capture loss released keys only while the session menu was open. The
other places that take input from the guest (on-screen keyboard,
achievements, input settings, WoW chat, touch editor, picture in picture)
have the same problem, so keys are now kept only when the session still
owns input and the game itself dropped its pointer lock.
* fix(input): release held keys when the session menu opens
While the menu is open it takes every key event, including the key-up,
so a key held when it opened stayed pressed in the guest. The keys are
now released when the menu opens, whether or not pointer capture was
active.
---------
Co-authored-by: zdennisz <15849186+zdennisz@users.noreply.github.com>
* feat(library): add a Hidden tab synced with Steam's hidden games
* fix(library): mark the hidden collection as modified so Steam uploads it
* fix(library): make the Steam hidden write atomic and refresh the hidden set on resume
* feat(library): show an error when hiding a game can't sync with Steam
* fix(library): keep the pending Steam upload list safe when it can't be read
* fix(library): report a failed Steam sync when the upload marker can't be saved
* fix(library): refresh the Steam hidden set from disk after a failed sync
* fix(library): keep the cached hidden set when Steam's file can't be read
* fix(library): handle a malformed hidden entry and a changed Steam root
* fix(library): refuse to overwrite a malformed hidden entry
* fix(library): validate the hidden collection, keep Steam saves separate from upload status and cover cloud and uninstalled games
* fix(library): accept numeric-string ids, keep uninstalled games in the Hidden tab and hide games from the Steam home list
* fix(library): keep hidden Steam games out of the home capsule wall
* fix(library): refresh home art when the hidden set changes and track the Steam account
* fix(library): clear the cached hidden set when the Steam account changes
* fix(library): ignore a previous account's art while the home scan runs
* fix(library): conceal home art for hidden added games that carry a Steam app id
* fix(library): move an unarranged store order to the new default at startup
* fix(stores): forget an unarranged store order instead of saving the default
* feat(agent): OpenCode takes keys for Ollama Cloud and other providers, or any OpenAI-compatible server
* fix(agent): write OpenCode config atomically, check auth first, Done hides the keyboard
* fix(agent): serialize OpenCode key saves and removals
* fix(agent): restore the OpenCode config atomically when the key write fails
* fix(agent): report a failed OpenCode config rollback
* fix(agent): delete the OpenCode config with deleteIfExists, lead with the rollback warning
* fix(xbox): log and show why an Xbox launch could not be prepared
* fix(xbox): hand Xodus a launch folder the Linux guest can see
---------
Co-authored-by: Zeratullo <8460122+Zeratullo@users.noreply.github.com>
* chore: consistent code branding
* fix: carry legacy overrides, queues and live sync rows across the rename
* fix: keep queued legacy overrides on a failed rename and separate merged queue entries
* fix(input): detect controller right stick axes
* fix(input): read dedicated triggers when the right stick is on RX/RY
* fix(input): stabilize mixed controller axis detection
* fix(input): normalize ranged signed triggers
* fix(input): reuse the event's device for stick and trigger detection
Pass the InputDevice captured at the start of motion handling into the
right-stick and trigger helpers instead of looking it up again, so a pad
that disappears mid-event keeps its axis layout. Write each pad-mouse
trigger only when that side has an analog source, so a held digital
trigger on the other side is not cleared.
* fix(input): align layout editor with game controls
* fix(input): regenerate built-in layouts for measured surfaces
* fix(input): preserve edge stick travel bounds
* Drop editor strings now translated on main
* fix(input): keep editor frame, undo state and previous edge sticks consistent
* fix(input): preserve draft controls and aspect in editor
* fix(dual-screen): allow game swap to external second screens
* fix(dual-screen): name external screens primary and secondary in the game screen choice
* Keep the selected game screen in step with display changes and session mode
* Refresh the game output when the screen is picked automatically
* Keep the game screen when controls open automatically
* Keep the external game screen across restarts and show the controls on the primary screen
* fix(dual-screen): preserve selected game display
* fix(dual-screen): retain fallback selection and keyboard access
* fix(input): clear field keyboard ownership on transitions
* fix(dual-screen): route keyboard when selected display is occupied
* fix(dual-screen): enable keyboard fallback routing
* fix(dual-screen): defer keyboard requests until focus returns
* fix(dual-screen): recheck keyboard destination on focus return and ignore stale hide events
* fix(dual-screen): end guest keyboard ownership on real dismissals
* fix(input): screen taps move the pointer without clicking while an on-screen mouse button is shown
* fix(input): keep tap-to-click with the stock Desktop on-screen gamepad
Only mouse buttons the user bound, not the profile base's stock bindings, make screen taps pointer-only.
* fix(input): count mouse buttons on the second-screen and foldable panels
* fix(input): keep tap clicks off on the swapped panel, copied stock panels and the scroll layer
* fix(input): move the pointer before the desktop gesture buffer when an on-screen mouse button clicks
* fix(input): the on-screen button gate leaves uninitialized gestures alone
---------
Co-authored-by: Kurt Himebauch <136133082+xXJSONDeruloXx@users.noreply.github.com>
* fix(shortcuts): the companion no longer keeps DroidDeck running, so Cocoon stops calling it open
* fix(shortcuts): reset the relay holder when it is installed again
* feat(hud): long press collapses or expands the HUD, locking becomes an option
* fix(hud): keep the collapsing tab at the press point while layout runs
The layout listener could run keepOnScreen during the dock animation and replace the press-centered target with the anchored edge position, which was then saved as the tab position. The dock animation now holds the same guard the other HUD animations use until the dock is committed.
* feat(agent): OpenCode as a third agent, with free Zen models or a pasted Zen key
* fix(agent): OpenCode's key check reads only its own entry and never overwrites an unreadable auth.json
* fix(proot): ship proot in the APK and run it from the native library directory
* fix(native): start rootfs programs through an APK-shipped exec loader where data files may not be executed
* fix(audio): start PulseAudio through the system linker where data files may not be executed
* feat: target Android 14 (SDK 34) so Play Protect installs DroidDeck without a warning
* fix: ask for all-files access on Android 11+, reject unloadable ELFs before exec, keep checking the proot pack
* fix: hand host scripts to the kernel, skip startup launch while all-files Settings is open
* fix: run proot from its add-on pack through Android's linker, so the APK carries no proot
* fix: declare high sampling rate sensors for controller gyro at target 34
* fix: offer the rootfs proot fallback only where Android lets proot exec its loader
* test: install the runtime's proot pack in GameFileSync fixtures
* feat: add per-game Steam Input override
* fix(steam-input): keep overrides in Steam's root apps section and keep the panel in step with the session
* fix(steam-input): reload the override after every session
* feat(boot-movie): add a Shuffle switch backed by Steam's startup movie shuffle setting
* test(boot-movie): keep unrelated settings unchanged
* fix(boot-movie): keep the Shuffle choice across reset, unreadable configs and running sessions
* fix(boot-movie): track the Shuffle baseline per Steam home and only for values staging wrote
* fix(boot-movie): adopt a Shuffle change made in the inactive Steam home before staging overwrites it
* fix(boot-movie): keep the 60 fps note under the Installed movies row
* fix(session): record the guest start steps in events.jsonl
* fix(session): keep both start step events in the session where the step began
* fix(session): record the controller and pre-guest start steps in events.jsonl
* fix(audio): restart host after stalled wake recovery
* Serialize audio host restarts and make the stall counter race-free
* fix(audio): an overtaken restart leaves the socket alone and a sink guard only acts on its own host
---------
Co-authored-by: Kurt Himebauch <136133082+xXJSONDeruloXx@users.noreply.github.com>
* fix(input): read a USB mouse directly where Android won't capture it, as in REDMAGIC cast mode
REDMAGIC's cast mode shows the session on a virtual display while its own
mirror activity keeps focus on the phone's display. Android only grants
pointer capture on the focused display, so the session got a cursor that
stopped at the screen edge and the camera stopped turning in mouse-look games.
When capture is wanted but not granted within 500 ms, the session claims the
mouse's USB HID boot interface instead, reads its reports with the layout from
its report descriptor (boot protocol if that can't be read), and sends motion,
five buttons and the wheel the way captured motion goes. The mouse goes back
to Android whenever capture is no longer wanted.
* feat(input): hold the middle button still to open the session menu while the mouse is read over USB
While the session reads the mouse over USB, Android has no cursor, so the
mouse can't reach REDMAGIC's cast-mode panel or anything else outside the
session. Holding the middle button for two seconds without moving opens the
session menu, which hands the mouse back to Android until the menu closes.
Quick middle clicks and middle-button drags still reach the game.
Relative motion sent to a locked pointer now hides the session's own cursor,
as pad pointer motion already does. A desktop keeps its grab while the menu
is open, so a click made with Android's cursor on the way back showed the
session cursor, which then stayed where the click was.
* fix(input): give a USB mouse back to Android when the app starts after being killed while reading it
A process killed while it reads the mouse, by a crash or an update, leaves
the mouse's interface with no driver, so Android has no mouse until it is
replugged. At startup, any permitted USB boot mouse that can be claimed
without force has no driver, and releasing it reattaches Android's.
* fix(input): only sessions shown off the phone's screen fall back to reading the USB mouse
Android grants pointer capture on the focused display, which the phone's own
screen is, so a session there never needs the USB mouse. Checking the display
keeps a slow capture grant from ever asking for the mouse's USB access there.
* fix(input): reject out-of-range HID report sizes and counts, and release USB buttons when the mouse is lost
A report descriptor can give any 32-bit report count or size, which drove the
field loop for billions of iterations and could overflow bit offsets into a
negative index on the reader thread. Items outside 32 bits and 4096 fields are
rejected, so such a mouse falls back to the boot protocol.
A mouse that stops responding while it is read over USB now releases the
buttons it held and cancels a pending menu hold, through a separate onLost
callback, since the session's own release path only runs while the grab is
still active.
* test(input): an oversized report count is rejected even when X and Y follow it
The earlier case had no X or Y fields, so it returned null without the count
check too. A 4097-field item before valid X and Y is now rejected, and a
4096-field one still parses.
* fix(input): Auto relative mouse sees a mouse read over USB, and re-checks when it is lost
The Auto relative-mouse decision from #2305 reads mouseAttached() on the
relative-mouse thread, which counts a mouse read over USB, so the grab's
state is now volatile. Losing that mouse re-runs the decision, as removing
an Android mouse does.
* fix(input): set up the USB mouse on its reader thread, select report protocol, and retry a failed permission request
Opening the receiver and its descriptor and protocol transfers now run on the
reader thread, so an unresponsive receiver can't stall the session for up to
two seconds. A release during setup lets the thread clean up as it already
did. A mouse that fails setup isn't tried again until it is replugged, since
releasing it re-adds Android's mouse, which would start another attempt.
A parsed report descriptor now selects report protocol first: HID 1.11
7.2.6 says a host must not assume the protocol a device is in.
A permission request that throws no longer marks the mouse as asked.
* fix(input): a cancelled or overlapping USB reader can't claim the mouse or mark it failed
A new reader waits for the previous one to let go of the interface, so a
quick release and grab, like toggling the session menu, no longer fails the
claim. A reader released before or during setup doesn't claim the interface
or keeps going. Failing to open or claim leaves the mouse with Android without
marking it failed, since nothing re-adds it to start another attempt; only a
failure after claiming, or a lost mouse, marks it, and only for the current
reader.
If SET_PROTOCOL(report) fails, setup tries boot protocol before giving up,
so the decoder always matches the protocol the mouse was set to.
* fix(input): release can't slip between the USB reader's cancellation check and its claim
The reader checked `running` and then claimed the interface, so a release in
between still let it claim and start setup. The check and claim now share a
lock with release, so the reader either never claims or claims first and sees
the release before setup.
* fix(input): hand the mouse back to Android while the WoW chat bar is on top
appUiOwnsPointer() already shows Android's pointer for the on-top chat bar,
but pointer capture, and now the USB mouse, stayed on, so that pointer
couldn't reach the bar. The bar on top now ends capture and USB reading like
the session menus do, and its changes re-run the capture decision.
* fix(dual-screen): the bottom-screen terminal keeps its shell when another app covers that screen
* fix(session): sweep the kept terminal shell's process tree when the session ends