The rebase onto master (switch-code feature) added an 8th request_relay
parameter; pass the interface's switch code from both WebRTC->relay
fallback paths so a role-swap session survives the fallback. Also drop
a duplicate bindgen 0.72.1 entry the Cargo.lock merge produced.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ExUfAkYbq8UC9pQCiLy8TQ
- treat ICMP-driven UDP socket errors (WSAECONNRESET 10054 on Windows,
ECONNREFUSED on Linux) as packet loss in punch_udp and the KCP pump
instead of tearing the session down; KCP retransmits through them and a
truly dead link is still reaped by the pong/app-level timeouts
- resolve STUN hostnames via tokio::net::lookup_host so DNS never blocks a
runtime worker; fix the inverted non-IPv4 error message
- add enable-kcp-congestion-control option (default on): switch the turbo
profile to nc=0 so brief loss on constrained links no longer spirals into
stalls; sender-side only, no wire negotiation
- pin kcp-sys to the rustdesk-patches branch: upstream main lost the
RustDesk patches on the EasyTier sync, and this branch also wires
set_kcp_config_factory into connection setup, making the option effective
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- the WebRTC offer now rides any punch request; only an offer-less request
may close and reuse the rendezvous socket for TCP punching
(request_allows_tcp_punch replaces the udp_port-based invariant), with a
separate offer-less request racing as the TCP fallback
- WebSocket mode no longer disables WebRTC — ws only tunnels the
signaling/relay legs while ICE stays the only P2P path there; SOCKS proxy
still disables it (ICE would bypass the proxy and leak the real IP)
- controlled side: WebRTC-only punch replies and trickled ICE candidates go
over dedicated TCP connections to the rendezvous server instead of the UDP
mediator channel, for ws/TCP-only hbbs deployments; drop the now-redundant
rz_sender plumbing and the 400ms candidate re-send on that leg
- guard is_udp handling against responses to requests that advertised no
udp_port; skip the IPv6 socket bind under force-relay
- test_udp_uat: drop the STUN port race — the punch port must come from the
rendezvous server's TestNatResponse observing this socket's mapping, a
STUN probe from another socket can advertise an unreachable port
- bump hbb_common (webrtc 0.13 MSRV pin rationale + upgrade checklist docs)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- prefer-P2P racing (race_transports_prefer_webrtc) across punch and RelayResponse; ICE bridge with 400ms candidate resend
- controlled-side answerer and ICE routing; sign local DTLS fingerprint into SignedId, controller verifies the binding fail-closed
- fix pc leaks: close_webrtc() on insecure-decline paths (io_loop, port_forward); compute direct before disarming the offerer guard
- point hbb_common to the WebRTC data-plane commit 9f5a296
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* AGENTS.md: require a regression-surface check before a change is done
The minimal-invasiveness rules say what to prefer; nothing made an
agent check the final diff against them, so a feature could still route
the old path through its new code while every principle was "followed".
This adds the gate: audit every modified existing path, keep feature-off
on the old code, report the regression surface, and treat an
unnecessarily rewritten legacy path as a review finding whatever the
tests say.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
* AGENTS.md: a scope check before shared code is touched
The minimal-invasiveness rules are principles; this adds the stop
condition that makes them mechanical. A fix for one path stays in that
path, and an unrelated caller needing a placeholder argument to satisfy
a changed signature is the signal that it did not.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* port forward: mappings take turns at the window's login slots
A window's mappings log in concurrently, and each login is built from
the shared `LoginConfigHandler`: `create_login_msg` reads
`port_forward`, which `listen()` set before connecting, and
`handle_login_from_ui` reads `hash`, which the last `Hash` to arrive
set. Two mappings logging in at once could swap targets, bridging a
local socket to the other's target, and answer each other's challenge,
failing one login. The window's password prompt is broadcast to every
mapping, so one whose `Hash` had not arrived answered with whatever the
handler held.
Each mapping now fills `port_forward` and `hash` and sends its login
under a per-window turn lock, and keeps its own `Hash` beside the
connection: a password typed before it arrived is left to the mapping
that prompted, which stores the salted password in the shared handler
for the others to log in with.
The fix stays in `port_forward.rs`. `LoginConfigHandler` gains the lock
and a setter for its private `hash`; `Interface`, `Session` and the
login functions keep their signatures.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
* port forward: the lock and the hash setter are crate-private; test the hash that arrives late on its real path
Both exist only so `port_forward.rs` can reach the handler's private
`hash`; neither is API.
The test for a password typed before a connection's hash ended by
answering the prompt again once the hash was there. What happens in the
code is that the hash's arrival runs `handle_hash`, which logs in with
the password the prompting mapping stored; the test now ends there, with
no preset password. Answering the prompt with one's own challenge while
the handler holds another's is a test of its own.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
* port forward: a password typed before the connection's hash answers it when it comes
The previous commit dropped such a password, counting on the mapping
that prompted having stored it in the shared handler by the time this
connection's `Hash` arrived. The broadcast wakes both mappings at once
and `select!` picks between a ready `Hash` and a ready password at
random, so this one could reach `handle_hash` first, find the handler
empty, and prompt again.
The connection keeps the password until its `Hash` arrives and answers
with it then. `login_from_ui` takes the challenge it answers; the wait
is `connect_and_login`'s, in `hash_arrived`.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
The minimal-invasiveness rules say what to prefer; nothing made an
agent check the final diff against them, so a feature could still route
the old path through its new code while every principle was "followed".
This adds the gate: audit every modified existing path, keep feature-off
on the old code, report the regression surface, and treat an
unnecessarily rewritten legacy path as a review finding whatever the
tests say.
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* connection: apply the non-video send timeout once the type is known
`Connection::start` set the send timeout before the login request had
arrived, when `file_transfer`, `port_forward_socket` and `terminal` were
all still unset, so every connection got `SEND_TIMEOUT_VIDEO` (12 s) and
the `SEND_TIMEOUT_OTHER` branch never ran. A file transfer, terminal or
port forward whose peer stopped draining for 12 s — a Wi-Fi roam, a VPN
reconnect — was dropped.
The type-specific timeout is now set in `on_message` right after the
login request's union has been matched; `start` keeps the video figure
for the login phase.
`SEND_TIMEOUT_OTHER` also drops from 120 s to 30 s, the same horizon as
the 30 s read timeout: the timeout wraps a single `send`, so it only
fires when the peer makes no progress at all for that long, and beyond
30 s the read check would declare the same peer dead anyway. The raw
port-forward pipe's write to its local target shares the constant and
moves with it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
* connection: keep the raw port-forward local write at 120 s
`SEND_TIMEOUT_OTHER` also bounded `forward.send` in
`try_port_forward_loop`, the write to the local target, whose own idle
timeout is an hour. Lowering it to 30 s made a target that stops
draining for half a minute drop the whole tunnel. That write gets its
own constant at the value it always had.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
* connection: keep the non-video send timeout at its long-standing 120 s
Lowering `SEND_TIMEOUT_OTHER` to 30 s was a policy change on top of the
bug fix, argued from the 30 s read timeout, which measures something
else and cannot even run while a send is blocked. The constant goes
back to `SEND_TIMEOUT_VIDEO * 10`, where it has been since 2021, and
the raw port-forward loop's local write shares it again. What remains
is the fix alone: the type-specific timeout is chosen once the login
request has said what the connection is.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EZ49AbZJYfm8NTp5yDPMab
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Keep the configured app name casing when renaming
the updated executable so legacy MSI custom actions
can terminate custom client processes.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(drm): deliver a rotated output upright instead of sideways (#15886)
the compositor draws a rotated desktop sideways into the landscape
scanout and the physically turned monitor straightens it locally, so the
raw scanout the drm path ships reads sideways in the viewer, and nothing
rebroadcasts on rotation because the framebuffer size never changes.
the capturer now resolves the output transform once per session from the
wayland enumeration, turns accepted frames upright into its own buffer,
and sizes the session in rotated dimensions. the advertised list swaps
width and height for 90/270 outputs, which also makes a mid-session
rotation a topology change that restarts the service, and computes scale
from the post-swap width so a rotated 1:1 monitor no longer advertises
scale 16/9. a non 4-byte format on a rotated session is a hard error and
degrades through the existing health path.
the greeter path where no compositor answers keeps today's behavior:
there is no transform source there. hbb_common carries the new transform
field (submodule bump).
* fix(drm): drop the wayland snapshot when the live layout drifts (#15886)
the advertised list is augmented from the cached wayland snapshot and
nothing invalidated it mid-session, so a rotation the 1.5 s live poll
plainly saw never reached check_changed: the poll reads live, the
advertise kept serving the pre-rotation snapshot. measured before this
commit: transform applied and held, 'desktop layout changed' logged,
zero new encoders. cleared only when the poll saw an actual change, so
the probe cost stays tied to real layout events; after it, the same
stimulus rebuilds into a 1080x1920 encoder within a poll turn.
* refactor: trim comment density to the file norm
* chore: bump hbb_common to the transform field from rustdesk/hbb_common#586
pinned to the #586 commits atop the current pin rather than main tip:
main also carries an unrelated config-keys refactor the app has not
adopted yet, and both #586 commits are reachable upstream through the
merge.
* fix: advertise a lone rotated output at delivered size, one snapshot per session
review findings, both real: the 90/270 swap sat below the origin-only
cut, so a single rotated output advertised unrotated dimensions while
the capturer delivered rotated frames; and transform and origin came
from two get_displays() reads that could straddle a cache invalidation.
the swap now precedes the cut (logical-scale adoption stays multi
output), and new() resolves one snapshot for transform, origin and the
session size, with tests for both. comments trimmed to the three-line
guideline.
* fix(drm): rotate every space the rotation touches, not just the pixels
review findings on #15889, all verified against the code first.
the uinput rect's single-display branches now serve the delivered
orientation, so the pointer reaches the whole of a rotated screen (1).
DisplayRect carries the transform, making 0/180 and 90/270 flips
visible to the drift comparison (5), and the drift poll is an edge on
live-vs-previous rather than a level against the baseline, so the cache
clear fires once per real layout event instead of every 300 ms
forever (9). on the drm path the baseline promotes together with the
clear, so the remap and the client rebase never correct the same origin
delta twice (7), and the poll now runs above the login-screen return,
which was the one place with no other invalidation trigger (6).
a snapshot generation gives a rotation a rebuild path at last (3, 4):
clears bump it, the capturer records it at build, and a stale
generation asks for a rebuild without counting against display health.
the cursor bitmap and hotspot turn with the same session transform the
frames use (11). original_resolution follows the 90/270 swap (12). the
transform comes only from an identity match, never the layout-order
fallback (13), and a missing wayland snapshot at build logs the degrade
instead of silently pinning an unrotated session (10).
unrotate_bgra's body is now libyuv's ARGBRotate, which the existing
direction tests pin to the measured anchor (14). 180 stays master
behavior: i915 advertises hardware rotate-180 and wl_output cannot tell
hardware from software rotation, so undoing it blind would invert an
already-upright frame; it needs the plane rotation property on the
wire (2). the pipewire fallback guard's comment now states the rotated
reality it compares (8).
* fix(drm): one owner for the layout generation, one identity rule for rotation
adversarial pass over the previous commit, three structural findings.
the generation bump rode on the cache clear, which every video-service
start also executes, so any session init or restart tore down every
other live capturer, with no damping against a ping-pong between two
displays. the bump now has a single owner: the edge-detected layout
change in the display-service poll. cache clears are side-effect free
again, and a two-display session survives a third session's init with
zero spurious rebuilds.
the advertise side swapped dimensions for a layout-order-fallback match
while the capturer's transform refused such matches, splitting
advertised size from delivered frames into a black screen. both sides
now key off the same identity-match pass (identity_matches), so a
guessed assignment rotates nothing anywhere.
an edge observed while the drm verdict was transiently non-available
was consumed unpromoted, leaving a rotation sideways for the session;
it now stays owed until the verdict returns. an enumeration that failed
at build pinned transform 0 forever with a warn promising a retry that
did not exist; a missing snapshot now makes the first successful poll
an edge, so the degrade is bounded by the outage. the multi-display
missing-logical-size fallback serves delivered orientation, stale docs
zhou named are updated, and the resolutions list stays mode-space on
purpose: resolution changes ride xrandr, which is inert on this path.
* fix: transpose-tolerant fallback size check, log a rejected rotate geometry
whether a portal stream's caps arrive rotated on a 90/270 output is
unmeasured either way (pipewiresrc does not apply
SPA_META_VideoTransform), and this guard has already broken two readers
who reasoned from its comment - so the size half now accepts either
orientation instead of gambling a permanent offline on one. a source
stride shorter than a row logs the rejected geometry instead of
publishing a silent black frame. comments trimmed to the guideline and
the stale sole-test claim updated.
* fix(wayland): never serve a transposed PipeWire stream
The fallback accepted a stream whose dimensions were the advertised
display's transposed, but CapturerInfo keeps the stream dimensions,
nothing on the wayland side ever reconciles the client afterwards,
and the flutter renderer drops every frame whose size differs from
the advertised display - a permanently blank fallback. Accept only
the exact orientation; a transposed pair now falls into the existing
bail, the display is advertised offline, and the client recovers by
re-enumerating.
* fix(drm): keep the cursor consistent with the session transform
Two holes from the same review pass. The wire cursor id hashes only
the plane pixels and geometry, so a stream rebuilt under a new
transform resent the SAME id and the client's by-id cursor cache kept
the old orientation until the shape itself changed; fold the session
transform into the served id. And a cursor racing new()'s transform
store was processed with transform 0 and never corrected, since the
producer resends only on a shape change; hold that cursor and replay
it once the transform is in - the receive loop wakes at least every
200 ms, so the replay is prompt even on an idle wire.
* fix(wayland): the single-display carve-out must not forgive a transposed stream
The carve-out forgives a size difference (a Full Workspace stream may
report the workspace rather than the mode), but a transposed pair is
the same served-vs-advertised orientation split the previous commit
rejects, and it blanks the client the same way.
* fix(drm): a lone display with a rejected fallback is honestly offline
The transposed rejection promised 'advertised offline', but the
lone-display carve-out in mark_demoted_displays kept the display
online on the grounds that the whole-desktop fallback remains usable
- which is exactly what the rejection just refuted. The video service
then restart-looped against a stream nothing can serve, rebuilding
the portal session about once a second, while the client saw a
display list that lied.
Record the geometry rejection in the display health and let it end
the carve-out; a delivered frame or the demote-cooldown re-arm clears
it, so a recovered output comes back on its own.
* ci: retrigger, the previous run died in the actions outage (all root jobs at exactly 8m)
* fix(drm): a blind capturer owes a rebuild, and name matches reserve globally
Two of the review's findings. A capturer built during a failed wayland
enumeration recorded nothing durable: a later successful enumeration
refills the cache, wayland_snapshot_missing goes false, and the first
live poll sees no edge - the session stays sideways until an unrelated
change. The build now latches that it ran blind and the layout poll
consumes the latch into the existing owed-promotion machinery.
And the identity matcher ran per-connector, so a resolution guess for
an earlier connector could steal a later connector's exact name match
and pin its rotation on the wrong output. Names now reserve in a
global first pass; resolution pairing runs on the remainder only when
forced - one free output and one unmatched connector at that size.
* fix(drm): consume the blind-build latch even on a live-changed poll
Adversarial pass on the previous commit: the short-circuit left the
latch set on exactly the poll where live_changed fired (the common
blind-recovery ordering, since a failed enumeration is not cached and
failed_init makes the first successful poll an edge), and the stale
latch then bought a second, spurious promotion one poll later,
tearing down the freshly rebuilt capturer. The latch is now taken
unconditionally so both edge sources merge into one promotion.
* fix(wayland): hand over a layout change the poll has not seen yet
set_wayland_layout_baseline clears live, which is the edge detector's only
memory of the previous layout. ensure_inited calls it at the top of every video
service start, so a second monitor service starting between a rotation and the
next 1.5s poll recorded the rotated layout as the baseline: the poll then found
baseline == live_rects, owed no promotion, and the first capturer kept its old
transform. Under mutter's software rotation the framebuffer size does not
change and the wayland display-change check is disabled, so the stream stayed
sideways until the next layout event.
The setter now arms the promotion itself when the outgoing live differs from
the incoming baseline, which is the one choke point every caller goes through.
An empty incoming baseline is the DRM-union fallback and proves nothing.
* fix(wayland): the edge detector needs a memory a session init cannot erase
The baseline reset was also the edge detector's memory, so two session inits
straddling a rotation left nothing to compare the next poll against. Keep the
observed layout separate from the per-session input baseline; before the first
poll the outgoing baseline seeds it.
* fix(wayland): a capturer records the layout it was built on
ensure_inited() runs the wayland query before the capturer exists, and a failure
there saves an empty baseline. The capturer's own retry can succeed a moment
later and build on that layout, and because the build was not blind nothing
latched it, so a rotation before the first poll had no memory to be an edge
against and the stream stayed at the old transform.
The build now seeds the edge detector when nothing else has, and only then, so a
capturer built later cannot overwrite what the poll is keeping.
* fix(wayland): keep a capturer record that lost the race with the first poll
The constructor reads its wayland snapshot and records it in the edge
detector in two steps, and the layout poll can land between them. After a
failed session init (empty baseline) the constructor takes layout A and
publishes it, the output rotates, and the poll reads B live: nothing is
recorded yet and the snapshot is present, so it is no edge, and observe()
sets seen=B. The late note_capturer(A) then met a non-empty memory and was
dropped, so the capturer showed A while the detector held B, and B against
B never bumped the generation.
note_capturer now flags a build layout that disagrees with the poll's
memory instead of dropping it (overwriting is still wrong: on a
multi-display session that memory is what the other capturers were built
against). edge() reports the flag as an edge whatever the live layout is,
observe() consumes it right after, and a session init's baseline reset
leaves it alone. Regression test for the interleaving, with the promotion
consuming it, a baseline reset in between, and an agreeing late record as
the control.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011ZwSrP3DFA6ZiPKVHkU5dL
* fix(wayland): a late capturer record from a promoted generation is not a second edge
The record can also land after the poll consumed an edge but before the
bump it promotes, or after the bump with a snapshot taken before it. That
capturer is stale by generation and rebuilds on its own, but the flag it
raised survived the promotion, and the next poll spent a second promotion
on the freshly rebuilt capturers.
Tag the record with the generation the capturer read before taking its
snapshot and count it as an edge only while that generation is current;
the newest generation wins when two records land. Regression test for the
consumed-edge interleaving, with a disagreeing record at the promoted
generation and a stale record after a fresh one as controls.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011ZwSrP3DFA6ZiPKVHkU5dL
* chore: bump hbb_common to main tip
dc95b4f -> 05ed68f, a fast-forward: the flipped-transform warning and the
wlroots xdg-output positions (rustdesk/hbb_common#591, #592), 90-day logs,
the webrtc session cleanup deadlock fix and the hide-general-settings
option. No public API changes and no dependency changes.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011ZwSrP3DFA6ZiPKVHkU5dL
---------
Co-authored-by: rustdesk <71636191+rustdesk@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
When Windows is set to "Show only on 1/2", DXGI still enumerates
detached outputs. Preferring that unfiltered list could select a
zero-size display as primary and hang clients waiting for video.
* feat(portable): load per-customer payload from a PE resource
Customizing a Windows client recompiled the packer for every customer,
because data.bin was baked in with include_bytes!. The generic payload is
identical across customers, so only the small per-customer delta needs to
vary: the branded runner exe, custom.txt and the icons.
The packer now also reads an RDPKG RCDATA resource holding a second blob in
the same format, and folds it over the compiled-in payload. A build can then
inject that resource into a prebuilt template instead of running cargo.
The executable to launch comes from the package trailer, and the extraction
directory follows its stem, which replaces the sed of APP_PREFIX. Where the
executable itself is not customized (sciter x86) it stays in the generic
payload and is only renamed, so the merge covers both shapes.
custom.txt keeps being written to disk next to the app: that is what the
client reads at startup and what the updater stages so a customization
survives an upgrade to a stock build.
Also fixes generate.py restoring os.curdir (the literal ".") instead of the
previous working directory, which left it inside the source folder.
CI: ship windows-aarch64 in the unsigned tarball, so ARM custom clients have
a template to build from.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* ci: publish msi templates for custom client builds
Custom clients rebuild the msi through WiX for every customer, though the
package only differs by the app name, a few GUIDs and four files.
Build the msi once more per release with a __RDAPPNAME__ placeholder and ship
it unsigned in the unsigned tarball, so a customer's build can patch it rather
than run msbuild. It stays unsigned because patching would invalidate a
signature anyway.
Doing this in CI is what makes ARM custom clients possible: preprocess.py runs
the packaged exe to read its version and build date, so an arm64 msi can only
be produced on a native arm64 machine, which the runner already is and the
build agents are not. Patching runs no exe, so an x64 agent can then patch the
arm64 template.
preprocess.py rewrites res/msi in place and locates the app as <app-name>.exe
inside the dist, so the tree is reset around the second build and the dist copy
is renamed to match. Sciter x86 ships no msi and is untouched.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* refactor(msi): pass the app name to the printer custom actions
preprocess.py rewrote the CustomActions sources per customer so the printer
carried the app name, which meant the dll was recompiled for every custom
client and, worse, left the app name baked into a compiled binary.
Pass it through CustomActionData instead. Only the printer and its port ever
varied: the INF path and the driver name ship under their stock names and
preprocess.py already forced the driver name back to RustDesk, so a single
build of the dll now serves every custom client.
Both actions treat the name as optional and fall back to the stock name, so a
package built before this still installs and uninstalls its printer.
This also unblocks patching a prebuilt msi template, which cannot work while a
compiled dll contains the app name: replacing a string inside a PE would shift
everything after it.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* ci: use an 8.3-safe placeholder for the msi template
WiX derives a short name for any name that is not valid 8.3, and a patch
cannot rewrite a truncated placeholder, so a long placeholder would leave the
package's short names pointing at it. RDAPPNAM is eight characters like
"RustDesk" and needs no short name, keeping the template as close to the
shipped package as the mechanism allows.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* feat(msi): give a template its own cabinet for per-customer files
Rebranding recompressed the whole ~100MB payload because one cabinet held
everything. In template mode preprocess.py puts the handful of files a custom
client replaces on a second cabinet, so a patch rebuilds a few hundred KB and
leaves the payload cabinet alone. The shipped msi is built without template
mode and keeps its single cabinet.
The branding assets need conditional components. A stock build ships none of
them -- there is no icon.ico, icon.png or logo*.png, only icon.svg -- so the
template has to carry placeholders for the File rows to exist, and a customer
supplies whichever they want. Installing a placeholder unconditionally would
give a customer with no logo a placeholder image, where today a missing asset
means no logo at all: the client tries each candidate and treats the failure as
absence. So each optional asset installs only when its property says the
customer supplied one.
CI creates those placeholders and builds the template with the new mode.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* ci: build the msi template with a sentinel revision
preprocess.py appends a build-time revision as the fourth version field, so a
template built without one would bake the CI clock into every customer's
package. Revision 0 marks the field as the patcher's to fill in, and makes the
template deterministic.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* fix(portable): delete files a later package no longer carries
The extraction directory is wiped only when the packer's compiled-in timestamp
changes. That used to be per customer, because generate.py ran for each build;
now the packer is compiled once per release, so every customer and every
rebuild within a release share one timestamp and nothing is ever wiped.
A customer who removes their logo and rebuilds would therefore keep showing it:
the new package simply omits logo.png, and md5 skipping only covers files that
are still present. Record the package's paths in the extraction's meta file and
delete the ones a later package drops.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* fix(portable): build the dropped-file path from plain components
meta.toml lives in a user-writable directory and now drives deletion, but the
traversal guard tested the normalised string while the join used the raw one.
Path::join replaces the base outright when handed an absolute path, so an
edited meta.toml could point remove_file anywhere.
The path is now rebuilt from Normal components only. A colon is rejected
explicitly rather than left to the host's parser: a drive-relative "C:x" parses
as a Normal component everywhere, and only a Windows host reads "C:/..." as a
prefix, so the same input escaped when the logic was exercised off-Windows --
which is what the new test catches.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* fix(msi): pass the printer name in a format the custom action can read
[~] is MSI's escape for a NUL character, not the delimiter WcaReadStringFromCaData
splits on -- that is a literal wide char 128, which a Formatted property value
cannot carry -- and WcaGetProperty returns a null-terminated string anyway. So
the second field was unreachable: InstallPrinter always fell back to the stock
name and installed a printer and port called "RustDesk Printer" inside a
customer's branded package, while UninstallPrinter, whose data is a single field
and parsed fine, went looking for "Acme Printer" and left the real one behind
for good.
Both actions now read CustomActionData directly and split on a character that
cannot occur in a Windows path or in a validated app name. A package built
before this carries no separator and keeps the stock name, as it did.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q7fBdTwziR5BHTkSz7Tzcm
* fix(portable): retry failed stale branding cleanup
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(portable): reject malformed RDPKG resources
Distinguish an absent customer package from an invalid resource and
propagate package errors instead of launching the stock payload.
Signed-off-by: fufesou <linlong1266@gmail.com>
* refact: format 2 files
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): match process names case-insensitively during uninstall
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(custom-client): validate portable exclusion and MSI action data
Fail when --exclude-exe does not match a file, and propagate MSI
CustomActionData read failures while preserving legacy fallback behavior.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix: generate.py, exclude-exe
Signed-off-by: fufesou <linlong1266@gmail.com>
* Revert "fix: generate.py, exclude-exe"
This reverts commit 5104664e95.
* fix: simple path fix in generate.py
Signed-off-by: fufesou <linlong1266@gmail.com>
* Remove useless comments
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(portable): remove expect() anyway
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(portable): validate executable path boundaries
Reject executables outside the source folder and
reuse the package path normalization logic during
stale file cleanup.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix, remove useless file
Signed-off-by: fufesou <linlong1266@gmail.com>
---------
Signed-off-by: fufesou <linlong1266@gmail.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: fufesou <linlong1266@gmail.com>
Re. export and import.
Without detailed information, reference or examples from en.rs, de.rs or fr.rs, I have simply translated the three strings verbatim
* feat(clipboard): optionally sync clipboard between connected sessions
Clipboard content received from a remote session is written to the local
clipboard with an owner marker, so the client clipboard loop deliberately
skips re-broadcasting it to avoid echo loops. As a result, text copied in
one remote window could not be pasted in another connected remote window.
Add an opt-in local option (allow-sync-clipboard-between-sessions) that
relays Clipboard/MultiClipboards messages received from one session to
all other connected sessions, excluding the source session. Per-session
clipboard permissions and view-only mode are still respected via the
existing send path, and the owner marker on the receiving peers prevents
any echo back.
Desktop (flutter) only; file clipboard is not affected.
* fix(lang): propagate sync-clipboard-between-sessions-tip to all locale files
Add the new key to template.rs and every locale file per the localization
convention, move the en.rs entry to the end of the list, and drop comments
that only restated the names next to them.
* fix(lang): add the 'Sync clipboard between sessions' label to the localization catalog
The checkbox label goes through translate(), so add it to template.rs
and every locale file so non-English locales can translate it. en.rs is
skipped since the English display text is identical to the key.
* fix(clipboard): check the source session's full clipboard permission before relaying
The relay was gated only by the incoming clipboard_allowed check
(!disable_clipboard && !view_only). Gate it with
is_text_clipboard_required() instead, which additionally respects the
source session's server_clipboard_enabled and server_keyboard_enabled
state, matching the predicate already applied to destination sessions.
A message arriving after the source permission was revoked (or from a
non-conforming peer) is no longer propagated to other sessions. The
existing local update_clipboard behavior is unchanged.
* fix(lang): translate the new clipboard sync entries in all locale files
Fill the 'Sync clipboard between sessions' label and its tooltip in
every locale file instead of leaving them blank, following each file's
existing terminology. template.rs keeps the empty master entries.
* fix: android: target API 35
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: handle API 35 foreground service types
Integrate the foreground-service and MediaProjection lifecycle changes
from fufesou/rustdesk#68 while leaving storage permission handling to
#15602.
Co-authored-by: fufesou <linlong1266@gmail.com>
Signed-off-by: fufesou <linlong1266@gmail.com>
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: bump required android sdk version to 36, per recent google requirement change.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix(android): clear microphone FGS type when capture stops
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): harden API 36 capture service lifecycle
- isolate MediaProjection callbacks per session
- keep foreground service types in sync with capture state
- handle audio startup failures and shared frame ownership
- upgrade AGP to 8.10.1 for API 36 support
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): reset capture state on FGS update failure
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): recover capture after projection failure
Propagate virtual display startup failures, clean up partial video
resources, and resume capture after media projection is reauthorized.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): preserve voice call during projection replacement
Keep the existing capture active until
a new projection is acquired, and restore the
voice-call audio source when capture restarts.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): use JDK 17 in playground workflow
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): clear pending capture restart on denial
Notify MainService when a recovery projection
request is canceled so a later projection grant
cannot restart stale capture state.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): handle audio and projection recovery failures
Verify AudioRecord startup, propagate voice-call restoration failures,
and clear stale capture recovery state when projection setup fails.
Signed-off-by: fufesou <linlong1266@gmail.com>
---------
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
Signed-off-by: fufesou <linlong1266@gmail.com>
Co-authored-by: fufesou <linlong1266@gmail.com>
* perf(linux): stop the service loop from forking a shell per environment variable
The service loop re-derives the desktop every 500 ms, and every lookup on that
path forks. A healthy GNOME session spends ~104 process spawns a second, 8 full
`ps -u <uid>` scans and 2 full `ps aux` scans, to re-answer a question whose
answer has not changed. `get_env` alone is a `sh -c` pipeline of ~12 processes
per variable.
`get_envs` already reads `/proc` directly and was documented as the intended
replacement, so move the remaining `get_env` callers to it and delete it. The
xwayland probe drops from 4 pipelines (~48 processes) to one `/proc` walk, and
the pathological walk that #15952 was about drops from ~2900 processes to at
most 60 `/proc` walks. `get_cm` and `is_xwayland_running` read `/proc` instead
of forking `ps aux` and `pgrep -a`; `get_cm` also called `current_exe()` once
per line of `ps` output.
Selection semantics are preserved where they were load-bearing:
* `get_envs_of_newest` reproduces the `ps ... | tail -1` the removed pipelines
used, so a variable the newest matching process does not have means moving on
to the next pattern, never on to an older process that may belong to a session
which has since logged out.
* `get_envs` keeps its own order (readdir) and its all-process ranking, so the
existing `get_display_xauth_wayland` caller is unaffected. Only its handling
of an exported-but-empty value changes: `DISPLAY=` no longer counts as found,
where it used to satisfy a single-name query and return the empty value before
a process holding a real one was examined.
* `get_envs_where` lets the caller state what a complete answer is. Ranking by
how many of the requested names a process carries cannot know that `DISPLAY`
is mandatory and the rest interchangeable, so it could rank a process holding
three optional values above the one holding the pair that matters.
`is_xwayland_running` is scoped to the session's uid. The compositor starts
Xwayland as the session user, so another user's Xwayland -- a switched-away
session, a second seat -- used to route a pure-Wayland session into the Xwayland
probe, which has no display for it to find there.
Not addressed: this discovery path has never had any notion of the active
session, and filters by uid alone. Constraining candidates to the active session
is not possible for the most important one, since `xdg-desktop-portal` and its
backends run under `user@<uid>.service`, which spans sessions and carries no
`XDG_SESSION_ID`, no session cgroup and no audit sessionid.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Q5egQpH4q4GoXJiuMoTJ5t
* fix(linux): the newest-process walk must not answer with a grep or an older PID
Three findings from review of the commit before this one.
`/proc/<pid>/environ` failing to read left the walk on to the next PID, which in
`newest_first` mode is an older process -- possibly of a session that has since
logged out -- where the `ps ... | tail -1` pipeline this replaces stopped at the
one PID it had already picked. A read that fails is a process carrying none of
the requested names, not a process to skip. The `seen` latch that was meant to
hold the newest process is deleted: `accept` is reached once per matching
process, so returning on the first is what it already did.
The regex is matched against the whole `/proc/<pid>/cmdline`, where the pipeline
had a `grep -v 'grep'`. A user running `grep Xwayland` is otherwise the newest
match for that pattern and answers with whatever environment their shell had --
an X forwarding endpoint over ssh, say. This is the one place the walk still
differs from the `get_envs` it grew out of, which never had that filter and
could take an ssh `grep` over the portal it was looking for.
`get_envs` is left exactly as it was. Its completeness test was every requested
name *present*; stating it through `accept` turned it into every name *non-empty*
and, with the empty-value change that went with it, moved which process the
existing `get_display_xauth_wayland` caller settles on. `accept` is now told the
count and asks the question the loop it replaced asked. This supersedes the
`get_envs` bullet of the previous commit message: an exported-but-empty value
counts as found again, as it always did.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019QgsYAUYKDei1AM5yHJsMX
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
The root workspace lists libs/hbb_common as a member, so without the
submodule cargo cannot load the workspace and `cargo update` exits 101.
The job has failed on every scheduled run since it was added.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gecc6fgEeSxs6VRiQmAeof
* fix: android: replace all-files access with scoped storage + system picker
Remove MANAGE_EXTERNAL_STORAGE, READ_EXTERNAL_STORAGE, and
WRITE_EXTERNAL_STORAGE from the Android manifest. Remove
requestLegacyExternalStorage. Replace broad external storage with
app-scoped external storage for the file-transfer workspace.
File import uses the system file_picker. File export uses Android's
SAF ACTION_CREATE_DOCUMENT with path validation that restricts
export sources to app-owned directories.
Remove the external_path dependency.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: refine file import feedback
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: use SAF for file imports
Replace file_picker imports with Android's Storage Access Framework to avoid legacy storage permissions, stale cached files, and duplicate staging of large imports. Stream selected documents into app-scoped storage with failure-safe replacement, keep exports restricted to validated app storage roots, use filesDir for the internal fallback workspace, and remove legacy permissions contributed during manifest merging.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: keep file imports in the selected directory
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: reset projection and constrain file workspace
Release capture resources when media projection is revoked externally. Keep Android local file navigation within the app-scoped workspace.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: handle scoped storage start-up regressions. Allow zero digits in POSIX filenames by rejecting NUL explicitly, and initialise the app-specific home directory before the Android service starts the native server.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: update content resolver mode to use 'wt' instead of 'w' to prevent trailing bytes from old document whilst reporting sucess
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android, enforce file workspace boundary on the server, and unblock the ui thread.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: android: validate rename destinations against the app workspace bound file-operation paths. report rename failures, general import failures, and unregister / reregister projection when its onStop callback fires.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: reconnect was refreshing the directory with net entry instances, while selected items retained the old instances, it was reporting a selected item, but checkbox statue used object identity, and appeared unchecked. Fixed by reconciling by path and entry type before replacing the directory snapshot, rebinding valid selections, and dropping missing ones.
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix: (android) add SAF folder import and multi item export - import directories using ACTION_OPEN_DOCUMENT_TREE. Export multiple files, logs, and screen recordings via export buttons, add localisation keys for new actions
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
* fix(android): harden scoped storage file handling
- create new SAF documents instead of overwriting export sources
- reject empty peer paths except for home directory reads
- report directory backup restore and cleanup failures
- resolve log export paths from the configured app name
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): harden scoped-storage file operations
- snapshot directory exports before writing to the destination
- query document provider metadata off the main thread
- reject invalid remote directories without read timeouts
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): handle SAF directory name collisions
- reject dot-segment folder names during import
- fail imports with duplicate document display names
- only reuse matching directories during export
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(android): handle SAF folder import collisions
Reject filesystem-equivalent destination names and
avoid showing a failure when folder overwrite is skipped.
Signed-off-by: fufesou <linlong1266@gmail.com>
---------
Signed-off-by: michaeljclarkk <104532890+michaeljclarkk@users.noreply.github.com>
Signed-off-by: fufesou <linlong1266@gmail.com>
Co-authored-by: fufesou <linlong1266@gmail.com>
Pick up rustdesk-org/rustdesk_desktop_multi_window#37, which re-arms the existing bounded redraw timer whenever a secondary window is shown, including when its first frame was generated while hidden but not presented.
This may perform one delayed child refresh on each show. It intentionally does not add a presentation-complete flag: Flutter reports frame generation rather than successful presentation, so recording success after a synthetic refresh could suppress later self-recovery without a reliable success signal.
Fixes#15952.
Hyprland runs Xwayland without exporting `XAUTHORITY`, and
`get_display_xauth_xwayland` only returns once it has both `DISPLAY` and
`XAUTHORITY`. On such a session that condition is never met, so every refresh
runs the retry loop to the end: 10 rounds x 6 process patterns x 4 variables =
240 `get_env` calls, each a `sh -c` pipeline of ~12 processes starting with a
full `ps -u <uid> -f`. That is ~2900 fork/exec per refresh, and the service loop
repeats every 500 ms. The reporter measured a full core on a low-end laptop and
~60% of a core on a 13600KF.
The Wayland side answers for such a session, so accept `DISPLAY` together with
either `XAUTHORITY` or `WAYLAND_DISPLAY` + `DBUS_SESSION_BUS_ADDRESS`. The
portal answers on the first pattern, which ends the walk there, as it already
did on desktops that do export an xauth.
The loop also assigned all four variables unconditionally per pattern, so the
patterns that do not run on a given desktop blanked out what an earlier one had
answered with -- the portal's valid `DISPLAY=:1` included. That is why the
`--server` was then started with no `WAYLAND_DISPLAY` and no
`DBUS_SESSION_BUS_ADDRESS`. Candidates are now taken from one pattern as a whole
and ranked, so a later pattern replaces an earlier answer only by being better,
and a session that can only offer a compositor and a bus still keeps them.
A compositor that starts Xwayland on demand shows the same shape from the other
side: the portal came up before Xwayland did, so its environment carries a valid
`WAYLAND_DISPLAY` and `DBUS_SESSION_BUS_ADDRESS` but no `DISPLAY`, and no pattern
here may ever produce one. That pair alone is a session the child server can be
started against -- it is exactly what `get_display_xauth_wayland` returns on --
so it outranks a bare `DISPLAY` and ends the retrying, while the rest of the
round still looks for something that completes the session.
Not specific to the drm build: the function is not feature-gated, and the commit
the report points at does not touch it.
Claude-Session: https://claude.ai/code/session_01Q5egQpH4q4GoXJiuMoTJ5t
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix(linux): a session logout should hand the peer to the login screen
Logging out closes every window in the session, the connection manager's
included, and its close handler kicks every peer with the reason a person
gets when they disconnect one by hand. That reason is the one thing the
client never retries on, so the remote session dies on a frozen frame
instead of reconnecting to the greeter that is already there.
The close carries nothing to tell the two apart: measured on KDE, the CM
receives no signal and logind still reports the session active at that
instant, and the server is killed within a few hundred ms either way, so
neither a state check nor a grace period can decide it. What is
distinguishable is the ACTION: disconnecting a peer is not the same event
as this window going away. So the window-close path now says so, and the
server ends the session without poisoning the retry; the Disconnect
button and the app's own close control keep kicking exactly as before.
Linux only, since that is where a logout closes the window.
Verified on plasma/sddm with a client attached: a logout now reconnects
to the greeter with no dialog, while closing the manager window still
shows Closed manually by the peer.
* fix(linux): close the tunnel too, and keep the web build compiling
Three seams the first pass missed. The web bridge is hand written, not
generated, so the new call needs its stub there or flutter build web
stops compiling - and that job is disabled in CI, so it would have gone
green. try_port_forward_loop is a second consumer of the same channel
and only knew Close, so a forwarded tunnel outlived the window it was
supposed to die with. And the variant had landed inside the DRM section,
whose comment says everything below it is drm-gated.
* Add Urdu language support for UI strings till 329 line
Co-authored-by: Copilot <copilot@github.com>
* Add Urdu translations for additional UI strings
* Add Urdu language support in lang.rs
* Fix Urdu translations and remove unused keys in ur.rs
---------
Co-authored-by: Copilot <copilot@github.com>
* fix(msi): keep only native ProductCode uninstall entry
Move installer state outside the Uninstall registry path,
clean up legacy duplicate entries, and use the MSI ProductCode
for updates and uninstalling.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): harden update and uninstall handling
- handle legacy EXE updates without an MSI ProductCode
- propagate MsiExec uninstall failures
- validate and XML-quote custom ARP values
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): validate registry state before update and uninstall
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): pass WindowsInstaller state to elevated sequence
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): block unsupported MSI-to-EXE upgrades
- resolve native MSI state and ProductCode safely
- suppress reboot while preserving MSI uninstall results
- publish the resolved ARP install location
- skip invalid unrelated MSI uninstall entries
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): fail uninstall when ProductCode is missing
Prevent known MSI installations from falling back to
EXE cleanup when the ProductCode cannot be resolved.
Signed-off-by: fufesou <linlong1266@gmail.com>
* fix(msi): do not abort update on ARP version write failure
Signed-off-by: fufesou <linlong1266@gmail.com>
---------
Signed-off-by: fufesou <linlong1266@gmail.com>
