macOS input methods can expose non-user-selectable keyboard layouts bundled with the input method rather than the user's real selectable keyboard layout. Translating remote key events through those layouts can send IME punctuation semantics to clients instead of the expected ASCII punctuation.
Use the current ASCII-capable keyboard layout when the active input source has no Unicode key layout data and the reported keyboard layout is not directly selectable. Normal selectable keyboard layouts continue to use the existing translation path.
This affects macOS servers forwarding keyboard input to remote clients while an input-method-bundled layout is active.
Remapping the cursor coordinates to the portal edge size makes the mouse
position being relative to the edge size, in a multi-monitor environment
this edge can be in only one of the monitors and the cursor coordinates
got wrong.
This changes uses the full height and the full width of the combined
monitors to set the mouse cursor position, cropped by the portal edge
size.
A BITMAPINFOHEADER with negative biHeight (top-down DIB) caused
std::length_error crash in toIClipboard because 4 * w * h went
negative and wrapped to a huge size_t in string::append.
Fix: use abs(height) for buffer size and SetDIBitsToDevice, and
add null check for CreateDIBSection.
fixes: #9869
TISGetInputSourceProperty returns NULL for an input source that has no
kTISPropertyInputSourceLanguages value, but both loops evaluate
CFArrayGetCount(layoutLanguages) before the "&& layoutLanguages" guard.
The guard is therefore dead and CFArrayGetCount dereferences NULL.
Test the pointer first.
Signed-off-by: rNoz <rnoz.commits@gmail.com>
doWrite()'s retry/buffer state was held in function-local static
variables, shared across every SecureSocket instance in the process
rather than per connection. Move it to instance members, and simplify
secureRead()/secureWrite()'s retry counters (also static, but not
load-bearing across calls) to plain locals.
pollSocket() caches a socket's writability and skips waiting once it's
known writable, relying on writeSocket() to clear the cache on
WSAEWOULDBLOCK. SecureSocket writes via SSL_write(), bypassing
writeSocket() entirely, so the cache never clears and the multiplexer
spins at a zero timeout instead of blocking whenever TLS hits
backpressure.
Add IArchNetwork::resetPollWriteOnSocket() and call it from
checkResult()'s SSL_ERROR_WANT_WRITE case, mirroring writeSocket()'s
existing WSAEWOULDBLOCK handling.
On the server, while the cursor is on a client, every physical mouse move
warped the local cursor back to screen center and the tap callback returned
the mouse-move event instead of consuming it. That leaked continuous motion
to whatever local app sat under the centered cursor (e.g. a fullscreen game
or a centered video reacting to the phantom movement).
Capture the mouse the way macOS games do instead: dissociate the hardware
mouse from the cursor in leave() so the cursor freezes and hides, read raw
deltas from the event in onMouseMove(), consume off-screen moves in the tap
callback, and re-associate in enter() (disable() already does via
showCursor()). This retires the center-warp and the "bogus motion" edge
heuristic that only existed to paper over the warp.
fixes: #5871
updateShape() unconditionally recentered the cursor and notified the
server on every call, even mid-churn when no device reports a region
yet. Only touch tracked state once a real region is found, and only
notify the server when the shape actually changed.
