Files
voice-cat/clients/windows/VoiceCat.App/Native/RawInput.cs
Talon 2baefddbe4 feat(windows): system-wide push-to-talk via Raw Input
PTT was focus-scoped (WinForms KeyDown/KeyUp) so it died the moment the
window lost focus. Add an optional system-wide path using the Raw Input
API (RegisterRawInputDevices + WM_INPUT with RIDEV_INPUTSINK) instead of
a WH_KEYBOARD_LL low-level hook -- the latter is the keylogger pattern AV
heuristics flag, which is worse for our unsigned MinGW binary. Raw Input
involves no DLL injection or global hook and passes keys through.

- New VoiceCat.App/Native/RawInput.cs: P/Invoke + structs; register the
  keyboard sink, decode WM_INPUT to vkey/up-down, GetAsyncKeyState helper.
- MainForm overrides OnHandleCreated/OnHandleDestroyed/WndProc to manage
  the sink and route WM_INPUT to PTT; gates the focus-scoped KeyDown/KeyUp
  off when system-wide is on; makes the Deactivate force-release
  conditional; adds a GetAsyncKeyState watchdog on the pump timer so a
  missed key-up (RDP/lock-screen) cannot leave PTT stuck.
- VoiceSettings.SystemWidePtt (default ON) + system-wide checkbox in the
  Audio settings PTT section.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-24 13:14:01 +02:00

148 lines
6.5 KiB
C#

using System.Runtime.InteropServices;
namespace VoiceCat.App.Native;
/// <summary>
/// Thin wrapper over the Win32 Raw Input API (user32) used to observe the push-to-talk key
/// system-wide — i.e. while another application is focused.
///
/// We deliberately use Raw Input (<c>RegisterRawInputDevices</c> + <c>WM_INPUT</c> with
/// <c>RIDEV_INPUTSINK</c>) rather than a low-level keyboard hook (<c>SetWindowsHookEx</c> /
/// <c>WH_KEYBOARD_LL</c>). A low-level hook is the textbook keylogger pattern and is exactly
/// what antivirus heuristics flag — especially for an unsigned, statically linked MinGW binary
/// like ours. Raw Input involves no DLL injection and no global hook: the OS simply posts
/// <c>WM_INPUT</c> messages to our own window's message queue (it is the same input path games
/// use), so it does not trip those heuristics. It also passes keystrokes through to the focused
/// app rather than swallowing them.
///
/// Only the keyboard device is registered; <see cref="TryParseKey"/> filters to the single PTT
/// virtual-key code one level up (MainForm).
/// </summary>
internal static partial class RawInput
{
// ── Constants ──────────────────────────────────────────────────────────────────────────
public const int WM_INPUT = 0x00FF;
private const uint RID_INPUT = 0x10000003; // GetRawInputData: get the raw data
private const uint RIM_TYPEKEYBOARD = 1; // RAWINPUTHEADER.dwType for a keyboard
private const uint RIDEV_INPUTSINK = 0x00000100; // receive input even when not in foreground
private const uint RIDEV_REMOVE = 0x00000001; // stop receiving input from the device
private const ushort RI_KEY_BREAK = 0x01; // RAWKEYBOARD.Flags bit set on key-up
private const ushort HID_USAGE_PAGE_GENERIC = 0x01;
private const ushort HID_USAGE_GENERIC_KEYBOARD = 0x06;
// ── Structs (must match the Win32 layout exactly) ──────────────────────────────────────
[StructLayout(LayoutKind.Sequential)]
private struct RAWINPUTDEVICE
{
public ushort usUsagePage;
public ushort usUsage;
public uint dwFlags;
public IntPtr hwndTarget;
}
[StructLayout(LayoutKind.Sequential)]
private struct RAWINPUTHEADER
{
public uint dwType;
public uint dwSize;
public IntPtr hDevice;
public IntPtr wParam;
}
[StructLayout(LayoutKind.Sequential)]
private struct RAWKEYBOARD
{
public ushort MakeCode;
public ushort Flags;
public ushort Reserved;
public ushort VKey;
public uint Message;
public uint ExtraInformation;
}
// ── P/Invoke (source-generated via LibraryImport, matching VoiceCat.Interop) ────────────
[LibraryImport("user32.dll", SetLastError = true)]
[return: MarshalAs(UnmanagedType.Bool)]
private static partial bool RegisterRawInputDevices(
ref RAWINPUTDEVICE pRawInputDevices, uint uiNumDevices, uint cbSize);
[LibraryImport("user32.dll")]
private static partial uint GetRawInputData(
IntPtr hRawInput, uint uiCommand, IntPtr pData, ref uint pcbSize, uint cbSizeHeader);
[LibraryImport("user32.dll")]
private static partial short GetAsyncKeyState(int vKey);
// ── Public helpers ─────────────────────────────────────────────────────────────────────
/// <summary>Register a background keyboard sink targeting <paramref name="hwnd"/>. With
/// RIDEV_INPUTSINK the window receives WM_INPUT even when it is not in the foreground.</summary>
public static bool RegisterKeyboardSink(IntPtr hwnd)
{
var rid = new RAWINPUTDEVICE
{
usUsagePage = HID_USAGE_PAGE_GENERIC,
usUsage = HID_USAGE_GENERIC_KEYBOARD,
dwFlags = RIDEV_INPUTSINK,
hwndTarget = hwnd,
};
return RegisterRawInputDevices(ref rid, 1, (uint)Marshal.SizeOf<RAWINPUTDEVICE>());
}
/// <summary>Tear down the keyboard sink. For RIDEV_REMOVE the target handle must be NULL.</summary>
public static bool UnregisterKeyboardSink()
{
var rid = new RAWINPUTDEVICE
{
usUsagePage = HID_USAGE_PAGE_GENERIC,
usUsage = HID_USAGE_GENERIC_KEYBOARD,
dwFlags = RIDEV_REMOVE,
hwndTarget = IntPtr.Zero,
};
return RegisterRawInputDevices(ref rid, 1, (uint)Marshal.SizeOf<RAWINPUTDEVICE>());
}
/// <summary>Decode a WM_INPUT message's <paramref name="hRawInput"/> (the message's LParam).
/// Returns false for anything that isn't a keyboard event. On success, <paramref name="vkey"/>
/// is the virtual-key code and <paramref name="keyUp"/> distinguishes a release from a press
/// (auto-repeat arrives as repeated presses).</summary>
public static bool TryParseKey(IntPtr hRawInput, out ushort vkey, out bool keyUp)
{
vkey = 0;
keyUp = false;
uint headerSize = (uint)Marshal.SizeOf<RAWINPUTHEADER>();
uint size = 0;
// First call (pData == NULL) returns 0 on success and fills `size` with the buffer length.
if (GetRawInputData(hRawInput, RID_INPUT, IntPtr.Zero, ref size, headerSize) != 0 || size == 0)
return false;
IntPtr buf = Marshal.AllocHGlobal((int)size);
try
{
if (GetRawInputData(hRawInput, RID_INPUT, buf, ref size, headerSize) != size)
return false;
var header = Marshal.PtrToStructure<RAWINPUTHEADER>(buf);
if (header.dwType != RIM_TYPEKEYBOARD)
return false;
// The keyboard payload immediately follows the (8-byte-aligned) header.
var kb = Marshal.PtrToStructure<RAWKEYBOARD>(buf + (int)headerSize);
vkey = kb.VKey;
keyUp = (kb.Flags & RI_KEY_BREAK) != 0;
return true;
}
finally
{
Marshal.FreeHGlobal(buf);
}
}
/// <summary>True if the given virtual-key code is currently held down. Used as a watchdog to
/// catch a missed key-up (e.g. across an RDP / lock-screen focus switch) so PTT can't stick.</summary>
public static bool IsKeyDown(int vKey) => (GetAsyncKeyState(vKey) & 0x8000) != 0;
}