I have a basic keystroke converter app in development. The conversion works with the following:
CFRunLoopSourceRef runLoopSource = NULL;
CFMachPortRef eventTap = CGEventTapCreate(kCGHIDEventTap, kCGHeadInsertEventTap, kCGEventTapOptionDefault, kCGEventMaskForAllEvents, myCGEventCallback, NULL);
runLoopSource = CFMachPortCreateRunLoopSource(kCFAllocatorDefault, eventTap, 0);
CFRunLoopAddSource(CFRunLoopGetCurrent(), runLoopSource, kCFRunLoopCommonModes);
CGEventTapEnable(eventTap, true);
As you might expect, kCGEventMaskForAllEvents is constantly firing for any mouse movement or click in addition to the keyboard, and I suspect tying up system resources. I tried substituting CGEventMaskBit(kCGEventKeyDown), which best I can tell from Quartz Event doc on Event Types is what I want, and would weed out mouse movements and clicks. Unfortunately, using this seems to just eat the keystrokes, rather than convert them.
What am I doing wrong?
The following works, but I still don't understand why CGEventMaskBit(kCGEventKeyUp) by itself isn't the correct implementation.
CGEventMaskBit(kCGEventKeyUp) | CGEventMaskBit(kCGEventKeyDown) | CGEventMaskBit(NX_SYSDEFINED)
because a keystroke key press consists of a keydown and a key up
The discussion section of the CGEventTapCreate doc page says:
Event taps receive key up and key down events if one of the following conditions is true:
The current process is running as the root user.
Access for assistive devices is enabled. In OS X v10.4, you can enable this feature using System Preferences, Universal Access panel, Keyboard view.
Running as the root user definitely worked for me (MacOS Sierra.) I didn't try the assistive devices approach.
To run as root inside XCode (I have 8.3.3 at this time), choose Product/Scheme/Edit Scheme.../Run/Info/Debug Process As: root
In the CGEventTapCreate call, replace the kCGEventMaskForAllEvents argument with CGEventMaskBit(kCGEventKeyDown) | CGEventMaskBit(kCGEventKeyUp). Your callback will now get invoked for most key presses, except for modifier keys: shift, ctrl, cmd, some of teh function keys.
To get the callback invoked for the modifier keys, add CGEventMaskBit(kCGEventKeyDown) | CGEventMaskBit(kCGEventKeyUp) | CGEventMaskBit(NX_SYSDEFINED). For some reason, with this change I also get the callback invoked for mouse button presses. This might be a side effect of how the Logitech mouse driver works -- I didn't investigate. But the volume of calls is much lower than before and doesn't include mouse moves.
Dave Keck's response to this CocoaBuilder thread gets credit for figuring this out.
Related
I have an old C program for displaying caller ID called YAC. Fortunately, the author Jensen Harris provided the source.
15 years ago, I modified the source to turn on the monitor if the computer was awake but the monitor was off. The code below worked well, turning on the monitor and making the caller ID message visible on the screen.
// TG - add a call to turn on the monitor if it is sleeping.....
SendMessage(hwnd, WM_SYSCOMMAND, SC_MONITORPOWER, -1);
Recently the behavior has changed (presumably a Windows update changed something)
Now when a Caller ID message should be displayed, the monitor turns on (as evidenced by the LED), but the screen remains black. The monitor remains in the black-screen condition for a few seconds, then turns off again.
What additional or different call is now required to cause Windows to activate the display and show the desktop? Possibly this could be forced by sending a mouse move, but is there a better way?
EDIT:
I have implemented the following additional code to press and release ESC. I was unable to find a good example of a relative mouse move of 1 pixel, so I used an example for keyboard. I will test and see if it is effective.
INPUT inputs[2];
UINT uSent;
// reference: https://learn.microsoft.com/en-us/windows/win32/api/winuser/nf-winuser-sendinput
ZeroMemory(inputs, sizeof(inputs));
inputs[0].type = INPUT_KEYBOARD;
inputs[0].ki.wVk = VK_ESCAPE;
inputs[1].type = INPUT_KEYBOARD;
inputs[1].ki.wVk = VK_ESCAPE;
inputs[1].ki.dwFlags = KEYEVENTF_KEYUP;
uSent = SendInput(ARRAYSIZE(inputs), inputs, sizeof(INPUT));
EDIT2 - I can confirm this approach does work to cause the monitor to display video, but of course has the potential for side-effects as any keyboard or mouse action would. I would still be interested in learning of a pure API function that works to fully wake the system like SC_MONITORPOWER used to.
As an experiment, I am trying to achieve the following:
Let spacebar work as a modifier key - like the Shift key - where holding the spacebar key down and typing keys print different letters. Releasing the spacebar would set the state back to normal, and just pressing it behaves like a normal space key.
I was thinking of handling the keydown and keyup event, but apparently handleEvent:client: in IMKServerInput Protocol seems to only catch key down and mouse events.
Without much experience with cocoa, I’ve tried some methods with no success:
went through the Technical Note 2128 via internet archive, which gave me the suitable explanations of plist items. Still, nothing about keyup.
tried adding NSKeyUpMask to recognizedEvents: in IMKStateSetting Protocol, but that didn’t seem to catch the event either.
tested a bit with addLocalMonitorForEventsMatchingMask:handler: but nothing happens.
failed to find a way to make NSFlagsChanged event fire with spacebar.
read about Quartz Event Service and CGEventTap which seems to handle user inputs in lower level. Didn’t go further to this route, yet.
IOHIDManager?
I reached to a conclusion that IMKit is only capable of passively receiving events.
Since it is not an application, there is no keyUp: method to override - AFAIK, IMKit does not inherit NSResponder class.
Unfortunately cocoa is way too broad and has much less (or overflowed with non-helping) documentations for a novice like me to dive in.
Can anyone help me to the right direction?
I tried all possible alternatives one by one, and eventually achieved it by creating a global EventTap
with CGEventTap.
The code basically looks like this:
// Create an event tap.
CGEventMask eventMask = ((1 << kCGEventKeyDown) | (1 << kCGEventKeyUp));
CFMachPortRef eventTap = CGEventTapCreate(kCGSessionEventTap,
kCGHeadInsertEventTap,
0,
eventMask,
myCGEventCallback,
NULL);
if (!eventTap) {
NSLog(#"failed to create event tap\n");
return NO;
} else {
// Create a run loop source.
runLoopSource = CFMachPortCreateRunLoopSource(kCFAllocatorDefault, eventTap, 0);
// Add to the current run loop.
CFRunLoopAddSource(CFRunLoopGetCurrent(), runLoopSource, kCFRunLoopCommonModes);
// Enable the event tap.
CGEventTapEnable(eventTap, true);
return YES;
}
where myCGEventCallback handles the global states.
Meanwhile here are some of what I've found out:
According to The Key-Input Message Sequence document, the application only passes the keydown event to the Input Method Kit, after trying other bunch of handlers in the chain. You cannot let IMKServerInput 'catch' the NSKeyUp event. Just adding an NSKeyUpMask to recognizedEvents: would not work.
addLocalMonitorForEventsMatchingMask:handler: and CGEventTapCreateForPSN would not catch the event. I suppose this is because though an Input Method may run as a separate process, the event itself is fired from the application, like TextEdit, and handed over to the Input `Method.
IOHIDManager: is for adding new hardware devices and making drivers.
Creating a global EventTap requires either running the process with sudo privilege -- copying the Input Method to /Library/Input Methods does not run with sudo privilege --, or registering the application to the Accessibility control. That's in System Preferences → Security & Privacy → Privacy tab → Accessibility, in Mavericks.
IMKStateSetting protocol's recognizedEvents: method should enable NSKeyUp events:
- (NSUInteger)recognizedEvents:(id)sender {
return NSKeyDownMask | NSKeyUpMask;
}
A client calls this method to check whether an input method supports an event. The default implementation returns NSKeyDownMask.
However, in testing my Input Method still does not catch NSKeyUp events. This seems like a bug.
I've filed the following Radar, and hope others will duplicate it:
rdar://21376535
I have a system with no keyboard. I can connect a keyboard, but ultimately the inputs come from a custom keypad which is not a HID device, it sends serial data which I can interpret and decode to determine if the user pressed Up, Down, Left, Right, or Enter.
Right now all I have is a Fl_Window, with two Fl_Button widgets. Focus is set for one of the buttons and callbacks are defined for these buttons. I know that if I attach a real keyboard, and use the arrow keys I can change focus from button to button. I do have to hit SPACE to activate a button.
My problem is determining how to cause these key presses using code when I decode the outcome form the embedded key pad. Because in deployment, there will be no actual keyboard.
What I've tried is to invoke int Fl_Window::handle(int) and not really had success. I've also tried to invoke int Fl::handle(int, Fl_Window *) and not had success.
Here are code examples:
if((ret = Fl::handle(FL_Left, window)) == 0)
That compiles, but I find that I get zero back, implying that it did not process the event.
if((ret = Fl_Window::handle(FL_Right)) == 0)
That does not compile, informing me that it "cannot call member function virtual int Fl_Window::handle(int) without object"
I'm thinking that the "int event" actually ought to be FL_KEYDOWN.
That logic leaves me to wonder though how I "set" event_key(). For instance, there are API functions to get that when one has a handler, but I do not wish to get that; I wish to cause that event to occur.
Is my only option here to figure out how to emulate a HID or make some type of virtual HID where I then cause the keyboard events to occur?
I do not feel I require a handler function in my application, I'm fine with the default behaviors which occur and cause my callback functions to be invoked. My problem is that I can't "cause" these events to occur.
You need to assign the desired key to e_keysym, then dispatch a FL_KEYDOWN event using Fl::handle_(). (Fl::handle() will not generate the followup FL_SHORTCUT event.)
Fl::e_keysym = FL_Left;
Fl::handle_(FL_KEYDOWN, window);
// sleep() and/or Fl::wait() as appropriate
Fl::handle_(FL_KEYUP, window);
I need to programmatically disable/suppress system-wide touch gestures on Mac OS. I'm referring to gestures such as the 4-finger swipe between spaces, etc.
I've looked to EventTap but that doesn't appear to be an option (despite previous reports here - perhaps it's changed under 10.8)
I've also tried numerous ways of changing the the system preferences programatically. For example, I've tried using IOConnectSetCFProperties on the service having located it using IORegistryEntryCreateCFProperties.
I've also delved into the trackpad preference pane to see how they do it, and I tried to reproduce it (ignore any create/release inconsistencies, this is just test code):
NSInteger zero = 0;
CFNumberRef numberWith0 = CFNumberCreate(kCFAllocatorDefault, kCFNumberNSIntegerType, &zero);
CFMutableDictionaryRef propertyDict = CFDictionaryCreateMutable(kCFAllocatorDefault, 0,
&kCFTypeDictionaryKeyCallBacks,
&kCFTypeDictionaryValueCallBacks);
CFDictionarySetValue(propertyDict, #"TrackpadFourFingerHorizSwipeGesture", numberWith0);
io_connect_t connect = getEVSHandle(); // Found in the MachineSettings framework
if (!connect)
{
NSLog(#"Unable to get EVS handle");
}
kern_return_t status = IOConnectSetCFProperties(connect, propertyDict);
if (status != KERN_SUCCESS)
{
NSLog(#"Unable to get set IO properties");
}
CFRelease(propertyDict);
CFPreferencesSetValue(CFSTR("com.apple.trackpad.fourFingerHorizSwipeGesture"), _numberWith0, kCFPreferencesAnyApplication, kCFPreferencesCurrentUser, kCFPreferencesCurrentHost);
CFPreferencesSetValue(CFSTR("TrackpadFourFingerHorizSwipeGesture"), _numberWith0, CFSTR("com.apple.driver.AppleBluetoothMultitouch.trackpad"), kCFPreferencesCurrentUser, kCFPreferencesCurrentHost);
CFPreferencesSynchronize(kCFPreferencesAnyApplication, kCFPreferencesCurrentUser, kCFPreferencesCurrentHost);
status = BSKernelPreferenceChanged(CFSTR("com.apple.driver.AppleBluetoothMultitouch.trackpad"));
In this case it appears to work, there are no errors and the option becomes disabled in the system preference pane, however the four finger gesture continues to work. I suspect that logging out then in will have an effect, but I haven't tried because that's not good enough in any case.
It's worth noting that the Pref Pane itself also calls BSKernelPreferenceChanged, but I don't know which framework that might be in order to link to it. Perhaps that's the key to the problem...
UPDATE: Actually I've now found it and linked it to. Adding that call made no difference, although it returns 1 which may indicate an error. I've added the call to the code above.
Finally I tried this from the terminal:
defaults write -globalDomain com.apple.trackpad.fourFingerHorizSwipeGesture 0
defaults write com.apple.driver.AppleBluetoothMultitouch.trackpad TrackpadFourFingerHorizSwipeGesture 0
That doesn't have an immediate effect either.
I don't believe that this isn't possible, there must be a way...
MAS compatibility is not required.
I'm also trying to do this.
Event taps does not work, neither does having a view that is first responder.
From Apple docs:
However, there are certain system-wide gestures, such as a four-finger swipe. for which the system implementation takes precedence over any gesture handling an application performs.
The only way i've been able to stop the system wide gestures is using CGDisplayCapture. This gives my application exclusive access to all events... but also a fullscreen drawing context.
Perhaps it's possible to see what calls are made to the quartz event services when entering this mode
https://developer.apple.com/library/mac/#documentation/graphicsimaging/Conceptual/QuartzDisplayServicesConceptual/Articles/DisplayCapture.html
I think you are looking in the wrong spot for disabling the touch events. The way OSX (and many other systems) is that the first responder in the view chain to handle the event will stop the event from propagating. You will need to write event handlers in your views for each of the touch events you want to handle, and if they exist, the OS will stop sending the events all the way to finder or whatever other application is next in line to handle the touch events.
See: http://developer.apple.com/library/mac/#documentation/cocoa/conceptual/EventOverview/HandlingTouchEvents/HandlingTouchEvents.html
Specifically: Handling Multi-Touch Events (call setAcceptsTouchEvents, then implement touches...WithEvent...)
Hope that helps!
Ola Folks,
Once again I want to drink from the pool of knowledge shared by people using SO.
I have written a small app for OSX that sends key events to an application. I am targeting OSX 10.5.x and newer. However, the problem exists when I build for 10.6.x as well. Everything works fine except when I send only the modifier keys; Alt, Command, Control and Shift.
The problem is that on the two MacBooks, the events for the modifier keys appear to be cleared as soon as the testers move the cursor using the mouse or touch the touchpad.
On a desktop with XCode installed, everything works fine. Just like it should. On two different MacBooks, the problem occurs. The desktop has a standard 101 key keyboard and a multi-button mouse attached.
When a mouse is connected to the MacBooks, a two button mouse with scrollwheel is used. However, the problem exists when no peripherals are attached and the touchpad is used.
What I expect to happen is that the Modifier Key event is sent to the target application, the user moves the cursor using the mouse / touchpad, press buttons on the mouse / touchpad and/or press keys on the keyboard with the modifier key down event 'active'. Then, when they finish, the key up event for the modifier key is sent.
Here is how I am sending key down events, the Shift key for this example:
case ModKeyShiftDown:
xEventSource = CGEventSourceCreate(kCGEventSourceStatePrivate);
xTheCommand = CGEventCreateKeyboardEvent(xEventSource, kVK_Shift, true);
CGEventSetFlags(xTheCommand, kCGEventFlagMaskAlternate);
CGEventPost(kCGHIDEventTap, xTheCommand);
//CGEventPost(kCGSessionEventTap, xTheCommand);
//CGEventPost(kCGAnnotatedSessionEventTap, xTheCommand);
CFRelease(xTheCommand);
CFRelease(xEventSource);
break;
I have used all three flags for creating the event source (kCGEventSourceStatePrivate, kCGEventSourceStateCombinedSessionState and kCGEventSourceStateHIDSystemState).
I have tried Creating the Keyboard Event with the Event Source as well as null as the first parameter.
I have tried with and without the appropriate flags on the event.
I have tried various combinations of posting the event; kCGHIDEventTap, kCGSessionEventTap and kCGAnnotatedSessionEventTap.
For completeness, here is how I send an up event for the Shift key:
case ModKeyShiftUp:
xEventSource = CGEventSourceCreate(kCGEventSourceStatePrivate);
xTheCommand = CGEventCreateKeyboardEvent(xEventSource, kVK_Shift, false);
CGEventSetFlags(xTheCommand, 0);
CGEventPost(kCGHIDEventTap, xTheCommand);
//CGEventPost(kCGSessionEventTap, xTheCommand);
//CGEventPost(kCGAnnotatedSessionEventTap, xTheCommand);
CFRelease(xTheCommand);
CFRelease(xEventSource);
break;
When the testers trigger a modifier key down event, they can see the cursor change as expected. This lets me know the event is being processed by the target application. However, as soon as they touch the mouse or touchpad, the cursor changes back to a standard cursor and the mouse events are processed as if no modifier key events are active.
I would like to know if there is a problem with the way I am sending the events. I would also like to know if there is an alternate way to send Modifier Key events that is Going To Work.
Sorry if I overtalked this. My excuse is that I only slept a couple of hours. :P
Thanx
-isdi-
Ola,
Interestingly enough, using AXUIElementRef and AXUIElementPostKeyboardEvent seem to work.
For whatever reason, sending key events using the Accessibility object and method above solves the problem for the testers.