python win32 simulate click

Question

Let's say I've got a window for which I want to simulate a mouse click at a specific x, y coordinate. I already have the hwnd but I'm not sure how to construct the lParam. I've used SendMessage in the past to click on buttons, etc., but I knew their hwnds. Any help would be greatly appreciated. I also can't help but wonder if I'm going about this the right way. My end goal is clicking on a certain user on skype main window (for example). I used EnumChildWindows to find all the main window's children, but couldn't find the right one. So figured I would try to 'click' on it using coordinates.

Answer 1

I think this is good one for you, you can use by direct or import this to your python program.

"""mousemacro.py defines the following functions:

click() -- calls left mouse click
hold() -- presses and holds left mouse button
release() -- releases left mouse button

rightclick() -- calls right mouse click
righthold() -- calls right mouse hold
rightrelease() -- calls right mouse release

middleclick() -- calls middle mouse click
middlehold() -- calls middle mouse hold
middlerelease() -- calls middle mouse release

move(x,y) -- moves mouse to x/y coordinates (in pixels)
getpos() -- returns mouse x/y coordinates (in pixels)
slide(x,y) -- slides mouse to x/y coodinates (in pixels)
              also supports optional speed='slow', speed='fast'
"""

from ctypes import*
from ctypes.wintypes import *
from time import sleep
import win32ui

__all__ = ['click', 'hold', 'release', 'rightclick', 'righthold', 'rightrelease', 'middleclick', 'middlehold', 'middlerelease', 'move', 'slide', 'getpos']

# START SENDINPUT TYPE DECLARATIONS
PUL = POINTER(c_ulong)

class KeyBdInput(Structure):
    _fields_ = [("wVk", c_ushort),
             ("wScan", c_ushort),
             ("dwFlags", c_ulong),
             ("time", c_ulong),
             ("dwExtraInfo", PUL)]

class HardwareInput(Structure):
    _fields_ = [("uMsg", c_ulong),
             ("wParamL", c_short),
             ("wParamH", c_ushort)]

class MouseInput(Structure):
    _fields_ = [("dx", c_long),
             ("dy", c_long),
             ("mouseData", c_ulong),
             ("dwFlags", c_ulong),
             ("time",c_ulong),
             ("dwExtraInfo", PUL)]

class Input_I(Union):
    _fields_ = [("ki", KeyBdInput),
              ("mi", MouseInput),
              ("hi", HardwareInput)]

class Input(Structure):
    _fields_ = [("type", c_ulong),
             ("ii", Input_I)]

class POINT(Structure):
    _fields_ = [("x", c_ulong),
             ("y", c_ulong)]
# END SENDINPUT TYPE DECLARATIONS

  #  LEFTDOWN   = 0x00000002,
  #  LEFTUP     = 0x00000004,
  #  MIDDLEDOWN = 0x00000020,
  #  MIDDLEUP   = 0x00000040,
  #  MOVE       = 0x00000001,
  #  ABSOLUTE   = 0x00008000,
  #  RIGHTDOWN  = 0x00000008,
  #  RIGHTUP    = 0x00000010

MIDDLEDOWN = 0x00000020
MIDDLEUP   = 0x00000040
MOVE       = 0x00000001
ABSOLUTE   = 0x00008000
RIGHTDOWN  = 0x00000008
RIGHTUP    = 0x00000010


FInputs = Input * 2
extra = c_ulong(0)

click = Input_I()
click.mi = MouseInput(0, 0, 0, 2, 0, pointer(extra))
release = Input_I()
release.mi = MouseInput(0, 0, 0, 4, 0, pointer(extra))

x = FInputs( (0, click), (0, release) )
#user32.SendInput(2, pointer(x), sizeof(x[0])) CLICK & RELEASE

x2 = FInputs( (0, click) )
#user32.SendInput(2, pointer(x2), sizeof(x2[0])) CLICK & HOLD

x3 = FInputs( (0, release) )
#user32.SendInput(2, pointer(x3), sizeof(x3[0])) RELEASE HOLD


def move(x,y):
    windll.user32.SetCursorPos(x,y)

def getpos():
    global pt
    pt = POINT()
    windll.user32.GetCursorPos(byref(pt))
    return pt.x, pt.y

def slide(a,b,speed=0):
    while True:
        if speed == 'slow':
            sleep(0.005)
            Tspeed = 2
        if speed == 'fast':
            sleep(0.001)
            Tspeed = 5
        if speed == 0:
            sleep(0.001)
            Tspeed = 3

        x = getpos()[0]
        y = getpos()[1]
        if abs(x-a) < 5:
            if abs(y-b) < 5:
                break

        if a < x:
            x -= Tspeed
        if a > x:
            x += Tspeed
        if b < y:
            y -= Tspeed
        if b > y:
            y += Tspeed
        move(x,y)


def click():
    windll.user32.SendInput(2,pointer(x),sizeof(x[0]))

def hold():
    windll.user32.SendInput(2, pointer(x2), sizeof(x2[0]))

def release():
    windll.user32.SendInput(2, pointer(x3), sizeof(x3[0]))


def rightclick():
    windll.user32.mouse_event(RIGHTDOWN,0,0,0,0)
    windll.user32.mouse_event(RIGHTUP,0,0,0,0)

def righthold():
    windll.user32.mouse_event(RIGHTDOWN,0,0,0,0)

def rightrelease():
    windll.user32.mouse_event(RIGHTUP,0,0,0,0)


def middleclick():
    windll.user32.mouse_event(MIDDLEDOWN,0,0,0,0)
    windll.user32.mouse_event(MIDDLEUP,0,0,0,0)

def middlehold():
    windll.user32.mouse_event(MIDDLEDOWN,0,0,0,0)

def middlerelease():
    windll.user32.mouse_event(MIDDLEUP,0,0,0,0)


if __name__ == "__main__":
    speed = 10
    for i in range(1000//speed):
        move(i*speed, i*speed)
        sleep(0.01)

Answer 2

You can use the low level windows api thanks to ctypes. See an example below (adapted from something i didn't test it but it should be ok)

import ctypes
MOUSEEVENTF_MOVE = 0x0001 # mouse move
MOUSEEVENTF_ABSOLUTE = 0x8000 # absolute move
MOUSEEVENTF_MOVEABS = MOUSEEVENTF_MOVE + MOUSEEVENTF_ABSOLUTE

MOUSEEVENTF_LEFTDOWN = 0x0002 # left button down 
MOUSEEVENTF_LEFTUP = 0x0004 # left button up 
MOUSEEVENTF_CLICK = MOUSEEVENTF_LEFTDOWN + MOUSEEVENTF_LEFTUP

def click(x, y):
    #move first
    x = 65536L * x / ctypes.windll.user32.GetSystemMetrics(0) + 1
    y = 65536L * y / ctypes.windll.user32.GetSystemMetrics(1) + 1
    ctypes.windll.user32.mouse_event(MOUSEEVENTF_MOVEABS, x, y, 0, 0)

    #then click
    ctypes.windll.user32.mouse_event(MOUSEEVENTF_CLICK, 0, 0, 0, 0)

Update: I didn't test the code below but I think that it should help you to write something to get the child position. Then you can try to click at the correct position.

CHILD= None

def the_callback(child_hwnd, regex):
    '''Pass to win32gui.EnumWindows() to check all the opened windows'''
    if re.match(regex, win32gui.GetWindowText(child_hwnd)):
        CHILD= child_hwnd

win32gui.EnumChildWindows(hwnd, the_callback, regex)


if CHILD:
    (x_tl, y_tl, x_br, y_br) = win32gui.GetWindowRect(CHILD)