System Information

OpenCV version: build from https://github.com/opencv/opencv/tree/d9a139f9e85a17b4c47dbca559ed90aef517c279
Operating System / Platform: macOS 15.2
Python version: 3.12

Detailed description

There is a comment left in highgui/src/window_cocoa.mm:

According to Apple support page, there are two ways to right-click in macOS:

1. Control-click an item
   Press and hold the Control key while clicking an item using mouse or trackpad.
2. Control-click items using only mouse/trackpad
   Need to change its settings to open "Secondary click". Details in Apple support page.

After testing, opencv incorrectly maps the first type of right-click as EVENT_LBUTTONDOWN/EVENT_LBUTTONUP.

This is because NSEvent.EventType does not automatically map control-click into a right-click. Developers need to handle this mapping themselves. Similar codes can be found in chromium.

I can submit a pull request to fix it, or update the documentation to give users the option. However, please note that the second type of right-click mentioned above is disabled by default and must be enabled manually in the settings. The mapping of control-click to right-click is likely a common understanding among most macOS users.

Steps to reproduce

import cv2
import numpy as np

EVENTS_MAPPING = {
    cv2.EVENT_LBUTTONDOWN: "Left Button Down",
    cv2.EVENT_LBUTTONUP: "Left Button Up",
    cv2.EVENT_RBUTTONDOWN: "Right Button Down",
    cv2.EVENT_RBUTTONUP: "Right Button Up"
}

def mouse_callback(event, x, y, flags, param):
    global display_text
    event_name = EVENTS_MAPPING.get(event, "")
    if event_name:
        display_text = f"event: {event_name}\nx: {x}\ny: {y}\nflags: {flags}\nparam: {param}"
    else:
        display_text = ""

def main():
    global display_text
    display_text = ""

    width, height = 800, 600
    image = 255 * np.ones((height, width, 3), dtype=np.uint8)

    window_name = "Mouse Event"
    cv2.namedWindow(window_name)

    cv2.setMouseCallback(window_name, mouse_callback)

    while True:
        display_image = image.copy()
        if display_text:
            y0, dy = 30, 30
            for i, line in enumerate(display_text.split('\n')):
                y = y0 + i * dy
                cv2.putText(display_image, line, (12, y), cv2.FONT_HERSHEY_DUPLEX, 1, (0, 0, 0), 1)
        cv2.imshow(window_name, display_image)

        key = cv2.waitKey(1) & 0xFF
        if key == ord('q'):
            break

    cv2.destroyAllWindows()

if __name__ == "__main__":
    main()

Issue submission checklist

• I report the issue, it's not a question
• I checked the problem with documentation, FAQ, open issues, forum.opencv.org, Stack Overflow, etc and have not found any solution
• I updated to the latest OpenCV version and the issue is still there
• There is reproducer code and related data files (videos, images, onnx, etc)