[Bug] Nothing happens when I try to manipulate the mouse report in `pointing_device_task_user()`

[Anomalocaridid]

In the keymap for my Ploopy Classic Trackball, I have a custom key that is intended to make scrolling go horizontally rather than vertically. However, when I try to set mouse_report.h and mouse_report.v, in pointing_device_task_user(), it appears to have no effect and my mouse continues to scroll vertically after pressing the key.

Describe the Bug

Nothing seems to happen when I try to set mouse_report.h and mouse_report.v in pointing_device_task_user().

System Information

Keyboard: Ploopy Classic Trackball Revision (if applicable): rev. 1.005 Operating system: EndeavourOS Sway Edition qmk doctor output:

Ψ QMK Doctor is checking your environment.
Ψ CLI version: 1.1.0
Ψ QMK home: /home/anomalocaris/qmk_firmware
Ψ Detected Linux.
Ψ Git branch: personal-keymap
Ψ Repo version: 0.14.23
Ψ All dependencies are installed.
Ψ Found arm-none-eabi-gcc version 12.1.0
Ψ Found avr-gcc version 8.3.0
Ψ Found avrdude version 7.0
Ψ Found dfu-util version 0.11
Ψ Found dfu-programmer version 0.7.2
Ψ Submodules are up to date.
Ψ QMK is ready to go

Any keyboard related software installed?

• [ ] AutoHotKey (Windows)
• [ ] Karabiner (macOS)
• [ ] Other:

Additional Context

Here is a link to my keymap. The code that implements the custom keycode, HORI_SCROLL is located at lines 70-91. pointing_device_task_user(), where I attempt to manipulate the mouse report is located at lines 83-91.

[drashna]

Are you returning said mouse report?

And could you post the code that you're using?

[Anomalocaridid]

Yes, I am returning the mouse report.

Here's my pointing_device_task_user():

report_mouse_t pointing_device_task_user(report_mouse_t mouse_report) {
    if (horizontal_scroll) {
        int8_t h = mouse_report.h;
        mouse_report.h = -mouse_report.v;
        mouse_report.v = h;
    }

    return mouse_report;
}

horizontal_scroll is a boolean that gets flipped when I press a custom keycode, HORI_SCROLL. Just in case it's important, here's its implementation:

static bool horizontal_scroll = false;

bool process_record_keymap(uint16_t keycode, keyrecord_t *record) {
    switch (keycode) {
        case HORI_SCROLL:
            if (record->event.pressed) {
                horizontal_scroll ^= 1;
            }
            return false;
    }
    return true;
}

[drashna]

process_record_keymap isn't supported by default, so there is that.

But as for pointing_device_task_user, that does appear to be getting called. And the code looks like it should work.

Are you using PLOOPY_SAFE_RANGE rather than SAFE_RANGE for your keycode enum?

[Anomalocaridid]

Sorry about that and sorry for taking so long to respond. I completely forgot that I use some custom stuff I defined in userspace:

__attribute__((weak)) bool process_record_keymap(uint16_t keycode, keyrecord_t *record) {
    return true;
}

bool process_record_user(uint16_t keycode, keyrecord_t *record) {
    switch (keycode) {
        case VRSN:
            if (record->event.pressed) {
                send_version();
            }
            return false;
        case FLASH:
            if (record->event.pressed) {
                send_flash_command();
            }
            return false;
    }

    return process_record_keymap(keycode, record);
}

For my keycode enum, I have this in my keymap:

enum custom_keycodes_keymap {
    HORI_SCROLL = KEYMAP_SAFE_RANGE,
};

And this in userspace:

enum custom_keycodes_user {
#if defined(KEYBOARD_ergodox_ez)
    VRSN = EZ_SAFE_RANGE,
#elif defined(KEYBOARD_ploopyco_trackball_rev1_005)
    VRSN = PLOOPY_SAFE_RANGE,
#else
    VRSN = SAFE_RANGE,
#endif
    FLASH, // Type qmk command for flashing firmware
    KEYMAP_SAFE_RANGE,
};

Here are links to my keymap and userspace in case I forgot anything else that is pertinent:

• https://github.com/Anomalocaridid/qmk_firmware/tree/personal-keymap/keyboards/ploopyco/trackball/keymaps/anomalocaridid
• https://github.com/Anomalocaridid/qmk_firmware/tree/personal-keymap/users/anomalocaridid

Rebasing onto the newest commits did not seem to fix anything.