win_monitor/backend/app/automation/nodes/media.py

from __future__ import annotations

from typing import Any

from ... import windows_automation
from ..context import WorkflowContext
from ..registry import control_ports, field_def, register_node


MEDIA_KEY_MAP = {
    "play_pause": "space",
    "site_fullscreen": "f",
    "mute": "m",
    "volume_up": "volumeup",
    "volume_down": "volumedown",
    "system_mute": "volumemute",
    "next": "down",
    "previous": "up",
    "escape": "escape",
}


def media_control_node(node: dict[str, Any], inputs: dict[str, Any], context: WorkflowContext) -> dict[str, Any]:
    params = node.get("params", {})
    action = str(inputs.get("action", params.get("action")) or "play_pause")
    key = MEDIA_KEY_MAP.get(action, "space")
    result = windows_automation.keyboard_action("press", key=key)
    return {"action": action, "key": key, "keyboard": result}


register_node(
    {
        "type": "media.control",
        "category": "media",
        "label": "媒体遥控",
        "description": "对当前播放页面执行播放暂停、网页全屏、静音、音量和上下条等客厅遥控动作。",
        "params": {
            "action": field_def(
                "select",
                "动作",
                "play_pause",
                options=[
                    "play_pause",
                    "site_fullscreen",
                    "mute",
                    "volume_up",
                    "volume_down",
                    "system_mute",
                    "next",
                    "previous",
                    "escape",
                ],
            )
        },
        "inputs": {"action": field_def("string", "动作")},
        "outputs": {
            "action": {"type": "string", "label": "动作"},
            "key": {"type": "string", "label": "按键"},
            "keyboard": {"type": "object", "label": "键盘结果"},
        },
        "control_ports": control_ports(),
    },
    media_control_node,
)