autoglm-gui 1.4.1__tar.gz → 1.5.0__tar.gz

{autoglm_gui-1.4.1 → autoglm_gui-1.5.0}/AutoGLM_GUI/__init__.py

@@ -5,6 +5,9 @@ import sys
 from functools import wraps
 from importlib import metadata
 
+from AutoGLM_GUI.config import AgentConfig, ModelConfig, StepResult
+from AutoGLM_GUI.logger import logger
+
 # 修复 Windows 编码问题 - 必须在所有其他导入之前
 if sys.platform == "win32":
     import codecs
@@ -58,3 +61,11 @@ try:
     __version__ = metadata.version("autoglm-gui")
 except metadata.PackageNotFoundError:
     __version__ = "unknown"
+
+__all__ = [
+    "__version__",
+    "ModelConfig",
+    "AgentConfig",
+    "StepResult",
+    "logger",
+]

{autoglm_gui-1.4.1 → autoglm_gui-1.5.0}/AutoGLM_GUI/__main__.py

@@ -44,19 +44,25 @@ def find_available_port(
     )
 
 
-def open_browser(host: str, port: int, delay: float = 1.5) -> None:
+def open_browser(
+    host: str, port: int, use_ssl: bool = False, delay: float = 1.5
+) -> None:
     """Open browser after a delay to ensure server is ready.
 
     Args:
         host: Server host
         port: Server port
+        use_ssl: Whether to use HTTPS
         delay: Delay in seconds before opening browser
     """
 
     def _open():
         time.sleep(delay)
+        protocol = "https" if use_ssl else "http"
         url = (
-            f"http://127.0.0.1:{port}" if host == "0.0.0.0" else f"http://{host}:{port}"
+            f"{protocol}://127.0.0.1:{port}"
+            if host == "0.0.0.0"
+            else f"{protocol}://{host}:{port}"
         )
         try:
             webbrowser.open(url)
@@ -125,6 +131,16 @@ def main() -> None:
         action="store_true",
         help="Disable file logging",
     )
+    parser.add_argument(
+        "--ssl-keyfile",
+        default=None,
+        help="SSL key file path (for HTTPS)",
+    )
+    parser.add_argument(
+        "--ssl-certfile",
+        default=None,
+        help="SSL certificate file path (for HTTPS)",
+    )
 
     args = parser.parse_args()
 
@@ -172,6 +188,9 @@ def main() -> None:
     # 获取配置来源
     config_source = config_manager.get_config_source()
 
+    # Determine if SSL is enabled
+    use_ssl = args.ssl_keyfile is not None and args.ssl_certfile is not None
+
     # Display startup banner
     print()
     print("=" * 50)
@@ -179,7 +198,8 @@ def main() -> None:
     print("=" * 50)
     print(f"  Version:    {__version__}")
     print()
-    print(f"  Server:     http://{args.host}:{args.port}")
+    protocol = "https" if use_ssl else "http"
+    print(f"  Server:     {protocol}://{args.host}:{args.port}")
     print()
     print("  Model Configuration:")
     print(f"    Source:   {config_source.value}")
@@ -202,13 +222,15 @@ def main() -> None:
 
     # Open browser automatically unless disabled
     if not args.no_browser:
-        open_browser(args.host, args.port)
+        open_browser(args.host, args.port, use_ssl=use_ssl)
 
     uvicorn.run(
         server.app if not args.reload else "AutoGLM_GUI.server:app",
         host=args.host,
         port=args.port,
         reload=args.reload,
+        ssl_keyfile=args.ssl_keyfile,
+        ssl_certfile=args.ssl_certfile,
     )
 
 

autoglm_gui-1.5.0/AutoGLM_GUI/actions/__init__.py

@@ -0,0 +1,6 @@
+"""Action system for executing phone operations."""
+
+from .handler import ActionHandler
+from .types import ActionResult
+
+__all__ = ["ActionHandler", "ActionResult"]

autoglm_gui-1.5.0/AutoGLM_GUI/actions/handler.py

@@ -0,0 +1,196 @@
+"""Action handler for executing phone operations."""
+
+import time
+from typing import Any, Callable
+
+from AutoGLM_GUI.device_protocol import DeviceProtocol
+
+from .types import ActionResult
+
+
+class ActionHandler:
+    def __init__(
+        self,
+        device: DeviceProtocol,
+        confirmation_callback: Callable[[str], bool] | None = None,
+        takeover_callback: Callable[[str], None] | None = None,
+    ):
+        self.device = device
+        self.confirmation_callback = confirmation_callback or self._default_confirmation
+        self.takeover_callback = takeover_callback or self._default_takeover
+
+    def execute(
+        self, action: dict[str, Any], screen_width: int, screen_height: int
+    ) -> ActionResult:
+        action_type = action.get("_metadata")
+
+        if action_type == "finish":
+            return ActionResult(
+                success=True, should_finish=True, message=action.get("message")
+            )
+
+        if action_type != "do":
+            return ActionResult(
+                success=False,
+                should_finish=True,
+                message=f"Unknown action type: {action_type}",
+            )
+
+        action_name = action.get("action")
+        if not isinstance(action_name, str) or not action_name:
+            return ActionResult(
+                success=False,
+                should_finish=False,
+                message=f"Unknown action: {action_name}",
+            )
+
+        handler_method = self._get_handler(action_name)
+
+        if handler_method is None:
+            return ActionResult(
+                success=False,
+                should_finish=False,
+                message=f"Unknown action: {action_name}",
+            )
+
+        try:
+            return handler_method(action, screen_width, screen_height)
+        except Exception as e:
+            return ActionResult(
+                success=False, should_finish=False, message=f"Action failed: {e}"
+            )
+
+    def _get_handler(self, action_name: str) -> Callable | None:
+        handlers = {
+            "Launch": self._handle_launch,
+            "Tap": self._handle_tap,
+            "Type": self._handle_type,
+            "Type_Name": self._handle_type,
+            "Swipe": self._handle_swipe,
+            "Back": self._handle_back,
+            "Home": self._handle_home,
+            "Double Tap": self._handle_double_tap,
+            "Long Press": self._handle_long_press,
+            "Wait": self._handle_wait,
+            "Take_over": self._handle_takeover,
+            "Note": self._handle_note,
+        }
+        return handlers.get(action_name)
+
+    def _convert_relative_to_absolute(
+        self, element: list[int], screen_width: int, screen_height: int
+    ) -> tuple[int, int]:
+        x = int(element[0] / 1000 * screen_width)
+        y = int(element[1] / 1000 * screen_height)
+        return x, y
+
+    def _handle_launch(self, action: dict, width: int, height: int) -> ActionResult:
+        app_name = action.get("app")
+        if not app_name:
+            return ActionResult(False, False, "No app name specified")
+
+        success = self.device.launch_app(app_name)
+        if success:
+            return ActionResult(True, False)
+        return ActionResult(False, False, f"App not found: {app_name}")
+
+    def _handle_tap(self, action: dict, width: int, height: int) -> ActionResult:
+        element = action.get("element")
+        if not element:
+            return ActionResult(False, False, "No element coordinates")
+
+        x, y = self._convert_relative_to_absolute(element, width, height)
+
+        if "message" in action:
+            if not self.confirmation_callback(action["message"]):
+                return ActionResult(
+                    success=False,
+                    should_finish=True,
+                    message="User cancelled sensitive operation",
+                )
+
+        self.device.tap(x, y)
+        return ActionResult(True, False)
+
+    def _handle_type(self, action: dict, width: int, height: int) -> ActionResult:
+        text = action.get("text", "")
+
+        original_ime = self.device.detect_and_set_adb_keyboard()
+        time.sleep(0.5)
+
+        self.device.clear_text()
+        time.sleep(0.3)
+
+        self.device.type_text(text)
+        time.sleep(0.5)
+
+        self.device.restore_keyboard(original_ime)
+        time.sleep(0.3)
+
+        return ActionResult(True, False)
+
+    def _handle_swipe(self, action: dict, width: int, height: int) -> ActionResult:
+        start = action.get("start")
+        end = action.get("end")
+
+        if not start or not end:
+            return ActionResult(False, False, "Missing swipe coordinates")
+
+        start_x, start_y = self._convert_relative_to_absolute(start, width, height)
+        end_x, end_y = self._convert_relative_to_absolute(end, width, height)
+
+        self.device.swipe(start_x, start_y, end_x, end_y)
+        return ActionResult(True, False)
+
+    def _handle_back(self, action: dict, width: int, height: int) -> ActionResult:
+        self.device.back()
+        return ActionResult(True, False)
+
+    def _handle_home(self, action: dict, width: int, height: int) -> ActionResult:
+        self.device.home()
+        return ActionResult(True, False)
+
+    def _handle_double_tap(self, action: dict, width: int, height: int) -> ActionResult:
+        element = action.get("element")
+        if not element:
+            return ActionResult(False, False, "No element coordinates")
+
+        x, y = self._convert_relative_to_absolute(element, width, height)
+        self.device.double_tap(x, y)
+        return ActionResult(True, False)
+
+    def _handle_long_press(self, action: dict, width: int, height: int) -> ActionResult:
+        element = action.get("element")
+        if not element:
+            return ActionResult(False, False, "No element coordinates")
+
+        x, y = self._convert_relative_to_absolute(element, width, height)
+        self.device.long_press(x, y)
+        return ActionResult(True, False)
+
+    def _handle_wait(self, action: dict, width: int, height: int) -> ActionResult:
+        duration_str = action.get("duration", "1 seconds")
+        try:
+            duration = float(duration_str.replace("seconds", "").strip())
+        except ValueError:
+            duration = 1.0
+
+        time.sleep(duration)
+        return ActionResult(True, False)
+
+    def _handle_takeover(self, action: dict, width: int, height: int) -> ActionResult:
+        message = action.get("message", "User intervention required")
+        self.takeover_callback(message)
+        return ActionResult(True, False)
+
+    def _handle_note(self, action: dict, width: int, height: int) -> ActionResult:
+        return ActionResult(True, False)
+
+    @staticmethod
+    def _default_confirmation(message: str) -> bool:
+        response = input(f"\n⚠️  Confirm action: {message} (y/n): ")
+        return response.lower() in ("y", "yes")
+
+    @staticmethod
+    def _default_takeover(message: str) -> None:
+        input(f"\n🤚 {message}. Press Enter to continue...")

autoglm_gui-1.5.0/AutoGLM_GUI/actions/types.py

@@ -0,0 +1,15 @@
+"""Type definitions for actions."""
+
+from dataclasses import dataclass
+from typing import Any
+
+
+@dataclass
+class ActionResult:
+    success: bool
+    should_finish: bool
+    message: str | None = None
+    requires_confirmation: bool = False
+
+
+Action = dict[str, Any]

autoglm_gui-1.5.0/AutoGLM_GUI/adb/__init__.py

@@ -0,0 +1,53 @@
+from AutoGLM_GUI.adb.apps import APP_PACKAGES, get_app_name, get_package_name
+from AutoGLM_GUI.adb.connection import (
+    ADBConnection,
+    ConnectionType,
+    DeviceInfo,
+    list_devices,
+    quick_connect,
+)
+from AutoGLM_GUI.adb.device import (
+    back,
+    double_tap,
+    get_current_app,
+    home,
+    launch_app,
+    long_press,
+    swipe,
+    tap,
+)
+from AutoGLM_GUI.adb.input import (
+    clear_text,
+    detect_and_set_adb_keyboard,
+    restore_keyboard,
+    type_text,
+)
+from AutoGLM_GUI.adb.screenshot import Screenshot, get_screenshot
+from AutoGLM_GUI.adb.timing import TIMING_CONFIG, TimingConfig
+
+__all__ = [
+    "ADBConnection",
+    "ConnectionType",
+    "DeviceInfo",
+    "list_devices",
+    "quick_connect",
+    "TIMING_CONFIG",
+    "TimingConfig",
+    "APP_PACKAGES",
+    "get_package_name",
+    "get_app_name",
+    "tap",
+    "double_tap",
+    "long_press",
+    "swipe",
+    "back",
+    "home",
+    "launch_app",
+    "get_current_app",
+    "type_text",
+    "clear_text",
+    "detect_and_set_adb_keyboard",
+    "restore_keyboard",
+    "Screenshot",
+    "get_screenshot",
+]

autoglm_gui-1.5.0/AutoGLM_GUI/adb/connection.py

@@ -0,0 +1,323 @@
+"""ADB connection management for local and remote devices."""
+
+import subprocess
+import time
+from dataclasses import dataclass
+from enum import Enum
+
+from AutoGLM_GUI.adb.timing import TIMING_CONFIG
+from AutoGLM_GUI.adb_plus.ip import get_wifi_ip
+
+
+class ConnectionType(Enum):
+    """Type of ADB connection."""
+
+    USB = "usb"
+    WIFI = "wifi"
+    REMOTE = "remote"
+
+
+@dataclass
+class DeviceInfo:
+    """Information about a connected device."""
+
+    device_id: str
+    status: str
+    connection_type: ConnectionType
+    model: str | None = None
+    android_version: str | None = None
+
+
+class ADBConnection:
+    """
+    Manages ADB connections to Android devices.
+
+    Supports USB, WiFi, and remote TCP/IP connections.
+
+    Example:
+        >>> conn = ADBConnection()
+        >>> # Connect to remote device
+        >>> conn.connect("192.168.1.100:5555")
+        >>> # List devices
+        >>> devices = conn.list_devices()
+        >>> # Disconnect
+        >>> conn.disconnect("192.168.1.100:5555")
+    """
+
+    def __init__(self, adb_path: str = "adb"):
+        """
+        Initialize ADB connection manager.
+
+        Args:
+            adb_path: Path to ADB executable.
+        """
+        self.adb_path = adb_path
+
+    def connect(self, address: str, timeout: int = 10) -> tuple[bool, str]:
+        """
+        Connect to a remote device via TCP/IP.
+
+        Args:
+            address: Device address in format "host:port" (e.g., "192.168.1.100:5555").
+            timeout: Connection timeout in seconds.
+
+        Returns:
+            Tuple of (success, message).
+
+        Note:
+            The remote device must have TCP/IP debugging enabled.
+            On the device, run: adb tcpip 5555
+        """
+        # Validate address format
+        if ":" not in address:
+            address = f"{address}:5555"  # Default ADB port
+
+        try:
+            result = subprocess.run(
+                [self.adb_path, "connect", address],
+                capture_output=True,
+                text=True,
+                timeout=timeout,
+            )
+
+            output = result.stdout + result.stderr
+
+            if "connected" in output.lower():
+                return True, f"Connected to {address}"
+            elif "already connected" in output.lower():
+                return True, f"Already connected to {address}"
+            else:
+                return False, output.strip()
+
+        except subprocess.TimeoutExpired:
+            return False, f"Connection timeout after {timeout}s"
+        except Exception as e:
+            return False, f"Connection error: {e}"
+
+    def disconnect(self, address: str | None = None) -> tuple[bool, str]:
+        """
+        Disconnect from a remote device.
+
+        Args:
+            address: Device address to disconnect. If None, disconnects all.
+
+        Returns:
+            Tuple of (success, message).
+        """
+        try:
+            cmd = [self.adb_path, "disconnect"]
+            if address:
+                cmd.append(address)
+
+            result = subprocess.run(
+                cmd, capture_output=True, text=True, encoding="utf-8", timeout=5
+            )
+
+            output = result.stdout + result.stderr
+            return True, output.strip() or "Disconnected"
+
+        except Exception as e:
+            return False, f"Disconnect error: {e}"
+
+    def list_devices(self) -> list[DeviceInfo]:
+        """
+        List all connected devices.
+
+        Returns:
+            List of DeviceInfo objects.
+        """
+        try:
+            result = subprocess.run(
+                [self.adb_path, "devices", "-l"],
+                capture_output=True,
+                text=True,
+                timeout=5,
+            )
+
+            devices = []
+            for line in result.stdout.strip().split("\n")[1:]:  # Skip header
+                if not line.strip():
+                    continue
+
+                parts = line.split()
+                if len(parts) >= 2:
+                    device_id = parts[0]
+                    status = parts[1]
+
+                    # Determine connection type
+                    if ":" in device_id:
+                        conn_type = ConnectionType.REMOTE
+                    elif "emulator" in device_id:
+                        conn_type = ConnectionType.USB  # Emulator via USB
+                    else:
+                        conn_type = ConnectionType.USB
+
+                    # Parse additional info
+                    model = None
+                    for part in parts[2:]:
+                        if part.startswith("model:"):
+                            model = part.split(":", 1)[1]
+                            break
+
+                    devices.append(
+                        DeviceInfo(
+                            device_id=device_id,
+                            status=status,
+                            connection_type=conn_type,
+                            model=model,
+                        )
+                    )
+
+            return devices
+
+        except Exception as e:
+            print(f"Error listing devices: {e}")
+            return []
+
+    def get_device_info(self, device_id: str | None = None) -> DeviceInfo | None:
+        """
+        Get detailed information about a device.
+
+        Args:
+            device_id: Device ID. If None, uses first available device.
+
+        Returns:
+            DeviceInfo or None if not found.
+        """
+        devices = self.list_devices()
+
+        if not devices:
+            return None
+
+        if device_id is None:
+            return devices[0]
+
+        for device in devices:
+            if device.device_id == device_id:
+                return device
+
+        return None
+
+    def is_connected(self, device_id: str | None = None) -> bool:
+        """
+        Check if a device is connected.
+
+        Args:
+            device_id: Device ID to check. If None, checks if any device is connected.
+
+        Returns:
+            True if connected, False otherwise.
+        """
+        devices = self.list_devices()
+
+        if not devices:
+            return False
+
+        if device_id is None:
+            return any(d.status == "device" for d in devices)
+
+        return any(d.device_id == device_id and d.status == "device" for d in devices)
+
+    def enable_tcpip(
+        self, port: int = 5555, device_id: str | None = None
+    ) -> tuple[bool, str]:
+        """
+        Enable TCP/IP debugging on a USB-connected device.
+
+        This allows subsequent wireless connections to the device.
+
+        Args:
+            port: TCP port for ADB (default: 5555).
+            device_id: Device ID. If None, uses first available device.
+
+        Returns:
+            Tuple of (success, message).
+
+        Note:
+            The device must be connected via USB first.
+            After this, you can disconnect USB and connect via WiFi.
+        """
+        try:
+            cmd = [self.adb_path]
+            if device_id:
+                cmd.extend(["-s", device_id])
+            cmd.extend(["tcpip", str(port)])
+
+            result = subprocess.run(
+                cmd, capture_output=True, text=True, encoding="utf-8", timeout=10
+            )
+
+            output = result.stdout + result.stderr
+
+            if "restarting" in output.lower() or result.returncode == 0:
+                time.sleep(TIMING_CONFIG.connection.adb_restart_delay)
+                return True, f"TCP/IP mode enabled on port {port}"
+            else:
+                return False, output.strip()
+
+        except Exception as e:
+            return False, f"Error enabling TCP/IP: {e}"
+
+    def get_device_ip(self, device_id: str | None = None) -> str | None:
+        """
+        Get the IP address of a connected device.
+
+        Delegates to adb_plus.ip.get_wifi_ip() for better WiFi interface detection.
+
+        Args:
+            device_id: Device ID. If None, uses first available device.
+
+        Returns:
+            IP address string or None if not found.
+        """
+        return get_wifi_ip(adb_path=self.adb_path, device_id=device_id)
+
+    def restart_server(self) -> tuple[bool, str]:
+        """
+        Restart the ADB server.
+
+        Returns:
+            Tuple of (success, message).
+        """
+        try:
+            # Kill server
+            subprocess.run(
+                [self.adb_path, "kill-server"], capture_output=True, timeout=5
+            )
+
+            time.sleep(TIMING_CONFIG.connection.server_restart_delay)
+
+            # Start server
+            subprocess.run(
+                [self.adb_path, "start-server"], capture_output=True, timeout=5
+            )
+
+            return True, "ADB server restarted"
+
+        except Exception as e:
+            return False, f"Error restarting server: {e}"
+
+
+def quick_connect(address: str) -> tuple[bool, str]:
+    """
+    Quick helper to connect to a remote device.
+
+    Args:
+        address: Device address (e.g., "192.168.1.100" or "192.168.1.100:5555").
+
+    Returns:
+        Tuple of (success, message).
+    """
+    conn = ADBConnection()
+    return conn.connect(address)
+
+
+def list_devices() -> list[DeviceInfo]:
+    """
+    Quick helper to list connected devices.
+
+    Returns:
+        List of DeviceInfo objects.
+    """
+    conn = ADBConnection()
+    return conn.list_devices()