Add window management commands

libs/python/computer-server/computer_server/handlers/base.py

@@ -105,6 +105,55 @@ class BaseDesktopHandler(ABC):
         pass
 
 
+class BaseWindowHandler(ABC):
+    """Abstract class for OS-specific window management handlers.
+
+    Categories:
+    - Window Management: Methods for application/window control
+    """
+
+    # Window Management
+    @abstractmethod
+    async def open(self, target: str) -> Dict[str, Any]:
+        """Open a file or URL with the default application."""
+        pass
+
+    @abstractmethod
+    async def launch(self, app: str, args: Optional[List[str]] = None) -> Dict[str, Any]:
+        """Launch an application with optional arguments."""
+        pass
+
+    @abstractmethod
+    async def get_current_window_id(self) -> Dict[str, Any]:
+        """Get the currently active window ID."""
+        pass
+
+    @abstractmethod
+    async def get_application_windows(self, app: str) -> Dict[str, Any]:
+        """Get windows belonging to an application (by name or bundle)."""
+        pass
+
+    @abstractmethod
+    async def get_window_name(self, window_id: str) -> Dict[str, Any]:
+        """Get the title/name of a window by ID."""
+        pass
+
+    @abstractmethod
+    async def get_window_size(self, window_id: str) -> Dict[str, Any]:
+        """Get the size of a window by ID as {width, height}."""
+        pass
+
+    @abstractmethod
+    async def activate_window(self, window_id: str) -> Dict[str, Any]:
+        """Bring a window to the foreground by ID."""
+        pass
+
+    @abstractmethod
+    async def close_window(self, window_id: str) -> Dict[str, Any]:
+        """Close a window by ID."""
+        pass
+
+
 class BaseAutomationHandler(ABC):
     """Abstract base class for OS-specific automation handlers.
 

libs/python/computer-server/computer_server/handlers/factory.py

@@ -9,6 +9,7 @@ from .base import (
     BaseAutomationHandler,
     BaseDesktopHandler,
     BaseFileHandler,
+    BaseWindowHandler,
 )
 
 # Conditionally import platform-specific handlers
@@ -22,7 +23,7 @@ elif system == "linux":
 elif system == "windows":
     from .windows import WindowsAccessibilityHandler, WindowsAutomationHandler
 
-from .generic import GenericDesktopHandler, GenericFileHandler
+from .generic import GenericDesktopHandler, GenericFileHandler, GenericWindowHandler
 
 
 class HandlerFactory:
@@ -61,6 +62,7 @@ class HandlerFactory:
             BaseDioramaHandler,
             BaseFileHandler,
             BaseDesktopHandler,
+            BaseWindowHandler,
         ]
     ):
         """Create and return appropriate handlers for the current OS.
@@ -82,6 +84,7 @@ class HandlerFactory:
                 MacOSDioramaHandler(),
                 GenericFileHandler(),
                 GenericDesktopHandler(),
+                GenericWindowHandler(),
             )
         elif os_type == "linux":
             return (
@@ -90,6 +93,7 @@ class HandlerFactory:
                 BaseDioramaHandler(),
                 GenericFileHandler(),
                 GenericDesktopHandler(),
+                GenericWindowHandler(),
             )
         elif os_type == "windows":
             return (
@@ -98,6 +102,7 @@ class HandlerFactory:
                 BaseDioramaHandler(),
                 GenericFileHandler(),
                 GenericDesktopHandler(),
+                GenericWindowHandler(),
             )
         else:
             raise NotImplementedError(f"OS '{os_type}' is not supported")

libs/python/computer-server/computer_server/handlers/generic.py

@@ -8,11 +8,20 @@ Includes:
 """
 
 import base64
+import os
+import platform
+import subprocess
+import webbrowser
 from pathlib import Path
 from typing import Any, Dict, Optional
 
 from ..utils import wallpaper
-from .base import BaseDesktopHandler, BaseFileHandler
+from .base import BaseDesktopHandler, BaseFileHandler, BaseWindowHandler
+
+try:
+    import pywinctl as pwc
+except Exception:  # pragma: no cover
+    pwc = None  # type: ignore
 
 
 def resolve_path(path: str) -> Path:
@@ -27,6 +36,9 @@ def resolve_path(path: str) -> Path:
     return Path(path).expanduser().resolve()
 
 
+# ===== Cross-platform Desktop command handlers =====
+
+
 class GenericDesktopHandler(BaseDesktopHandler):
     """
     Generic desktop handler providing desktop-related operations.
@@ -67,6 +79,125 @@ class GenericDesktopHandler(BaseDesktopHandler):
             return {"success": False, "error": str(e)}
 
 
+# ===== Cross-platform window control command handlers =====
+
+
+class GenericWindowHandler(BaseWindowHandler):
+    """
+    Cross-platform window management using pywinctl where possible.
+    """
+
+    async def open(self, target: str) -> Dict[str, Any]:
+        try:
+            if target.startswith("http://") or target.startswith("https://"):
+                ok = webbrowser.open(target)
+                return {"success": bool(ok)}
+            path = str(resolve_path(target))
+            sys = platform.system().lower()
+            if sys == "darwin":
+                subprocess.Popen(["open", path])
+            elif sys == "linux":
+                subprocess.Popen(["xdg-open", path])
+            elif sys == "windows":
+                os.startfile(path)  # type: ignore[attr-defined]
+            else:
+                return {"success": False, "error": f"Unsupported OS: {sys}"}
+            return {"success": True}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def launch(self, app: str, args: Optional[list[str]] = None) -> Dict[str, Any]:
+        try:
+            if args:
+                proc = subprocess.Popen([app, *args])
+            else:
+                # allow shell command like "libreoffice --writer"
+                proc = subprocess.Popen(app, shell=True)
+            return {"success": True, "pid": proc.pid}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    def _get_window_by_id(self, window_id: int | str):
+        if pwc is None:
+            raise RuntimeError("pywinctl not available")
+        try:
+            windows = pwc.getAllWindowsDict()
+            return windows.get(window_id) or windows.get(int(window_id))
+        except Exception:
+            return None
+
+    async def get_current_window_id(self) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            win = pwc.getActiveWindow()
+            if not win:
+                return {"success": False, "error": "No active window"}
+            return {"success": True, "window_id": win.getHandle()}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def get_application_windows(self, app: str) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            wins = pwc.getWindowsWithTitle(app, condition=pwc.Re.CONTAINS, flags=pwc.Re.IGNORECASE)
+            ids = [w.getHandle() for w in wins]
+            return {"success": True, "windows": ids}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def get_window_name(self, window_id: int | str) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            w = self._get_window_by_id(window_id)
+            if not w:
+                return {"success": False, "error": "Window not found"}
+            return {"success": True, "name": w.title}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def get_window_size(self, window_id: int | str) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            w = self._get_window_by_id(window_id)
+            if not w:
+                return {"success": False, "error": "Window not found"}
+            width, height = w.size
+            return {"success": True, "width": int(width), "height": int(height)}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def activate_window(self, window_id: int | str) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            w = self._get_window_by_id(window_id)
+            if not w:
+                return {"success": False, "error": "Window not found"}
+            ok = w.activate()
+            return {"success": bool(ok)}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+    async def close_window(self, window_id: int | str) -> Dict[str, Any]:
+        try:
+            if pwc is None:
+                return {"success": False, "error": "pywinctl not available"}
+            w = self._get_window_by_id(window_id)
+            if not w:
+                return {"success": False, "error": "Window not found"}
+            ok = w.close()
+            return {"success": bool(ok)}
+        except Exception as e:
+            return {"success": False, "error": str(e)}
+
+
+# ===== Cross-platform file system command handlers =====
+
+
 class GenericFileHandler(BaseFileHandler):
     """
     Generic file handler that provides file system operations for all operating systems.

libs/python/computer-server/computer_server/main.py

@@ -75,7 +75,7 @@ except Exception:
     except Exception:
         package_version = "unknown"
 
-accessibility_handler, automation_handler, diorama_handler, file_handler, desktop_handler = (
+accessibility_handler, automation_handler, diorama_handler, file_handler, desktop_handler, window_handler = (
     HandlerFactory.create_handlers()
 )
 handlers = {
@@ -102,6 +102,15 @@ handlers = {
     # Desktop commands
     "get_desktop_environment": desktop_handler.get_desktop_environment,
     "set_wallpaper": desktop_handler.set_wallpaper,
+    # Window management
+    "open": window_handler.open,
+    "launch": window_handler.launch,
+    "get_current_window_id": window_handler.get_current_window_id,
+    "get_application_windows": window_handler.get_application_windows,
+    "get_window_name": window_handler.get_window_name,
+    "get_window_size": window_handler.get_window_size,
+    "activate_window": window_handler.activate_window,
+    "close_window": window_handler.close_window,
     # Mouse commands
     "mouse_down": automation_handler.mouse_down,
     "mouse_up": automation_handler.mouse_up,