Merge branch 'main' into feature/computer/typescript

libs/python/computer/computer/interface/base.py

@@ -3,8 +3,7 @@
 from abc import ABC, abstractmethod
 from typing import Optional, Dict, Any, Tuple, List
 from ..logger import Logger, LogLevel
-from .models import MouseButton
-
+from .models import MouseButton, CommandResult
 
 class BaseComputerInterface(ABC):
     """Base class for computer control interfaces."""
@@ -234,8 +233,31 @@ class BaseComputerInterface(ABC):
         pass
     
     @abstractmethod
-    async def run_command(self, command: str) -> Tuple[str, str]:
-        """Run shell command."""
+    async def run_command(self, command: str) -> CommandResult:
+        """Run shell command and return structured result.
+        
+        Executes a shell command using subprocess.run with shell=True and check=False.
+        The command is run in the target environment and captures both stdout and stderr.
+        
+        Args:
+            command (str): The shell command to execute
+            
+        Returns:
+            CommandResult: A structured result containing:
+                - stdout (str): Standard output from the command
+                - stderr (str): Standard error from the command  
+                - returncode (int): Exit code from the command (0 indicates success)
+                
+        Raises:
+            RuntimeError: If the command execution fails at the system level
+            
+        Example:
+            result = await interface.run_command("ls -la")
+            if result.returncode == 0:
+                print(f"Output: {result.stdout}")
+            else:
+                print(f"Error: {result.stderr}, Exit code: {result.returncode}")
+        """
         pass
 
     # Accessibility Actions

libs/python/computer/computer/interface/linux.py

@@ -9,8 +9,7 @@ import websockets
 from ..logger import Logger, LogLevel
 from .base import BaseComputerInterface
 from ..utils import decode_base64_image, encode_base64_image, bytes_to_image, draw_box, resize_image
-from .models import Key, KeyType, MouseButton
-
+from .models import Key, KeyType, MouseButton, CommandResult
 
 class LinuxComputerInterface(BaseComputerInterface):
     """Interface for Linux."""
@@ -616,11 +615,15 @@ class LinuxComputerInterface(BaseComputerInterface):
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to delete directory"))
 
-    async def run_command(self, command: str) -> Tuple[str, str]:
+    async def run_command(self, command: str) -> CommandResult:
         result = await self._send_command("run_command", {"command": command})
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to run command"))
-        return result.get("stdout", ""), result.get("stderr", "")
+        return CommandResult(
+            stdout=result.get("stdout", ""),
+            stderr=result.get("stderr", ""),
+            returncode=result.get("return_code", 0)
+        )
 
     # Accessibility Actions
     async def get_accessibility_tree(self) -> Dict[str, Any]:

libs/python/computer/computer/interface/macos.py

@@ -9,8 +9,7 @@ import websockets
 from ..logger import Logger, LogLevel
 from .base import BaseComputerInterface
 from ..utils import decode_base64_image, encode_base64_image, bytes_to_image, draw_box, resize_image
-from .models import Key, KeyType, MouseButton
-
+from .models import Key, KeyType, MouseButton, CommandResult
 
 class MacOSComputerInterface(BaseComputerInterface):
     """Interface for macOS."""
@@ -623,11 +622,15 @@ class MacOSComputerInterface(BaseComputerInterface):
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to delete directory"))
 
-    async def run_command(self, command: str) -> Tuple[str, str]:
+    async def run_command(self, command: str) -> CommandResult:
         result = await self._send_command("run_command", {"command": command})
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to run command"))
-        return result.get("stdout", ""), result.get("stderr", "")
+        return CommandResult(
+            stdout=result.get("stdout", ""),
+            stderr=result.get("stderr", ""),
+            returncode=result.get("return_code", 0)
+        )
 
     # Accessibility Actions
     async def get_accessibility_tree(self) -> Dict[str, Any]:

libs/python/computer/computer/interface/models.py

@@ -1,5 +1,17 @@
 from enum import Enum
 from typing import Dict, List, Any, TypedDict, Union, Literal
+from dataclasses import dataclass
+
+@dataclass
+class CommandResult:
+    stdout: str
+    stderr: str  
+    returncode: int
+    
+    def __init__(self, stdout: str, stderr: str, returncode: int):
+        self.stdout = stdout
+        self.stderr = stderr
+        self.returncode = returncode
 
 # Navigation key literals
 NavigationKey = Literal['pagedown', 'pageup', 'home', 'end', 'left', 'right', 'up', 'down']

libs/python/computer/computer/interface/windows.py

@@ -9,8 +9,7 @@ import websockets
 from ..logger import Logger, LogLevel
 from .base import BaseComputerInterface
 from ..utils import decode_base64_image, encode_base64_image, bytes_to_image, draw_box, resize_image
-from .models import Key, KeyType, MouseButton
-
+from .models import Key, KeyType, MouseButton, CommandResult
 
 class WindowsComputerInterface(BaseComputerInterface):
     """Interface for Windows."""
@@ -615,11 +614,15 @@ class WindowsComputerInterface(BaseComputerInterface):
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to delete directory"))
 
-    async def run_command(self, command: str) -> Tuple[str, str]:
+    async def run_command(self, command: str) -> CommandResult:
         result = await self._send_command("run_command", {"command": command})
         if not result.get("success", False):
             raise RuntimeError(result.get("error", "Failed to run command"))
-        return result.get("stdout", ""), result.get("stderr", "")
+        return CommandResult(
+            stdout=result.get("stdout", ""),
+            stderr=result.get("stderr", ""),
+            returncode=result.get("return_code", 0)
+        )
 
     # Accessibility Actions
     async def get_accessibility_tree(self) -> Dict[str, Any]:

libs/python/computer/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "pdm.backend"
 
 [project]
 name = "cua-computer"
-version = "0.2.0"
+version = "0.3.0"
 description = "Computer-Use Interface (CUI) framework powering Cua"
 readme = "README.md"
 authors = [