cua-computer 0.4.4__py3-none-any.whl → 0.4.6__py3-none-any.whl

computer/computer.py

@@ -154,8 +154,8 @@ class Computer:
         self.interface_logger = Logger("computer.interface", verbosity)
 
         if not use_host_computer_server:
-            if ":" not in image or len(image.split(":")) != 2:
-                raise ValueError("Image must be in the format <image_name>:<tag>")
+            if ":" not in image:
+                image = f"{image}:latest"
 
             if not name:
                 # Normalize the name to be used for the VM

computer/diorama_computer.py

@@ -6,16 +6,35 @@ class DioramaComputer:
     A Computer-compatible proxy for Diorama that sends commands over the ComputerInterface.
     """
     def __init__(self, computer, apps):
+        """
+        Initialize the DioramaComputer with a computer instance and list of apps.
+        
+        Args:
+            computer: The computer instance to proxy commands through
+            apps: List of applications available in the diorama environment
+        """
         self.computer = computer
         self.apps = apps
         self.interface = DioramaComputerInterface(computer, apps)
         self._initialized = False
 
     async def __aenter__(self):
+        """
+        Async context manager entry point.
+        
+        Returns:
+            self: The DioramaComputer instance
+        """
         self._initialized = True
         return self
 
     async def run(self):
+        """
+        Initialize and run the DioramaComputer if not already initialized.
+        
+        Returns:
+            self: The DioramaComputer instance
+        """
         if not self._initialized:
             await self.__aenter__()
         return self
@@ -25,11 +44,31 @@ class DioramaComputerInterface:
     Diorama Interface proxy that sends diorama_cmds via the Computer's interface.
     """
     def __init__(self, computer, apps):
+        """
+        Initialize the DioramaComputerInterface.
+        
+        Args:
+            computer: The computer instance to send commands through
+            apps: List of applications available in the diorama environment
+        """
         self.computer = computer
         self.apps = apps
         self._scene_size = None
 
     async def _send_cmd(self, action, arguments=None):
+        """
+        Send a command to the diorama interface through the computer.
+        
+        Args:
+            action (str): The action/command to execute
+            arguments (dict, optional): Additional arguments for the command
+            
+        Returns:
+            The result from the diorama command execution
+            
+        Raises:
+            RuntimeError: If the computer interface is not initialized or command fails
+        """
         arguments = arguments or {}
         arguments = {"app_list": self.apps, **arguments}
         # Use the computer's interface (must be initialized)
@@ -42,6 +81,15 @@ class DioramaComputerInterface:
         return result.get("result")
 
     async def screenshot(self, as_bytes=True):
+        """
+        Take a screenshot of the diorama scene.
+        
+        Args:
+            as_bytes (bool): If True, return image as bytes; if False, return PIL Image object
+            
+        Returns:
+            bytes or PIL.Image: Screenshot data in the requested format
+        """
         from PIL import Image
         import base64
         result = await self._send_cmd("screenshot")
@@ -53,41 +101,122 @@ class DioramaComputerInterface:
         return img_bytes if as_bytes else img
 
     async def get_screen_size(self):
+        """
+        Get the dimensions of the diorama scene.
+        
+        Returns:
+            dict: Dictionary containing 'width' and 'height' keys with pixel dimensions
+        """
         if not self._scene_size:
             await self.screenshot(as_bytes=False)
         return {"width": self._scene_size[0], "height": self._scene_size[1]}
 
     async def move_cursor(self, x, y):
+        """
+        Move the cursor to the specified coordinates.
+        
+        Args:
+            x (int): X coordinate to move cursor to
+            y (int): Y coordinate to move cursor to
+        """
         await self._send_cmd("move_cursor", {"x": x, "y": y})
 
     async def left_click(self, x=None, y=None):
+        """
+        Perform a left mouse click at the specified coordinates or current cursor position.
+        
+        Args:
+            x (int, optional): X coordinate to click at. If None, clicks at current cursor position
+            y (int, optional): Y coordinate to click at. If None, clicks at current cursor position
+        """
         await self._send_cmd("left_click", {"x": x, "y": y})
 
     async def right_click(self, x=None, y=None):
+        """
+        Perform a right mouse click at the specified coordinates or current cursor position.
+        
+        Args:
+            x (int, optional): X coordinate to click at. If None, clicks at current cursor position
+            y (int, optional): Y coordinate to click at. If None, clicks at current cursor position
+        """
         await self._send_cmd("right_click", {"x": x, "y": y})
 
     async def double_click(self, x=None, y=None):
+        """
+        Perform a double mouse click at the specified coordinates or current cursor position.
+        
+        Args:
+            x (int, optional): X coordinate to double-click at. If None, clicks at current cursor position
+            y (int, optional): Y coordinate to double-click at. If None, clicks at current cursor position
+        """
         await self._send_cmd("double_click", {"x": x, "y": y})
 
     async def scroll_up(self, clicks=1):
+        """
+        Scroll up by the specified number of clicks.
+        
+        Args:
+            clicks (int): Number of scroll clicks to perform upward. Defaults to 1
+        """
         await self._send_cmd("scroll_up", {"clicks": clicks})
 
     async def scroll_down(self, clicks=1):
+        """
+        Scroll down by the specified number of clicks.
+        
+        Args:
+            clicks (int): Number of scroll clicks to perform downward. Defaults to 1
+        """
         await self._send_cmd("scroll_down", {"clicks": clicks})
 
     async def drag_to(self, x, y, duration=0.5):
+        """
+        Drag from the current cursor position to the specified coordinates.
+        
+        Args:
+            x (int): X coordinate to drag to
+            y (int): Y coordinate to drag to
+            duration (float): Duration of the drag operation in seconds. Defaults to 0.5
+        """
         await self._send_cmd("drag_to", {"x": x, "y": y, "duration": duration})
 
     async def get_cursor_position(self):
+        """
+        Get the current cursor position.
+        
+        Returns:
+            dict: Dictionary containing the current cursor coordinates
+        """
         return await self._send_cmd("get_cursor_position")
 
     async def type_text(self, text):
+        """
+        Type the specified text at the current cursor position.
+        
+        Args:
+            text (str): The text to type
+        """
         await self._send_cmd("type_text", {"text": text})
 
     async def press_key(self, key):
+        """
+        Press a single key.
+        
+        Args:
+            key: The key to press
+        """
         await self._send_cmd("press_key", {"key": key})
 
     async def hotkey(self, *keys):
+        """
+        Press multiple keys simultaneously as a hotkey combination.
+        
+        Args:
+            *keys: Variable number of keys to press together. Can be Key enum instances or strings
+            
+        Raises:
+            ValueError: If any key is not a Key enum or string type
+        """
         actual_keys = []
         for key in keys:
             if isinstance(key, Key):
@@ -101,4 +230,14 @@ class DioramaComputerInterface:
         await self._send_cmd("hotkey", {"keys": actual_keys})
 
     async def to_screen_coordinates(self, x, y):
+        """
+        Convert coordinates to screen coordinates.
+        
+        Args:
+            x (int): X coordinate to convert
+            y (int): Y coordinate to convert
+            
+        Returns:
+            dict: Dictionary containing the converted screen coordinates
+        """
         return await self._send_cmd("to_screen_coordinates", {"x": x, "y": y})

computer/providers/winsandbox/provider.py

@@ -5,6 +5,7 @@ import asyncio
 import logging
 import time
 from typing import Dict, Any, Optional, List
+from pathlib import Path
 
 from ..base import BaseVMProvider, VMProviderType
 
@@ -242,8 +243,15 @@ class WinSandboxProvider(BaseVMProvider):
             
             networking = run_opts.get("networking", self.networking)
             
-            # Create folder mappers if shared directories are specified
+            # Create folder mappers; always map a persistent venv directory on host for caching packages
             folder_mappers = []
+            # Ensure host side persistent venv directory exists (Path.home()/wsb_venv)
+            host_wsb_env = Path.home() / ".cua" / "wsb_cache"
+            try:
+                host_wsb_env.mkdir(parents=True, exist_ok=True)
+            except Exception:
+                # If cannot create, continue without persistent mapping
+                host_wsb_env = None
             shared_directories = run_opts.get("shared_directories", [])
             for shared_dir in shared_directories:
                 if isinstance(shared_dir, dict):
@@ -255,6 +263,15 @@ class WinSandboxProvider(BaseVMProvider):
                     
                 if host_path and os.path.exists(host_path):
                     folder_mappers.append(winsandbox.FolderMapper(host_path))
+
+            # Add mapping for the persistent venv directory (read/write) so it appears in Sandbox Desktop
+            if host_wsb_env is not None and host_wsb_env.exists():
+                try:
+                    folder_mappers.append(
+                        winsandbox.FolderMapper(str(host_wsb_env), read_only=False)
+                    )
+                except Exception as e:
+                    self.logger.warning(f"Failed to map host winsandbox_venv: {e}")
             
             self.logger.info(f"Creating Windows Sandbox: {name}")
             self.logger.info(f"Memory: {memory_mb}MB, Networking: {networking}")
@@ -290,8 +307,10 @@ class WinSandboxProvider(BaseVMProvider):
             
             self.logger.info(f"Windows Sandbox {name} created successfully")
             
+            venv_exists = (host_wsb_env / "venv" / "Lib" / "site-packages" / "computer_server").exists() if host_wsb_env else False
+
             # Setup the computer server in the sandbox
-            await self._setup_computer_server(sandbox, name)
+            await self._setup_computer_server(sandbox, name, wait_for_venv=(not venv_exists))
             
             return {
                 "success": True,
@@ -423,7 +442,7 @@ class WinSandboxProvider(BaseVMProvider):
             if total_attempts % 10 == 0:
                 self.logger.info(f"Still waiting for Windows Sandbox {name} IP after {total_attempts} attempts...")
     
-    async def _setup_computer_server(self, sandbox, name: str, visible: bool = False):
+    async def _setup_computer_server(self, sandbox, name: str, visible: bool = False, wait_for_venv: bool = True):
         """Setup the computer server in the Windows Sandbox using RPyC.
         
         Args:
@@ -471,10 +490,12 @@ class WinSandboxProvider(BaseVMProvider):
                 creationflags=creation_flags,
                 shell=False
             )
-            
-            # # Sleep for 30 seconds
-            # await asyncio.sleep(30)
 
+            if wait_for_venv:
+                print("Waiting for venv to be created for the first time setup of Windows Sandbox...")
+                print("This may take a minute...")
+                await asyncio.sleep(120)
+            
             ip = await self.get_ip(name)
             self.logger.info(f"Sandbox IP: {ip}")
             self.logger.info(f"Setup script started in background in sandbox {name} with PID: {process.pid}")

computer/providers/winsandbox/setup_script.ps1

@@ -79,23 +79,48 @@ try {
     $pythonVersion = & $pythonExe --version 2>&1
     Write-Host "Python version: $pythonVersion"
 
-    # Step 2: Install cua-computer-server directly
-    Write-Host "Step 2: Installing cua-computer-server..."
+    # Step 2: Create a dedicated virtual environment in mapped Desktop folder (persistent)
+    Write-Host "Step 2: Creating virtual environment (if needed)..."
+    $cachePath = "C:\Users\WDAGUtilityAccount\Desktop\wsb_cache"
+    $venvPath = "C:\Users\WDAGUtilityAccount\Desktop\wsb_cache\venv"
+    if (!(Test-Path $venvPath)) {
+        Write-Host "Creating venv at: $venvPath"
+        & $pythonExe -m venv $venvPath
+    } else {
+        Write-Host "Venv already exists at: $venvPath"
+    }
+    # Hide the folder to keep Desktop clean
+    try {
+        $item = Get-Item $cachePath -ErrorAction SilentlyContinue
+        if ($item) {
+            if (-not ($item.Attributes -band [IO.FileAttributes]::Hidden)) {
+                $item.Attributes = $item.Attributes -bor [IO.FileAttributes]::Hidden
+            }
+        }
+    } catch { }
+    $venvPython = Join-Path $venvPath "Scripts\python.exe"
+    if (!(Test-Path $venvPython)) {
+        throw "Virtual environment Python not found at $venvPython"
+    }
+    Write-Host "Using venv Python: $venvPython"
+
+    # Step 3: Install cua-computer-server into the venv
+    Write-Host "Step 3: Installing cua-computer-server..."
     
     Write-Host "Upgrading pip..."
-    & $pythonExe -m pip install --upgrade pip --quiet
+    & $venvPython -m pip install --upgrade pip --quiet
     
     Write-Host "Installing cua-computer-server..."
-    & $pythonExe -m pip install cua-computer-server --quiet
+    & $venvPython -m pip install cua-computer-server
     
     Write-Host "cua-computer-server installation completed."
 
-    # Step 3: Start computer server in background
-    Write-Host "Step 3: Starting computer server in background..."
-    Write-Host "Starting computer server with: $pythonExe"
+    # Step 4: Start computer server in background using the venv Python
+    Write-Host "Step 4: Starting computer server in background..."
+    Write-Host "Starting computer server with: $venvPython"
     
     # Start the computer server in the background
-    $serverProcess = Start-Process -FilePath $pythonExe -ArgumentList "-m", "computer_server.main" -WindowStyle Hidden -PassThru
+    $serverProcess = Start-Process -FilePath $venvPython -ArgumentList "-m", "computer_server.main" -WindowStyle Hidden -PassThru
     Write-Host "Computer server started in background with PID: $($serverProcess.Id)"
     
     # Give it a moment to start

{cua_computer-0.4.4.dist-info → cua_computer-0.4.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cua-computer
-Version: 0.4.4
+Version: 0.4.6
 Summary: Computer-Use Interface (CUI) framework powering Cua
 Author-Email: TryCua <gh@trycua.com>
 Requires-Python: >=3.11
@@ -26,8 +26,8 @@ Description-Content-Type: text/markdown
 <h1>
   <div class="image-wrapper" style="display: inline-block;">
     <picture>
-      <source media="(prefers-color-scheme: dark)" alt="logo" height="150" srcset="../../img/logo_white.png" style="display: block; margin: auto;">
-      <source media="(prefers-color-scheme: light)" alt="logo" height="150" srcset="../../img/logo_black.png" style="display: block; margin: auto;">
+      <source media="(prefers-color-scheme: dark)" alt="logo" height="150" srcset="https://raw.githubusercontent.com/trycua/cua/main/img/logo_white.png" style="display: block; margin: auto;">
+      <source media="(prefers-color-scheme: light)" alt="logo" height="150" srcset="https://raw.githubusercontent.com/trycua/cua/main/img/logo_black.png" style="display: block; margin: auto;">
       <img alt="Shows my svg">
     </picture>
   </div>
@@ -44,7 +44,7 @@ Description-Content-Type: text/markdown
 ### Get started with Computer
 
 <div align="center">
-    <img src="../../img/computer.png"/>
+    <img src="https://raw.githubusercontent.com/trycua/cua/main/img/computer.png"/>
 </div>
 
 ```python
@@ -87,82 +87,11 @@ The `cua-computer` PyPi package pulls automatically the latest executable versio
 
 Refer to this notebook for a step-by-step guide on how to use the Computer-Use Interface (CUI):
 
-- [Computer-Use Interface (CUI)](../../notebooks/computer_nb.ipynb)
+- [Computer-Use Interface (CUI)](https://github.com/trycua/cua/blob/main/notebooks/computer_nb.ipynb)
 
-## Using the Gradio Computer UI
-
-The computer module includes a Gradio UI for creating and sharing demonstration data. We make it easy for people to build community datasets for better computer use models with an upload to Huggingface feature.
-
-```bash
-# Install with UI support
-pip install "cua-computer[ui]"
-```
-
-> **Note:** For precise control of the computer, we recommend using VNC or Screen Sharing instead of the Computer Gradio UI.
-
-### Building and Sharing Demonstrations with Huggingface
-
-Follow these steps to contribute your own demonstrations:
-
-#### 1. Set up Huggingface Access
-
-Set your HF_TOKEN in a .env file or in your environment variables:
-
-```bash
-# In .env file
-HF_TOKEN=your_huggingface_token
-```
-
-#### 2. Launch the Computer UI
-
-```python
-# launch_ui.py
-from computer.ui.gradio.app import create_gradio_ui
-from dotenv import load_dotenv
-load_dotenv('.env')
-
-app = create_gradio_ui()
-app.launch(share=False)
-```
-
-For examples, see [Computer UI Examples](../../examples/computer_ui_examples.py)
-
-#### 3. Record Your Tasks
-
-<details open>
-<summary>View demonstration video</summary>
-<video src="https://github.com/user-attachments/assets/de3c3477-62fe-413c-998d-4063e48de176" controls width="600"></video>
-</details>
-
-Record yourself performing various computer tasks using the UI.
-
-#### 4. Save Your Demonstrations
-
-<details open>
-<summary>View demonstration video</summary>
-<video src="https://github.com/user-attachments/assets/5ad1df37-026a-457f-8b49-922ae805faef" controls width="600"></video>
-</details>
-
-Save each task by picking a descriptive name and adding relevant tags (e.g., "office", "web-browsing", "coding").
-
-#### 5. Record Additional Demonstrations
-
-Repeat steps 3 and 4 until you have a good amount of demonstrations covering different tasks and scenarios.
-
-#### 6. Upload to Huggingface
-
-<details open>
-<summary>View demonstration video</summary>
-<video src="https://github.com/user-attachments/assets/c586d460-3877-4b5f-a736-3248886d2134" controls width="600"></video>
-</details>
-
-Upload your dataset to Huggingface by:
-- Naming it as `{your_username}/{dataset_name}`
-- Choosing public or private visibility
-- Optionally selecting specific tags to upload only tasks with certain tags
-
-#### Examples and Resources
-
-- Example Dataset: [ddupont/test-dataset](https://huggingface.co/datasets/ddupont/test-dataset)
-- Find Community Datasets: 🔍 [Browse CUA Datasets on Huggingface](https://huggingface.co/datasets?other=cua)
+## Docs
 
+- [Computers](https://trycua.com/docs/computer-sdk/computers)
+- [Commands](https://trycua.com/docs/computer-sdk/commands)
+- [Computer UI](https://trycua.com/docs/computer-sdk/computer-ui)
+- [Sandboxed Python](https://trycua.com/docs/computer-sdk/sandboxed-python)

{cua_computer-0.4.4.dist-info → cua_computer-0.4.6.dist-info}/RECORD

@@ -1,6 +1,6 @@
 computer/__init__.py,sha256=HG8dhCmSPjuQ4G-NGAoiXEhzhO37kwrHHmyboNhGWOA,1159
-computer/computer.py,sha256=AjPqUAv1hF2AgK1eODRrQCvMikyxeuE6gxlwsnwNB-s,42701
-computer/diorama_computer.py,sha256=jOP7_eXxxU6SMIoE25ni0YXPK0E7p5sZeLKmkYLh6G8,3871
+computer/computer.py,sha256=TBryDea-XM76PdVv4DH85MTSaOOTYHC8BOPrqeKRkeY,42630
+computer/diorama_computer.py,sha256=3JaXKpcSi_OAVXtwlmNwQgrcnvqP1AxdlKEQ0XRJ0aQ,8569
 computer/helpers.py,sha256=iHkO2WhuCLc15g67kfMnpQWxfNRlz2YeJNEvYaL9jlM,1826
 computer/interface/__init__.py,sha256=xQvYjq5PMn9ZJOmRR5mWtONTl_0HVd8ACvW6AQnzDdw,262
 computer/interface/base.py,sha256=1beR4T0z5anb9NaNgKJrMJTF0BFIKyiHlokMLesOV5Q,15131
@@ -25,14 +25,14 @@ computer/providers/lume_api.py,sha256=i9dXJGrUhfA49VSY4p6_O6_AzeLNlRppG7jbM3jIJm
 computer/providers/lumier/__init__.py,sha256=qz8coMA2K5MVoqNC12SDXJe6lI7z2pn6RHssUOMY5Ug,212
 computer/providers/lumier/provider.py,sha256=BDgnTuik42H9OuCmnd-1TxM8p4vl_ahfrhNbi0FNCMM,46644
 computer/providers/winsandbox/__init__.py,sha256=WsMVBBa_qFfqVHPQzg6j4PegQwLiIudkzUedpYkrfXU,244
-computer/providers/winsandbox/provider.py,sha256=vduDKUB1OuimvjJdUus9RQY8gv3XQfFgVS8fKOzHVME,19306
-computer/providers/winsandbox/setup_script.ps1,sha256=8aGwR7PEvqnYzCNyXTDKIwJ6pYrwyWYLRjmNT_jYIwQ,4623
+computer/providers/winsandbox/provider.py,sha256=rWLsvdgVd4DTTqWS3_smj-bnuEXEkbliRz5GpoFsdaY,20543
+computer/providers/winsandbox/setup_script.ps1,sha256=_fj4DPegmMjDbKx6bdSrJy0vtk_cICI1HPSrI9D1fys,5697
 computer/ui/__init__.py,sha256=pmo05ek9qiB_x7DPeE6Vf_8RsIOqTD0w1dBLMHfoOnY,45
 computer/ui/__main__.py,sha256=Jwy2oC_mGZLN0fX7WLqpjaQkbXMeM3ISrUc8WSRUG0c,284
 computer/ui/gradio/__init__.py,sha256=5_KimixM48-X74FCsLw7LbSt39MQfUMEL8-M9amK3Cw,117
 computer/ui/gradio/app.py,sha256=_V6FI-g0GJGMEk-C2iPFtxPO1Gn0juCaeCrWsBtjC4E,70395
 computer/utils.py,sha256=zY50NXB7r51GNLQ6l7lhG_qv0_ufpQ8n0-SDhCei8m4,2838
-cua_computer-0.4.4.dist-info/METADATA,sha256=17a2ljQZ5SZzcx_krdjSXDzoePmFvEInJA3gHzgOUbo,5802
-cua_computer-0.4.4.dist-info/WHEEL,sha256=9P2ygRxDrTJz3gsagc0Z96ukrxjr-LFBGOgv3AuKlCA,90
-cua_computer-0.4.4.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
-cua_computer-0.4.4.dist-info/RECORD,,
+cua_computer-0.4.6.dist-info/METADATA,sha256=MUzK45YA8YvzCdfDOgIt8Ko0qYAp7_uPYxz19N2hB2s,3776
+cua_computer-0.4.6.dist-info/WHEEL,sha256=9P2ygRxDrTJz3gsagc0Z96ukrxjr-LFBGOgv3AuKlCA,90
+cua_computer-0.4.6.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
+cua_computer-0.4.6.dist-info/RECORD,,