webscout 8.3__py3-none-any.whl → 8.3.2__py3-none-any.whl

webscout/Provider/deepseek_assistant.py

@@ -0,0 +1,378 @@
+from curl_cffi.requests import Session
+from curl_cffi import CurlError
+import json
+import re
+from typing import Any, Dict, Optional, Generator, Union, List
+
+from webscout.AIutel import Optimizers
+from webscout.AIutel import Conversation
+from webscout.AIutel import AwesomePrompts, sanitize_stream
+from webscout.AIbase import Provider
+from webscout import exceptions
+from webscout.litagent import LitAgent
+
+
+class DeepSeekAssistant(Provider):
+    """
+    A class to interact with the DeepSeek Assistant API.
+    
+    This provider interfaces with the deepseek-assistant.com API to provide
+    AI chat completions using the V3 model.
+    
+    Attributes:
+        AVAILABLE_MODELS (list): List of available models for the provider.
+    
+    Examples:
+        >>> from webscout.Provider.deepseek_assistant import DeepSeekAssistant
+        >>> ai = DeepSeekAssistant()
+        >>> response = ai.chat("What's the weather today?")
+        >>> print(response)
+        'I can help you with weather information...'
+    """
+
+    AVAILABLE_MODELS = ["V3 model", "R1 model"]
+
+    @staticmethod
+    def _deepseek_assistant_extractor(chunk: Union[str, Dict[str, Any]]) -> Optional[str]:
+        """Extracts content from DeepSeek Assistant stream JSON objects."""
+        if isinstance(chunk, dict):
+            return chunk.get("choices", [{}])[0].get("delta", {}).get("content")
+        return None
+
+    def __init__(
+        self,
+        is_conversation: bool = True,
+        max_tokens: int = 2049,
+        timeout: int = 30,
+        intro: str = None,
+        filepath: str = None,
+        update_file: bool = True,
+        proxies: dict = {},
+        history_offset: int = 10250,
+        act: str = None,
+        model: str = "V3 model",
+        system_prompt: str = "You are a helpful assistant.",
+        browser: str = "chrome"
+    ):
+        """
+        Initializes the DeepSeek Assistant API client.
+        
+        Args:
+            is_conversation (bool): Whether the provider is in conversation mode.
+            max_tokens (int): Maximum number of tokens to sample.
+            timeout (int): Timeout for API requests.
+            intro (str): Introduction message for the conversation.
+            filepath (str): Filepath for storing conversation history.
+            update_file (bool): Whether to update the conversation history file.
+            proxies (dict): Proxies for the API requests.
+            history_offset (int): Offset for conversation history.
+            act (str): Act for the conversation.
+            model (str): The model to use for completions.
+            system_prompt (str): The system prompt to define the assistant's role.
+            browser (str): Browser type for fingerprinting.
+        
+        Examples:
+            >>> ai = DeepSeekAssistant(model="V3 model")
+            >>> print(ai.model)
+            'V3 model'
+        """
+        if model not in self.AVAILABLE_MODELS:
+            raise ValueError(f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}")
+
+        self.url = "https://deepseek-assistant.com/api/search-stream-deep-chat-testing.php"
+
+        # Initialize LitAgent for user agent generation
+        self.agent = LitAgent()
+        self.fingerprint = self.agent.generate_fingerprint(browser)
+
+        # Headers based on the JavaScript code
+        self.headers = {
+            "accept": "*/*",
+            "accept-language": "id-ID,id;q=0.9",
+            "cache-control": "no-cache",
+            "content-type": "application/json",
+            "cookie": "click_id=OS3Hz0E1yKfu4YnZNwedESMEdKEgMTzL; organic_user_deepseek_assistant_ch=%7B%22pixel%22%3A%22OS3Hz0E1yKfu4YnZNwedESMEdKEgMTzL%22%2C%22cc%22%3A%22ID%22%2C%22channel%22%3A%22organic_flag%22%7D",
+            "origin": "https://deepseek-assistant.com",
+            **self.fingerprint
+
+        }
+
+        # Initialize curl_cffi Session
+        self.session = Session()
+        self.session.headers.update(self.headers)
+        self.session.proxies = proxies
+        
+        self.system_prompt = system_prompt
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.timeout = timeout
+        self.last_response = {}
+        self.model = model
+
+        self.__available_optimizers = (
+            method
+            for method in dir(Optimizers)
+            if callable(getattr(Optimizers, method)) and not method.startswith("__")
+        )
+        
+        Conversation.intro = (
+            AwesomePrompts().get_act(
+                act, raise_not_found=True, default=None, case_insensitive=True
+            )
+            if act
+            else intro or Conversation.intro
+        )
+
+        self.conversation = Conversation(
+            is_conversation, self.max_tokens_to_sample, filepath, update_file
+        )
+        self.conversation.history_offset = history_offset
+
+    def refresh_identity(self, browser: str = None):
+        """
+        Refreshes the browser identity fingerprint.
+
+        Args:
+            browser: Specific browser to use for the new fingerprint
+        """
+        browser = browser or self.fingerprint.get("browser_type", "chrome")
+        self.fingerprint = self.agent.generate_fingerprint(browser)
+
+        # Update user-agent header with new fingerprint
+        self.headers.update({
+            "user-agent": self.fingerprint.get("user_agent", self.headers["user-agent"])
+        })
+
+        # Update session headers
+        self.session.headers.update(self.headers)
+
+        return self.fingerprint
+
+    def _parse_chat_response(self, input_text: str) -> str:
+        """
+        Parses the chat response from the API, similar to the JavaScript parseChatResponse method.
+        
+        Args:
+            input_text (str): The raw response text from the API
+            
+        Returns:
+            str: The parsed content from the response
+        """
+        lines = input_text.strip().split("\n")
+        result = ""
+        
+        for line in lines:
+            trimmed_line = line.strip()
+            if trimmed_line.startswith("data: {") and trimmed_line.endswith("}"):
+                try:
+                    # Extract JSON from the line
+                    json_start = trimmed_line.find("{")
+                    if json_start != -1:
+                        json_str = trimmed_line[json_start:]
+                        parsed_data = json.loads(json_str)
+                        
+                        # Extract content from the parsed data
+                        content = parsed_data.get("choices", [{}])[0].get("delta", {}).get("content")
+                        if content is not None:
+                            result += content
+                except (json.JSONDecodeError, KeyError, IndexError):
+                    # Skip malformed JSON or missing keys
+                    continue
+        
+        return result.strip()
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> Union[Dict[str, Any], Generator]:
+        """
+        Sends a prompt to the DeepSeek Assistant API and returns the response.
+
+        Args:
+            prompt (str): The prompt to send to the API.
+            stream (bool): Whether to stream the response.
+            raw (bool): Whether to return the raw response.
+            optimizer (str): Optimizer to use for the prompt.
+            conversationally (bool): Whether to generate the prompt conversationally.
+
+        Returns:
+            Union[Dict[str, Any], Generator]: The API response.
+
+        Examples:
+            >>> ai = DeepSeekAssistant()
+            >>> response = ai.ask("Tell me a joke!")
+            >>> print(response)
+            {'text': 'Why did the scarecrow win an award? Because he was outstanding in his field!'}
+        """
+        conversation_prompt = self.conversation.gen_complete_prompt(prompt)
+        if optimizer:
+            if optimizer in self.__available_optimizers:
+                conversation_prompt = getattr(Optimizers, optimizer)(
+                    conversation_prompt if conversationally else prompt
+                )
+            else:
+                raise Exception(f"Optimizer is not one of {self.__available_optimizers}")
+
+        payload = {
+            "model": self.model,
+            "messages": [
+                {"role": "system", "content": self.system_prompt},  # Add system role
+                {"role": "user", "content": conversation_prompt}
+            ]
+        }
+
+        def for_stream():
+            streaming_text = ""
+            try:
+                response = self.session.post(
+                    self.url,
+                    data=json.dumps(payload),
+                    stream=True,
+                    timeout=self.timeout,
+                    impersonate="chrome110"
+                )
+                response.raise_for_status()
+
+                # Use sanitize_stream to process the response
+                processed_stream = sanitize_stream(
+                    data=response.iter_content(chunk_size=None),
+                    intro_value="data:",
+                    to_json=True,
+                    skip_markers=["[DONE]"],
+                    content_extractor=self._deepseek_assistant_extractor,
+                    yield_raw_on_error=False
+                )
+
+                for content_chunk in processed_stream:
+                    if content_chunk and isinstance(content_chunk, str):
+                        streaming_text += content_chunk
+                        resp = dict(text=content_chunk)
+                        yield resp if not raw else content_chunk
+
+            except CurlError as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {str(e)}") from e
+            except Exception as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed ({type(e).__name__}): {str(e)}") from e
+            finally:
+                # Update history after stream finishes or fails
+                if streaming_text:
+                    self.last_response = {"text": streaming_text}
+                    self.conversation.update_chat_history(prompt, streaming_text)
+
+        def for_non_stream():
+            try:
+                response = self.session.post(
+                    self.url,
+                    data=json.dumps(payload),
+                    timeout=self.timeout,
+                    impersonate="chrome110"
+                )
+                response.raise_for_status()
+
+                # Parse the response using the custom parser
+                content = self._parse_chat_response(response.text)
+
+                self.last_response = {"text": content}
+                self.conversation.update_chat_history(prompt, content)
+                return self.last_response if not raw else content
+
+            except CurlError as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}") from e
+            except Exception as e:
+                err_text = getattr(e, 'response', None) and getattr(e.response, 'text', '')
+                raise exceptions.FailedToGenerateResponseError(f"Request failed ({type(e).__name__}): {e} - {err_text}") from e
+
+        return for_stream() if stream else for_non_stream()
+
+    def chat(
+        self,
+        prompt: str,
+        stream: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> Union[str, Generator[str, None, None]]:
+        """
+        Initiates a chat with the DeepSeek Assistant API using the provided prompt.
+
+        Args:
+            prompt (str): The prompt to send to the API.
+            stream (bool): Whether to stream the response.
+            optimizer (str): Optimizer to use for the prompt.
+            conversationally (bool): Whether to generate the prompt conversationally.
+
+        Returns:
+            Union[str, Generator[str, None, None]]: The chat response.
+
+        Examples:
+            >>> ai = DeepSeekAssistant()
+            >>> response = ai.chat("Tell me a joke")
+            >>> print(response)
+            'Why did the scarecrow win an award? Because he was outstanding in his field!'
+        """
+        def for_stream_chat():
+            gen = self.ask(
+                prompt, stream=True, raw=False,
+                optimizer=optimizer, conversationally=conversationally
+            )
+            for response_dict in gen:
+                yield self.get_message(response_dict)
+
+        def for_non_stream_chat():
+            response_data = self.ask(
+                prompt, stream=False, raw=False,
+                optimizer=optimizer, conversationally=conversationally
+            )
+            return self.get_message(response_data)
+
+        return for_stream_chat() if stream else for_non_stream_chat()
+
+    def get_message(self, response: dict) -> str:
+        """
+        Extracts the message content from the API response.
+
+        Args:
+            response (dict): The API response.
+
+        Returns:
+            str: The message content.
+
+        Examples:
+            >>> ai = DeepSeekAssistant()
+            >>> response = ai.ask("Tell me a joke!")
+            >>> message = ai.get_message(response)
+            >>> print(message)
+            'Why did the scarecrow win an award? Because he was outstanding in his field!'
+        """
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["text"]
+
+
+if __name__ == "__main__":
+    # Test the provider
+    print("-" * 80)
+    print(f"{'Model':<50} {'Status':<10} {'Response'}")
+    print("-" * 80)
+
+    for model in DeepSeekAssistant.AVAILABLE_MODELS:
+        try:
+            test_ai = DeepSeekAssistant(model=model, timeout=60)
+            response = test_ai.chat("Say 'Hello' in one word", stream=True)
+            response_text = ""
+            for chunk in response:
+                response_text += chunk
+
+            if response_text and len(response_text.strip()) > 0:
+                status = "✓"
+                # Clean and truncate response
+                clean_text = response_text.strip().encode('utf-8', errors='ignore').decode('utf-8')
+                display_text = clean_text[:50] + "..." if len(clean_text) > 50 else clean_text
+            else:
+                status = "✗"
+                display_text = "Empty or invalid response"
+            print(f"\r{model:<50} {status:<10} {display_text}")
+        except Exception as e:
+            print(f"\r{model:<50} {'✗':<10} {str(e)}")

webscout/Provider/scira_chat.py

@@ -20,12 +20,13 @@ class SciraAI(Provider):
     AVAILABLE_MODELS = {
         "scira-default": "Grok3-mini", # thinking model
         "scira-grok-3": "Grok3",
-        "scira-anthropic": "Sonnet 3.7 thinking",
+        "scira-anthropic": "Claude 4 Sonnet",
+        "scira-anthropic-thinking": "Claude 4 Sonnet Thinking", # thinking model
         "scira-vision" : "Grok2-Vision", # vision model
         "scira-4o": "GPT4o",
         "scira-qwq": "QWQ-32B",
         "scira-o4-mini": "o4-mini",
-        "scira-google": "gemini 2.5 flash",
+        "scira-google": "gemini 2.5 flash Thinking", # thinking model
         "scira-google-pro": "gemini 2.5 pro",
         "scira-llama-4": "llama 4 Maverick",
     }

webscout/Provider/toolbaz.py

@@ -34,7 +34,6 @@ class Toolbaz(Provider):
         "Llama-4-Scout",
         "Llama-3.3-70B",
         "Qwen2.5-72B",
-        "Qwen2-72B",
         "grok-2-1212",
         "grok-3-beta",
         "toolbaz_v3.5_pro",

webscout/auth/__init__.py

@@ -0,0 +1,44 @@
+# webscout/auth/__init__.py
+
+from .models import User, APIKey
+from .database import DatabaseManager
+from .api_key_manager import APIKeyManager
+from .rate_limiter import RateLimiter
+from .middleware import AuthMiddleware
+from .schemas import (
+    APIKeyCreateRequest,
+    APIKeyCreateResponse,
+    APIKeyValidationResponse,
+    UserCreateRequest,
+    UserResponse,
+    HealthCheckResponse
+)
+from .server import create_app, run_api, start_server
+from .routes import Api
+from .config import ServerConfig, AppConfig
+from .exceptions import APIError
+from .providers import initialize_provider_map, initialize_tti_provider_map
+
+__all__ = [
+    "User",
+    "APIKey",
+    "DatabaseManager",
+    "APIKeyManager",
+    "RateLimiter",
+    "AuthMiddleware",
+    "APIKeyCreateRequest",
+    "APIKeyCreateResponse",
+    "APIKeyValidationResponse",
+    "UserCreateRequest",
+    "UserResponse",
+    "HealthCheckResponse",
+    "create_app",
+    "run_api",
+    "start_server",
+    "Api",
+    "ServerConfig",
+    "AppConfig",
+    "APIError",
+    "initialize_provider_map",
+    "initialize_tti_provider_map"
+]

webscout/auth/api_key_manager.py

@@ -0,0 +1,189 @@
+# webscout/auth/api_key_manager.py
+
+import secrets
+import string
+from datetime import datetime, timezone, timedelta
+from typing import Optional, Tuple
+import hashlib
+import logging
+
+from .models import User, APIKey
+from .database import DatabaseManager
+
+logger = logging.getLogger(__name__)
+
+
+class APIKeyManager:
+    """Manages API key generation, validation, and lifecycle."""
+    
+    def __init__(self, database_manager: DatabaseManager):
+        self.db = database_manager
+        self.key_prefix = "ws_"  # webscout prefix
+        self.key_length = 32  # Length of the random part
+    
+    def generate_api_key(self) -> str:
+        """Generate a secure API key."""
+        # Generate random string
+        alphabet = string.ascii_letters + string.digits
+        random_part = ''.join(secrets.choice(alphabet) for _ in range(self.key_length))
+        
+        # Add prefix
+        api_key = f"{self.key_prefix}{random_part}"
+        
+        return api_key
+    
+    def hash_api_key(self, api_key: str) -> str:
+        """Hash an API key for secure storage (optional, for extra security)."""
+        return hashlib.sha256(api_key.encode()).hexdigest()
+    
+    async def create_api_key(
+        self,
+        username: str,
+        telegram_id: str,
+        name: Optional[str] = None,
+        rate_limit: int = 10,
+        expires_in_days: Optional[int] = None
+    ) -> Tuple[APIKey, User]:
+        """Create a new API key and associated user if needed. Only one API key per user allowed."""
+
+        # Check if user already exists by telegram_id
+        user = await self.db.get_user_by_telegram_id(telegram_id)
+
+        if user:
+            # Check if user already has an API key
+            existing_keys = await self.db.get_api_keys_by_user(user.id)
+            active_keys = [key for key in existing_keys if key.is_active and not key.is_expired()]
+
+            if active_keys:
+                raise ValueError(f"User with Telegram ID {telegram_id} already has an active API key. Only one API key per user is allowed.")
+        else:
+            # Check if username is already taken
+            existing_user = await self.db.get_user_by_username(username)
+            if existing_user:
+                raise ValueError(f"Username '{username}' is already taken")
+
+            # Create new user
+            user = User(
+                username=username,
+                telegram_id=telegram_id,
+                created_at=datetime.now(timezone.utc),
+                updated_at=datetime.now(timezone.utc)
+            )
+            try:
+                user = await self.db.create_user(user)
+                logger.info(f"Created new user: {user.username} (Telegram ID: {telegram_id})")
+            except ValueError as e:
+                # User might already exist, try to get it
+                if "already exists" in str(e):
+                    user = await self.db.get_user_by_telegram_id(telegram_id)
+                    if not user:
+                        raise e
+                else:
+                    raise e
+        
+        # Generate API key
+        api_key_value = self.generate_api_key()
+        
+        # Calculate expiration
+        expires_at = None
+        if expires_in_days:
+            expires_at = datetime.now(timezone.utc) + timedelta(days=expires_in_days)
+        
+        # Create API key object
+        api_key = APIKey(
+            key=api_key_value,
+            user_id=user.id,
+            name=name,
+            created_at=datetime.now(timezone.utc),
+            expires_at=expires_at,
+            rate_limit=rate_limit,
+            is_active=True
+        )
+        
+        # Store in database
+        try:
+            api_key = await self.db.create_api_key(api_key)
+            logger.info(f"Created API key for user {user.username}: {api_key.id}")
+            return api_key, user
+        except ValueError as e:
+            # Key collision (very unlikely), try again
+            if "already exists" in str(e):
+                logger.warning("API key collision detected, regenerating...")
+                return await self.create_api_key(username, telegram_id, name, rate_limit, expires_in_days)
+            raise e
+    
+    async def validate_api_key(self, api_key: str) -> Tuple[bool, Optional[APIKey], Optional[str]]:
+        """
+        Validate an API key.
+        
+        Returns:
+            Tuple of (is_valid, api_key_object, error_message)
+        """
+        if not api_key:
+            return False, None, "API key is required"
+        
+        if not api_key.startswith(self.key_prefix):
+            return False, None, "Invalid API key format"
+        
+        try:
+            # Get API key from database
+            key_obj = await self.db.get_api_key(api_key)
+            
+            if not key_obj:
+                return False, None, "API key not found"
+            
+            if not key_obj.is_active:
+                return False, None, "API key is inactive"
+            
+            if key_obj.is_expired():
+                return False, None, "API key has expired"
+            
+            # Update last used timestamp
+            key_obj.last_used_at = datetime.now(timezone.utc)
+            key_obj.usage_count += 1
+            
+            try:
+                await self.db.update_api_key(key_obj)
+            except Exception as e:
+                logger.warning(f"Failed to update API key usage: {e}")
+            
+            return True, key_obj, None
+            
+        except Exception as e:
+            logger.error(f"Error validating API key: {e}")
+            return False, None, "Internal error during validation"
+    
+    async def revoke_api_key(self, api_key: str) -> bool:
+        """Revoke an API key by marking it as inactive."""
+        try:
+            key_obj = await self.db.get_api_key(api_key)
+            
+            if not key_obj:
+                return False
+            
+            key_obj.is_active = False
+            key_obj.updated_at = datetime.now(timezone.utc)
+            
+            await self.db.update_api_key(key_obj)
+            logger.info(f"Revoked API key: {key_obj.id}")
+            return True
+            
+        except Exception as e:
+            logger.error(f"Error revoking API key: {e}")
+            return False
+    
+    async def get_user_api_keys(self, user_id: str) -> list[APIKey]:
+        """Get all API keys for a user."""
+        try:
+            return await self.db.get_api_keys_by_user(user_id)
+        except Exception as e:
+            logger.error(f"Error getting user API keys: {e}")
+            return []
+    
+    async def cleanup_expired_keys(self) -> int:
+        """Clean up expired API keys (mark as inactive)."""
+        # This would require a method to get all API keys, which we haven't implemented
+        # For now, we'll just return 0
+        # In a production system, you'd want to implement this properly
+        logger.info("Cleanup expired keys called (not implemented)")
+        return 0