webscout 8.2.9__py3-none-any.whl → 2026.1.19__py3-none-any.whl

webscout/Provider/yep.py

@@ -1,389 +0,0 @@
-import uuid
-import json
-from curl_cffi import CurlError
-from curl_cffi.requests import Session
-
-from typing import Any, Dict, Optional, Generator, Union, List, TypeVar
-
-from webscout.AIutel import Optimizers
-from webscout.AIutel import AwesomePrompts, sanitize_stream # Import sanitize_stream
-from webscout.AIbase import Provider
-from webscout import exceptions
-from webscout.litagent import LitAgent
-from webscout.conversation import Conversation, Fn
-
-T = TypeVar('T')
-
-
-class YEPCHAT(Provider):
-    """
-    YEPCHAT is a provider class for interacting with the Yep API.
-
-    Attributes:
-        AVAILABLE_MODELS (list): List of available models for the provider.
-    """
-
-    AVAILABLE_MODELS = ["DeepSeek-R1-Distill-Qwen-32B", "Mixtral-8x7B-Instruct-v0.1"]
-
-    def __init__(
-        self,
-        is_conversation: bool = True,
-        max_tokens: int = 1280,
-        timeout: int = 30,
-        intro: str = None,
-        filepath: str = None,
-        update_file: bool = True,
-        proxies: dict = {},
-        history_offset: int = 10250,
-        act: str = None,
-        model: str = "DeepSeek-R1-Distill-Qwen-32B",
-        temperature: float = 0.6,
-        top_p: float = 0.7,
-        browser: str = "chrome",
-        tools: Optional[List[Fn]] = None
-    ):
-        """
-        Initializes the YEPCHAT provider with the specified parameters.
-
-        Examples:
-            >>> ai = YEPCHAT()
-            >>> ai.ask("What's the weather today?")
-            Sends a prompt to the Yep API and returns the response.
-
-            >>> ai.chat("Tell me a joke", stream=True)
-            Initiates a chat with the Yep API using the provided prompt.
-            
-            >>> weather_tool = Fn(name="get_weather", description="Get the current weather", parameters={"location": "string"})
-            >>> ai = YEPCHAT(tools=[weather_tool])
-            >>> ai.chat("What's the weather in New York?")
-            Uses the weather tool to provide weather information.
-        """
-        if model not in self.AVAILABLE_MODELS:
-            raise ValueError(
-                f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}"
-            )
-
-        # Initialize curl_cffi Session instead of cloudscraper
-        self.session = Session() 
-        self.is_conversation = is_conversation
-        self.max_tokens_to_sample = max_tokens
-        self.chat_endpoint = "https://api.yep.com/v1/chat/completions"
-        self.stream_chunk_size = 64
-        self.timeout = timeout
-        self.last_response = {}
-        self.model = model
-        self.temperature = temperature
-        self.top_p = top_p
-
-        # Initialize LitAgent for user agent generation
-        self.agent = LitAgent()
-        # Use fingerprinting to create a consistent browser identity
-        self.fingerprint = self.agent.generate_fingerprint(browser)
-
-        # Use the fingerprint for headers
-        self.headers = {
-            "Accept": self.fingerprint["accept"],
-            "Accept-Encoding": "gzip, deflate, br, zstd",
-            "Accept-Language": self.fingerprint["accept_language"],
-            "Content-Type": "application/json; charset=utf-8",
-            "DNT": "1",
-            "Origin": "https://yep.com",
-            "Referer": "https://yep.com/",
-            "Sec-CH-UA": self.fingerprint["sec_ch_ua"] or '"Not)A;Brand";v="99", "Microsoft Edge";v="127", "Chromium";v="127"',
-            "Sec-CH-UA-Mobile": "?0",
-            "Sec-CH-UA-Platform": f'"{self.fingerprint["platform"]}"',
-            "User-Agent": self.fingerprint["user_agent"],
-        }
-        
-        # Create session cookies with unique identifiers
-        self.cookies = {"__Host-session": uuid.uuid4().hex, '__cf_bm': uuid.uuid4().hex}
-
-        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, tools=tools
-        )
-        self.conversation.history_offset = history_offset
-        # Set consistent headers and proxies for the curl_cffi session
-        self.session.headers.update(self.headers)
-        self.session.proxies = proxies
-        # Note: curl_cffi handles cookies differently, passed directly in requests
-
-    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 headers with new fingerprint
-        self.headers.update({
-            "Accept": self.fingerprint["accept"],
-            "Accept-Language": self.fingerprint["accept_language"],
-            "Sec-CH-UA": self.fingerprint["sec_ch_ua"] or self.headers["Sec-CH-UA"],
-            "Sec-CH-UA-Platform": f'"{self.fingerprint["platform"]}"',
-            "User-Agent": self.fingerprint["user_agent"],
-        })
-        
-        # Update session headers
-        self.session.headers.update(self.headers)
-        
-        # Generate new cookies (will be passed in requests)
-        self.cookies = {"__Host-session": uuid.uuid4().hex, '__cf_bm': uuid.uuid4().hex}
-        
-        return self.fingerprint
-
-    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 Yep API and returns the response.
-        Now supports tool calling functionality.
-
-        Examples:
-            >>> ai = YEPCHAT()
-            >>> ai.ask("What's the weather today?")
-            Returns the response from the Yep API.
-
-            >>> ai.ask("Tell me a joke", stream=True)
-            Streams the response from the Yep API.
-            
-            >>> weather_tool = Fn(name="get_weather", description="Get the current weather", parameters={"location": "string"})
-            >>> ai = YEPCHAT(tools=[weather_tool])
-            >>> ai.ask("What's the weather in New York?")
-            Will use the weather tool to provide response.
-        """
-        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}"
-                )
-
-        data = {
-            "stream": stream,
-            "max_tokens": self.max_tokens_to_sample,
-            "top_p": self.top_p,
-            "temperature": self.temperature,
-            "messages": [{"content": conversation_prompt, "role": "user"}],
-            "model": self.model,
-        }
-
-        def for_stream():
-            try:
-                # buffer = b"" # No longer needed here
-                # Use curl_cffi session post, pass cookies explicitly
-                response = self.session.post(self.chat_endpoint, headers=self.headers, cookies=self.cookies, json=data, stream=True, timeout=self.timeout, impersonate=self.fingerprint.get("browser_type", "chrome110"))
-                
-                if not response.ok:
-                    # If we get a non-200 response, try refreshing our identity once
-                    if response.status_code in [403, 429]:
-                        self.refresh_identity()
-                        # Retry with new identity
-                        # Use curl_cffi session post, pass cookies explicitly
-                        retry_response = self.session.post(self.chat_endpoint, headers=self.headers, cookies=self.cookies, json=data, stream=True, timeout=self.timeout, impersonate=self.fingerprint.get("browser_type", "chrome110"))
-                        if not retry_response.ok:
-                            raise exceptions.FailedToGenerateResponseError(
-                                f"Failed to generate response after identity refresh - ({retry_response.status_code}, {retry_response.reason}) - {retry_response.text}"
-                            )
-                        response = retry_response # Use the successful retry response
-                    else:
-                        raise exceptions.FailedToGenerateResponseError(
-                            f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                        )
-
-                # --- Start of stream processing block (should be outside the 'if not response.ok' block) ---
-                streaming_text = ""
-
-                # Use sanitize_stream to process the lines
-                processed_stream = sanitize_stream(
-                    data=response.iter_content(chunk_size=None), # Pass the byte iterator directly
-                    intro_value="data:",
-                    to_json=True, # Yep sends JSON after 'data:'
-                    skip_markers=["[DONE]"], # Skip the final marker
-                    yield_raw_on_error=False, # Only process valid JSON data
-                    # --- Add the content extractor ---
-                    content_extractor=lambda chunk: chunk.get('choices', [{}])[0].get('delta', {}).get('content') if isinstance(chunk, dict) else None
-                )
-                # The loop now yields the final extracted string content directly
-                for content_chunk in processed_stream:
-                    # --- TEMPORARY DEBUG PRINT ---
-                    # print(f"\nDEBUG: Received extracted content: {content_chunk!r}\n", flush=True) # Keep or remove debug print as needed
-                    if content_chunk and isinstance(content_chunk, str): # Ensure it's a non-empty string
-                        streaming_text += content_chunk
-                        # Yield dict or raw string chunk based on 'raw' flag
-                        yield dict(text=content_chunk) if not raw else content_chunk
-                # --- End of stream processing block ---
-
-                # Check if the response contains a tool call (This should happen *after* processing the stream)
-                response_data = self.conversation.handle_tool_response(streaming_text)
-                
-                if response_data["is_tool_call"]:
-                    # Handle tool call results
-                    if response_data["success"]:
-                        for tool_call in response_data.get("tool_calls", []):
-                            tool_name = tool_call.get("name", "unknown_tool")
-                            result = response_data["result"]
-                            self.conversation.update_chat_history_with_tool(prompt, tool_name, result)
-                    else:
-                        # If tool call failed, update history with error
-                        self.conversation.update_chat_history(prompt, 
-                            f"Error executing tool call: {response_data['result']}")
-                else:
-                    # Normal response handling
-                    self.conversation.update_chat_history(prompt, streaming_text)
-                        
-            except CurlError as e: # Catch CurlError
-                 raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}")
-            except Exception as e:
-                raise exceptions.FailedToGenerateResponseError(f"Request failed: {e}")
-
-        def for_non_stream():
-            try:
-                # Use curl_cffi session post, pass cookies explicitly
-                response = self.session.post(self.chat_endpoint, headers=self.headers, cookies=self.cookies, json=data, timeout=self.timeout, impersonate=self.fingerprint.get("browser_type", "chrome110"))
-                if not response.ok:
-                    if response.status_code in [403, 429]:
-                        self.refresh_identity()
-                        # Use curl_cffi session post, pass cookies explicitly
-                        response = self.session.post(self.chat_endpoint, headers=self.headers, cookies=self.cookies, json=data, timeout=self.timeout, impersonate=self.fingerprint.get("browser_type", "chrome110"))
-                        if not response.ok:
-                            raise exceptions.FailedToGenerateResponseError(
-                                f"Failed to generate response after identity refresh - ({response.status_code}, {response.reason}) - {response.text}"
-                            )
-                    else:
-                        raise exceptions.FailedToGenerateResponseError(
-                            f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
-                        )
-
-                # ... existing non-stream response handling code ...
-                response_data = response.json()
-                if 'choices' in response_data and len(response_data['choices']) > 0:
-                    content = response_data['choices'][0].get('message', {}).get('content', '')
-                    
-                    # Check if the response contains a tool call
-                    tool_response = self.conversation.handle_tool_response(content)
-                    
-                    if tool_response["is_tool_call"]:
-                        # Process tool call
-                        if tool_response["success"]:
-                            # Get the first tool call for simplicity
-                            if "tool_calls" in tool_response and len(tool_response["tool_calls"]) > 0:
-                                tool_call = tool_response["tool_calls"][0]
-                                tool_name = tool_call.get("name", "unknown_tool")
-                                tool_result = tool_response["result"]
-                                
-                                # Update chat history with tool call
-                                self.conversation.update_chat_history_with_tool(prompt, tool_name, tool_result)
-                                
-                                # Return tool result
-                                return {"text": tool_result, "is_tool_call": True, "tool_name": tool_name}
-                        
-                        # If tool call processing failed
-                        return {"text": tool_response["result"], "is_tool_call": True, "error": True}
-                    else:
-                        # Normal response handling
-                        self.conversation.update_chat_history(prompt, content)
-                        return {"text": content}
-                else:
-                    raise exceptions.FailedToGenerateResponseError("No response content found")
-            except CurlError as e: # Catch CurlError
-                 raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}")
-            except Exception as e:
-                raise exceptions.FailedToGenerateResponseError(f"Request failed: {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 Yep API using the provided prompt.
-
-        Examples:
-            >>> ai = YEPCHAT()
-            >>> ai.chat("Tell me a joke")
-            Returns the chat response from the Yep API.
-
-            >>> ai.chat("What's the weather today?", stream=True)
-            Streams the chat response from the Yep API.
-        """
-        def for_stream():
-            for response in self.ask(
-                prompt, True, optimizer=optimizer, conversationally=conversationally
-            ):
-                yield self.get_message(response)
-
-        def for_non_stream():
-            return self.get_message(
-                self.ask(
-                    prompt,
-                    False,
-                    optimizer=optimizer,
-                    conversationally=conversationally,
-                )
-            )
-
-        return for_stream() if stream else for_non_stream()
-
-    def get_message(self, response: dict) -> str:
-        """
-        Extracts the message content from the API response.
-
-        Examples:
-            >>> ai = YEPCHAT()
-            >>> response = ai.ask("Tell me a joke")
-            >>> ai.get_message(response)
-            Extracts and returns the message content from the response.
-        """
-        assert isinstance(response, dict)
-        return response["text"]
-
-
-if __name__ == "__main__":
-    print("-" * 80)
-    print(f"{'Model':<50} {'Status':<10} {'Response'}")
-    print("-" * 80)
-
-    for model in YEPCHAT.AVAILABLE_MODELS:
-        try:
-            test_ai = YEPCHAT(model=model, timeout=60)
-            response = test_ai.chat("Say 'Hello' in one word")
-            response_text = response
-            
-            if response_text and len(response_text.strip()) > 0:
-                status = "✓"
-                # Truncate response if too long
-                display_text = response_text.strip()[:50] + "..." if len(response_text.strip()) > 50 else response_text.strip()
-            else:
-                status = "✗"
-                display_text = "Empty or invalid response"
-            print(f"{model:<50} {status:<10} {display_text}")
-        except Exception as e:
-            print(f"{model:<50} {'✗':<10} {str(e)}")
-

webscout/litagent/Readme.md

@@ -1,276 +0,0 @@
-# 🔥 LitAgent - The Lit User Agent Generator
-
-LitAgent is a powerful and modern user agent generator that keeps your requests fresh and undetectable! Built with style and packed with features, it's your go-to solution for managing user agents in your web scraping projects.
-
-## 🚀 Quick Start
-
-```python
-from webscout import LitAgent
-
-# Create a LitAgent instance
-agent = LitAgent()
-
-# Get a random user agent
-ua = agent.random()
-print(ua)  # Mozilla/5.0 (Windows NT 11.0) AppleWebKit/537.36 ...
-```
-
-## 🎯 Features
-
-### Browser-Specific Agents
-
-```python
-# Get agents for specific browsers
-chrome_ua = agent.chrome()    # Latest Chrome agent
-firefox_ua = agent.firefox()  # Latest Firefox agent
-safari_ua = agent.safari()    # Latest Safari agent
-edge_ua = agent.edge()       # Latest Edge agent
-opera_ua = agent.opera()     # Latest Opera agent
-```
-
-### Device-Specific Agents
-
-```python
-# Get mobile or desktop agents
-mobile_ua = agent.mobile()    # Mobile device agent
-desktop_ua = agent.desktop()  # Desktop device agent
-
-# New - Get agents for specific device types
-tablet_ua = agent.tablet()    # Tablet device agent
-tv_ua = agent.smart_tv()      # Smart TV agent
-console_ua = agent.gaming()   # Gaming console agent
-```
-
-### OS-Specific Agents
-
-```python
-# New - Get agents for specific operating systems
-windows_ua = agent.windows()  # Windows agent
-mac_ua = agent.macos()        # macOS agent
-linux_ua = agent.linux()      # Linux agent
-android_ua = agent.android()  # Android agent
-ios_ua = agent.ios()          # iOS agent
-```
-
-### Custom Agent Generation
-
-```python
-# New - Generate custom user agents with specific attributes
-custom_ua = agent.custom(
-    browser="chrome",
-    version="91.0",
-    os="windows",
-    os_version="10",
-    device_type="desktop"
-)
-```
-
-### Keep It Fresh
-
-```python
-# Refresh your agents pool anytime
-agent.refresh()  # Generates new set of agents
-
-# New - Schedule automatic refreshes
-agent.auto_refresh(interval_minutes=30)  # Auto-refresh every 30 minutes
-```
-
-## 💫 Real-World Examples
-
-### With Requests
-
-```python
-import requests
-from webscout import LitAgent
-
-agent = LitAgent()
-
-def make_request(url):
-    headers = {
-        'User-Agent': agent.random(),
-        'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
-        'Accept-Language': 'en-US,en;q=0.5',
-        'Connection': 'keep-alive',
-    }
-    return requests.get(url, headers=headers)
-
-# Make requests with different agents
-response1 = make_request('https://api.example.com')  # Random agent
-response2 = make_request('https://mobile.example.com')  # Another random agent
-```
-
-### With aiohttp
-
-```python
-import aiohttp
-import asyncio
-from webscout import LitAgent
-
-agent = LitAgent()
-
-async def fetch(url):
-    headers = {'User-Agent': agent.random()}
-    async with aiohttp.ClientSession() as session:
-        async with session.get(url, headers=headers) as response:
-            return await response.text()
-
-# Use it in your async code
-async def main():
-    urls = [
-        'https://api1.example.com',
-        'https://api2.example.com',
-        'https://api3.example.com'
-    ]
-    tasks = [fetch(url) for url in urls]
-    results = await asyncio.gather(*tasks)
-    return results
-```
-
-### With Selenium
-
-```python
-from selenium import webdriver
-from webscout import LitAgent
-
-agent = LitAgent()
-
-def create_driver():
-    options = webdriver.ChromeOptions()
-    options.add_argument(f'user-agent={agent.chrome()}')
-    return webdriver.Chrome(options=options)
-
-# Use it with Selenium
-driver = create_driver()
-driver.get('https://example.com')
-```
-
-### New - With Playwright
-
-```python
-from playwright.sync_api import sync_playwright
-from webscout import LitAgent
-
-agent = LitAgent()
-
-def browse_with_playwright():
-    with sync_playwright() as p:
-        browser_options = {
-            "user_agent": agent.chrome(),
-            "viewport": {"width": 1280, "height": 720}
-        }
-        browser = p.chromium.launch()
-        context = browser.new_context(**browser_options)
-        page = context.new_page()
-        page.goto('https://example.com')
-        # Continue with your scraping logic
-        browser.close()
-```
-
-## 🌟 Pro Tips
-
-1. **Rotate Agents**: Refresh your agents pool periodically to avoid detection
-   ```python
-   agent = LitAgent()
-   for _ in range(10):
-       response = requests.get(url, headers={'User-Agent': agent.random()})
-       if _ % 3 == 0:  # Refresh every 3 requests
-           agent.refresh()
-   ```
-
-2. **Device-Specific Scraping**: Use appropriate agents for different platforms
-   ```python
-   # Mobile site scraping
-   mobile_response = requests.get(
-       'https://m.example.com',
-       headers={'User-Agent': agent.mobile()}
-   )
-
-   # Desktop site scraping
-   desktop_response = requests.get(
-       'https://example.com',
-       headers={'User-Agent': agent.desktop()}
-   )
-   ```
-
-3. **Browser Consistency**: Stick to one browser type per session
-   ```python
-   chrome_agent = agent.chrome()
-   headers = {
-       'User-Agent': chrome_agent,
-       'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
-   }
-   # Use these headers for all requests in this session
-   ```
-
-4. **New - Browser Fingerprinting Defense**:
-   ```python
-   # Create consistent browser fingerprinting
-   fingerprint = agent.generate_fingerprint(browser="chrome")
-   
-   headers = {
-       'User-Agent': fingerprint['user_agent'],
-       'Accept-Language': fingerprint['accept_language'],
-       'Accept': fingerprint['accept'],
-       'Sec-Ch-Ua': fingerprint['sec_ch_ua'],
-       'Sec-Ch-Ua-Platform': fingerprint['platform']
-   }
-   
-   # Use this consistent set for all session requests
-   ```
-
-5. **New - Multi-threading Support**:
-   ```python
-   import concurrent.futures
-   from webscout import LitAgent
-   
-   agent = LitAgent(thread_safe=True)  # Thread-safe instance
-   
-   def fetch_url(url):
-       headers = {'User-Agent': agent.random()}
-       return requests.get(url, headers=headers).text
-   
-   urls = ['https://example1.com', 'https://example2.com', 'https://example3.com']
-   
-   with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:
-       results = list(executor.map(fetch_url, urls))
-   ```
-
-## 🔧 Supported Browsers & Devices
-
-- **Browsers**: Chrome, Firefox, Safari, Edge, Opera, Brave, Vivaldi
-- **Operating Systems**: Windows, macOS, Linux, Android, iOS, Chrome OS
-- **Devices**: Mobile phones, Tablets, Desktops, Game consoles, Smart TVs, Wearables
-
-## 🎨 Why LitAgent?
-
-- 🚀 Modern and up-to-date user agents
-- 💪 Easy to use, hard to detect
-- 🔄 Fresh agents on demand
-- 📱 Device-specific agents
-- 🌐 All major browsers supported
-- ⚡ Lightweight and fast
-- 🧩 Advanced fingerprinting protection
-- 🔄 Seamless proxy integration
-- 🧵 Thread-safe operation
-- 🕰️ Automatic refresh scheduling
-
-## 📊 New - Analytics and Reporting
-
-```python
-# Get statistics on your agent usage
-stats = agent.get_stats()
-print(f"Agents generated: {stats.total_generated}")
-print(f"Most used browser: {stats.top_browser}")
-print(f"Detection avoidance rate: {stats.avoidance_rate}%")
-
-# Export your usage data
-agent.export_stats('agent_usage.json')
-```
-
-## 📋 Installation
-
-```bash
-pip install webscout
-```
-
-Made with 💖 by the HelpingAI team