webscout 8.2.7__py3-none-any.whl → 8.2.9__py3-none-any.whl

webscout/Provider/samurai.py

@@ -0,0 +1,224 @@
+from typing import *
+from webscout.AIutel import Conversation
+from webscout.AIutel import Optimizers
+from webscout.AIutel import AwesomePrompts, sanitize_stream
+from webscout.AIbase import Provider
+from webscout import exceptions
+from curl_cffi.requests import Session
+from curl_cffi import CurlError
+import json
+
+class samurai(Provider):
+    """
+    A class to interact with a custom API endpoint.
+    """
+    AVAILABLE_MODELS = [
+        "openai/gpt-4.1",
+        "openai/gpt-4o-search-preview",
+        "openai/gpt-4o-mini-search-preview",
+        "openai/gpt-4.1-nano",
+        "openai/chatgpt-4o-latest",
+        "openai/gpt-4.1-mini",
+        "gpt-4o",
+        "o3-mini",
+        "Claude-sonnet-3.7",
+        "uncensored-r1",
+        "anthropic/claude-3.5-sonnet",
+        "gemini-1.5-pro",
+        "gemini-1.5-pro-latest",
+        "gemini-flash-2.0",
+        "gemini-1.5-flash",
+        "llama-3.1-405b",
+        "Meta-Llama-3.1-405B-Instruct-Turbo",
+        "Meta-Llama-3.3-70B-Instruct-Turbo",
+        "chutesai/Llama-4-Maverick-17B-128E-Instruct-FP8",
+        "chutesai/Llama-4-Scout-17B-16E-Instruct",
+        "Qwen-QwQ-32B-Preview",
+        "Qwen/Qwen3-235B-A22B-fp8-tput",
+        "deepseek-r1",
+        "deepseek-v3",
+        "deepseek-ai/DeepSeek-V3-0324",
+        "dbrx-instruct",
+        "x-ai/grok-3-",
+        "perplexity-ai/r1-1776"
+    ]
+    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 = "openai/gpt-4.1",
+        system_prompt: str = "You are a helpful assistant."
+    ):
+        """Initializes the Custom API client."""
+        self.url = "https://newapi-9qln.onrender.com/v1/chat/completions"
+        self.headers = {
+            "Authorization": "Bearer Samurai-AP1-Fr33",
+            "Content-Type": "application/json"
+        }
+        self.session = Session()
+        self.session.headers.update(self.headers)
+        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
+
+    @staticmethod
+    def _extractor(chunk: dict) -> Optional[str]:
+        """Extracts content from stream JSON objects."""
+        return chunk.get("choices", [{}])[0].get("delta", {}).get("content")
+
+    def ask(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> Union[Dict[str, Any], Generator]:
+        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},
+                {"role": "user", "content": conversation_prompt},
+            ],
+            "stream": stream
+        }
+
+        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()
+
+                processed_stream = sanitize_stream(
+                    data=response.iter_lines(),
+                    intro_value="data:",
+                    to_json=True,
+                    skip_markers=["[DONE]"],
+                    content_extractor=self._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 Exception as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed: {str(e)}") from e
+            finally:
+                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()
+                json_response = response.json()
+                content = json_response.get("choices", [{}])[0].get("message", {}).get("content", "")
+
+                self.last_response = {"text": content}
+                self.conversation.update_chat_history(prompt, content)
+                return self.last_response if not raw else content
+
+            except Exception as e:
+                raise exceptions.FailedToGenerateResponseError(f"Request failed: {str(e)}") 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]]:
+        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:
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        return response["text"]
+
+if __name__ == "__main__":
+    # Ensure curl_cffi is installed
+    print("-" * 80)
+    print(f"{'Model':<50} {'Status':<10} {'Response'}")
+    print("-" * 80)
+
+    for model in samurai.AVAILABLE_MODELS:
+        try:
+            test_ai = samurai(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

@@ -0,0 +1,299 @@
+from os import system
+from curl_cffi import CurlError
+from curl_cffi.requests import Session
+import json
+import uuid
+import re
+from typing import Any, Dict, Optional, Union, List
+from webscout.AIutel import Optimizers
+from webscout.AIutel import Conversation, sanitize_stream # Import sanitize_stream
+from webscout.AIutel import AwesomePrompts
+from webscout.AIbase import Provider
+from webscout import exceptions
+from webscout.litagent import LitAgent
+
+class SciraAI(Provider):
+    """
+    A class to interact with the Scira AI chat API.
+    """
+
+    AVAILABLE_MODELS = {
+        "scira-default": "Grok3-mini", # thinking model
+        "scira-grok-3": "Grok3",
+        "scira-anthropic": "Sonnet 3.7 thinking",
+        "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-pro": "gemini 2.5 pro",
+        "scira-llama-4": "llama 4 Maverick",
+    }
+
+    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 = "scira-default",
+        chat_id: str = None,
+        user_id: str = None,
+        browser: str = "chrome",
+        system_prompt: str = "You are a helpful assistant.",
+    ):
+        """Initializes the Scira AI API client.
+
+        Args:
+            is_conversation (bool): Whether to maintain conversation history.
+            max_tokens (int): Maximum number of tokens to generate.
+            timeout (int): Request timeout in seconds.
+            intro (str): Introduction text for the conversation.
+            filepath (str): Path to save conversation history.
+            update_file (bool): Whether to update the conversation history file.
+            proxies (dict): Proxy configuration for requests.
+            history_offset (int): Maximum history length in characters.
+            act (str): Persona for the AI to adopt.
+            model (str): Model to use, must be one of AVAILABLE_MODELS.
+            chat_id (str): Unique identifier for the chat session.
+            user_id (str): Unique identifier for the user.
+            browser (str): Browser to emulate in requests.
+            system_prompt (str): System prompt for the AI.
+
+        """
+        if model not in self.AVAILABLE_MODELS:
+            raise ValueError(f"Invalid model: {model}. Choose from: {self.AVAILABLE_MODELS}")
+
+        self.url = "https://scira.ai/api/search"
+
+        # Initialize LitAgent for user agent generation
+        self.agent = LitAgent()
+        # Use fingerprinting to create a consistent browser identity
+        self.fingerprint = self.agent.generate_fingerprint(browser)
+        self.system_prompt = system_prompt
+        
+        # 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",
+            "Origin": "https://scira.ai",
+            "Referer": "https://scira.ai/",
+            "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"],
+            "Sec-Fetch-Dest": "empty",
+            "Sec-Fetch-Mode": "cors",
+            "Sec-Fetch-Site": "same-origin"
+        }
+
+        self.session = Session() # Use curl_cffi Session
+        self.session.headers.update(self.headers)
+        self.session.proxies = proxies # Assign proxies directly
+
+        self.is_conversation = is_conversation
+        self.max_tokens_to_sample = max_tokens
+        self.timeout = timeout
+        self.last_response = {}
+        self.model = model
+        self.chat_id = chat_id or str(uuid.uuid4())
+        self.user_id = user_id or f"user_{str(uuid.uuid4())[:8].upper()}"
+
+        # Always use chat mode (no web search)
+        self.search_mode = "chat"
+
+        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 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
+        for header, value in self.headers.items():
+            self.session.headers[header] = value
+
+        return self.fingerprint
+
+    @staticmethod
+    def _scira_extractor(chunk: Union[str, Dict[str, Any]]) -> Optional[str]:
+        """Extracts content from the Scira stream format '0:"..."'."""
+        if isinstance(chunk, str):
+            match = re.search(r'0:"(.*?)"(?=,|$)', chunk) # Look for 0:"...", possibly followed by comma or end of string
+            if match:
+                # Decode potential unicode escapes like \u00e9 and handle escaped quotes/backslashes
+                content = match.group(1).encode().decode('unicode_escape')
+                return content.replace('\\\\', '\\').replace('\\"', '"')
+        return None
+
+    def ask(
+        self,
+        prompt: str,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> Dict[str, Any]: # Note: Stream parameter removed as API doesn't seem to support it
+        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}")
+
+        messages = [
+            {"role": "system", "content": self.system_prompt},
+            {"role": "user", "content": conversation_prompt, "parts": [{"type": "text", "text": conversation_prompt}]}
+        ]
+
+        # Prepare the request payload
+        payload = {
+            "id": self.chat_id,
+            "messages": messages,
+            "model": self.model,
+            "group": self.search_mode,
+            "user_id": self.user_id,
+            "timezone": "Asia/Calcutta"
+        }
+
+        try:
+            # Use curl_cffi post with impersonate
+            response = self.session.post(
+                self.url,
+                json=payload,
+                timeout=self.timeout,
+                impersonate="chrome120" # Add impersonate
+            )
+            if response.status_code != 200:
+                # Try to get response content for better error messages
+                try: # Use try-except for reading response content
+                    error_content = response.text
+                except:
+                    error_content = "<could not read response content>"
+
+                if response.status_code in [403, 429]:
+                    print(f"Received status code {response.status_code}, refreshing identity...")
+                    self.refresh_identity()
+                    response = self.session.post(
+                        self.url, json=payload, timeout=self.timeout,
+                        impersonate="chrome120" # Add impersonate to retry
+                    )
+                    if not response.ok:
+                        raise exceptions.FailedToGenerateResponseError(
+                            f"Failed to generate response after identity refresh - ({response.status_code}, {response.reason}) - {error_content}"
+                        )
+                    print("Identity refreshed successfully.")
+                else:
+                    raise exceptions.FailedToGenerateResponseError(
+                        f"Request failed with status code {response.status_code}. Response: {error_content}"
+                    )
+
+            response_text_raw = response.text # Get raw response text
+
+            # Process the text using sanitize_stream line by line
+            processed_stream = sanitize_stream(
+                data=response_text_raw.splitlines(), # Split into lines
+                intro_value=None, # No simple prefix
+                to_json=False,    # Content is not JSON
+                content_extractor=self._scira_extractor # Use the specific extractor
+            )
+
+            # Aggregate the results from the generator
+            full_response = ""
+            for content in processed_stream:
+                if content and isinstance(content, str):
+                    full_response += content
+
+            self.last_response = {"text": full_response}
+            self.conversation.update_chat_history(prompt, full_response)
+            return {"text": full_response}
+        except CurlError as e: # Catch CurlError
+            raise exceptions.FailedToGenerateResponseError(f"Request failed (CurlError): {e}") from e
+        except Exception as e:
+            raise exceptions.FailedToGenerateResponseError(f"Request failed: {e}")
+
+    def chat(
+        self,
+        prompt: str,
+        optimizer: str = None,
+        conversationally: bool = False,
+    ) -> str:
+        return self.get_message(
+            self.ask(
+                prompt, optimizer=optimizer, conversationally=conversationally
+            )
+        )
+
+    def get_message(self, response: dict) -> str:
+        assert isinstance(response, dict), "Response should be of dict data-type only"
+        # Extractor handles formatting
+        return response.get("text", "").replace('\\n', '\n').replace('\\n\\n', '\n\n')
+
+if __name__ == "__main__":
+    print("-" * 100)
+    print(f"{'Model':<50} {'Status':<10} {'Response'}")
+    print("-" * 100)
+
+    test_prompt = "Say 'Hello' in one word"
+
+    # Test each model
+    for model in SciraAI.AVAILABLE_MODELS:
+        print(f"\rTesting {model}...", end="")
+
+        try:
+            test_ai = SciraAI(model=model, timeout=120)  # Increased timeout
+            response = test_ai.chat(test_prompt)
+
+            if response and len(response.strip()) > 0:
+                status = "✓"
+                # Clean and truncate response
+                clean_text = response.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:
+            error_msg = str(e)
+            # Truncate very long error messages
+            if len(error_msg) > 100:
+                error_msg = error_msg[:97] + "..."
+            print(f"\r{model:<50} {'✗':<10} Error: {error_msg}")