webscout 7.8__py3-none-any.whl → 8.0__py3-none-any.whl

webscout/Provider/AISEARCH/hika_search.py

@@ -0,0 +1,194 @@
+import requests
+import hashlib
+import json
+import random
+import time
+import re
+from typing import Dict, Optional, Generator, Union, Any
+
+from webscout.AIbase import AISearch
+from webscout import exceptions
+from webscout import LitAgent
+
+
+class Response:
+    """A wrapper class for Hika API responses."""
+    def __init__(self, text: str):
+        self.text = text
+    
+    def __str__(self):
+        return self.text
+    
+    def __repr__(self):
+        return self.text
+
+
+class Hika(AISearch):
+    """A class to interact with the Hika AI search API."""
+
+    def __init__(
+        self,
+        timeout: int = 60,
+        proxies: Optional[dict] = None,
+        language: str = "en",
+    ):
+        self.session = requests.Session()
+        self.base_url = "https://api.hika.fyi/api/"
+        self.endpoint = "kbase/web"
+        self.timeout = timeout
+        self.language = language
+        self.last_response = {}
+        
+        self.headers = {
+            "Content-Type": "application/json",
+            "Origin": "https://hika.fyi",
+            "Referer": "https://hika.fyi/",
+            "User-Agent": LitAgent().random()
+        }
+        
+        self.session.headers.update(self.headers)
+        self.proxies = proxies
+
+    def generate_id(self):
+        """Generate a unique ID and hash for the request."""
+        uid = ''.join(random.choice('0123456789abcdefghijklmnopqrstuvwxyz') for _ in range(10))
+        uid += hex(int(time.time()))[2:]
+        hash_id = hashlib.sha256(f"#{uid}*".encode()).hexdigest()
+        return {"uid": uid, "hashId": hash_id}
+
+    def clean_text(self, text):
+        """Clean all XML tags and control markers from text.
+        
+        Args:
+            text (str): The text to clean
+            
+        Returns:
+            str: Cleaned text
+        """
+        if not text:
+            return ""
+            
+        # Remove XML tags and special markers
+        # First remove <r> tag at the beginning
+        text = text.lstrip("<r>")
+        
+        # Remove any remaining XML tags
+        text = re.sub(r'<[^>]+>', '', text)
+        
+        # Remove [DONE] marker at the end
+        text = re.sub(r'\[DONE\]\s*$', '', text)
+        
+        return text
+
+    def search(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+    ) -> Union[Response, Generator[Union[Dict[str, str], Response], None, None]]:
+        """Search using the Hika API and get AI-generated responses."""
+        if not prompt or len(prompt) < 2:
+            raise exceptions.APIConnectionError("Search query must be at least 2 characters long")
+        
+        # Generate ID for this request
+        id_data = self.generate_id()
+        uid, hash_id = id_data["uid"], id_data["hashId"]
+        
+        # Update headers with request-specific values
+        request_headers = {
+            **self.headers,
+            "x-hika": hash_id,
+            "x-uid": uid
+        }
+        
+        # Prepare payload
+        payload = {
+            "keyword": prompt,
+            "language": self.language,
+            "stream": True  # Always request streaming for consistent handling
+        }
+
+        def for_stream():
+            try:
+                with self.session.post(
+                    f"{self.base_url}{self.endpoint}",
+                    json=payload,
+                    headers=request_headers,
+                    stream=True,
+                    timeout=self.timeout,
+                    proxies=self.proxies
+                ) as response:
+                    if not response.ok:
+                        raise exceptions.APIConnectionError(
+                            f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                        )
+                    
+                    for line in response.iter_lines(decode_unicode=True):
+                        if line and line.startswith("data: "):
+                            try:
+                                data = json.loads(line[6:])
+                                if "chunk" in data:
+                                    chunk = data["chunk"]
+                                    
+                                    # Skip [DONE] markers completely
+                                    if "[DONE]" in chunk:
+                                        continue
+                                    
+                                    # Clean the chunk
+                                    clean_chunk = self.clean_text(chunk)
+                                    
+                                    if clean_chunk:  # Only yield if there's content after cleaning
+                                        if raw:
+                                            yield {"text": clean_chunk}
+                                        else:
+                                            yield Response(clean_chunk)
+                            except json.JSONDecodeError:
+                                pass
+                                
+            except requests.exceptions.RequestException as e:
+                raise exceptions.APIConnectionError(f"Request failed: {e}")
+
+        def for_non_stream():
+            full_response = ""
+            for chunk in for_stream():
+                if raw:
+                    yield chunk
+                else:
+                    full_response += str(chunk)
+            
+            if not raw:
+                # Clean up the response text one final time
+                cleaned_response = self.format_response(full_response)
+                self.last_response = Response(cleaned_response)
+                return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+    
+    def format_response(self, text: str) -> str:
+        """Format the response text for better readability."""
+        if not text:
+            return ""
+            
+        # First clean any tags or markers
+        cleaned_text = self.clean_text(text)
+        
+        # Remove any empty lines
+        cleaned_text = re.sub(r'\n\s*\n', '\n\n', cleaned_text)
+        
+        # Remove any trailing whitespace
+        cleaned_text = cleaned_text.strip()
+        
+        return cleaned_text
+
+
+if __name__ == "__main__":
+    from rich import print
+    ai = Hika()
+    try:
+        response = ai.search(input(">>> "), stream=True, raw=False)
+        for chunk in response:
+            print(chunk, end="", flush=True)
+    except KeyboardInterrupt:
+        print("\nSearch interrupted by user.")
+    except Exception as e:
+        print(f"\nError: {e}")

webscout/Provider/AISEARCH/monica_search.py

@@ -0,0 +1,246 @@
+import requests
+import json
+import re
+import uuid
+from typing import Dict, Optional, Generator, Union, Any
+
+from webscout.AIbase import AISearch
+from webscout import exceptions
+from webscout.litagent import LitAgent
+
+
+class Response:
+    """A wrapper class for Monica API responses.
+    
+    This class automatically converts response objects to their text representation
+    when printed or converted to string.
+    
+    Attributes:
+        text (str): The text content of the response
+        
+    Example:
+        >>> response = Response("Hello, world!")
+        >>> print(response)
+        Hello, world!
+        >>> str(response)
+        'Hello, world!'
+    """
+    def __init__(self, text: str):
+        self.text = text
+    
+    def __str__(self):
+        return self.text
+    
+    def __repr__(self):
+        return self.text
+
+
+class Monica(AISearch):
+    """A class to interact with the Monica AI search API.
+    
+    Monica provides a powerful search interface that returns AI-generated responses
+    based on web content. It supports both streaming and non-streaming responses.
+    
+    Basic Usage:
+        >>> from webscout import Monica
+        >>> ai = Monica()
+        >>> # Non-streaming example
+        >>> response = ai.search("What is Python?")
+        >>> print(response)
+        Python is a high-level programming language...
+        
+        >>> # Streaming example
+        >>> for chunk in ai.search("Tell me about AI", stream=True):
+        ...     print(chunk, end="", flush=True)
+        Artificial Intelligence is...
+        
+        >>> # Raw response format
+        >>> for chunk in ai.search("Hello", stream=True, raw=True):
+        ...     print(chunk)
+        {'text': 'Hello'}
+        {'text': ' there!'}
+    
+    Args:
+        timeout (int, optional): Request timeout in seconds. Defaults to 60.
+        proxies (dict, optional): Proxy configuration for requests. Defaults to None.
+    """
+
+    def __init__(
+        self,
+        timeout: int = 60,
+        proxies: Optional[dict] = None,
+    ):
+        """Initialize the Monica API client.
+        
+        Args:
+            timeout (int, optional): Request timeout in seconds. Defaults to 60.
+            proxies (dict, optional): Proxy configuration for requests. Defaults to None.
+        """
+        self.session = requests.Session()
+        self.api_endpoint = "https://monica.so/api/search_v1/search"
+        self.stream_chunk_size = 64
+        self.timeout = timeout
+        self.last_response = {}
+        self.client_id = str(uuid.uuid4())
+        self.session_id = ""
+        
+        self.headers = {
+            "accept": "*/*",
+            "accept-encoding": "gzip, deflate, br, zstd",
+            "accept-language": "en-US,en;q=0.9",
+            "content-type": "application/json",
+            "dnt": "1",
+            "origin": "https://monica.so",
+            "referer": "https://monica.so/answers",
+            "sec-ch-ua": '"Microsoft Edge";v="135", "Not-A.Brand";v="8", "Chromium";v="135"',
+            "sec-ch-ua-mobile": "?0",
+            "sec-ch-ua-platform": '"Windows"',
+            "sec-fetch-dest": "empty",
+            "sec-fetch-mode": "cors",
+            "sec-fetch-site": "same-origin",
+            "sec-gpc": "1",
+            "user-agent": LitAgent().random(),
+            "x-client-id": self.client_id,
+            "x-client-locale": "en",
+            "x-client-type": "web",
+            "x-client-version": "5.4.3",
+            "x-from-channel": "NA",
+            "x-product-name": "Monica-Search",
+            "x-time-zone": "Asia/Calcutta;-330"
+        }
+        
+        self.cookies = {
+            "monica_home_theme": "auto",
+        }
+        
+        self.session.headers.update(self.headers)
+        self.proxies = proxies
+
+    def search(
+        self,
+        prompt: str,
+        stream: bool = False,
+        raw: bool = False,
+    ) -> Union[Response, Generator[Union[Dict[str, str], Response], None, None]]:
+        """Search using the Monica API and get AI-generated responses.
+        
+        This method sends a search query to Monica and returns the AI-generated response.
+        It supports both streaming and non-streaming modes, as well as raw response format.
+        
+        Args:
+            prompt (str): The search query or prompt to send to the API.
+            stream (bool, optional): If True, yields response chunks as they arrive.
+                                   If False, returns complete response. Defaults to False.
+            raw (bool, optional): If True, returns raw response dictionaries with 'text' key.
+                                If False, returns Response objects that convert to text automatically.
+                                Defaults to False.
+        
+        Returns:
+            Union[Response, Generator[Union[Dict[str, str], Response], None, None]]: 
+                - If stream=False: Returns complete response as Response object
+                - If stream=True: Yields response chunks as either Dict or Response objects
+        
+        Raises:
+            APIConnectionError: If the API request fails
+        """
+        task_id = str(uuid.uuid4())
+        payload = {
+            "pro": False,
+            "query": prompt,
+            "round": 1,
+            "session_id": self.session_id,
+            "language": "auto",
+            "task_id": task_id
+        }
+
+        def for_stream():
+            try:
+                with self.session.post(
+                    self.api_endpoint,
+                    json=payload,
+                    stream=True,
+                    cookies=self.cookies,
+                    timeout=self.timeout,
+                    proxies=self.proxies
+                ) as response:
+                    if not response.ok:
+                        raise exceptions.APIConnectionError(
+                            f"Failed to generate response - ({response.status_code}, {response.reason}) - {response.text}"
+                        )
+                    
+                    # Process the Server-Sent Events (SSE) stream
+                    for line in response.iter_lines(decode_unicode=True):
+                        if line and line.startswith("data: "):
+                            try:
+                                data = json.loads(line[6:])  # Remove 'data: ' prefix
+                                
+                                # Save session_id for future requests if present
+                                if "session_id" in data and data["session_id"]:
+                                    self.session_id = data["session_id"]
+                                
+                                # Only process chunks with text content
+                                if "text" in data and data["text"]:
+                                    text_chunk = data["text"]
+                                    
+                                    if raw:
+                                        yield {"text": text_chunk}
+                                    else:
+                                        yield Response(text_chunk)
+                                        
+                                # Check if stream is finished
+                                if "finished" in data and data["finished"]:
+                                    break
+                                    
+                            except json.JSONDecodeError:
+                                continue
+                    
+            except requests.exceptions.RequestException as e:
+                raise exceptions.APIConnectionError(f"Request failed: {e}")
+
+        def for_non_stream():
+            full_response = ""
+            search_results = []
+            
+            for chunk in for_stream():
+                if raw:
+                    yield chunk
+                else:
+                    full_response += str(chunk)
+            
+            if not raw:
+                # Process the full response to clean up formatting
+                formatted_response = self.format_response(full_response)
+                self.last_response = Response(formatted_response)
+                return self.last_response
+
+        return for_stream() if stream else for_non_stream()
+
+    @staticmethod
+    def format_response(text: str) -> str:
+        """Format the response text for better readability.
+        
+        Args:
+            text (str): The raw response text
+        
+        Returns:
+            str: Formatted text
+        """
+        # Clean up markdown formatting
+        cleaned_text = text.replace('**', '')
+        
+        # Remove any empty lines
+        cleaned_text = re.sub(r'\n\s*\n', '\n\n', cleaned_text)
+        
+        # Remove any trailing whitespace
+        cleaned_text = cleaned_text.strip()
+        
+        return cleaned_text
+
+
+if __name__ == "__main__":
+    from rich import print
+
+    ai = Monica()
+    response = ai.search(input(">>> "), stream=True, raw=False)
+    for chunk in response:
+        print(chunk, end="", flush=True)