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

webscout/AIauto.py

@@ -6,7 +6,7 @@ API keys or cookies.
 
 from webscout.AIbase import Provider
 from webscout.exceptions import AllProvidersFailure
-from typing import Union, Any, Dict, Generator
+from typing import Union, Any, Dict, Generator, Optional
 import importlib
 import pkgutil
 import random
@@ -71,7 +71,7 @@ class AUTO(Provider):
         proxies: dict = {},
         history_offset: int = 10250,
         act: str = None,
-        exclude: list[str] = [],
+        exclude: Optional[list[str]] = None,
         print_provider_info: bool = False,
     ):
         """
@@ -90,7 +90,7 @@ class AUTO(Provider):
             proxies (dict): Proxies for requests. Defaults to {}.
             history_offset (int): History character offset limit. Defaults to 10250.
             act (str, optional): Awesome prompt key. Defaults to None.
-            exclude (list[str]): List of provider names (uppercase) to exclude. Defaults to [].
+            exclude (Optional[list[str]]): List of provider names (uppercase) to exclude. Defaults to None.
             print_provider_info (bool): Whether to print the name of the successful provider. Defaults to False.
         """
         self.provider = None  # type: Provider
@@ -104,7 +104,7 @@ class AUTO(Provider):
         self.proxies: dict = proxies
         self.history_offset: int = history_offset
         self.act: str = act
-        self.exclude: list[str] = [e.upper() for e in exclude]
+        self.exclude: list[str] = [e.upper() for e in exclude] if exclude else []
         self.print_provider_info: bool = print_provider_info
 
 

webscout/AIbase.py

@@ -1,6 +1,6 @@
 from abc import ABC, abstractmethod
 from pathlib import Path
-from typing import AsyncGenerator, Dict, List, Union, Generator, Optional
+from typing import AsyncGenerator, Dict, List, Union, Generator, Optional, Any
 from typing_extensions import TypeAlias
 
 # Type aliases for better readability
@@ -178,6 +178,66 @@ class AsyncTTSProvider(ABC):
         """
         raise NotImplementedError("Method needs to be implemented in subclass")
 
+
+class STTProvider(ABC):
+    """Abstract base class for Speech-to-Text providers."""
+
+    @abstractmethod
+    def transcribe(self, audio_path: Union[str, Path], **kwargs) -> Dict[str, Any]:
+        """Transcribe audio file to text.
+
+        Args:
+            audio_path (Union[str, Path]): Path to the audio file
+            **kwargs: Additional provider-specific parameters
+
+        Returns:
+            Dict[str, Any]: Transcription result in OpenAI Whisper format
+        """
+        raise NotImplementedError("Method needs to be implemented in subclass")
+
+    @abstractmethod
+    def transcribe_from_url(self, audio_url: str, **kwargs) -> Dict[str, Any]:
+        """Transcribe audio from URL to text.
+
+        Args:
+            audio_url (str): URL of the audio file
+            **kwargs: Additional provider-specific parameters
+
+        Returns:
+            Dict[str, Any]: Transcription result in OpenAI Whisper format
+        """
+        raise NotImplementedError("Method needs to be implemented in subclass")
+
+
+class AsyncSTTProvider(ABC):
+    """Abstract base class for asynchronous Speech-to-Text providers."""
+
+    @abstractmethod
+    async def transcribe(self, audio_path: Union[str, Path], **kwargs) -> Dict[str, Any]:
+        """Transcribe audio file to text asynchronously.
+
+        Args:
+            audio_path (Union[str, Path]): Path to the audio file
+            **kwargs: Additional provider-specific parameters
+
+        Returns:
+            Dict[str, Any]: Transcription result in OpenAI Whisper format
+        """
+        raise NotImplementedError("Method needs to be implemented in subclass")
+
+    @abstractmethod
+    async def transcribe_from_url(self, audio_url: str, **kwargs) -> Dict[str, Any]:
+        """Transcribe audio from URL to text asynchronously.
+
+        Args:
+            audio_url (str): URL of the audio file
+            **kwargs: Additional provider-specific parameters
+
+        Returns:
+            Dict[str, Any]: Transcription result in OpenAI Whisper format
+        """
+        raise NotImplementedError("Method needs to be implemented in subclass")
+
     async def save_audio(self, audio_file: str, destination: str = None, verbose: bool = False) -> str:
         """Save audio to a specific destination asynchronously.
 

webscout/AIutel.py

@@ -277,69 +277,50 @@ def _sanitize_stream_sync(
     processing_active = start_marker is None
     buffer = ""
     found_start = False if start_marker else True
+    line_iterator: Iterable[str]
 
-    # Fast path for single string processing
     if isinstance(data, str):
-        processed_item = None
-        if processing_active:
-            if to_json:
-                try:
-                    data = data.strip()
-                    if data:
-                        processed_item = json.loads(data)
-                except Exception as e:
-                    if error_handler:
-                        try:
-                            handled = error_handler(e, data)
-                            if handled is not None:
-                                processed_item = handled
-
-                        except Exception:
-                            pass
-                    if processed_item is None:
-                        processed_item = data if yield_raw_on_error else None
+        # If data is a string, decide whether to split it into lines
+        # or treat it as an iterable containing a single chunk.
+        temp_lines: List[str]
+        if line_delimiter is None:  # Default: split by newlines if present
+            if '\n' in data or '\r' in data:
+                temp_lines = data.splitlines()
             else:
-                processed_item = _process_chunk(
-                    data, intro_value, False, effective_skip_markers,
-                    strip_chars, yield_raw_on_error, error_handler
-                )
-            if processed_item is not None:
-                if content_extractor:
-                    try:
-                        final_content = content_extractor(processed_item)
-                        if final_content is not None:
-                            yield final_content
-                    except Exception:
-                        pass
-                else:
-                    yield processed_item
-        return
-
-    # Stream processing path
-    if not hasattr(data, '__iter__'):
-        raise TypeError(f"Input must be a string or an iterable, not {type(data).__name__}")
-
-    try:
-        iterator = iter(data)
-        first_item = next(iterator, None)
-        if first_item is None:
+                temp_lines = [data]  # Treat as a single line/chunk
+        elif line_delimiter in data:  # Custom delimiter found in string
+            temp_lines = data.split(line_delimiter)
+        else:  # Custom delimiter not found, or string is effectively a single segment
+            temp_lines = [data]
+        line_iterator = iter(temp_lines)
+    elif hasattr(data, '__iter__'):  # data is an iterable (but not a string)
+        _iter = iter(data)
+        first_item = next(_iter, None)
+
+        if first_item is None:  # Iterable was empty
             return
+
         from itertools import chain
-        stream = chain([first_item], iterator)
+        # Reconstruct the full iterable including the first_item
+        stream_input_iterable = chain([first_item], _iter)
 
-        # Determine if we're dealing with bytes or strings
         if isinstance(first_item, bytes):
+            # Ensure stream_input_iterable is typed as Iterable[bytes] for _decode_byte_stream
             line_iterator = _decode_byte_stream(
-                stream,
+                stream_input_iterable, # type: ignore
                 encoding=encoding,
                 errors=encoding_errors,
                 buffer_size=buffer_size
             )
         elif isinstance(first_item, str):
-            line_iterator = stream
+            # Ensure stream_input_iterable is typed as Iterable[str]
+            line_iterator = stream_input_iterable # type: ignore
         else:
-            raise TypeError(f"Stream must yield strings or bytes, not {type(first_item).__name__}")
+            raise TypeError(f"Iterable must yield strings or bytes, not {type(first_item).__name__}")
+    else:  # Not a string and not an iterable
+        raise TypeError(f"Input must be a string or an iterable, not {type(data).__name__}")
 
+    try:
         for line in line_iterator:
             if not line:
                 continue
@@ -683,20 +664,32 @@ def sanitize_stream(
         Union[Generator[Any, None, None], AsyncGenerator[Any, None]]:
             A generator or an asynchronous generator yielding the processed data.
     """
+    # Determine the actual data payload to process
+    payload: Any  # The type of payload can change based on data's attributes
 
-    if hasattr(data, "__aiter__"):
+    text_attr = getattr(data, "text", None)
+    content_attr = getattr(data, "content", None)
+
+    if isinstance(text_attr, str):
+        payload = text_attr
+    elif isinstance(content_attr, bytes):
+        payload = content_attr.decode(encoding, encoding_errors)
+    else:
+        # Use the original data if .text or .content are not applicable or not found
+        payload = data
+
+    # Dispatch to sync or async worker based on the nature of the 'payload'
+    if hasattr(payload, "__aiter__"):
         return _sanitize_stream_async(
-            data, intro_value, to_json, skip_markers, strip_chars,
+            payload, intro_value, to_json, skip_markers, strip_chars,
             start_marker, end_marker, content_extractor, yield_raw_on_error,
             encoding, encoding_errors, buffer_size, line_delimiter, error_handler,
         )
     return _sanitize_stream_sync(
-        data, intro_value, to_json, skip_markers, strip_chars,
+        payload, intro_value, to_json, skip_markers, strip_chars,
         start_marker, end_marker, content_extractor, yield_raw_on_error,
         encoding, encoding_errors, buffer_size, line_delimiter, error_handler,
     )
-
-
 from .conversation import Conversation  # noqa: E402,F401
 from .Extra.autocoder import AutoCoder  # noqa: E402,F401
 from .optimizers import Optimizers  # noqa: E402,F401

webscout/Bing_search.py

@@ -0,0 +1,418 @@
+"""
+BingSearch - A Bing search library with advanced features
+"""
+from time import sleep
+from curl_cffi.requests import Session
+from urllib.parse import urlencode, unquote, urlparse, parse_qs
+import base64
+from typing import List, Dict, Optional, Any
+from concurrent.futures import ThreadPoolExecutor
+from webscout.litagent import LitAgent
+class BingSearchResult:
+    """Class to represent a Bing search result with metadata."""
+    def __init__(self, url: str, title: str, description: str):
+        self.url = url
+        self.title = title
+        self.description = description
+        self.metadata: Dict[str, Any] = {}
+
+    def __repr__(self) -> str:
+        return f"BingSearchResult(url={self.url}, title={self.title}, description={self.description})"
+
+class BingImageResult:
+    """Class to represent a Bing image search result."""
+    def __init__(self, title: str, image: str, thumbnail: str, url: str, source: str):
+        self.title = title
+        self.image = image
+        self.thumbnail = thumbnail
+        self.url = url
+        self.source = source
+    def __repr__(self):
+        return f"BingImageResult(title={self.title}, image={self.image}, url={self.url}, source={self.source})"
+
+class BingNewsResult:
+    """Class to represent a Bing news search result."""
+    def __init__(self, title: str, url: str, description: str, source: str = ""):
+        self.title = title
+        self.url = url
+        self.description = description
+        self.source = source
+    def __repr__(self):
+        return f"BingNewsResult(title={self.title}, url={self.url}, source={self.source})"
+
+class BingSearch:
+    """Bing search implementation with configurable parameters and advanced features."""
+    _executor: ThreadPoolExecutor = ThreadPoolExecutor()
+
+    def __init__(
+        self,
+        timeout: int = 10,
+        proxies: Optional[Dict[str, str]] = None,
+        verify: bool = True,
+        lang: str = "en-US",
+        sleep_interval: float = 0.0,
+        impersonate: str = "chrome110"
+    ):
+        self.timeout = timeout
+        self.proxies = proxies if proxies else {}
+        self.verify = verify
+        self.lang = lang
+        self.sleep_interval = sleep_interval
+        self._base_url = "https://www.bing.com"
+        self.session = Session(
+            proxies=self.proxies,
+            verify=self.verify,
+            timeout=self.timeout,
+            impersonate=impersonate
+        )
+        self.session.headers.update(LitAgent().generate_fingerprint())
+
+    def _selectors(self, element):
+        selectors = {
+            'url': 'h2 a',
+            'title': 'h2',
+            'text': 'p',
+            'links': 'ol#b_results > li.b_algo',
+            'next': 'div#b_content nav[role="navigation"] a.sb_pagN'
+        }
+        return selectors[element]
+
+    def _first_page(self, query):
+        url = f'{self._base_url}/search?q={query}&search=&form=QBLH'
+        return {'url': url, 'data': None}
+
+    def _next_page(self, soup):
+        selector = self._selectors('next')
+        next_page_tag = soup.select_one(selector)
+        url = None
+        if next_page_tag and next_page_tag.get('href'):
+            url = self._base_url + next_page_tag['href']
+        return {'url': url, 'data': None}
+
+    def _get_url(self, tag):
+        url = tag.get('href', '')
+        resp = url
+        try:
+            parsed_url = urlparse(url)
+            query_params = parse_qs(parsed_url.query)
+            if "u" in query_params:
+                encoded_url = query_params["u"][0][2:]
+                try:
+                    decoded_bytes = base64.urlsafe_b64decode(encoded_url + '===')
+                except base64.binascii.Error as e:
+                    print(f"Error decoding Base64 string: {e}")
+                    return url
+                resp = decoded_bytes.decode('utf-8')
+        except Exception as e:
+            print(f"Error decoding Base64 string: {e}")
+        return resp
+
+    def _make_request(self, term: str, results: int, start: int = 0) -> str:
+        params = {
+            "q": term,
+            "count": results,
+            "first": start + 1,
+            "setlang": self.lang,
+        }
+        url = self._base_url + "/search"
+        try:
+            resp = self.session.get(
+                url=url,
+                params=params,
+            )
+            resp.raise_for_status()
+            return resp.text
+        except Exception as e:
+            if hasattr(e, 'response') and e.response is not None:
+                raise Exception(f"Bing search failed with status {e.response.status_code}: {str(e)}")
+            else:
+                raise Exception(f"Bing search failed: {str(e)}")
+
+    def text(
+        self,
+        keywords: str,
+        region: str = None,
+        safesearch: str = "moderate",
+        max_results: int = 10,
+        unique: bool = True
+    ) -> List[BingSearchResult]:
+        """
+        Perform a text search on Bing.
+
+        Args:
+            keywords (str): The search keywords.
+            region (str, optional): The region for the search. Defaults to None.
+            safesearch (str): The safe search level ("on", "moderate", "off"). Defaults to "moderate".
+            max_results (int): The maximum number of results to fetch. Defaults to 10.
+            unique (bool): Whether to exclude duplicate URLs from the results. Defaults to True.
+
+        Returns:
+            List[BingSearchResult]: A list of Bing search results.
+        """
+        if not keywords:
+            raise ValueError("Search keywords cannot be empty")
+        from bs4 import BeautifulSoup
+        safe_map = {
+            "on": "Strict",
+            "moderate": "Moderate",
+            "off": "Off"
+        }
+        safe = safe_map.get(safesearch.lower(), "Moderate")
+        fetched_results = []
+        fetched_links = set()
+        def fetch_page(url):
+            try:
+                resp = self.session.get(url)
+                resp.raise_for_status()
+                return resp.text
+            except Exception as e:
+                if hasattr(e, 'response') and e.response is not None:
+                    raise Exception(f"Bing search failed with status {e.response.status_code}: {str(e)}")
+                else:
+                    raise Exception(f"Bing search failed: {str(e)}")
+
+        # Fix: get the first page URL
+        url = self._first_page(keywords)['url']
+        urls_to_fetch = [url]
+        while len(fetched_results) < max_results and urls_to_fetch:
+            with self._executor as executor:
+                html_pages = list(executor.map(fetch_page, urls_to_fetch))
+            urls_to_fetch = []
+            for html in html_pages:
+                soup = BeautifulSoup(html, "html.parser")
+                selector_links = self._selectors('links')
+                result_blocks = soup.select(selector_links)
+                for result in result_blocks:
+                    link_tag = result.select_one(self._selectors('url'))
+                    if not link_tag:
+                        continue
+                    url_val = self._get_url(link_tag)
+                    title_tag = result.select_one(self._selectors('title'))
+                    title = title_tag.get_text(strip=True) if title_tag else ''
+                    desc_tag = result.select_one(self._selectors('text'))
+                    description = desc_tag.get_text(strip=True) if desc_tag else ''
+                    if url_val and title:
+                        if unique and url_val in fetched_links:
+                            continue
+                        fetched_results.append(BingSearchResult(url_val, title, description))
+                        fetched_links.add(url_val)
+                        if len(fetched_results) >= max_results:
+                            break
+                if len(fetched_results) >= max_results:
+                    break
+                next_page_info = self._next_page(soup)
+                if next_page_info['url']:
+                    urls_to_fetch.append(next_page_info['url'])
+                sleep(self.sleep_interval)
+            next_page_info = self._next_page(soup)
+            url = next_page_info['url']
+            sleep(self.sleep_interval)
+        return fetched_results[:max_results]
+
+    def suggestions(self, query: str, region: str = None) -> List[str]:
+        """
+        Fetches search suggestions for a given query.
+
+        Args:
+            query (str): The search query for which suggestions are needed.
+            region (str, optional): The region code (e.g., "en-US") for localized suggestions.
+
+        Returns:
+            List[str]: A list of suggestion strings related to the query.
+        """
+        if not query:
+            raise ValueError("Search query cannot be empty")
+        params = {
+            "query": query,
+            "mkt": region if region else "en-US"
+        }
+        url = f"https://api.bing.com/osjson.aspx?{urlencode(params)}"
+        try:
+            resp = self.session.get(url)
+            resp.raise_for_status()
+            data = resp.json()
+            if isinstance(data, list) and len(data) > 1 and isinstance(data[1], list):
+                return data[1]
+            return []
+        except Exception as e:
+            if hasattr(e, 'response') and e.response is not None:
+                raise Exception(f"Bing suggestions failed with status {e.response.status_code}: {str(e)}")
+            else:
+                raise Exception(f"Bing suggestions failed: {str(e)}")
+
+    def images(
+        self,
+        keywords: str,
+        region: str = None,
+        safesearch: str = "moderate",
+        max_results: int = 10
+    ) -> List[BingImageResult]:
+        """
+        Perform an image search on Bing.
+
+        Args:
+            keywords (str): The search keywords.
+            region (str, optional): The region for the search. Defaults to None.
+            safesearch (str): The safe search level ("on", "moderate", "off"). Defaults to "moderate".
+            max_results (int): The maximum number of results to fetch. Defaults to 10.
+
+        Returns:
+            List[BingImageResult]: A list of Bing image search results.
+        """
+        if not keywords:
+            raise ValueError("Search keywords cannot be empty")
+        from bs4 import BeautifulSoup
+        safe_map = {
+            "on": "Strict",
+            "moderate": "Moderate",
+            "off": "Off"
+        }
+        safe = safe_map.get(safesearch.lower(), "Moderate")
+        params = {
+            "q": keywords,
+            "count": max_results,
+            "setlang": self.lang,
+            "safeSearch": safe,
+        }
+        if region:
+            params["mkt"] = region
+        url = f"{self._base_url}/images/search?{urlencode(params)}"
+        try:
+            resp = self.session.get(url)
+            resp.raise_for_status()
+            html = resp.text
+        except Exception as e:
+            if hasattr(e, 'response') and e.response is not None:
+                raise Exception(f"Bing image search failed with status {e.response.status_code}: {str(e)}")
+            else:
+                raise Exception(f"Bing image search failed: {str(e)}")
+        soup = BeautifulSoup(html, "html.parser")
+        results = []
+        for item in soup.select("a.iusc"):
+            try:
+                m = item.get("m")
+                import json
+                meta = json.loads(m) if m else {}
+                image_url = meta.get("murl", "")
+                thumb_url = meta.get("turl", "")
+                title = meta.get("t", "")
+                page_url = meta.get("purl", "")
+                source = meta.get("surl", "")
+                if image_url:
+                    results.append(BingImageResult(title, image_url, thumb_url, page_url, source))
+                    if len(results) >= max_results:
+                        break
+            except Exception:
+                continue
+        return results[:max_results]
+
+    def news(
+        self,
+        keywords: str,
+        region: str = None,
+        safesearch: str = "moderate",
+        max_results: int = 10,
+    ) -> List['BingNewsResult']:
+        """Bing news search."""
+        if not keywords:
+            raise ValueError("Search keywords cannot be empty")
+        from bs4 import BeautifulSoup
+        safe_map = {
+            "on": "Strict",
+            "moderate": "Moderate",
+            "off": "Off"
+        }
+        safe = safe_map.get(safesearch.lower(), "Moderate")
+        params = {
+            "q": keywords,
+            "form": "QBNH",
+            "safeSearch": safe,
+        }
+        if region:
+            params["mkt"] = region
+        url = f"{self._base_url}/news/search?{urlencode(params)}"
+        try:
+            resp = self.session.get(url)
+            resp.raise_for_status()
+        except Exception as e:
+            if hasattr(e, 'response') and e.response is not None:
+                raise Exception(f"Bing news search failed with status {e.response.status_code}: {str(e)}")
+            else:
+                raise Exception(f"Bing news search failed: {str(e)}")
+        soup = BeautifulSoup(resp.text, "html.parser")
+        results = []
+        for item in soup.select("div.news-card, div.card, div.newsitem, div.card-content, div.t_s_main"):
+            a_tag = item.find("a")
+            title = a_tag.get_text(strip=True) if a_tag else ''
+            url_val = a_tag['href'] if a_tag and a_tag.has_attr('href') else ''
+            desc_tag = item.find("div", class_="snippet") or item.find("div", class_="news-card-snippet") or item.find("div", class_="snippetText")
+            description = desc_tag.get_text(strip=True) if desc_tag else ''
+            source_tag = item.find("div", class_="source")
+            source = source_tag.get_text(strip=True) if source_tag else ''
+            if url_val and title:
+                results.append(BingNewsResult(title, url_val, description, source))
+                if len(results) >= max_results:
+                    break
+        # Fallback: try main news list if above selectors fail
+        if not results:
+            for item in soup.select("a.title"):
+                title = item.get_text(strip=True)
+                url_val = item['href'] if item.has_attr('href') else ''
+                description = ''
+                source = ''
+                if url_val and title:
+                    results.append(BingNewsResult(title, url_val, description, source))
+                    if len(results) >= max_results:
+                        break
+        return results[:max_results]
+
+if __name__ == "__main__":
+    from rich import print
+    bing = BingSearch(
+        timeout=10,
+        proxies=None,
+        verify=True
+    )
+    print("TEXT SEARCH RESULTS:")
+    text_results = bing.text(
+        keywords="Python programming",
+        region="us",
+        safesearch="moderate",
+        max_results=30
+    )
+    for result in text_results:
+        print(f"Title: {result.title}")
+        print(f"URL: {result.url}")
+        print(f"Description: {result.description}")
+        print("---")
+    print("\nSEARCH SUGGESTIONS:")
+    suggestions = bing.suggestions("how to")
+    print(suggestions)
+
+    print("\nIMAGE SEARCH RESULTS:")
+    image_results = bing.images(
+        keywords="Python programming",
+        region="us",
+        safesearch="moderate",
+        max_results=10
+    )
+    for result in image_results:
+        print(f"Title: {result.title}")
+        print(f"Image URL: {result.image}")
+        print(f"Page URL: {result.url}")
+        print(f"Source: {result.source}")
+        print("---")
+
+    print("\nNEWS SEARCH RESULTS:")
+    news_results = bing.news(
+        keywords="Python programming",
+        region="us",
+        safesearch="moderate",
+        max_results=10
+    )
+    for result in news_results:
+        print(f"Title: {result.title}")
+        print(f"URL: {result.url}")
+        print(f"Description: {result.description}")
+        print(f"Source: {result.source}")
+        print("---")