# main.py (Revised: Apify 201 fix + Supadata verify=False TEST) import os import re import logging import asyncio import json from flask import Flask, request, Response # For web server from telegram import Update, InlineKeyboardButton, InlineKeyboardMarkup from telegram.ext import ( Application, CommandHandler, MessageHandler, filters, ContextTypes, CallbackQueryHandler, ApplicationBuilder # Import ApplicationBuilder ) from telegram.constants import ParseMode # Import ParseMode explicitly # Import specific libraries (Ensure these are covered in requirements.txt) from youtube_transcript_api import YouTubeTranscriptApi import requests from bs4 import BeautifulSoup # Only import ApifyClient if you might use it (i.e., have the secret) _apify_token_exists = bool(os.environ.get('APIFY_API_TOKEN')) if _apify_token_exists: from apify_client import ApifyClient else: ApifyClient = None # --- Logging Setup --- logging.basicConfig( format='%(asctime)s - %(name)s - %(levelname)s - %(message)s', level=logging.DEBUG # Keep DEBUG ) logging.getLogger("httpx").setLevel(logging.WARNING) if ApifyClient: logging.getLogger("apify_client").setLevel(logging.WARNING) logging.getLogger("telegram.ext").setLevel(logging.DEBUG) logging.getLogger('telegram.bot').setLevel(logging.DEBUG) logging.getLogger("urllib3").setLevel(logging.INFO) # Reduce requests noise slightly logging.getLogger('gunicorn.error').setLevel(logging.INFO) logger = logging.getLogger(__name__) logger.info("Logging configured (DEBUG level).") # --- Environment Variable Loading --- logger.info("Attempting to load secrets from environment variables...") def get_secret(secret_name): logger.debug(f"Attempting to read secret: {secret_name}") value = os.environ.get(secret_name) if value: logger.info(f"Secret '{secret_name}': Found (Value length: {len(value)})") else: logger.warning(f"Secret '{secret_name}': Not Found") return value TELEGRAM_TOKEN = get_secret('TELEGRAM_TOKEN') OPENROUTER_API_KEY = get_secret('OPENROUTER_API_KEY') URLTOTEXT_API_KEY = get_secret('URLTOTEXT_API_KEY') SUPADATA_API_KEY = get_secret('SUPADATA_API_KEY') APIFY_API_TOKEN = get_secret('APIFY_API_TOKEN') logger.info("Secret loading attempt finished.") # --- Bot Logic Functions --- # Helper Functions def is_youtube_url(url): """Checks if the URL is a valid YouTube video or shorts URL.""" youtube_regex = r'(https?://)?(www\.)?(youtube\.com/(watch\?v=|shorts/)|youtu\.be/)([\w-]{11})' return bool(re.search(youtube_regex, url)) def extract_youtube_id(url): """Extracts the YouTube video ID from a URL.""" youtube_id_regex = r'(?:youtube\.com/(?:watch\?v=|shorts/)|youtu\.be/)([\w-]{11})' match = re.search(youtube_id_regex, url) if match: return match.group(1) logger.warning(f"Could not extract YouTube ID from URL: {url}") return None # Supadata Transcript Fetching async def get_transcript_via_supadata(video_id: str, api_key: str): """Fetches YouTube transcript via Supadata API.""" if not video_id: logger.error("[Supadata] get_transcript_via_supadata called with no video_id"); return None if not api_key: logger.error("[Supadata] API key is missing."); return None logger.info(f"[Supadata] Attempting fetch for video ID: {video_id}") api_endpoint = f"https://api.supadata.net/v1/youtube/transcript" params = {"videoId": video_id, "format": "text"} headers = {"X-API-Key": api_key} try: # ---!!! INSECURE TEST - DISABLES SSL VERIFICATION !!!--- logger.warning("[Supadata] Making request with verify=False (INSECURE TEST)") response = await asyncio.to_thread(requests.get, api_endpoint, headers=headers, params=params, timeout=30, verify=False) # ---!!! END INSECURE TEST --- (Remember to remove verify=False later) --- logger.debug(f"[Supadata] Received status code {response.status_code} for {video_id}") if response.status_code == 200: # (Rest of the success handling code remains the same) try: data = response.json() content = data if isinstance(data, str) else data.get("transcript") or data.get("text") or data.get("data") if content and isinstance(content, str): logger.info(f"[Supadata] Successfully fetched transcript for {video_id}. Length: {len(content)}") return content.strip() else: logger.warning(f"[Supadata] API success but content empty/invalid for {video_id}. Response: {data}") return None except json.JSONDecodeError: if response.text: logger.info(f"[Supadata] Successfully fetched transcript (plain text) for {video_id}. Length: {len(response.text)}") return response.text.strip() else: logger.error(f"[Supadata] Failed to decode JSON response (and no text body) for {video_id}. Response: {response.text[:200]}...") return None except Exception as e: logger.error(f"[Supadata] Error processing successful response for {video_id}: {e}", exc_info=True) return None elif response.status_code in [401, 403]: logger.error(f"[Supadata] Authentication error ({response.status_code}). Check API key.") return None elif response.status_code == 404: logger.warning(f"[Supadata] Transcript not found ({response.status_code}) for {video_id}.") return None else: logger.error(f"[Supadata] Unexpected status code {response.status_code} for {video_id}. Response: {response.text[:200]}...") return None except requests.exceptions.Timeout: logger.error(f"[Supadata] Timeout error connecting to API for {video_id}") return None except requests.exceptions.RequestException as e: logger.error(f"[Supadata] Request error connecting to API for {video_id}: {e}") # Log the specific SSLError if verify=False wasn't the only issue if isinstance(e, requests.exceptions.SSLError): logger.error(f"[Supadata] SSL Error details: {e}") return None except Exception as e: logger.error(f"[Supadata] Unexpected error during API call for {video_id}: {e}", exc_info=True) return None # Apify Transcript Fetching async def get_transcript_via_apify(video_url: str, api_token: str): """Fetches YouTube transcript via Apify API.""" if not video_url: logger.error("[Apify] get_transcript_via_apify called with no video_url"); return None if not api_token: logger.error("[Apify] API token is missing."); return None if not ApifyClient: logger.error("[Apify] ApifyClient not available/imported."); return None logger.info(f"[Apify] Attempting fetch for URL: {video_url}") actor_id = "karamelo~youtube-transcripts" api_endpoint = f"https://api.apify.com/v2/acts/{actor_id}/run-sync-get-dataset-items" params = {"token": api_token} payload = json.dumps({ "urls": [video_url], "outputFormat": "singleStringText", "maxRetries": 5, "channelHandleBoolean": False, "channelNameBoolean": False, "datePublishedBoolean": False, "relativeDateTextBoolean": False, }) headers = {"Content-Type": "application/json"} try: logger.debug(f"[Apify] Sending request to run actor {actor_id} synchronously for {video_url}") response = await asyncio.to_thread(requests.post, api_endpoint, headers=headers, params=params, data=payload, timeout=90) logger.debug(f"[Apify] Received status code {response.status_code} for {video_url}") # --- MODIFIED STATUS CODE CHECK --- if response.status_code in [200, 201]: # Accept 200 OK or 201 Created # --- END MODIFIED STATUS CODE CHECK --- try: results = response.json() if isinstance(results, list) and len(results) > 0: item = results[0] content = item.get("text") or item.get("transcript") or item.get("captions_concatenated") if not content and item.get("captions") and isinstance(item["captions"], list): logger.info("[Apify] Processing 'captions' format.") content = " ".join(cap.get("text", "") for cap in item["captions"] if cap.get("text")) if content and isinstance(content, str): logger.info(f"[Apify] Successfully fetched transcript for {video_url} (Status: {response.status_code}). Length: {len(content)}") return content.strip() else: logger.warning(f"[Apify] Actor run successful ({response.status_code}) but content not found/empty for {video_url}. Item: {item}") return None else: logger.warning(f"[Apify] Actor run successful ({response.status_code}) but dataset empty for {video_url}. Response: {results}") return None except json.JSONDecodeError: logger.error(f"[Apify] Failed to decode JSON response for {video_url}. Status: {response.status_code}. Resp: {response.text[:200]}...") return None except Exception as e: logger.error(f"[Apify] Error processing successful response ({response.status_code}) for {video_url}: {e}", exc_info=True) return None elif response.status_code == 400: logger.error(f"[Apify] Bad Request (400) for {video_url}. Check payload. Response: {response.text[:200]}...") return None elif response.status_code == 401: logger.error("[Apify] Authentication error (401). Check API token.") return None else: # Catch other non-200/201 codes here logger.error(f"[Apify] Unexpected status code {response.status_code} for {video_url}. Response: {response.text[:200]}...") return None except requests.exceptions.Timeout: logger.error(f"[Apify] Timeout error running actor for {video_url}") return None except requests.exceptions.RequestException as e: logger.error(f"[Apify] Request error running actor for {video_url}: {e}") return None except Exception as e: logger.error(f"[Apify] Unexpected error during Apify call for {video_url}: {e}", exc_info=True) return None # (The rest of the functions: get_youtube_transcript, get_website_content_via_requests, # get_website_content_via_urltotext_api, generate_summary, start, help_command, # handle_potential_url, handle_summary_type_callback, error_handler, setup_bot, # webhook, index, and the main execution block remain EXACTLY THE SAME as in the # previous complete code block. Ensure they are included below this point.) # Combined YouTube Transcript Function (with Fallbacks) async def get_youtube_transcript(video_id: str, video_url: str, supadata_key: str | None, apify_token: str | None): """Fetches YouTube transcript using library, then Supadata, then Apify.""" if not video_id: logger.error("get_youtube_transcript called with no video_id"); return None logger.info(f"Fetching transcript for video ID: {video_id} (URL: {video_url})") transcript_text = None # 1. Primary Method: youtube-transcript-api logger.info("[Primary YT] Attempting youtube-transcript-api...") try: transcript_list = await asyncio.to_thread( YouTubeTranscriptApi.get_transcript, video_id, languages=['en', 'en-GB', 'en-US'] ) if transcript_list: transcript_text = " ".join([item['text'] for item in transcript_list if 'text' in item]) if transcript_text: logger.info(f"[Primary YT] Successfully fetched transcript via library for {video_id} (length: {len(transcript_text)})") return transcript_text else: logger.warning(f"[Primary YT] Joined transcript text is empty for {video_id}") transcript_text = None else: logger.warning(f"[Primary YT] Transcript list empty for {video_id}") transcript_text = None except Exception as e: logger.warning(f"[Primary YT] Error getting transcript via library for {video_id}: {e}") if "YouTube is blocking requests" in str(e): logger.warning("[Primary YT] IP likely blocked by YouTube.") elif "No transcript found" in str(e): logger.warning(f"[Primary YT] No transcript found for {video_id}.") elif "disabled" in str(e): logger.warning(f"[Primary YT] Transcripts disabled for {video_id}.") transcript_text = None # 2. Fallback 1: Supadata API if transcript_text is None: logger.info("[Fallback YT 1] Primary method failed. Trying Supadata API...") if supadata_key: transcript_text = await get_transcript_via_supadata(video_id, supadata_key) if transcript_text: logger.info(f"[Fallback YT 1] Successfully fetched transcript via Supadata for {video_id}") return transcript_text else: logger.warning(f"[Fallback YT 1] Supadata API failed or returned no content for {video_id}.") else: logger.warning("[Fallback YT 1] Supadata API key not available. Skipping.") # 3. Fallback 2: Apify API if transcript_text is None: logger.info("[Fallback YT 2] Primary & Supadata failed. Trying Apify API...") if apify_token: transcript_text = await get_transcript_via_apify(video_url, apify_token) if transcript_text: logger.info(f"[Fallback YT 2] Successfully fetched transcript via Apify for {video_url}") return transcript_text else: logger.warning(f"[Fallback YT 2] Apify API failed or returned no content for {video_url}.") else: logger.warning("[Fallback YT 2] Apify API token not available. Skipping.") # If all methods failed if transcript_text is None: logger.error(f"All methods failed to fetch transcript for video ID: {video_id}") return None return transcript_text # Website Content via Requests/BS4 (Primary Method for Simplified Bot) async def get_website_content_via_requests(url): """Attempts to scrape website content using requests/BeautifulSoup (Primary Method).""" if not url: logger.error("[Web Scraper - Requests/BS4] called with no URL"); return None logger.info(f"[Web Scraper - Requests/BS4] Fetching website content for: {url}") try: headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/108.0.0.0 Safari/537.36', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8', 'Accept-Language': 'en-US,en;q=0.9', 'Connection': 'keep-alive', 'DNT': '1', 'Upgrade-Insecure-Requests': '1' } logger.debug(f"[Web Scraper - Requests/BS4] Sending request to {url}") response = await asyncio.to_thread(requests.get, url, headers=headers, timeout=25, allow_redirects=True) response.raise_for_status() logger.debug(f"[Web Scraper - Requests/BS4] Received response {response.status_code} from {url}") content_type = response.headers.get('content-type', '').lower() if 'html' not in content_type: logger.warning(f"[Web Scraper - Requests/BS4] Non-HTML content type received from {url}: {content_type}") return None soup = BeautifulSoup(response.text, 'html.parser') for element in soup(["script", "style", "header", "footer", "nav", "aside", "form", "button", "input", "textarea", "select", "option", "label", "iframe", "img", "svg", "link", "meta", "noscript", "figure", "figcaption", "video", "audio"]): element.extract() main_content = soup.find('main') or soup.find('article') or soup.find(id='content') or soup.find(class_='content') or soup.find(id='main-content') or soup.find(class_='main-content') or soup.find(role='main') target_element = main_content if main_content else soup.body if not target_element: logger.warning(f"[Web Scraper - Requests/BS4] Could not find body or main content container for parsing {url}") return None lines = [line.strip() for line in target_element.get_text(separator='\n', strip=True).splitlines() if line.strip()] text = "\n".join(lines) if not text or len(text) < 50: logger.warning(f"[Web Scraper - Requests/BS4] Extracted text is very short or empty after cleaning for {url} (Length: {len(text)})") logger.info(f"[Web Scraper - Requests/BS4] Successfully scraped content for {url} (final length: {len(text)})") return text except requests.exceptions.Timeout: logger.error(f"[Web Scraper - Requests/BS4] Timeout error scraping website: {url}"); return None except requests.exceptions.TooManyRedirects: logger.error(f"[Web Scraper - Requests/BS4] Too many redirects error scraping website: {url}"); return None except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - Requests/BS4] Request error scraping website {url}: {e}"); return None except Exception as e: logger.error(f"[Web Scraper - Requests/BS4] Error scraping or parsing website {url}: {e}", exc_info=True); return None # Website Content via URLToText API (Fallback Method) async def get_website_content_via_urltotext_api(url: str, api_key: str): """Fetches website content using the URLToText API (Fallback).""" if not url: logger.error("[Web Scraper - URLToText API] called with no URL"); return None if not api_key: logger.error("[Web Scraper - URLToText API] API key is missing."); return None logger.info(f"[Web Scraper - URLToText API] Attempting to fetch content for: {url}") api_endpoint = "https://urltotext.com/api/v1/urltotext/" payload = json.dumps({ "url": url, "output_format": "text", "extract_main_content": True, "render_javascript": True, "residential_proxy": False }) headers = { "Authorization": f"Token {api_key}", "Content-Type": "application/json" } try: logger.debug(f"[Web Scraper - URLToText API] Sending request for {url}") response = await asyncio.to_thread(requests.post, api_endpoint, headers=headers, data=payload, timeout=45) logger.debug(f"[Web Scraper - URLToText API] Received status code {response.status_code} for {url}") if response.status_code == 200: try: data = response.json() content = data.get("data", {}).get("content") credits = data.get("credits_used", "N/A") warning = data.get("data", {}).get("warning") if warning: logger.warning(f"[Web Scraper - URLToText API] Warning for {url}: {warning}") if content: logger.info(f"[Web Scraper - URLToText API] Successfully fetched content via API for {url}. Length: {len(content)}. Credits: {credits}"); return content.strip() else: logger.warning(f"[Web Scraper - URLToText API] API returned success but content was empty for {url}. Response: {data}"); return None except json.JSONDecodeError: logger.error(f"[Web Scraper - URLToText API] Failed to decode JSON response for {url}. Response: {response.text[:500]}..."); return None except Exception as e: logger.error(f"[Web Scraper - URLToText API] Error processing successful API response for {url}: {e}", exc_info=True); return None elif response.status_code in [400, 402, 422, 500]: logger.error(f"[Web Scraper - URLToText API] Error {response.status_code} from API for {url}. Response: {response.text[:200]}..."); return None else: logger.error(f"[Web Scraper - URLToText API] Unexpected status {response.status_code} from API for {url}. Response: {response.text[:200]}..."); return None except requests.exceptions.Timeout: logger.error(f"[Web Scraper - URLToText API] Timeout error connecting to API for {url}"); return None except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - URLToText API] Request error connecting to API for {url}: {e}"); return None except Exception as e: logger.error(f"[Web Scraper - URLToText API] Unexpected error during API call for {url}: {e}", exc_info=True); return None # DeepSeek Summary Function (via OpenRouter) async def generate_summary(text: str, summary_type: str, api_key: str) -> str: """Generates summary using DeepSeek via OpenRouter API.""" logger.info(f"Generating {summary_type} summary using DeepSeek/OpenRouter. Input text length: {len(text)}") if not api_key: logger.error("OpenRouter API key was not provided."); return "Error: AI model config key missing." openrouter_api_endpoint = "https://openrouter.ai/api/v1/chat/completions" model_name = "deepseek/deepseek-chat:free" if summary_type == "paragraph": prompt = "..." # Keep prompt as before else: prompt = """...""" # Keep prompt as before MAX_INPUT_LENGTH = 500000 if len(text) > MAX_INPUT_LENGTH: logger.warning(f"Input text ({len(text)}) > limit ({MAX_INPUT_LENGTH}). Truncating."); text = text[:MAX_INPUT_LENGTH] + "... (Truncated)" full_prompt = f"{prompt}\n\n{text}" headers = { "Authorization": f"Bearer {api_key}", "Content-Type": "application/json", "HTTP-Referer": "https://huggingface.co/spaces/", "X-Title": "Telegram Summary Bot (HF Space)"} payload = json.dumps({ "model": model_name, "messages": [{"role": "user", "content": full_prompt}]}) try: logger.debug(f"Sending request to OpenRouter ({model_name})...") response = await asyncio.to_thread(requests.post, openrouter_api_endpoint, headers=headers, data=payload, timeout=60) logger.debug(f"Received status code {response.status_code} from OpenRouter.") if response.status_code == 200: try: data = response.json() if data.get("choices") and data["choices"]: message = data["choices"][0].get("message") if message and message.get("content"): summary = message["content"].strip() logger.info(f"Success generating summary. Len: {len(summary)}") return summary else: logger.warning(f"OpenRouter success but empty content. Resp: {data}"); return "Sorry, AI model returned empty summary." else: if data.get("error"): logger.error(f"OpenRouter API Error: {data['error']}") else: logger.error(f"Unexpected choices structure: {data.get('choices')}. Resp: {data}") return "Sorry, could not parse AI response (choices/error)." except json.JSONDecodeError: logger.error(f"Failed JSON decode from OpenRouter. Status: {response.status_code}. Resp: {response.text[:500]}..."); return "Sorry, failed to understand AI response." except Exception as e: logger.error(f"Error processing OpenRouter success resp: {e}", exc_info=True); return "Sorry, error processing AI response." elif response.status_code == 401: logger.error("OpenRouter API key invalid (401). Check HF Secrets."); return "Error: AI model config key invalid." elif response.status_code == 402: logger.error("OpenRouter Payment Required (402). Check OpenRouter account."); return "Sorry, issue with AI service limits/payment." elif response.status_code == 429: logger.warning("OpenRouter Rate Limit (429)."); return "Sorry, AI model busy. Try again." elif response.status_code == 500: logger.error(f"OpenRouter Internal Error (500). Resp: {response.text[:500]}..."); return "Sorry, AI model service error. Try again later." else: logger.error(f"Unexpected status {response.status_code} from OpenRouter. Resp: {response.text[:500]}...") try: error_data = response.json(); error_msg = error_data.get("error", {}).get("message", response.text[:100]); return f"Sorry, AI service error ({response.status_code}): {error_msg}" except: return f"Sorry, AI service returned status {response.status_code}." except requests.exceptions.Timeout: logger.error("Timeout connecting to OpenRouter."); return "Sorry, request to AI model timed out." except requests.exceptions.RequestException as e: logger.error(f"Request error connecting to OpenRouter: {e}"); return "Sorry, error connecting to AI model service." except Exception as e: logger.error(f"Unexpected error in generate_summary: {e}", exc_info=True); return "Sorry, unexpected error generating summary." # --- Telegram Bot Handlers --- async def start(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None: user = update.effective_user; logger.info(f"User {user.id} ({user.username or 'NoUsername'}) used /start.") mention = user.mention_html() if user.username else user.first_name await update.message.reply_html(f"👋 Hello {mention}! I can summarize YouTube links or website URLs.\n\nJust send me a link anytime!") async def help_command(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None: logger.info(f"User {update.effective_user.id} used /help.") await update.message.reply_text("...", parse_mode=ParseMode.MARKDOWN) # Keep help text async def handle_potential_url(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None: if not update.message or not update.message.text: return url = update.message.text.strip(); user = update.effective_user logger.info(f"User {user.id} ({user.username or 'NoUsername'}) sent potential URL: {url}") if not (url.startswith('http://') or url.startswith('https://')) or '.' not in url[8:]: logger.debug(f"Ignoring non-URL: {url}"); return context.user_data['url_to_summarize'] = url; logger.debug(f"Stored URL '{url}' for user {user.id}") keyboard = [[InlineKeyboardButton("Paragraph", callback_data="paragraph"), InlineKeyboardButton("Points", callback_data="points")]] reply_markup = InlineKeyboardMarkup(keyboard) await update.message.reply_text(f"Link detected:\n{url}\n\nChoose summary type:", reply_markup=reply_markup, disable_web_page_preview=True) async def handle_summary_type_callback(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None: query = update.callback_query; await query.answer() summary_type = query.data; user = update.effective_user or query.from_user url = context.user_data.get('url_to_summarize') logger.info(f"User {user.id} chose '{summary_type}' for URL '{url}'.") if not url: logger.warning(f"User {user.id} pressed button, NO URL in context."); await query.edit_message_text(text="Context lost. Send link again."); return context.user_data.pop('url_to_summarize', None); logger.debug(f"Cleared URL {url} for user {user.id}") current_openrouter_key = os.environ.get('OPENROUTER_API_KEY'); current_urltotext_key = os.environ.get('URLTOTEXT_API_KEY') current_supadata_key = os.environ.get('SUPADATA_API_KEY'); current_apify_token = os.environ.get('APIFY_API_TOKEN') if not current_openrouter_key: logger.error("OpenRouter key missing."); await context.bot.send_message(chat_id=user.id, text="Error: AI config missing."); await query.delete_message(); return processing_message = f"Working on '{summary_type}' summary for:\n{url}\n..."; message_to_delete_later = None try: await query.edit_message_text(processing_message); logger.debug(f"Edited message query {query.id}") except Exception as e: logger.warning(f"Could not edit message {query.id}: {e}. Sending new."); message_to_delete_later = await context.bot.send_message(chat_id=user.id, text=processing_message) content = None; user_feedback_message = None; success = False; is_youtube = is_youtube_url(url) try: logger.debug(f"Sending 'typing' action for chat {user.id}"); await context.bot.send_chat_action(chat_id=user.id, action='typing') if is_youtube: video_id = extract_youtube_id(url) if video_id: logger.info(f"Fetching YT transcript: {video_id}"); content = await get_youtube_transcript(video_id, url, current_supadata_key, current_apify_token) user_feedback_message = None if content else "Sorry, couldn't get YT transcript." logger.info(f"YT transcript fetch done. Found: {bool(content)}") else: user_feedback_message = "Sorry, couldn't parse YT URL." else: logger.info(f"Scraping website (Requests/BS4): {url}"); content = await get_website_content_via_requests(url) if content: logger.info("Website scrape (Requests/BS4) OK."); user_feedback_message = None else: logger.warning(f"Website scrape failed for {url}. Trying URLToText API."); if current_urltotext_key: await context.bot.send_chat_action(chat_id=user.id, action='typing'); content = await get_website_content_via_urltotext_api(url, current_urltotext_key) if content: logger.info("URLToText API scrape OK."); user_feedback_message = None else: user_feedback_message = "Sorry, couldn't fetch web content (both methods)." else: user_feedback_message = "Sorry, couldn't fetch web content (fallback not configured)." if content: logger.info("Content found, generating summary."); await context.bot.send_chat_action(chat_id=user.id, action='typing') summary = await generate_summary(content, summary_type, current_openrouter_key) if summary.startswith("Error:") or summary.startswith("Sorry,"): user_feedback_message = summary; logger.warning(f"Summary generation failed: {summary}") else: logger.info("Summary generated OK. Sending."); await context.bot.send_message(chat_id=user.id, text=summary, parse_mode=ParseMode.MARKDOWN, disable_web_page_preview=True); success = True; user_feedback_message = None elif not user_feedback_message: user_feedback_message = "Sorry, couldn't retrieve content." if user_feedback_message and not success: logger.warning(f"Sending failure feedback: {user_feedback_message}"); await context.bot.send_message(chat_id=user.id, text=user_feedback_message) except Exception as e: logger.error(f"Unexpected error in callback processing: {e}", exc_info=True); await context.bot.send_message(chat_id=user.id, text="Oops! Internal error.") finally: logger.debug("Cleaning up status message..."); try: if message_to_delete_later: await context.bot.delete_message(chat_id=user.id, message_id=message_to_delete_later.message_id); logger.debug("Deleted separate status msg.") elif query: await query.delete_message(); logger.debug(f"Deleted original message query {query.id}.") except Exception as del_e: logger.warning(f"Could not delete status/button message: {del_e}") async def error_handler(update: object, context: ContextTypes.DEFAULT_TYPE) -> None: logger.error(f"Exception while handling update: {context.error}", exc_info=context.error) # --- Bot Application Setup Function --- async def setup_bot(): logger.info("Setting up Telegram Application..."); if not TELEGRAM_TOKEN: logger.critical("Cannot initialize: TELEGRAM_TOKEN missing."); return None application = Application.builder().token(TELEGRAM_TOKEN).build() logger.info("Running application.initialize()..."); await application.initialize(); logger.info("Finished application.initialize().") application.add_handler(CommandHandler("start", start)); application.add_handler(CommandHandler("help", help_command)) application.add_handler(MessageHandler(filters.TEXT & ~filters.COMMAND, handle_potential_url)); application.add_handler(CallbackQueryHandler(handle_summary_type_callback)) application.add_error_handler(error_handler); logger.info("Telegram handlers registered.") return application logger.info("Running bot setup..."); ptb_app = asyncio.run(setup_bot()); logger.info(f"Bot setup finished. App instance: {'OK' if ptb_app else 'Failed'}") # --- Flask App Setup --- app = Flask(__name__); logger.info("Flask app created.") # --- Webhook Endpoint --- @app.route('/webhook', methods=['POST']) async def webhook() -> Response: logger.info("Webhook request received...") if not ptb_app: logger.error("PTB App not initialized."); return Response('Bot not configured', status=500) if request.is_json: try: update = Update.de_json(request.get_json(), ptb_app.bot); logger.debug(f"Processing update ID: {update.update_id}") logger.debug("Directly awaiting process_update..."); await ptb_app.process_update(update); logger.debug("Finished awaiting process_update.") return Response('ok', status=200) except json.JSONDecodeError: logger.error("Failed JSON decode from Telegram."); return Response('Bad Request', status=400) except Exception as e: logger.error(f"Error processing update in webhook: {e}", exc_info=True); return Response('Internal Server Error', status=500) else: logger.warning("Received non-JSON request to webhook."); return Response('Bad Request', status=400) @app.route('/') def index(): logger.debug("Health check '/' accessed."); bot_status = "Initialized" if ptb_app else "FAILED Init"; return f"TG Bot Webhook Listener ({bot_status}) running." # --- Main Execution Block --- if __name__ == '__main__': if not ptb_app: logger.critical("Aborting local Flask start: PTB App failed init.") else: logger.info("Starting Flask server directly (local testing?)..."); port = int(os.environ.get('PORT', 5000)); app.run(host='0.0.0.0', port=port, debug=True)