Spaces:
Running
Running
Update main.py
Browse files
main.py
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
# main.py (Revised: Starlette Lifespan
|
2 |
import os
|
3 |
import re
|
4 |
import logging
|
@@ -24,7 +24,7 @@ from telegram.ext import (
|
|
24 |
CallbackQueryHandler,
|
25 |
)
|
26 |
from telegram.constants import ParseMode
|
27 |
-
from telegram.error import NetworkError #
|
28 |
|
29 |
# --- Other Libraries ---
|
30 |
from youtube_transcript_api import YouTubeTranscriptApi
|
@@ -47,8 +47,8 @@ logging.getLogger("telegram.ext").setLevel(logging.INFO)
|
|
47 |
logging.getLogger('telegram.bot').setLevel(logging.INFO)
|
48 |
logging.getLogger("urllib3").setLevel(logging.INFO)
|
49 |
logging.getLogger('gunicorn.error').setLevel(logging.INFO)
|
50 |
-
logging.getLogger('uvicorn').setLevel(logging.INFO)
|
51 |
-
logging.getLogger('starlette').setLevel(logging.INFO)
|
52 |
logger = logging.getLogger(__name__)
|
53 |
logger.info("Logging configured.")
|
54 |
|
@@ -180,7 +180,6 @@ async def get_transcript_via_apify(video_url: str, api_token: str):
|
|
180 |
item = results[0]
|
181 |
content = item.get("text") or item.get("transcript") or item.get("captions_concatenated")
|
182 |
|
183 |
-
# --- !!! FIXED FALLBACK LOGIC !!! ---
|
184 |
if not content and item.get("captions"):
|
185 |
captions_data = item["captions"]
|
186 |
if isinstance(captions_data, str):
|
@@ -202,179 +201,135 @@ async def get_transcript_via_apify(video_url: str, api_token: str):
|
|
202 |
content = html.unescape(content) # Use imported html module
|
203 |
except Exception as unescape_err:
|
204 |
logger.warning(f"[Apify] Error during html unescaping: {unescape_err}")
|
205 |
-
# --- !!! END FIXED FALLBACK LOGIC !!! ---
|
206 |
|
207 |
if content and isinstance(content, str):
|
208 |
logger.info(f"[Apify] Successfully fetched transcript via run-sync for {video_url} (Status: {response.status_code}). Length: {len(content)}")
|
209 |
-
return content
|
210 |
else:
|
211 |
-
|
212 |
-
|
213 |
-
|
214 |
-
elif not item.get("captions"):
|
215 |
-
logger.warning(f"[Apify] Actor run successful ({response.status_code}) but no text/transcript/captions_concatenated/captions field found for {video_url}. Item: {item}")
|
216 |
-
else:
|
217 |
-
logger.warning(f"[Apify] Actor run successful ({response.status_code}), 'captions' field found but fallback parsing failed to extract content for {video_url}.")
|
218 |
return None
|
219 |
else:
|
220 |
-
logger.warning(f"[Apify] Actor
|
221 |
return None
|
222 |
except json.JSONDecodeError:
|
223 |
-
logger.error(f"[Apify] Failed
|
224 |
return None
|
225 |
except Exception as e:
|
226 |
logger.error(f"[Apify] Error processing successful response ({response.status_code}) for {video_url}: {e}", exc_info=True)
|
227 |
return None
|
228 |
-
elif response.status_code == 400:
|
229 |
-
|
230 |
-
|
231 |
-
elif response.status_code == 401:
|
232 |
-
logger.error("[Apify] Authentication error (401). Check API token.")
|
233 |
-
return None
|
234 |
-
else:
|
235 |
-
logger.error(f"[Apify] Unexpected status code {response.status_code} from Apify API for {video_url}. Response: {response.text[:200]}...")
|
236 |
-
return None
|
237 |
|
238 |
-
except requests.exceptions.Timeout:
|
239 |
-
|
240 |
-
|
241 |
-
except requests.exceptions.RequestException as e:
|
242 |
-
logger.error(f"[Apify] Request error running actor for {video_url}: {e}")
|
243 |
-
return None
|
244 |
-
except Exception as e:
|
245 |
-
logger.error(f"[Apify] Unexpected error during Apify call for {video_url}: {e}", exc_info=True)
|
246 |
-
return None
|
247 |
|
248 |
-
# Combined YouTube Transcript Function
|
249 |
async def get_youtube_transcript(video_id: str, video_url: str, supadata_key: str | None, apify_token: str | None):
|
250 |
"""Fetches YouTube transcript using library, then Supadata, then Apify."""
|
251 |
if not video_id: logger.error("get_youtube_transcript called with no video_id"); return None
|
252 |
logger.info(f"Fetching transcript for video ID: {video_id} (URL: {video_url})")
|
253 |
transcript_text = None
|
254 |
-
try: #
|
255 |
logger.info("[Primary YT] Attempting youtube-transcript-api...")
|
256 |
-
transcript_list = await asyncio.to_thread(
|
257 |
-
YouTubeTranscriptApi.get_transcript,
|
258 |
-
video_id,
|
259 |
-
languages=['en', 'en-GB', 'en-US']
|
260 |
-
)
|
261 |
if transcript_list:
|
262 |
transcript_text = " ".join([item['text'] for item in transcript_list if 'text' in item])
|
263 |
transcript_text = re.sub(r'\s+', ' ', transcript_text).strip()
|
264 |
-
if transcript_text:
|
265 |
-
|
266 |
-
|
267 |
-
else:
|
268 |
-
logger.warning(f"[Primary YT] Joined transcript text is empty after cleaning for {video_id}")
|
269 |
-
transcript_text = None
|
270 |
-
else:
|
271 |
-
logger.warning(f"[Primary YT] Transcript list was empty for {video_id}")
|
272 |
-
transcript_text = None
|
273 |
except Exception as e:
|
274 |
-
logger.warning(f"[Primary YT] Error
|
275 |
-
if "YouTube is blocking requests" in str(e) or "HTTP Error 429" in str(e): logger.warning("[Primary YT] IP likely blocked
|
276 |
-
elif "No transcript found" in str(e): logger.warning(
|
277 |
-
elif "TranscriptsDisabled" in str(e) or "disabled" in str(e): logger.warning(
|
278 |
transcript_text = None
|
279 |
|
280 |
if transcript_text is None: # Fallback 1: Supadata
|
281 |
-
logger.info("[Fallback YT 1]
|
282 |
if supadata_key:
|
283 |
transcript_text = await get_transcript_via_supadata(video_id, supadata_key)
|
284 |
-
if transcript_text:
|
285 |
-
|
286 |
-
|
287 |
-
else: logger.warning(f"[Fallback YT 1] Supadata API failed or returned no content for {video_id}.")
|
288 |
-
else: logger.warning("[Fallback YT 1] Supadata API key not available. Skipping.")
|
289 |
|
290 |
if transcript_text is None: # Fallback 2: Apify
|
291 |
-
logger.info("[Fallback YT 2]
|
292 |
if apify_token:
|
293 |
transcript_text = await get_transcript_via_apify(video_url, apify_token)
|
294 |
-
if transcript_text:
|
295 |
-
|
296 |
-
|
297 |
-
|
298 |
-
|
299 |
-
|
300 |
-
if transcript_text is None:
|
301 |
-
logger.error(f"All methods failed to fetch transcript for video ID: {video_id}")
|
302 |
-
return None
|
303 |
-
return transcript_text # Should technically be unreachable if logic is correct
|
304 |
|
305 |
# Website Content via Requests/BS4
|
306 |
async def get_website_content_via_requests(url):
|
307 |
"""Attempts to scrape website content using requests/BeautifulSoup."""
|
308 |
-
if not url: logger.error("[Web Scraper - Requests/BS4]
|
309 |
-
logger.info(f"[Web Scraper - Requests/BS4] Fetching
|
310 |
try:
|
311 |
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'}
|
312 |
-
logger.debug(f"[Web Scraper - Requests/BS4] Sending GET request to {url}")
|
313 |
response = await asyncio.to_thread(requests.get, url, headers=headers, timeout=25, allow_redirects=True)
|
314 |
response.raise_for_status()
|
315 |
-
logger.debug(f"[Web Scraper - Requests/BS4]
|
316 |
content_type = response.headers.get('content-type', '').lower()
|
317 |
if 'html' not in content_type:
|
318 |
-
logger.warning(f"[Web Scraper - Requests/BS4] Non-HTML
|
319 |
if 'text/plain' in content_type and response.text: return response.text.strip()
|
320 |
return None
|
321 |
soup = BeautifulSoup(response.text, 'html.parser')
|
322 |
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()
|
323 |
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')
|
324 |
target_element = main_content if main_content else soup.body
|
325 |
-
if not target_element:
|
326 |
-
logger.warning(f"[Web Scraper - Requests/BS4] Could not find body or main content for {url}")
|
327 |
-
return None
|
328 |
lines = [line.strip() for line in target_element.get_text(separator='\n', strip=True).splitlines() if line.strip()]
|
329 |
text = "\n".join(lines)
|
330 |
MIN_TEXT_LENGTH = 50
|
331 |
-
if not text or len(text) < MIN_TEXT_LENGTH:
|
332 |
-
|
333 |
-
logger.info(f"[Web Scraper - Requests/BS4] Successfully scraped content for {url} (final length: {len(text)})")
|
334 |
return text
|
335 |
-
except requests.exceptions.Timeout: logger.error(f"[Web Scraper - Requests/BS4] Timeout
|
336 |
-
except requests.exceptions.TooManyRedirects: logger.error(f"[Web Scraper - Requests/BS4]
|
337 |
-
except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - Requests/BS4] Request error
|
338 |
-
except Exception as e: logger.error(f"[Web Scraper - Requests/BS4]
|
339 |
|
340 |
# Website Content via URLToText API
|
341 |
async def get_website_content_via_urltotext_api(url: str, api_key: str):
|
342 |
"""Fetches website content using the URLToText API."""
|
343 |
-
if not url: logger.error("[Web Scraper - URLToText API]
|
344 |
-
if not api_key: logger.error("[Web Scraper - URLToText API] API key
|
345 |
-
logger.info(f"[Web Scraper - URLToText API] Attempting fetch
|
346 |
api_endpoint = "https://urltotext.com/api/v1/urltotext/"
|
347 |
payload = json.dumps({"url": url, "output_format": "text", "extract_main_content": True, "render_javascript": True, "residential_proxy": False})
|
348 |
headers = {"Authorization": f"Token {api_key}", "Content-Type": "application/json"}
|
349 |
try:
|
350 |
-
logger.debug(f"[Web Scraper - URLToText API] Sending POST request for {url}")
|
351 |
response = await asyncio.to_thread(requests.post, api_endpoint, headers=headers, data=payload, timeout=45)
|
352 |
-
logger.debug(f"[Web Scraper - URLToText API]
|
353 |
if response.status_code == 200:
|
354 |
try:
|
355 |
data = response.json()
|
356 |
-
content = data.get("data", {}).get("content")
|
357 |
-
credits = data.get("credits_used", "N/A")
|
358 |
-
warning = data.get("data", {}).get("warning")
|
359 |
if warning: logger.warning(f"[Web Scraper - URLToText API] Warning for {url}: {warning}")
|
360 |
-
if content:
|
361 |
-
|
362 |
-
|
363 |
-
|
364 |
-
|
365 |
-
|
366 |
-
|
367 |
-
|
368 |
-
elif response.status_code
|
369 |
-
|
370 |
-
elif response.status_code == 402: logger.error(f"[Web Scraper - URLToText API] Payment Required (402) for {url}. Check credits. Resp: {response.text[:200]}...")
|
371 |
-
elif response.status_code == 422: logger.warning(f"[Web Scraper - URLToText API] Unprocessable URL (422) for {url}. Resp: {response.text[:200]}...")
|
372 |
-
elif response.status_code >= 500: logger.error(f"[Web Scraper - URLToText API] Server Error ({response.status_code}) from API for {url}. Resp: {response.text[:200]}...")
|
373 |
-
else: logger.error(f"[Web Scraper - URLToText API] Unexpected status {response.status_code} from API for {url}. Resp: {response.text[:200]}...")
|
374 |
return None
|
375 |
-
except requests.exceptions.Timeout: logger.error(f"[Web Scraper - URLToText API] Timeout
|
376 |
-
except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - URLToText API] Request error
|
377 |
-
except Exception as e: logger.error(f"[Web Scraper - URLToText API] Unexpected error
|
378 |
|
379 |
# DeepSeek Summary Function
|
380 |
async def generate_summary(text: str, summary_type: str, api_key: str) -> str:
|
@@ -382,23 +337,21 @@ async def generate_summary(text: str, summary_type: str, api_key: str) -> str:
|
|
382 |
logger.info(f"Generating '{summary_type}' summary. Input length: {len(text)}")
|
383 |
if not api_key: logger.error("OpenRouter API key missing."); return "Error: AI config key missing."
|
384 |
if not text: logger.warning("generate_summary called with empty text."); return "Error: No content to summarize."
|
385 |
-
openrouter_api_endpoint = "https://openrouter.ai/api/v1/chat/completions"
|
386 |
-
model_name = "deepseek/deepseek-chat:free"
|
387 |
if summary_type == "paragraph": prompt = "Please provide a concise, well-written paragraph summarizing the key information and main points of the following text. Focus on capturing the essence of the content accurately."
|
388 |
elif summary_type == "points": prompt = "Please summarize the following text into clear, distinct bullet points. Each point should highlight a key piece of information, finding, or main topic discussed. Aim for clarity and conciseness."
|
389 |
else: logger.error(f"Invalid summary_type '{summary_type}'."); return f"Error: Invalid summary type ('{summary_type}')."
|
390 |
MAX_INPUT_LENGTH = 500000
|
391 |
if len(text) > MAX_INPUT_LENGTH: logger.warning(f"Input text ({len(text)}) > limit ({MAX_INPUT_LENGTH}). Truncating."); text = text[:MAX_INPUT_LENGTH] + "... (Truncated)"
|
392 |
full_prompt = f"{prompt}\n\n--- Start of Text ---\n\n{text}\n\n--- End of Text ---"
|
393 |
-
|
394 |
-
space_host = os.environ.get("SPACE_HOST", "huggingface.co/spaces/YOUR_SPACE_NAME") # Default if env var not set
|
395 |
referer_url = f"https://{space_host}" if not space_host.startswith("http") else space_host
|
396 |
headers = {"Authorization": f"Bearer {api_key}","Content-Type": "application/json","HTTP-Referer": referer_url,"X-Title": "Telegram URL Summarizer Bot"}
|
397 |
payload = json.dumps({"model": model_name, "messages": [{"role": "user", "content": full_prompt}]})
|
398 |
try:
|
399 |
logger.debug(f"Sending request to OpenRouter (Model: {model_name})...")
|
400 |
response = await asyncio.to_thread(requests.post, openrouter_api_endpoint, headers=headers, data=payload, timeout=90)
|
401 |
-
logger.debug(f"Received status
|
402 |
if response.status_code == 200:
|
403 |
try:
|
404 |
data = response.json()
|
@@ -413,7 +366,7 @@ async def generate_summary(text: str, summary_type: str, api_key: str) -> str:
|
|
413 |
else: logger.error(f"Unexpected OpenRouter choices structure. Resp: {data}"); return "Sorry, could not parse AI response (choices)."
|
414 |
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 format."
|
415 |
except Exception as e: logger.error(f"Error processing OpenRouter success resp: {e}", exc_info=True); return "Sorry, error processing AI response."
|
416 |
-
elif response.status_code == 401: logger.error("OpenRouter
|
417 |
elif response.status_code == 402: logger.error("OpenRouter Payment Required (402)."); return "Sorry, issue with AI service limits/payment."
|
418 |
elif response.status_code == 429: logger.warning("OpenRouter Rate Limit (429)."); return "Sorry, AI model busy. Try again."
|
419 |
elif response.status_code >= 500: logger.error(f"OpenRouter Internal Error ({response.status_code}). Resp: {response.text[:500]}..."); return "Sorry, AI model service error. Try again later."
|
@@ -429,43 +382,37 @@ async def generate_summary(text: str, summary_type: str, api_key: str) -> str:
|
|
429 |
# --- Telegram Bot Handlers ---
|
430 |
|
431 |
async def start(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
432 |
-
|
433 |
-
user = update.effective_user
|
434 |
if not user: return
|
435 |
-
logger.info(f"User {user.id} ({user.username or 'NoUsername'})
|
436 |
mention = user.mention_html() if user.username else user.first_name
|
437 |
-
await update.message.reply_html(f"👋 Hello {mention}! Send a
|
438 |
|
439 |
async def help_command(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
440 |
-
"
|
441 |
-
|
442 |
-
|
443 |
-
help_text = ( "**How I Work:**\n\n" # ... (keep your detailed help text) ...
|
444 |
-
"1. Send URL.\n2. Choose format (Paragraph/Points).\n3. Get summary!\n\n"
|
445 |
-
"**Troubleshooting:**\n- YT transcripts might fail.\n- Complex websites hard to scrape.\n- AI errors possible.\n\nSend link to start!" )
|
446 |
await update.message.reply_text(help_text, parse_mode=ParseMode.MARKDOWN)
|
447 |
|
448 |
async def handle_potential_url(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
449 |
-
"""Handles messages containing potential URLs."""
|
450 |
if not update.message or not update.message.text: return
|
451 |
message_text = update.message.text.strip(); user = update.effective_user;
|
452 |
if not user: return
|
453 |
url_pattern = r'https?://[^\s/$.?#].[^\s]*'; match = re.search(url_pattern, message_text)
|
454 |
if match:
|
455 |
-
url = match.group(0); logger.info(f"User {user.id} sent
|
456 |
context.user_data['url_to_summarize'] = url; logger.debug(f"Stored URL '{url}' for user {user.id}")
|
457 |
-
keyboard = [[
|
458 |
reply_markup = InlineKeyboardMarkup(keyboard)
|
459 |
-
await update.message.reply_text(f"Link
|
460 |
-
else: logger.debug(f"Ignoring non-URL
|
461 |
|
462 |
async def handle_summary_type_callback(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
463 |
-
"""Handles button presses for choosing the summary type."""
|
464 |
query = update.callback_query; user = query.from_user
|
465 |
-
try: await query.answer(); logger.debug(f"Answered
|
466 |
-
except Exception as e: logger.error(f"Failed
|
467 |
summary_type = query.data; url = context.user_data.get('url_to_summarize')
|
468 |
-
logger.info(f"User {user.id} chose '{summary_type}'. URL
|
469 |
if not url:
|
470 |
logger.warning(f"User {user.id} pressed button, NO URL in context.");
|
471 |
try: await query.edit_message_text(text="Context lost. Send link again.")
|
@@ -474,10 +421,10 @@ async def handle_summary_type_callback(update: Update, context: ContextTypes.DEF
|
|
474 |
context.user_data.pop('url_to_summarize', None); logger.debug(f"Cleared URL {url} for user {user.id}")
|
475 |
current_openrouter_key = os.environ.get('OPENROUTER_API_KEY'); current_urltotext_key = os.environ.get('URLTOTEXT_API_KEY')
|
476 |
current_supadata_key = os.environ.get('SUPADATA_API_KEY'); current_apify_token = os.environ.get('APIFY_API_TOKEN')
|
477 |
-
logger.debug(f"Keys
|
478 |
if not current_openrouter_key:
|
479 |
logger.error("OpenRouter key missing.");
|
480 |
-
try: await query.edit_message_text(text="⚠️ AI service config error (key missing).
|
481 |
except Exception as edit_err: logger.error(f"Failed edit on missing OR key: {edit_err}")
|
482 |
return
|
483 |
processing_message_text = f"Working on '{summary_type}' summary for:\n{url}\n..."; message_to_delete_later_id = None
|
@@ -514,9 +461,10 @@ async def handle_summary_type_callback(update: Update, context: ContextTypes.DEF
|
|
514 |
else: logger.info("Summary generated OK. Sending."); await context.bot.send_message(chat_id=user.id, text=summary, parse_mode=ParseMode.MARKDOWN, link_preview_options={'is_disabled': True}); success = True; user_feedback_message = None
|
515 |
elif not user_feedback_message: user_feedback_message = "Sorry, couldn't retrieve content from link."
|
516 |
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)
|
517 |
-
except Exception as e:
|
518 |
-
|
519 |
-
|
|
|
520 |
finally:
|
521 |
logger.debug("Cleaning up status message...");
|
522 |
try:
|
@@ -564,14 +512,20 @@ async def lifespan(app: Starlette): # app argument is the Starlette instance
|
|
564 |
WEBHOOK_URL_BASE = os.environ.get("SPACE_HOST")
|
565 |
if WEBHOOK_URL_BASE:
|
566 |
if not WEBHOOK_URL_BASE.startswith("https://"): WEBHOOK_URL_BASE = f"https://{WEBHOOK_URL_BASE}"
|
567 |
-
webhook_path = "/webhook"
|
568 |
full_webhook_url = f"{WEBHOOK_URL_BASE.rstrip('/')}{webhook_path}"
|
569 |
logger.info(f"Attempting to set webhook to: {full_webhook_url}")
|
570 |
try:
|
|
|
|
|
|
|
571 |
await ptb_app.bot.set_webhook(url=full_webhook_url, allowed_updates=Update.ALL_TYPES)
|
572 |
webhook_info = await ptb_app.bot.get_webhook_info()
|
573 |
logger.info(f"Webhook set successfully! Info: {webhook_info}")
|
574 |
-
except
|
|
|
|
|
|
|
575 |
else: logger.warning("SPACE_HOST env variable not found. Skipping webhook setup.")
|
576 |
|
577 |
logger.info("ASGI Lifespan: Startup complete. Application ready.")
|
@@ -579,12 +533,12 @@ async def lifespan(app: Starlette): # app argument is the Starlette instance
|
|
579 |
|
580 |
except Exception as startup_err:
|
581 |
logger.critical(f"CRITICAL ERROR during ASGI startup: {startup_err}", exc_info=True)
|
582 |
-
# Optionally re-raise or handle to prevent server from potentially running in bad state
|
583 |
raise
|
584 |
finally:
|
585 |
# --- Shutdown ---
|
586 |
logger.info("ASGI Lifespan: Shutdown commencing...")
|
587 |
-
if
|
|
|
588 |
try:
|
589 |
logger.info("Stopping PTB App...")
|
590 |
await ptb_app.stop()
|
@@ -610,15 +564,26 @@ def index():
|
|
610 |
"""Basic health check endpoint."""
|
611 |
logger.debug("Health check '/' accessed.")
|
612 |
bot_status = "UNKNOWN"
|
613 |
-
if ptb_app:
|
614 |
-
|
|
|
|
|
|
|
|
|
615 |
return f"Telegram Bot Webhook Listener ({bot_status}) running via Starlette."
|
616 |
|
617 |
@flask_core_app.route('/webhook', methods=['POST'])
|
618 |
async def webhook() -> Response:
|
619 |
"""Webhook endpoint for Telegram updates."""
|
620 |
-
if not ptb_app
|
621 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
622 |
logger.error(f"Webhook triggered, but PTB Application is {status}.")
|
623 |
return Response('Bot service not ready.', status=503)
|
624 |
|
|
|
1 |
+
# main.py (Revised: Starlette Lifespan + ptb_app._running fix)
|
2 |
import os
|
3 |
import re
|
4 |
import logging
|
|
|
24 |
CallbackQueryHandler,
|
25 |
)
|
26 |
from telegram.constants import ParseMode
|
27 |
+
from telegram.error import NetworkError, RetryAfter # Import RetryAfter
|
28 |
|
29 |
# --- Other Libraries ---
|
30 |
from youtube_transcript_api import YouTubeTranscriptApi
|
|
|
47 |
logging.getLogger('telegram.bot').setLevel(logging.INFO)
|
48 |
logging.getLogger("urllib3").setLevel(logging.INFO)
|
49 |
logging.getLogger('gunicorn.error').setLevel(logging.INFO)
|
50 |
+
logging.getLogger('uvicorn').setLevel(logging.INFO)
|
51 |
+
logging.getLogger('starlette').setLevel(logging.INFO)
|
52 |
logger = logging.getLogger(__name__)
|
53 |
logger.info("Logging configured.")
|
54 |
|
|
|
180 |
item = results[0]
|
181 |
content = item.get("text") or item.get("transcript") or item.get("captions_concatenated")
|
182 |
|
|
|
183 |
if not content and item.get("captions"):
|
184 |
captions_data = item["captions"]
|
185 |
if isinstance(captions_data, str):
|
|
|
201 |
content = html.unescape(content) # Use imported html module
|
202 |
except Exception as unescape_err:
|
203 |
logger.warning(f"[Apify] Error during html unescaping: {unescape_err}")
|
|
|
204 |
|
205 |
if content and isinstance(content, str):
|
206 |
logger.info(f"[Apify] Successfully fetched transcript via run-sync for {video_url} (Status: {response.status_code}). Length: {len(content)}")
|
207 |
+
return content
|
208 |
else:
|
209 |
+
if item.get("text") or item.get("transcript") or item.get("captions_concatenated"): logger.warning(f"[Apify] Actor success ({response.status_code}) but primary fields empty for {video_url}.")
|
210 |
+
elif not item.get("captions"): logger.warning(f"[Apify] Actor success ({response.status_code}) but no relevant fields found for {video_url}. Item: {item}")
|
211 |
+
else: logger.warning(f"[Apify] Actor success ({response.status_code}), 'captions' found but fallback parsing failed for {video_url}.")
|
|
|
|
|
|
|
|
|
212 |
return None
|
213 |
else:
|
214 |
+
logger.warning(f"[Apify] Actor success ({response.status_code}) but dataset result list empty for {video_url}. Response: {results}")
|
215 |
return None
|
216 |
except json.JSONDecodeError:
|
217 |
+
logger.error(f"[Apify] Failed JSON decode for {video_url}. Status: {response.status_code}. Resp: {response.text[:200]}...")
|
218 |
return None
|
219 |
except Exception as e:
|
220 |
logger.error(f"[Apify] Error processing successful response ({response.status_code}) for {video_url}: {e}", exc_info=True)
|
221 |
return None
|
222 |
+
elif response.status_code == 400: logger.error(f"[Apify] Bad Request (400) for {video_url}. Resp: {response.text[:200]}..."); return None
|
223 |
+
elif response.status_code == 401: logger.error("[Apify] Auth error (401). Check token."); return None
|
224 |
+
else: logger.error(f"[Apify] Unexpected status {response.status_code} for {video_url}. Resp: {response.text[:200]}..."); return None
|
|
|
|
|
|
|
|
|
|
|
|
|
225 |
|
226 |
+
except requests.exceptions.Timeout: logger.error(f"[Apify] Timeout error running actor for {video_url}"); return None
|
227 |
+
except requests.exceptions.RequestException as e: logger.error(f"[Apify] Request error running actor for {video_url}: {e}"); return None
|
228 |
+
except Exception as e: logger.error(f"[Apify] Unexpected error during Apify call for {video_url}: {e}", exc_info=True); return None
|
|
|
|
|
|
|
|
|
|
|
|
|
229 |
|
230 |
+
# Combined YouTube Transcript Function
|
231 |
async def get_youtube_transcript(video_id: str, video_url: str, supadata_key: str | None, apify_token: str | None):
|
232 |
"""Fetches YouTube transcript using library, then Supadata, then Apify."""
|
233 |
if not video_id: logger.error("get_youtube_transcript called with no video_id"); return None
|
234 |
logger.info(f"Fetching transcript for video ID: {video_id} (URL: {video_url})")
|
235 |
transcript_text = None
|
236 |
+
try: # Primary: Library
|
237 |
logger.info("[Primary YT] Attempting youtube-transcript-api...")
|
238 |
+
transcript_list = await asyncio.to_thread(YouTubeTranscriptApi.get_transcript, video_id, languages=['en', 'en-GB', 'en-US'])
|
|
|
|
|
|
|
|
|
239 |
if transcript_list:
|
240 |
transcript_text = " ".join([item['text'] for item in transcript_list if 'text' in item])
|
241 |
transcript_text = re.sub(r'\s+', ' ', transcript_text).strip()
|
242 |
+
if transcript_text: logger.info(f"[Primary YT] Success via library. Length: {len(transcript_text)}"); return transcript_text
|
243 |
+
else: logger.warning("[Primary YT] Joined text empty after cleaning."); transcript_text = None
|
244 |
+
else: logger.warning("[Primary YT] Transcript list empty."); transcript_text = None
|
|
|
|
|
|
|
|
|
|
|
|
|
245 |
except Exception as e:
|
246 |
+
logger.warning(f"[Primary YT] Error via library: {type(e).__name__} - {e}")
|
247 |
+
if "YouTube is blocking requests" in str(e) or "HTTP Error 429" in str(e): logger.warning("[Primary YT] IP likely blocked.")
|
248 |
+
elif "No transcript found" in str(e): logger.warning("[Primary YT] No transcript in languages.")
|
249 |
+
elif "TranscriptsDisabled" in str(e) or "disabled" in str(e): logger.warning("[Primary YT] Transcripts disabled.")
|
250 |
transcript_text = None
|
251 |
|
252 |
if transcript_text is None: # Fallback 1: Supadata
|
253 |
+
logger.info("[Fallback YT 1] Trying Supadata API...")
|
254 |
if supadata_key:
|
255 |
transcript_text = await get_transcript_via_supadata(video_id, supadata_key)
|
256 |
+
if transcript_text: logger.info(f"[Fallback YT 1] Success via Supadata. Length: {len(transcript_text)}"); return transcript_text
|
257 |
+
else: logger.warning("[Fallback YT 1] Supadata failed or no content.")
|
258 |
+
else: logger.warning("[Fallback YT 1] Supadata key not available.")
|
|
|
|
|
259 |
|
260 |
if transcript_text is None: # Fallback 2: Apify
|
261 |
+
logger.info("[Fallback YT 2] Trying Apify API...")
|
262 |
if apify_token:
|
263 |
transcript_text = await get_transcript_via_apify(video_url, apify_token)
|
264 |
+
if transcript_text: logger.info(f"[Fallback YT 2] Success via Apify. Length: {len(transcript_text)}"); return transcript_text
|
265 |
+
else: logger.warning("[Fallback YT 2] Apify failed or no content.")
|
266 |
+
else: logger.warning("[Fallback YT 2] Apify token not available.")
|
267 |
+
|
268 |
+
if transcript_text is None: logger.error(f"All methods failed for video ID: {video_id}")
|
269 |
+
return transcript_text
|
|
|
|
|
|
|
|
|
270 |
|
271 |
# Website Content via Requests/BS4
|
272 |
async def get_website_content_via_requests(url):
|
273 |
"""Attempts to scrape website content using requests/BeautifulSoup."""
|
274 |
+
if not url: logger.error("[Web Scraper - Requests/BS4] no URL"); return None
|
275 |
+
logger.info(f"[Web Scraper - Requests/BS4] Fetching: {url}")
|
276 |
try:
|
277 |
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'}
|
|
|
278 |
response = await asyncio.to_thread(requests.get, url, headers=headers, timeout=25, allow_redirects=True)
|
279 |
response.raise_for_status()
|
280 |
+
logger.debug(f"[Web Scraper - Requests/BS4] Status {response.status_code} for {url}")
|
281 |
content_type = response.headers.get('content-type', '').lower()
|
282 |
if 'html' not in content_type:
|
283 |
+
logger.warning(f"[Web Scraper - Requests/BS4] Non-HTML: {content_type}. Plain text?")
|
284 |
if 'text/plain' in content_type and response.text: return response.text.strip()
|
285 |
return None
|
286 |
soup = BeautifulSoup(response.text, 'html.parser')
|
287 |
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()
|
288 |
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')
|
289 |
target_element = main_content if main_content else soup.body
|
290 |
+
if not target_element: logger.warning(f"[Web Scraper - Requests/BS4] No body/main content for {url}"); return None
|
|
|
|
|
291 |
lines = [line.strip() for line in target_element.get_text(separator='\n', strip=True).splitlines() if line.strip()]
|
292 |
text = "\n".join(lines)
|
293 |
MIN_TEXT_LENGTH = 50
|
294 |
+
if not text or len(text) < MIN_TEXT_LENGTH: logger.warning(f"[Web Scraper - Requests/BS4] Text too short (<{MIN_TEXT_LENGTH} chars) for {url} (Len: {len(text)})")
|
295 |
+
logger.info(f"[Web Scraper - Requests/BS4] Success scraping {url} (Len: {len(text)})")
|
|
|
296 |
return text
|
297 |
+
except requests.exceptions.Timeout: logger.error(f"[Web Scraper - Requests/BS4] Timeout: {url}"); return None
|
298 |
+
except requests.exceptions.TooManyRedirects: logger.error(f"[Web Scraper - Requests/BS4] Redirects: {url}"); return None
|
299 |
+
except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - Requests/BS4] Request error {url}: {e}"); return None
|
300 |
+
except Exception as e: logger.error(f"[Web Scraper - Requests/BS4] Parsing error {url}: {e}", exc_info=True); return None
|
301 |
|
302 |
# Website Content via URLToText API
|
303 |
async def get_website_content_via_urltotext_api(url: str, api_key: str):
|
304 |
"""Fetches website content using the URLToText API."""
|
305 |
+
if not url: logger.error("[Web Scraper - URLToText API] no URL"); return None
|
306 |
+
if not api_key: logger.error("[Web Scraper - URLToText API] API key missing."); return None
|
307 |
+
logger.info(f"[Web Scraper - URLToText API] Attempting fetch: {url}")
|
308 |
api_endpoint = "https://urltotext.com/api/v1/urltotext/"
|
309 |
payload = json.dumps({"url": url, "output_format": "text", "extract_main_content": True, "render_javascript": True, "residential_proxy": False})
|
310 |
headers = {"Authorization": f"Token {api_key}", "Content-Type": "application/json"}
|
311 |
try:
|
|
|
312 |
response = await asyncio.to_thread(requests.post, api_endpoint, headers=headers, data=payload, timeout=45)
|
313 |
+
logger.debug(f"[Web Scraper - URLToText API] Status {response.status_code} for {url}")
|
314 |
if response.status_code == 200:
|
315 |
try:
|
316 |
data = response.json()
|
317 |
+
content = data.get("data", {}).get("content"); credits = data.get("credits_used", "N/A"); warning = data.get("data", {}).get("warning")
|
|
|
|
|
318 |
if warning: logger.warning(f"[Web Scraper - URLToText API] Warning for {url}: {warning}")
|
319 |
+
if content: logger.info(f"[Web Scraper - URLToText API] Success via API. Length: {len(content)}. Credits: {credits}"); return content.strip()
|
320 |
+
else: logger.warning(f"[Web Scraper - URLToText API] API success (200) but content empty. Resp: {data}"); return None
|
321 |
+
except json.JSONDecodeError: logger.error(f"[Web Scraper - URLToText API] Failed JSON decode. Status: {response.status_code}. Resp: {response.text[:500]}..."); return None
|
322 |
+
except Exception as e: logger.error(f"[Web Scraper - URLToText API] Error processing API response: {e}", exc_info=True); return None
|
323 |
+
elif response.status_code == 400: logger.error(f"[Web Scraper - URLToText API] Bad Request (400). Resp: {response.text[:200]}...")
|
324 |
+
elif response.status_code == 401: logger.error(f"[Web Scraper - URLToText API] Unauthorized (401). Check Key. Resp: {response.text[:200]}...")
|
325 |
+
elif response.status_code == 402: logger.error(f"[Web Scraper - URLToText API] Payment Required (402). Check credits. Resp: {response.text[:200]}...")
|
326 |
+
elif response.status_code == 422: logger.warning(f"[Web Scraper - URLToText API] Unprocessable URL (422). Resp: {response.text[:200]}...")
|
327 |
+
elif response.status_code >= 500: logger.error(f"[Web Scraper - URLToText API] Server Error ({response.status_code}). Resp: {response.text[:200]}...")
|
328 |
+
else: logger.error(f"[Web Scraper - URLToText API] Unexpected status {response.status_code}. Resp: {response.text[:200]}...")
|
|
|
|
|
|
|
|
|
329 |
return None
|
330 |
+
except requests.exceptions.Timeout: logger.error(f"[Web Scraper - URLToText API] Timeout: {url}"); return None
|
331 |
+
except requests.exceptions.RequestException as e: logger.error(f"[Web Scraper - URLToText API] Request error: {e}"); return None
|
332 |
+
except Exception as e: logger.error(f"[Web Scraper - URLToText API] Unexpected error: {e}", exc_info=True); return None
|
333 |
|
334 |
# DeepSeek Summary Function
|
335 |
async def generate_summary(text: str, summary_type: str, api_key: str) -> str:
|
|
|
337 |
logger.info(f"Generating '{summary_type}' summary. Input length: {len(text)}")
|
338 |
if not api_key: logger.error("OpenRouter API key missing."); return "Error: AI config key missing."
|
339 |
if not text: logger.warning("generate_summary called with empty text."); return "Error: No content to summarize."
|
340 |
+
openrouter_api_endpoint = "https://openrouter.ai/api/v1/chat/completions"; model_name = "deepseek/deepseek-chat:free"
|
|
|
341 |
if summary_type == "paragraph": prompt = "Please provide a concise, well-written paragraph summarizing the key information and main points of the following text. Focus on capturing the essence of the content accurately."
|
342 |
elif summary_type == "points": prompt = "Please summarize the following text into clear, distinct bullet points. Each point should highlight a key piece of information, finding, or main topic discussed. Aim for clarity and conciseness."
|
343 |
else: logger.error(f"Invalid summary_type '{summary_type}'."); return f"Error: Invalid summary type ('{summary_type}')."
|
344 |
MAX_INPUT_LENGTH = 500000
|
345 |
if len(text) > MAX_INPUT_LENGTH: logger.warning(f"Input text ({len(text)}) > limit ({MAX_INPUT_LENGTH}). Truncating."); text = text[:MAX_INPUT_LENGTH] + "... (Truncated)"
|
346 |
full_prompt = f"{prompt}\n\n--- Start of Text ---\n\n{text}\n\n--- End of Text ---"
|
347 |
+
space_host = os.environ.get("SPACE_HOST", "huggingface.co/spaces/YOUR_SPACE_NAME")
|
|
|
348 |
referer_url = f"https://{space_host}" if not space_host.startswith("http") else space_host
|
349 |
headers = {"Authorization": f"Bearer {api_key}","Content-Type": "application/json","HTTP-Referer": referer_url,"X-Title": "Telegram URL Summarizer Bot"}
|
350 |
payload = json.dumps({"model": model_name, "messages": [{"role": "user", "content": full_prompt}]})
|
351 |
try:
|
352 |
logger.debug(f"Sending request to OpenRouter (Model: {model_name})...")
|
353 |
response = await asyncio.to_thread(requests.post, openrouter_api_endpoint, headers=headers, data=payload, timeout=90)
|
354 |
+
logger.debug(f"Received status {response.status_code} from OpenRouter.")
|
355 |
if response.status_code == 200:
|
356 |
try:
|
357 |
data = response.json()
|
|
|
366 |
else: logger.error(f"Unexpected OpenRouter choices structure. Resp: {data}"); return "Sorry, could not parse AI response (choices)."
|
367 |
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 format."
|
368 |
except Exception as e: logger.error(f"Error processing OpenRouter success resp: {e}", exc_info=True); return "Sorry, error processing AI response."
|
369 |
+
elif response.status_code == 401: logger.error("OpenRouter key invalid (401)."); return "Error: AI model config key invalid."
|
370 |
elif response.status_code == 402: logger.error("OpenRouter Payment Required (402)."); return "Sorry, issue with AI service limits/payment."
|
371 |
elif response.status_code == 429: logger.warning("OpenRouter Rate Limit (429)."); return "Sorry, AI model busy. Try again."
|
372 |
elif response.status_code >= 500: logger.error(f"OpenRouter Internal Error ({response.status_code}). Resp: {response.text[:500]}..."); return "Sorry, AI model service error. Try again later."
|
|
|
382 |
# --- Telegram Bot Handlers ---
|
383 |
|
384 |
async def start(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
385 |
+
user = update.effective_user;
|
|
|
386 |
if not user: return
|
387 |
+
logger.info(f"User {user.id} ({user.username or 'NoUsername'}) /start.")
|
388 |
mention = user.mention_html() if user.username else user.first_name
|
389 |
+
await update.message.reply_html(f"👋 Hello {mention}! Send a URL to summarize.")
|
390 |
|
391 |
async def help_command(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
392 |
+
user = update.effective_user; logger.info(f"User {user.id if user else '?'} /help.")
|
393 |
+
help_text = ("**How:**\n1. Send URL.\n2. Choose Paragraph/Points.\n3. Get summary!\n\n"
|
394 |
+
"**Notes:**\n- YT transcripts can fail.\n- Complex sites hard to scrape.\n- AI errors possible.")
|
|
|
|
|
|
|
395 |
await update.message.reply_text(help_text, parse_mode=ParseMode.MARKDOWN)
|
396 |
|
397 |
async def handle_potential_url(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
|
|
398 |
if not update.message or not update.message.text: return
|
399 |
message_text = update.message.text.strip(); user = update.effective_user;
|
400 |
if not user: return
|
401 |
url_pattern = r'https?://[^\s/$.?#].[^\s]*'; match = re.search(url_pattern, message_text)
|
402 |
if match:
|
403 |
+
url = match.group(0); logger.info(f"User {user.id} sent URL: {url}")
|
404 |
context.user_data['url_to_summarize'] = url; logger.debug(f"Stored URL '{url}' for user {user.id}")
|
405 |
+
keyboard = [[InlineKeyboardButton("Paragraph", callback_data="paragraph"), InlineKeyboardButton("Points", callback_data="points")]]
|
406 |
reply_markup = InlineKeyboardMarkup(keyboard)
|
407 |
+
await update.message.reply_text(f"Link:\n{url}\n\nChoose summary type:", reply_markup=reply_markup, link_preview_options={'is_disabled': True})
|
408 |
+
else: logger.debug(f"Ignoring non-URL from {user.id}: {message_text[:100]}")
|
409 |
|
410 |
async def handle_summary_type_callback(update: Update, context: ContextTypes.DEFAULT_TYPE) -> None:
|
|
|
411 |
query = update.callback_query; user = query.from_user
|
412 |
+
try: await query.answer(); logger.debug(f"Answered query {query.id}")
|
413 |
+
except Exception as e: logger.error(f"Failed answer query {query.id}: {e}")
|
414 |
summary_type = query.data; url = context.user_data.get('url_to_summarize')
|
415 |
+
logger.info(f"User {user.id} chose '{summary_type}'. URL: '{url}'.")
|
416 |
if not url:
|
417 |
logger.warning(f"User {user.id} pressed button, NO URL in context.");
|
418 |
try: await query.edit_message_text(text="Context lost. Send link again.")
|
|
|
421 |
context.user_data.pop('url_to_summarize', None); logger.debug(f"Cleared URL {url} for user {user.id}")
|
422 |
current_openrouter_key = os.environ.get('OPENROUTER_API_KEY'); current_urltotext_key = os.environ.get('URLTOTEXT_API_KEY')
|
423 |
current_supadata_key = os.environ.get('SUPADATA_API_KEY'); current_apify_token = os.environ.get('APIFY_API_TOKEN')
|
424 |
+
logger.debug(f"Keys: OR={'Y' if current_openrouter_key else 'N'}, UTT={'Y' if current_urltotext_key else 'N'}, SD={'Y' if current_supadata_key else 'N'}, AP={'Y' if current_apify_token else 'N'}")
|
425 |
if not current_openrouter_key:
|
426 |
logger.error("OpenRouter key missing.");
|
427 |
+
try: await query.edit_message_text(text="⚠️ AI service config error (key missing).")
|
428 |
except Exception as edit_err: logger.error(f"Failed edit on missing OR key: {edit_err}")
|
429 |
return
|
430 |
processing_message_text = f"Working on '{summary_type}' summary for:\n{url}\n..."; message_to_delete_later_id = None
|
|
|
461 |
else: logger.info("Summary generated OK. Sending."); await context.bot.send_message(chat_id=user.id, text=summary, parse_mode=ParseMode.MARKDOWN, link_preview_options={'is_disabled': True}); success = True; user_feedback_message = None
|
462 |
elif not user_feedback_message: user_feedback_message = "Sorry, couldn't retrieve content from link."
|
463 |
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)
|
464 |
+
except Exception as e:
|
465 |
+
logger.error(f"Unexpected error in callback processing: {e}", exc_info=True);
|
466 |
+
try: await context.bot.send_message(chat_id=user.id, text="Oops! Internal error processing request.")
|
467 |
+
except Exception as final_err: logger.error(f"Failed sending final error message: {final_err}")
|
468 |
finally:
|
469 |
logger.debug("Cleaning up status message...");
|
470 |
try:
|
|
|
512 |
WEBHOOK_URL_BASE = os.environ.get("SPACE_HOST")
|
513 |
if WEBHOOK_URL_BASE:
|
514 |
if not WEBHOOK_URL_BASE.startswith("https://"): WEBHOOK_URL_BASE = f"https://{WEBHOOK_URL_BASE}"
|
515 |
+
webhook_path = "/webhook"
|
516 |
full_webhook_url = f"{WEBHOOK_URL_BASE.rstrip('/')}{webhook_path}"
|
517 |
logger.info(f"Attempting to set webhook to: {full_webhook_url}")
|
518 |
try:
|
519 |
+
# Add a small delay before setting webhook, especially with multiple workers
|
520 |
+
# This might help avoid the initial rate limit error, although one worker succeeding is enough.
|
521 |
+
await asyncio.sleep(1.5) # Wait 1.5 seconds
|
522 |
await ptb_app.bot.set_webhook(url=full_webhook_url, allowed_updates=Update.ALL_TYPES)
|
523 |
webhook_info = await ptb_app.bot.get_webhook_info()
|
524 |
logger.info(f"Webhook set successfully! Info: {webhook_info}")
|
525 |
+
except RetryAfter as e:
|
526 |
+
logger.warning(f"Webhook set failed due to rate limit (RetryAfter: {e.retry_after}s). Another worker likely succeeded.")
|
527 |
+
except Exception as e:
|
528 |
+
logger.error(f"Failed to set webhook: {e}", exc_info=True)
|
529 |
else: logger.warning("SPACE_HOST env variable not found. Skipping webhook setup.")
|
530 |
|
531 |
logger.info("ASGI Lifespan: Startup complete. Application ready.")
|
|
|
533 |
|
534 |
except Exception as startup_err:
|
535 |
logger.critical(f"CRITICAL ERROR during ASGI startup: {startup_err}", exc_info=True)
|
|
|
536 |
raise
|
537 |
finally:
|
538 |
# --- Shutdown ---
|
539 |
logger.info("ASGI Lifespan: Shutdown commencing...")
|
540 |
+
# Use the correct attribute to check if running before stopping/shutting down
|
541 |
+
if ptb_app and ptb_app._running: # <--- Use _running here too
|
542 |
try:
|
543 |
logger.info("Stopping PTB App...")
|
544 |
await ptb_app.stop()
|
|
|
564 |
"""Basic health check endpoint."""
|
565 |
logger.debug("Health check '/' accessed.")
|
566 |
bot_status = "UNKNOWN"
|
567 |
+
if ptb_app:
|
568 |
+
# --- CORRECTED CHECK ---
|
569 |
+
bot_status = "Running" if ptb_app._running else "Initialized/Stopped/Starting/Error"
|
570 |
+
# --- END CORRECTION ---
|
571 |
+
else:
|
572 |
+
bot_status = "Not Initialized"
|
573 |
return f"Telegram Bot Webhook Listener ({bot_status}) running via Starlette."
|
574 |
|
575 |
@flask_core_app.route('/webhook', methods=['POST'])
|
576 |
async def webhook() -> Response:
|
577 |
"""Webhook endpoint for Telegram updates."""
|
578 |
+
if not ptb_app: # Check if instance exists first
|
579 |
+
logger.error("Webhook triggered, but PTB Application instance (ptb_app) is None.")
|
580 |
+
return Response('Bot service not configured.', status=503)
|
581 |
+
|
582 |
+
# --- CORRECTED CHECK ---
|
583 |
+
# Use the internal _running attribute as suggested by the AttributeError
|
584 |
+
if not ptb_app._running:
|
585 |
+
# --- END CORRECTION ---
|
586 |
+
status = "Not Running" # If instance exists but not running
|
587 |
logger.error(f"Webhook triggered, but PTB Application is {status}.")
|
588 |
return Response('Bot service not ready.', status=503)
|
589 |
|