Chrunos commited on
Commit
fc58599
·
verified ·
1 Parent(s): 8a350b6

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +130 -379
app.py CHANGED
@@ -1,5 +1,5 @@
1
- from fastapi import FastAPI, HTTPException, status, BackgroundTasks, Request, Depends
2
- from fastapi.responses import StreamingResponse, JSONResponse
3
  import instaloader
4
  import requests
5
  import os
@@ -7,12 +7,7 @@ import time
7
  import logging
8
  import random
9
  from functools import wraps
10
- import json
11
  from datetime import datetime, timedelta
12
- from typing import Dict, List, Optional
13
- import hashlib
14
- import asyncio
15
- from starlette.concurrency import run_in_threadpool
16
 
17
  # Configure logging
18
  logging.basicConfig(
@@ -30,73 +25,21 @@ INSTAGRAM_PASSWORD = os.getenv('INSTAGRAM_PASSWORD')
30
  # Session configuration
31
  SESSION_DIR = "/tmp/instagram_sessions"
32
  SESSION_FILE = os.path.join(SESSION_DIR, f"session-{INSTAGRAM_USERNAME}") if INSTAGRAM_USERNAME else None
33
- CACHE_DIR = "/tmp/instagram_cache"
34
 
35
- # Create required directories
36
  os.makedirs(SESSION_DIR, exist_ok=True)
37
- os.makedirs(CACHE_DIR, exist_ok=True)
38
 
39
- # Rate limiting state
40
- LAST_REQUEST_TIME = {}
41
- COOLDOWN_PERIODS = {}
42
- RATE_LIMIT_LOCK = asyncio.Lock()
43
-
44
- # Cache state
45
- STORY_CACHE: Dict[str, Dict] = {}
46
- CACHE_EXPIRY: Dict[str, datetime] = {}
47
- CACHE_LOCK = asyncio.Lock()
48
 
49
  USER_AGENTS = [
50
  "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
51
  "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Safari/605.1.15",
52
- "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36",
53
- "Mozilla/5.0 (iPhone; CPU iPhone OS 17_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.1 Mobile/15E148 Safari/604.1",
54
- "Mozilla/5.0 (iPad; CPU OS 17_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.1 Mobile/15E148 Safari/604.1"
55
  ]
56
 
57
- # Calculate cache key for a username
58
- def get_cache_key(username: str) -> str:
59
- return hashlib.md5(username.lower().encode()).hexdigest()
60
-
61
- # Check if we have a valid cache for a username
62
- async def get_cached_stories(username: str) -> Optional[Dict]:
63
- cache_key = get_cache_key(username)
64
-
65
- async with CACHE_LOCK:
66
- if cache_key in STORY_CACHE and cache_key in CACHE_EXPIRY:
67
- if datetime.now() < CACHE_EXPIRY[cache_key]:
68
- logger.info(f"Cache hit for {username}")
69
- return STORY_CACHE[cache_key]
70
- else:
71
- # Expired cache
72
- logger.info(f"Cache expired for {username}")
73
- STORY_CACHE.pop(cache_key, None)
74
- CACHE_EXPIRY.pop(cache_key, None)
75
-
76
- return None
77
-
78
- # Save stories to cache
79
- async def cache_stories(username: str, stories_data: Dict, ttl_minutes: int = 15):
80
- cache_key = get_cache_key(username)
81
-
82
- async with CACHE_LOCK:
83
- STORY_CACHE[cache_key] = stories_data
84
- CACHE_EXPIRY[cache_key] = datetime.now() + timedelta(minutes=ttl_minutes)
85
-
86
- logger.info(f"Cached stories for {username} for {ttl_minutes} minutes")
87
-
88
- # Save to disk for persistence
89
- cache_file = os.path.join(CACHE_DIR, f"{cache_key}.json")
90
- try:
91
- with open(cache_file, 'w') as f:
92
- json.dump({
93
- "data": stories_data,
94
- "expires": CACHE_EXPIRY[cache_key].isoformat()
95
- }, f)
96
- except Exception as e:
97
- logger.warning(f"Failed to save cache to disk: {str(e)}")
98
-
99
- def get_instaloader(force_login=False) -> instaloader.Instaloader:
100
  """Create and configure Instaloader instance with proper parameters"""
101
  L = instaloader.Instaloader(
102
  sleep=True,
@@ -120,58 +63,27 @@ def get_instaloader(force_login=False) -> instaloader.Instaloader:
120
  )
121
 
122
  try:
123
- if not force_login and SESSION_FILE and os.path.exists(SESSION_FILE):
124
  logger.info(f"Attempting to load session from {SESSION_FILE}")
125
  try:
126
  L.load_session_from_file(INSTAGRAM_USERNAME, SESSION_FILE)
127
  logger.info("Session loaded successfully")
128
- # Test session with a lightweight call if possible
129
  return L
130
  except Exception as e:
131
  logger.warning(f"Session load failed: {str(e)}, performing fresh login")
132
 
133
- # Add delay before login to mimic human behavior
134
  time.sleep(random.uniform(1.0, 3.0))
135
 
136
  logger.info("Starting fresh login process")
137
  L.login(INSTAGRAM_USERNAME, INSTAGRAM_PASSWORD)
138
 
139
- # Set realistic headers
140
- L.context._session.headers.update({
141
- 'Accept-Language': 'en-US,en;q=0.9',
142
- 'Accept-Encoding': 'gzip, deflate, br',
143
- 'Referer': 'https://www.instagram.com/',
144
- 'Origin': 'https://www.instagram.com',
145
- 'DNT': '1',
146
- 'Connection': 'keep-alive',
147
- 'Sec-Fetch-Dest': 'empty',
148
- 'Sec-Fetch-Mode': 'cors',
149
- 'Sec-Fetch-Site': 'same-origin',
150
- 'Pragma': 'no-cache',
151
- 'Cache-Control': 'no-cache',
152
- })
153
-
154
  if SESSION_FILE:
155
  L.save_session_to_file(SESSION_FILE)
156
  logger.info(f"Saved new session to {SESSION_FILE}")
157
 
158
- # Add delay after login
159
- time.sleep(random.uniform(2.0, 4.0))
160
-
161
  return L
162
 
163
- except instaloader.exceptions.BadCredentialsException as e:
164
- logger.error("Invalid credentials: %s", str(e))
165
- raise HTTPException(
166
- status_code=status.HTTP_401_UNAUTHORIZED,
167
- detail="Invalid Instagram credentials"
168
- )
169
- except instaloader.exceptions.TwoFactorAuthRequiredException as e:
170
- logger.error("2FA required: %s", str(e))
171
- raise HTTPException(
172
- status_code=status.HTTP_401_UNAUTHORIZED,
173
- detail="Two-factor authentication required"
174
- )
175
  except Exception as e:
176
  logger.error("Login failed: %s", str(e))
177
  raise HTTPException(
@@ -179,75 +91,46 @@ def get_instaloader(force_login=False) -> instaloader.Instaloader:
179
  detail="Instagram login service unavailable"
180
  )
181
 
182
- async def is_rate_limited(username: str) -> bool:
183
- """Check if a given username is currently rate limited"""
184
- async with RATE_LIMIT_LOCK:
185
- # If in cooldown period, check if it's expired
186
- if username in COOLDOWN_PERIODS:
187
- if datetime.now() < COOLDOWN_PERIODS[username]:
188
- remaining = (COOLDOWN_PERIODS[username] - datetime.now()).seconds
189
- logger.warning(f"User {username} in cooldown for {remaining} more seconds")
190
- return True
191
- else:
192
- # Cooldown expired
193
- del COOLDOWN_PERIODS[username]
194
-
195
- # Check time since last request
196
- if username in LAST_REQUEST_TIME:
197
- elapsed = (datetime.now() - LAST_REQUEST_TIME[username]).seconds
198
- min_interval = 60 # Minimum 60 seconds between requests
199
- if elapsed < min_interval:
200
- logger.warning(f"Rate limiting {username}: {elapsed}s elapsed, need {min_interval}s")
201
- return True
202
-
203
- # Update last request time
204
- LAST_REQUEST_TIME[username] = datetime.now()
205
- return False
206
-
207
- async def handle_rate_limit_error(username: str, retry_after: Optional[int] = None):
208
- """Handle rate limit by setting cooldown period"""
209
- async with RATE_LIMIT_LOCK:
210
- # Set cooldown period based on retry_after or default logic
211
- if retry_after:
212
- # Use server-provided retry duration if available
213
- cooldown_minutes = max(retry_after / 60, 1) # At least 1 minute
214
  else:
215
- # Default exponential backoff
216
- cooldown_minutes = 5
217
- if username in COOLDOWN_PERIODS:
218
- cooldown_minutes = min(cooldown_minutes * 2, 30) # Up to 30 minutes
219
-
220
- COOLDOWN_PERIODS[username] = datetime.now() + timedelta(minutes=cooldown_minutes)
221
- logger.warning(f"Setting cooldown for {username} for {cooldown_minutes} minutes")
 
 
 
 
222
 
223
  def handle_instagram_errors(func):
224
  @wraps(func)
225
  async def wrapper(*args, **kwargs):
226
- # Extract username from request path
227
- request_username = ""
228
- if len(args) > 1 and hasattr(args[1], '__str__'):
229
- request_username = str(args[1])
230
-
231
  try:
232
- # Check for rate limiting
233
- if request_username and await is_rate_limited(request_username):
234
- cached_response = await get_cached_stories(request_username)
235
- if cached_response:
236
- # Return cached response with cache note
237
- logger.info(f"Returning cached response for rate-limited user {request_username}")
238
- return {**cached_response, "from_cache": True}
239
- else:
240
- # No cache available
241
- raise HTTPException(
242
- status_code=status.HTTP_429_TOO_MANY_REQUESTS,
243
- detail="Rate limit exceeded. Please try again later."
244
- )
245
 
 
 
 
 
 
 
 
246
  response = await func(*args, **kwargs)
247
 
248
- # Cache successful responses if they contain story data
249
  if request_username and isinstance(response, dict) and "data" in response:
250
- await cache_stories(request_username, response)
251
 
252
  return response
253
 
@@ -255,77 +138,31 @@ def handle_instagram_errors(func):
255
  error_message = str(e)
256
  logger.error("Connection error: %s", error_message)
257
 
258
- # Check for rate limit related messages
259
- retry_after = None
260
  if "429 Too Many Requests" in error_message:
261
- # Extract retry timer from error message if possible
 
262
  try:
263
  retry_text = error_message.split("retried in")[1].split("minutes")[0].strip()
264
- retry_after = int(retry_text) * 60 # Convert to seconds
265
- except (IndexError, ValueError):
266
- retry_after = 1800 # Default 30 minutes
267
-
268
- if request_username:
269
- await handle_rate_limit_error(request_username, retry_after)
270
-
271
- # Try to serve from cache if available
272
- if request_username:
273
- cached_response = await get_cached_stories(request_username)
274
- if cached_response:
275
- logger.info(f"Returning cached response for rate-limited request")
276
- return {**cached_response, "from_cache": True}
277
 
278
  raise HTTPException(
279
  status_code=status.HTTP_429_TOO_MANY_REQUESTS,
280
- detail=f"Instagram rate limit exceeded. Please try again in {retry_after//60} minutes."
281
  )
282
- elif "401 Unauthorized" in error_message and "Please wait a few minutes" in error_message:
283
- if request_username:
284
- await handle_rate_limit_error(request_username)
285
-
286
- # Try to serve from cache if available
287
- if request_username:
288
- cached_response = await get_cached_stories(request_username)
289
- if cached_response:
290
- logger.info(f"Returning cached response for rate-limited request")
291
- return {**cached_response, "from_cache": True}
292
-
293
  raise HTTPException(
294
  status_code=status.HTTP_429_TOO_MANY_REQUESTS,
295
  detail="Instagram rate limit exceeded. Please try again later."
296
  )
297
- elif "404 Not Found" in error_message:
298
- raise HTTPException(
299
- status_code=status.HTTP_404_NOT_FOUND,
300
- detail="Profile not found or no stories available"
301
- )
302
  else:
303
  raise HTTPException(
304
  status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
305
  detail="Instagram service unavailable"
306
  )
307
 
308
- except instaloader.exceptions.QueryReturnedBadRequestException as e:
309
- logger.error("API error 400: %s", str(e))
310
- if request_username:
311
- await handle_rate_limit_error(request_username)
312
-
313
- # Try to serve from cache if available
314
- if request_username:
315
- cached_response = await get_cached_stories(request_username)
316
- if cached_response:
317
- logger.info(f"Returning cached response for rate-limited request")
318
- return {**cached_response, "from_cache": True}
319
-
320
- raise HTTPException(
321
- status_code=status.HTTP_429_TOO_MANY_REQUESTS,
322
- detail="Instagram rate limit exceeded"
323
- )
324
-
325
- except HTTPException:
326
- # Re-raise HTTP exceptions without modification
327
- raise
328
-
329
  except Exception as e:
330
  logger.error("Unexpected error: %s", str(e))
331
  raise HTTPException(
@@ -340,119 +177,86 @@ async def get_stories(username: str, background_tasks: BackgroundTasks, cached:
340
  """Retrieve stories with enhanced anti-detection measures"""
341
  logger.info(f"Processing request for @{username}")
342
 
343
- # Check for cache first if requested
344
  if cached:
345
- cached_response = await get_cached_stories(username)
346
  if cached_response:
347
  return {**cached_response, "from_cache": True}
 
 
348
 
349
  try:
350
- # Run the Instagram operations in a thread pool to not block the event loop
351
- async def fetch_stories():
352
- # Get loader with session
353
- L = get_instaloader()
354
- logger.info("Instaloader instance configured")
355
-
356
- # Randomized delay before profile request (more natural)
357
- delay = random.uniform(1.5, 3.0)
358
- logger.debug(f"Applying initial delay of {delay:.2f}s")
359
- time.sleep(delay)
360
-
361
- # Profile lookup with retry
362
- profile = None
363
- for attempt in range(3):
364
- try:
365
- logger.info(f"Fetching profile (attempt {attempt+1}/3)")
366
- profile = instaloader.Profile.from_username(L.context, username)
367
- break
368
- except instaloader.exceptions.QueryReturnedBadRequestException:
369
- wait_time = (attempt + 1) * random.uniform(3.0, 5.0)
370
- logger.warning(f"Rate limited, waiting {wait_time:.2f}s")
371
- time.sleep(wait_time)
372
- except instaloader.exceptions.ConnectionException as e:
373
- if "401 Unauthorized" in str(e) and "Please wait a few minutes" in str(e):
374
- # Try with a fresh login if session might be expired
375
- if attempt < 2: # Only try this once
376
- logger.warning("Session may be expired, trying with fresh login")
377
- L = get_instaloader(force_login=True)
378
- time.sleep(random.uniform(4.0, 6.0))
379
- continue
380
- raise
381
-
382
- if profile is None:
383
- raise HTTPException(
384
- status.HTTP_429_TOO_MANY_REQUESTS,
385
- "Too many attempts to access Instagram"
386
- )
387
-
388
- logger.info(f"Access check for @{username}")
389
- if not profile.has_viewable_story:
390
- logger.warning("No viewable story")
391
- raise HTTPException(
392
- status.HTTP_404_NOT_FOUND,
393
- "No accessible stories for this profile"
394
- )
395
 
396
- # Additional delay before story fetch
397
- time.sleep(random.uniform(1.0, 2.5))
398
-
399
- logger.info("Fetching stories")
400
- stories = []
401
- try:
402
- for story in L.get_stories(userids=[profile.userid]):
403
- for item in story.get_items():
404
- # Create a story dict with safe attribute access
405
- story_data = {
406
- "id": str(item.mediaid),
407
- "url": item.url,
408
- "type": "video" if item.is_video else "image",
409
- "timestamp": item.date_utc.isoformat(),
410
- }
411
-
412
- # Add any available metadata safely
413
- if hasattr(item, 'owner_username'):
414
- story_data["username"] = item.owner_username
415
-
416
- # Only try to add view_count if it's a video
417
- if item.is_video:
418
- try:
419
- if hasattr(item, 'view_count'):
420
- story_data["views"] = item.view_count
421
- except AttributeError:
422
- pass
423
-
424
- stories.append(story_data)
425
-
426
- except instaloader.exceptions.QueryReturnedNotFoundException:
427
- logger.error("Stories not found")
428
- raise HTTPException(
429
- status.HTTP_404_NOT_FOUND,
430
- "Stories not available"
431
- )
432
 
433
- if not stories:
434
- logger.info("No active stories found")
435
- raise HTTPException(
436
- status.HTTP_404_NOT_FOUND,
437
- "No active stories available"
438
- )
439
 
440
- # Queue session save in background
441
- background_tasks.add_task(lambda: L.save_session_to_file(SESSION_FILE) if SESSION_FILE else None)
442
-
443
- # Final randomized delay
444
- time.sleep(random.uniform(0.3, 0.7))
445
-
446
- logger.info(f"Returning {len(stories)} stories")
447
- return {
448
- "data": stories,
449
- "count": len(stories),
450
- "username": username,
451
- "fetched_at": datetime.now().isoformat()
452
- }
453
 
454
- # Execute in thread pool to not block the event loop
455
- return await run_in_threadpool(fetch_stories)
 
 
 
 
 
 
 
456
 
457
  except Exception as e:
458
  logger.error(f"Critical failure: {str(e)}")
@@ -465,37 +269,28 @@ async def download_media(url: str):
465
  logger.info(f"Download request for: {url}")
466
 
467
  try:
468
- # Basic validation to prevent arbitrary URL access
469
  if not url.startswith(("https://instagram", "https://scontent")):
470
  raise HTTPException(
471
  status_code=status.HTTP_400_BAD_REQUEST,
472
  detail="Invalid URL format"
473
  )
474
 
475
- # Random delay to avoid detection patterns
476
- time.sleep(random.uniform(0.5, 1.5))
477
 
478
- # Configure headers to mimic a browser
479
  headers = {
480
  "User-Agent": random.choice(USER_AGENTS),
481
  "Accept": "*/*",
482
- "Accept-Language": "en-US,en;q=0.9",
483
- "Accept-Encoding": "gzip, deflate, br",
484
- "Referer": "https://www.instagram.com/",
485
- "Origin": "https://www.instagram.com",
486
- "Sec-Fetch-Dest": "empty",
487
- "Sec-Fetch-Mode": "cors",
488
- "Sec-Fetch-Site": "cross-site",
489
- "Pragma": "no-cache",
490
- "Cache-Control": "no-cache",
491
  }
492
 
493
- # Request the media with a session
494
- session = requests.Session()
495
- response = session.get(url, headers=headers, stream=True, timeout=10)
496
  response.raise_for_status()
497
 
498
- # Determine content type from response or URL
499
  if "Content-Type" in response.headers:
500
  content_type = response.headers["Content-Type"]
501
  elif url.endswith((".jpg", ".jpeg")):
@@ -511,15 +306,10 @@ async def download_media(url: str):
511
  return StreamingResponse(
512
  response.iter_content(chunk_size=8192),
513
  media_type=content_type,
514
- headers={
515
- "Content-Disposition": f"attachment; filename={url.split('/')[-1]}",
516
- "Cache-Control": "no-cache, no-store, must-revalidate",
517
- "Pragma": "no-cache",
518
- "Expires": "0",
519
- }
520
  )
521
 
522
- except requests.exceptions.RequestException as e:
523
  logger.error(f"Download failed: {str(e)}")
524
  raise HTTPException(
525
  status_code=status.HTTP_502_BAD_GATEWAY,
@@ -529,43 +319,4 @@ async def download_media(url: str):
529
  # Add a health check endpoint
530
  @app.get("/health")
531
  async def health_check():
532
- return {"status": "ok", "timestamp": datetime.now().isoformat()}
533
-
534
- # Add middleware to handle rate limit headers
535
- @app.middleware("http")
536
- async def add_rate_limit_headers(request: Request, call_next):
537
- response = await call_next(request)
538
-
539
- # Add custom headers for rate limiting info
540
- username = request.path_params.get("username", None)
541
- if username and username in COOLDOWN_PERIODS:
542
- remaining = max(0, int((COOLDOWN_PERIODS[username] - datetime.now()).total_seconds()))
543
- response.headers["X-RateLimit-Reset"] = str(remaining)
544
- response.headers["X-RateLimit-Remaining"] = "0" if remaining > 0 else "1"
545
-
546
- return response
547
-
548
- # Startup event to load cache from disk
549
- @app.on_event("startup")
550
- async def load_cache_from_disk():
551
- try:
552
- for filename in os.listdir(CACHE_DIR):
553
- if filename.endswith('.json'):
554
- file_path = os.path.join(CACHE_DIR, filename)
555
- try:
556
- with open(file_path, 'r') as f:
557
- cache_data = json.load(f)
558
-
559
- if "data" in cache_data and "expires" in cache_data:
560
- expire_time = datetime.fromisoformat(cache_data["expires"])
561
- if expire_time > datetime.now():
562
- cache_key = filename[:-5] # Remove .json
563
- async with CACHE_LOCK:
564
- STORY_CACHE[cache_key] = cache_data["data"]
565
- CACHE_EXPIRY[cache_key] = expire_time
566
- except Exception as e:
567
- logger.warning(f"Failed to load cache file {filename}: {str(e)}")
568
-
569
- logger.info(f"Loaded {len(STORY_CACHE)} items from cache")
570
- except Exception as e:
571
- logger.error(f"Error loading cache: {str(e)}")
 
1
+ from fastapi import FastAPI, HTTPException, status, BackgroundTasks
2
+ from fastapi.responses import StreamingResponse
3
  import instaloader
4
  import requests
5
  import os
 
7
  import logging
8
  import random
9
  from functools import wraps
 
10
  from datetime import datetime, timedelta
 
 
 
 
11
 
12
  # Configure logging
13
  logging.basicConfig(
 
25
  # Session configuration
26
  SESSION_DIR = "/tmp/instagram_sessions"
27
  SESSION_FILE = os.path.join(SESSION_DIR, f"session-{INSTAGRAM_USERNAME}") if INSTAGRAM_USERNAME else None
 
28
 
29
+ # Create session directory if not exists
30
  os.makedirs(SESSION_DIR, exist_ok=True)
 
31
 
32
+ # Simple memory cache
33
+ STORY_CACHE = {}
34
+ CACHE_EXPIRY = {}
 
 
 
 
 
 
35
 
36
  USER_AGENTS = [
37
  "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
38
  "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Safari/605.1.15",
39
+ "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/119.0.0.0 Safari/537.36"
 
 
40
  ]
41
 
42
+ def get_instaloader() -> instaloader.Instaloader:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
  """Create and configure Instaloader instance with proper parameters"""
44
  L = instaloader.Instaloader(
45
  sleep=True,
 
63
  )
64
 
65
  try:
66
+ if SESSION_FILE and os.path.exists(SESSION_FILE):
67
  logger.info(f"Attempting to load session from {SESSION_FILE}")
68
  try:
69
  L.load_session_from_file(INSTAGRAM_USERNAME, SESSION_FILE)
70
  logger.info("Session loaded successfully")
 
71
  return L
72
  except Exception as e:
73
  logger.warning(f"Session load failed: {str(e)}, performing fresh login")
74
 
75
+ # Add delay before login
76
  time.sleep(random.uniform(1.0, 3.0))
77
 
78
  logger.info("Starting fresh login process")
79
  L.login(INSTAGRAM_USERNAME, INSTAGRAM_PASSWORD)
80
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
81
  if SESSION_FILE:
82
  L.save_session_to_file(SESSION_FILE)
83
  logger.info(f"Saved new session to {SESSION_FILE}")
84
 
 
 
 
85
  return L
86
 
 
 
 
 
 
 
 
 
 
 
 
 
87
  except Exception as e:
88
  logger.error("Login failed: %s", str(e))
89
  raise HTTPException(
 
91
  detail="Instagram login service unavailable"
92
  )
93
 
94
+ # Check for cached stories
95
+ def get_cached_stories(username):
96
+ if username in STORY_CACHE and username in CACHE_EXPIRY:
97
+ if datetime.now() < CACHE_EXPIRY[username]:
98
+ logger.info(f"Cache hit for {username}")
99
+ return STORY_CACHE[username]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
100
  else:
101
+ # Expired cache
102
+ logger.info(f"Cache expired for {username}")
103
+ STORY_CACHE.pop(username, None)
104
+ CACHE_EXPIRY.pop(username, None)
105
+ return None
106
+
107
+ # Save stories to cache
108
+ def cache_stories(username, stories_data, ttl_minutes=15):
109
+ STORY_CACHE[username] = stories_data
110
+ CACHE_EXPIRY[username] = datetime.now() + timedelta(minutes=ttl_minutes)
111
+ logger.info(f"Cached stories for {username} for {ttl_minutes} minutes")
112
 
113
  def handle_instagram_errors(func):
114
  @wraps(func)
115
  async def wrapper(*args, **kwargs):
 
 
 
 
 
116
  try:
117
+ # Extract username if present
118
+ request_username = ""
119
+ if len(args) > 1 and hasattr(args[1], '__str__'):
120
+ request_username = str(args[1])
 
 
 
 
 
 
 
 
 
121
 
122
+ # Check cache if it's a username request
123
+ if request_username:
124
+ cached_data = get_cached_stories(request_username)
125
+ if cached_data:
126
+ return {**cached_data, "from_cache": True}
127
+
128
+ # If not in cache, proceed with request
129
  response = await func(*args, **kwargs)
130
 
131
+ # Cache successful username responses
132
  if request_username and isinstance(response, dict) and "data" in response:
133
+ cache_stories(request_username, response)
134
 
135
  return response
136
 
 
138
  error_message = str(e)
139
  logger.error("Connection error: %s", error_message)
140
 
141
+ # Check for rate limit error
 
142
  if "429 Too Many Requests" in error_message:
143
+ # Try to extract retry time
144
+ retry_minutes = 30 # Default
145
  try:
146
  retry_text = error_message.split("retried in")[1].split("minutes")[0].strip()
147
+ retry_minutes = int(retry_text)
148
+ except:
149
+ pass
 
 
 
 
 
 
 
 
 
 
150
 
151
  raise HTTPException(
152
  status_code=status.HTTP_429_TOO_MANY_REQUESTS,
153
+ detail=f"Instagram rate limit exceeded. Please try again in {retry_minutes} minutes."
154
  )
155
+ elif "401 Unauthorized" in error_message:
 
 
 
 
 
 
 
 
 
 
156
  raise HTTPException(
157
  status_code=status.HTTP_429_TOO_MANY_REQUESTS,
158
  detail="Instagram rate limit exceeded. Please try again later."
159
  )
 
 
 
 
 
160
  else:
161
  raise HTTPException(
162
  status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
163
  detail="Instagram service unavailable"
164
  )
165
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
166
  except Exception as e:
167
  logger.error("Unexpected error: %s", str(e))
168
  raise HTTPException(
 
177
  """Retrieve stories with enhanced anti-detection measures"""
178
  logger.info(f"Processing request for @{username}")
179
 
180
+ # If cached param is true, only return from cache
181
  if cached:
182
+ cached_response = get_cached_stories(username)
183
  if cached_response:
184
  return {**cached_response, "from_cache": True}
185
+ else:
186
+ logger.info("No cache available for requested username")
187
 
188
  try:
189
+ L = get_instaloader()
190
+ logger.info("Instaloader instance configured")
191
+
192
+ # Randomized delay before profile request
193
+ time.sleep(random.uniform(1.0, 2.0))
194
+
195
+ # Profile lookup
196
+ logger.info(f"Fetching profile for {username}")
197
+ try:
198
+ profile = instaloader.Profile.from_username(L.context, username)
199
+ except Exception as e:
200
+ logger.error(f"Profile fetch error: {str(e)}")
201
+ raise
202
+
203
+ logger.info(f"Profile found: {profile.username}")
204
+ if not profile.has_viewable_story:
205
+ logger.warning("No viewable story")
206
+ raise HTTPException(
207
+ status.HTTP_404_NOT_FOUND,
208
+ "No accessible stories for this profile"
209
+ )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
210
 
211
+ # Additional delay before story fetch
212
+ time.sleep(random.uniform(1.0, 2.0))
213
+
214
+ logger.info("Fetching stories")
215
+ stories = []
216
+ try:
217
+ for story in L.get_stories(userids=[profile.userid]):
218
+ for item in story.get_items():
219
+ # Create a story dict with safe attribute access
220
+ story_data = {
221
+ "id": str(item.mediaid),
222
+ "url": item.url,
223
+ "type": "video" if item.is_video else "image",
224
+ "timestamp": item.date_utc.isoformat(),
225
+ }
226
+
227
+ # Only try to add view_count if it's a video
228
+ if item.is_video:
229
+ try:
230
+ if hasattr(item, 'view_count'):
231
+ story_data["views"] = item.view_count
232
+ except AttributeError:
233
+ pass
234
+
235
+ stories.append(story_data)
236
+
237
+ except Exception as e:
238
+ logger.error(f"Story fetch error: {str(e)}")
239
+ raise
 
 
 
 
 
 
 
240
 
241
+ if not stories:
242
+ logger.info("No active stories found")
243
+ raise HTTPException(
244
+ status.HTTP_404_NOT_FOUND,
245
+ "No active stories available"
246
+ )
247
 
248
+ # Save session in background
249
+ background_tasks.add_task(lambda: L.save_session_to_file(SESSION_FILE) if SESSION_FILE else None)
 
 
 
 
 
 
 
 
 
 
 
250
 
251
+ logger.info(f"Returning {len(stories)} stories")
252
+ result = {
253
+ "data": stories,
254
+ "count": len(stories),
255
+ "username": username,
256
+ "fetched_at": datetime.now().isoformat()
257
+ }
258
+
259
+ return result
260
 
261
  except Exception as e:
262
  logger.error(f"Critical failure: {str(e)}")
 
269
  logger.info(f"Download request for: {url}")
270
 
271
  try:
272
+ # Basic validation
273
  if not url.startswith(("https://instagram", "https://scontent")):
274
  raise HTTPException(
275
  status_code=status.HTTP_400_BAD_REQUEST,
276
  detail="Invalid URL format"
277
  )
278
 
279
+ # Random delay
280
+ time.sleep(random.uniform(0.5, 1.0))
281
 
282
+ # Configure headers
283
  headers = {
284
  "User-Agent": random.choice(USER_AGENTS),
285
  "Accept": "*/*",
286
+ "Referer": "https://www.instagram.com/"
 
 
 
 
 
 
 
 
287
  }
288
 
289
+ # Request the media
290
+ response = requests.get(url, headers=headers, stream=True, timeout=10)
 
291
  response.raise_for_status()
292
 
293
+ # Determine content type
294
  if "Content-Type" in response.headers:
295
  content_type = response.headers["Content-Type"]
296
  elif url.endswith((".jpg", ".jpeg")):
 
306
  return StreamingResponse(
307
  response.iter_content(chunk_size=8192),
308
  media_type=content_type,
309
+ headers={"Content-Disposition": f"attachment; filename={url.split('/')[-1]}"}
 
 
 
 
 
310
  )
311
 
312
+ except Exception as e:
313
  logger.error(f"Download failed: {str(e)}")
314
  raise HTTPException(
315
  status_code=status.HTTP_502_BAD_GATEWAY,
 
319
  # Add a health check endpoint
320
  @app.get("/health")
321
  async def health_check():
322
+ return {"status": "ok", "timestamp": datetime.now().isoformat()}