yt_dlp/extractor/iqiyi.py

   1 import hashlib
   2 import itertools
   3 import re
   4 import time
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urllib_parse_urlencode,
  10     compat_urllib_parse_unquote
  11 )
  12 from .openload import PhantomJSwrapper
  13 from ..utils import (
  14     clean_html,
  15     decode_packed_codes,
  16     ExtractorError,
  17     float_or_none,
  18     format_field,
  19     get_element_by_id,
  20     get_element_by_attribute,
  21     int_or_none,
  22     js_to_json,
  23     ohdave_rsa_encrypt,
  24     parse_age_limit,
  25     parse_duration,
  26     parse_iso8601,
  27     parse_resolution,
  28     qualities,
  29     remove_start,
  30     str_or_none,
  31     traverse_obj,
  32     urljoin,
  33 )
  34
  35
  36 def md5_text(text):
  37     return hashlib.md5(text.encode('utf-8')).hexdigest()
  38
  39
  40 class IqiyiSDK:
  41     def __init__(self, target, ip, timestamp):
  42         self.target = target
  43         self.ip = ip
  44         self.timestamp = timestamp
  45
  46     @staticmethod
  47     def split_sum(data):
  48         return compat_str(sum(map(lambda p: int(p, 16), list(data))))
  49
  50     @staticmethod
  51     def digit_sum(num):
  52         if isinstance(num, int):
  53             num = compat_str(num)
  54         return compat_str(sum(map(int, num)))
  55
  56     def even_odd(self):
  57         even = self.digit_sum(compat_str(self.timestamp)[::2])
  58         odd = self.digit_sum(compat_str(self.timestamp)[1::2])
  59         return even, odd
  60
  61     def preprocess(self, chunksize):
  62         self.target = md5_text(self.target)
  63         chunks = []
  64         for i in range(32 // chunksize):
  65             chunks.append(self.target[chunksize * i:chunksize * (i + 1)])
  66         if 32 % chunksize:
  67             chunks.append(self.target[32 - 32 % chunksize:])
  68         return chunks, list(map(int, self.ip.split('.')))
  69
  70     def mod(self, modulus):
  71         chunks, ip = self.preprocess(32)
  72         self.target = chunks[0] + ''.join(map(lambda p: compat_str(p % modulus), ip))
  73
  74     def split(self, chunksize):
  75         modulus_map = {
  76             4: 256,
  77             5: 10,
  78             8: 100,
  79         }
  80
  81         chunks, ip = self.preprocess(chunksize)
  82         ret = ''
  83         for i in range(len(chunks)):
  84             ip_part = compat_str(ip[i] % modulus_map[chunksize]) if i < 4 else ''
  85             if chunksize == 8:
  86                 ret += ip_part + chunks[i]
  87             else:
  88                 ret += chunks[i] + ip_part
  89         self.target = ret
  90
  91     def handle_input16(self):
  92         self.target = md5_text(self.target)
  93         self.target = self.split_sum(self.target[:16]) + self.target + self.split_sum(self.target[16:])
  94
  95     def handle_input8(self):
  96         self.target = md5_text(self.target)
  97         ret = ''
  98         for i in range(4):
  99             part = self.target[8 * i:8 * (i + 1)]
 100             ret += self.split_sum(part) + part
 101         self.target = ret
 102
 103     def handleSum(self):
 104         self.target = md5_text(self.target)
 105         self.target = self.split_sum(self.target) + self.target
 106
 107     def date(self, scheme):
 108         self.target = md5_text(self.target)
 109         d = time.localtime(self.timestamp)
 110         strings = {
 111             'y': compat_str(d.tm_year),
 112             'm': '%02d' % d.tm_mon,
 113             'd': '%02d' % d.tm_mday,
 114         }
 115         self.target += ''.join(map(lambda c: strings[c], list(scheme)))
 116
 117     def split_time_even_odd(self):
 118         even, odd = self.even_odd()
 119         self.target = odd + md5_text(self.target) + even
 120
 121     def split_time_odd_even(self):
 122         even, odd = self.even_odd()
 123         self.target = even + md5_text(self.target) + odd
 124
 125     def split_ip_time_sum(self):
 126         chunks, ip = self.preprocess(32)
 127         self.target = compat_str(sum(ip)) + chunks[0] + self.digit_sum(self.timestamp)
 128
 129     def split_time_ip_sum(self):
 130         chunks, ip = self.preprocess(32)
 131         self.target = self.digit_sum(self.timestamp) + chunks[0] + compat_str(sum(ip))
 132
 133
 134 class IqiyiSDKInterpreter:
 135     def __init__(self, sdk_code):
 136         self.sdk_code = sdk_code
 137
 138     def run(self, target, ip, timestamp):
 139         self.sdk_code = decode_packed_codes(self.sdk_code)
 140
 141         functions = re.findall(r'input=([a-zA-Z0-9]+)\(input', self.sdk_code)
 142
 143         sdk = IqiyiSDK(target, ip, timestamp)
 144
 145         other_functions = {
 146             'handleSum': sdk.handleSum,
 147             'handleInput8': sdk.handle_input8,
 148             'handleInput16': sdk.handle_input16,
 149             'splitTimeEvenOdd': sdk.split_time_even_odd,
 150             'splitTimeOddEven': sdk.split_time_odd_even,
 151             'splitIpTimeSum': sdk.split_ip_time_sum,
 152             'splitTimeIpSum': sdk.split_time_ip_sum,
 153         }
 154         for function in functions:
 155             if re.match(r'mod\d+', function):
 156                 sdk.mod(int(function[3:]))
 157             elif re.match(r'date[ymd]{3}', function):
 158                 sdk.date(function[4:])
 159             elif re.match(r'split\d+', function):
 160                 sdk.split(int(function[5:]))
 161             elif function in other_functions:
 162                 other_functions[function]()
 163             else:
 164                 raise ExtractorError('Unknown function %s' % function)
 165
 166         return sdk.target
 167
 168
 169 class IqiyiIE(InfoExtractor):
 170     IE_NAME = 'iqiyi'
 171     IE_DESC = '爱奇艺'
 172
 173     _VALID_URL = r'https?://(?:(?:[^.]+\.)?iqiyi\.com|www\.pps\.tv)/.+\.html'
 174
 175     _NETRC_MACHINE = 'iqiyi'
 176
 177     _TESTS = [{
 178         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
 179         # MD5 checksum differs on my machine and Travis CI
 180         'info_dict': {
 181             'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
 182             'ext': 'mp4',
 183             'title': '美国德州空中惊现奇异云团 酷似UFO',
 184         }
 185     }, {
 186         'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
 187         'md5': 'b7dc800a4004b1b57749d9abae0472da',
 188         'info_dict': {
 189             'id': 'e3f585b550a280af23c98b6cb2be19fb',
 190             'ext': 'mp4',
 191             # This can be either Simplified Chinese or Traditional Chinese
 192             'title': r're:^(?:名侦探柯南 国语版：第752集 迫近灰原秘密的黑影 下篇|名偵探柯南 國語版：第752集 迫近灰原秘密的黑影 下篇)$',
 193         },
 194         'skip': 'Geo-restricted to China',
 195     }, {
 196         'url': 'http://www.iqiyi.com/w_19rt6o8t9p.html',
 197         'only_matching': True,
 198     }, {
 199         'url': 'http://www.iqiyi.com/a_19rrhbc6kt.html',
 200         'only_matching': True,
 201     }, {
 202         'url': 'http://yule.iqiyi.com/pcb.html',
 203         'info_dict': {
 204             'id': '4a0af228fddb55ec96398a364248ed7f',
 205             'ext': 'mp4',
 206             'title': '第2017-04-21期 女艺人频遭极端粉丝骚扰',
 207         },
 208     }, {
 209         # VIP-only video. The first 2 parts (6 minutes) are available without login
 210         # MD5 sums omitted as values are different on Travis CI and my machine
 211         'url': 'http://www.iqiyi.com/v_19rrny4w8w.html',
 212         'info_dict': {
 213             'id': 'f3cf468b39dddb30d676f89a91200dc1',
 214             'ext': 'mp4',
 215             'title': '泰坦尼克号',
 216         },
 217         'skip': 'Geo-restricted to China',
 218     }, {
 219         'url': 'http://www.iqiyi.com/a_19rrhb8ce1.html',
 220         'info_dict': {
 221             'id': '202918101',
 222             'title': '灌篮高手 国语版',
 223         },
 224         'playlist_count': 101,
 225     }, {
 226         'url': 'http://www.pps.tv/w_19rrbav0ph.html',
 227         'only_matching': True,
 228     }]
 229
 230     _FORMATS_MAP = {
 231         '96': 1,    # 216p, 240p
 232         '1': 2,     # 336p, 360p
 233         '2': 3,     # 480p, 504p
 234         '21': 4,    # 504p
 235         '4': 5,     # 720p
 236         '17': 5,    # 720p
 237         '5': 6,     # 1072p, 1080p
 238         '18': 7,    # 1080p
 239     }
 240
 241     @staticmethod
 242     def _rsa_fun(data):
 243         # public key extracted from http://static.iqiyi.com/js/qiyiV2/20160129180840/jobs/i18n/i18nIndex.js
 244         N = 0xab86b6371b5318aaa1d3c9e612a9f1264f372323c8c0f19875b5fc3b3fd3afcc1e5bec527aa94bfa85bffc157e4245aebda05389a5357b75115ac94f074aefcd
 245         e = 65537
 246
 247         return ohdave_rsa_encrypt(data, e, N)
 248
 249     def _perform_login(self, username, password):
 250
 251         data = self._download_json(
 252             'http://kylin.iqiyi.com/get_token', None,
 253             note='Get token for logging', errnote='Unable to get token for logging')
 254         sdk = data['sdk']
 255         timestamp = int(time.time())
 256         target = '/apis/reglogin/login.action?lang=zh_TW&area_code=null&email=%s&passwd=%s&agenttype=1&from=undefined&keeplogin=0&piccode=&fromurl=&_pos=1' % (
 257             username, self._rsa_fun(password.encode('utf-8')))
 258
 259         interp = IqiyiSDKInterpreter(sdk)
 260         sign = interp.run(target, data['ip'], timestamp)
 261
 262         validation_params = {
 263             'target': target,
 264             'server': 'BEA3AA1908656AABCCFF76582C4C6660',
 265             'token': data['token'],
 266             'bird_src': 'f8d91d57af224da7893dd397d52d811a',
 267             'sign': sign,
 268             'bird_t': timestamp,
 269         }
 270         validation_result = self._download_json(
 271             'http://kylin.iqiyi.com/validate?' + compat_urllib_parse_urlencode(validation_params), None,
 272             note='Validate credentials', errnote='Unable to validate credentials')
 273
 274         MSG_MAP = {
 275             'P00107': 'please login via the web interface and enter the CAPTCHA code',
 276             'P00117': 'bad username or password',
 277         }
 278
 279         code = validation_result['code']
 280         if code != 'A00000':
 281             msg = MSG_MAP.get(code)
 282             if not msg:
 283                 msg = 'error %s' % code
 284                 if validation_result.get('msg'):
 285                     msg += ': ' + validation_result['msg']
 286             self.report_warning('unable to log in: ' + msg)
 287             return False
 288
 289         return True
 290
 291     def get_raw_data(self, tvid, video_id):
 292         tm = int(time.time() * 1000)
 293
 294         key = 'd5fb4bd9d50c4be6948c97edd7254b0e'
 295         sc = md5_text(compat_str(tm) + key + tvid)
 296         params = {
 297             'tvid': tvid,
 298             'vid': video_id,
 299             'src': '76f90cbd92f94a2e925d83e8ccd22cb7',
 300             'sc': sc,
 301             't': tm,
 302         }
 303
 304         return self._download_json(
 305             'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id),
 306             video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='),
 307             query=params, headers=self.geo_verification_headers())
 308
 309     def _extract_playlist(self, webpage):
 310         PAGE_SIZE = 50
 311
 312         links = re.findall(
 313             r'<a[^>]+class="site-piclist_pic_link"[^>]+href="(http://www\.iqiyi\.com/.+\.html)"',
 314             webpage)
 315         if not links:
 316             return
 317
 318         album_id = self._search_regex(
 319             r'albumId\s*:\s*(\d+),', webpage, 'album ID')
 320         album_title = self._search_regex(
 321             r'data-share-title="([^"]+)"', webpage, 'album title', fatal=False)
 322
 323         entries = list(map(self.url_result, links))
 324
 325         # Start from 2 because links in the first page are already on webpage
 326         for page_num in itertools.count(2):
 327             pagelist_page = self._download_webpage(
 328                 'http://cache.video.qiyi.com/jp/avlist/%s/%d/%d/' % (album_id, page_num, PAGE_SIZE),
 329                 album_id,
 330                 note='Download playlist page %d' % page_num,
 331                 errnote='Failed to download playlist page %d' % page_num)
 332             pagelist = self._parse_json(
 333                 remove_start(pagelist_page, 'var tvInfoJs='), album_id)
 334             vlist = pagelist['data']['vlist']
 335             for item in vlist:
 336                 entries.append(self.url_result(item['vurl']))
 337             if len(vlist) < PAGE_SIZE:
 338                 break
 339
 340         return self.playlist_result(entries, album_id, album_title)
 341
 342     def _real_extract(self, url):
 343         webpage = self._download_webpage(
 344             url, 'temp_id', note='download video page')
 345
 346         # There's no simple way to determine whether an URL is a playlist or not
 347         # Sometimes there are playlist links in individual videos, so treat it
 348         # as a single video first
 349         tvid = self._search_regex(
 350             r'data-(?:player|shareplattrigger)-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid', default=None)
 351         if tvid is None:
 352             playlist_result = self._extract_playlist(webpage)
 353             if playlist_result:
 354                 return playlist_result
 355             raise ExtractorError('Can\'t find any video')
 356
 357         video_id = self._search_regex(
 358             r'data-(?:player|shareplattrigger)-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
 359
 360         formats = []
 361         for _ in range(5):
 362             raw_data = self.get_raw_data(tvid, video_id)
 363
 364             if raw_data['code'] != 'A00000':
 365                 if raw_data['code'] == 'A00111':
 366                     self.raise_geo_restricted()
 367                 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
 368
 369             data = raw_data['data']
 370
 371             for stream in data['vidl']:
 372                 if 'm3utx' not in stream:
 373                     continue
 374                 vd = compat_str(stream['vd'])
 375                 formats.append({
 376                     'url': stream['m3utx'],
 377                     'format_id': vd,
 378                     'ext': 'mp4',
 379                     'quality': self._FORMATS_MAP.get(vd, -1),
 380                     'protocol': 'm3u8_native',
 381                 })
 382
 383             if formats:
 384                 break
 385
 386             self._sleep(5, video_id)
 387
 388         title = (get_element_by_id('widget-videotitle', webpage)
 389                  or clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage))
 390                  or self._html_search_regex(r'<span[^>]+data-videochanged-title="word"[^>]*>([^<]+)</span>', webpage, 'title'))
 391
 392         return {
 393             'id': video_id,
 394             'title': title,
 395             'formats': formats,
 396         }
 397
 398
 399 class IqIE(InfoExtractor):
 400     IE_NAME = 'iq.com'
 401     IE_DESC = 'International version of iQiyi'
 402     _VALID_URL = r'https?://(?:www\.)?iq\.com/play/(?:[\w%-]*-)?(?P<id>\w+)'
 403     _TESTS = [{
 404         'url': 'https://www.iq.com/play/one-piece-episode-1000-1ma1i6ferf4',
 405         'md5': '2d7caf6eeca8a32b407094b33b757d39',
 406         'info_dict': {
 407             'ext': 'mp4',
 408             'id': '1ma1i6ferf4',
 409             'title': '航海王 第1000集',
 410             'description': 'Subtitle available on Sunday 4PM（GMT+8）.',
 411             'duration': 1430,
 412             'timestamp': 1637488203,
 413             'upload_date': '20211121',
 414             'episode_number': 1000,
 415             'episode': 'Episode 1000',
 416             'series': 'One Piece',
 417             'age_limit': 13,
 418             'average_rating': float,
 419         },
 420         'params': {
 421             'format': '500',
 422         },
 423         'expected_warnings': ['format is restricted']
 424     }, {
 425         # VIP-restricted video
 426         'url': 'https://www.iq.com/play/mermaid-in-the-fog-2021-gbdpx13bs4',
 427         'only_matching': True
 428     }]
 429     _BID_TAGS = {
 430         '100': '240P',
 431         '200': '360P',
 432         '300': '480P',
 433         '500': '720P',
 434         '600': '1080P',
 435         '610': '1080P50',
 436         '700': '2K',
 437         '800': '4K',
 438     }
 439     _LID_TAGS = {
 440         '1': 'zh_CN',
 441         '2': 'zh_TW',
 442         '3': 'en',
 443         '4': 'ko',
 444         '5': 'ja',
 445         '18': 'th',
 446         '21': 'my',
 447         '23': 'vi',
 448         '24': 'id',
 449         '26': 'es',
 450         '27': 'pt',
 451         '28': 'ar',
 452     }
 453
 454     _DASH_JS = '''
 455         console.log(page.evaluate(function() {
 456             var tvid = "%(tvid)s"; var vid = "%(vid)s"; var src = "%(src)s";
 457             var uid = "%(uid)s"; var dfp = "%(dfp)s"; var mode = "%(mode)s"; var lang = "%(lang)s";
 458             var bid_list = %(bid_list)s; var ut_list = %(ut_list)s; var tm = new Date().getTime();
 459             var cmd5x_func = %(cmd5x_func)s; var cmd5x_exporter = {}; cmd5x_func({}, cmd5x_exporter, {}); var cmd5x = cmd5x_exporter.cmd5x;
 460             var authKey = cmd5x(cmd5x('') + tm + '' + tvid);
 461             var k_uid = Array.apply(null, Array(32)).map(function() {return Math.floor(Math.random() * 15).toString(16)}).join('');
 462             var dash_paths = {};
 463             bid_list.forEach(function(bid) {
 464                 var query = {
 465                     'tvid': tvid,
 466                     'bid': bid,
 467                     'ds': 1,
 468                     'vid': vid,
 469                     'src': src,
 470                     'vt': 0,
 471                     'rs': 1,
 472                     'uid': uid,
 473                     'ori': 'pcw',
 474                     'ps': 1,
 475                     'k_uid': k_uid,
 476                     'pt': 0,
 477                     'd': 0,
 478                     's': '',
 479                     'lid': '',
 480                     'slid': 0,
 481                     'cf': '',
 482                     'ct': '',
 483                     'authKey': authKey,
 484                     'k_tag': 1,
 485                     'ost': 0,
 486                     'ppt': 0,
 487                     'dfp': dfp,
 488                     'prio': JSON.stringify({
 489                         'ff': 'f4v',
 490                         'code': 2
 491                     }),
 492                     'k_err_retries': 0,
 493                     'up': '',
 494                     'su': 2,
 495                     'applang': lang,
 496                     'sver': 2,
 497                     'X-USER-MODE': mode,
 498                     'qd_v': 2,
 499                     'tm': tm,
 500                     'qdy': 'a',
 501                     'qds': 0,
 502                     'k_ft1': '143486267424900',
 503                     'k_ft4': '1572868',
 504                     'k_ft7': '4',
 505                     'k_ft5': '1',
 506                     'bop': JSON.stringify({
 507                         'version': '10.0',
 508                         'dfp': dfp
 509                     }),
 510                 };
 511                 var enc_params = [];
 512                 for (var prop in query) {
 513                     enc_params.push(encodeURIComponent(prop) + '=' + encodeURIComponent(query[prop]));
 514                 }
 515                 ut_list.forEach(function(ut) {
 516                     enc_params.push('ut=' + ut);
 517                 })
 518                 var dash_path = '/dash?' + enc_params.join('&'); dash_path += '&vf=' + cmd5x(dash_path);
 519                 dash_paths[bid] = dash_path;
 520             });
 521             return JSON.stringify(dash_paths);
 522         }));
 523         saveAndExit();
 524     '''
 525
 526     def _extract_vms_player_js(self, webpage, video_id):
 527         player_js_cache = self.cache.load('iq', 'player_js')
 528         if player_js_cache:
 529             return player_js_cache
 530         webpack_js_url = self._proto_relative_url(self._search_regex(
 531             r'<script src="((?:https?:)?//stc\.iqiyipic\.com/_next/static/chunks/webpack-\w+\.js)"', webpage, 'webpack URL'))
 532         webpack_js = self._download_webpage(webpack_js_url, video_id, note='Downloading webpack JS', errnote='Unable to download webpack JS')
 533
 534         webpack_map = self._search_json(
 535             r'["\']\s*\+\s*', webpack_js, 'JS locations', video_id,
 536             contains_pattern=r'{\s*(?:\d+\s*:\s*["\'][\da-f]+["\']\s*,?\s*)+}',
 537             end_pattern=r'\[\w+\]\+["\']\.js', transform_source=js_to_json)
 538
 539         replacement_map = self._search_json(
 540             r'["\']\s*\+\(\s*', webpack_js, 'replacement map', video_id,
 541             contains_pattern=r'{\s*(?:\d+\s*:\s*["\'][\w.-]+["\']\s*,?\s*)+}',
 542             end_pattern=r'\[\w+\]\|\|\w+\)\+["\']\.', transform_source=js_to_json,
 543             fatal=False) or {}
 544
 545         for module_index in reversed(webpack_map):
 546             real_module = replacement_map.get(module_index) or module_index
 547             module_js = self._download_webpage(
 548                 f'https://stc.iqiyipic.com/_next/static/chunks/{real_module}.{webpack_map[module_index]}.js',
 549                 video_id, note=f'Downloading #{module_index} module JS', errnote='Unable to download module JS', fatal=False) or ''
 550             if 'vms request' in module_js:
 551                 self.cache.store('iq', 'player_js', module_js)
 552                 return module_js
 553         raise ExtractorError('Unable to extract player JS')
 554
 555     def _extract_cmd5x_function(self, webpage, video_id):
 556         return self._search_regex(r',\s*(function\s*\([^\)]*\)\s*{\s*var _qda.+_qdc\(\)\s*})\s*,',
 557                                   self._extract_vms_player_js(webpage, video_id), 'signature function')
 558
 559     def _update_bid_tags(self, webpage, video_id):
 560         extracted_bid_tags = self._search_json(
 561             r'function\s*\([^)]*\)\s*\{\s*"use strict";?\s*var \w\s*=\s*',
 562             self._extract_vms_player_js(webpage, video_id), 'video tags', video_id,
 563             contains_pattern=r'{\s*\d+\s*:\s*\{\s*nbid\s*:.+}\s*}',
 564             end_pattern=r'\s*,\s*\w\s*=\s*\{\s*getNewVd', fatal=False, transform_source=js_to_json)
 565         if not extracted_bid_tags:
 566             return
 567         self._BID_TAGS = {
 568             bid: traverse_obj(extracted_bid_tags, (bid, 'value'), expected_type=str, default=self._BID_TAGS.get(bid))
 569             for bid in extracted_bid_tags.keys()
 570         }
 571
 572     def _get_cookie(self, name, default=None):
 573         cookie = self._get_cookies('https://iq.com/').get(name)
 574         return cookie.value if cookie else default
 575
 576     def _real_extract(self, url):
 577         video_id = self._match_id(url)
 578         webpage = self._download_webpage(url, video_id)
 579         self._update_bid_tags(webpage, video_id)
 580
 581         next_props = self._search_nextjs_data(webpage, video_id)['props']
 582         page_data = next_props['initialState']['play']
 583         video_info = page_data['curVideoInfo']
 584
 585         uid = traverse_obj(
 586             self._parse_json(
 587                 self._get_cookie('I00002', '{}'), video_id, transform_source=compat_urllib_parse_unquote, fatal=False),
 588             ('data', 'uid'), default=0)
 589
 590         if uid:
 591             vip_data = self._download_json(
 592                 'https://pcw-api.iq.com/api/vtype', video_id, note='Downloading VIP data', errnote='Unable to download VIP data', query={
 593                     'batch': 1,
 594                     'platformId': 3,
 595                     'modeCode': self._get_cookie('mod', 'intl'),
 596                     'langCode': self._get_cookie('lang', 'en_us'),
 597                     'deviceId': self._get_cookie('QC005', '')
 598                 }, fatal=False)
 599             ut_list = traverse_obj(vip_data, ('data', 'all_vip', ..., 'vipType'), expected_type=str_or_none)
 600         else:
 601             ut_list = ['0']
 602
 603         # bid 0 as an initial format checker
 604         dash_paths = self._parse_json(PhantomJSwrapper(self, timeout=120_000).get(
 605             url, note2='Executing signature code (this may take a couple minutes)',
 606             html='<!DOCTYPE html>', video_id=video_id, jscode=self._DASH_JS % {
 607                 'tvid': video_info['tvId'],
 608                 'vid': video_info['vid'],
 609                 'src': traverse_obj(next_props, ('initialProps', 'pageProps', 'ptid'),
 610                                     expected_type=str, default='04022001010011000000'),
 611                 'uid': uid,
 612                 'dfp': self._get_cookie('dfp', ''),
 613                 'mode': self._get_cookie('mod', 'intl'),
 614                 'lang': self._get_cookie('lang', 'en_us'),
 615                 'bid_list': '[' + ','.join(['0', *self._BID_TAGS.keys()]) + ']',
 616                 'ut_list': '[' + ','.join(ut_list) + ']',
 617                 'cmd5x_func': self._extract_cmd5x_function(webpage, video_id),
 618             })[1].strip(), video_id)
 619
 620         formats, subtitles = [], {}
 621         initial_format_data = self._download_json(
 622             urljoin('https://cache-video.iq.com', dash_paths['0']), video_id,
 623             note='Downloading initial video format info', errnote='Unable to download initial video format info')['data']
 624
 625         preview_time = traverse_obj(
 626             initial_format_data, ('boss_ts', (None, 'data'), ('previewTime', 'rtime')), expected_type=float_or_none, get_all=False)
 627         if traverse_obj(initial_format_data, ('boss_ts', 'data', 'prv'), expected_type=int_or_none):
 628             self.report_warning('This preview video is limited%s' % format_field(preview_time, None, ' to %s seconds'))
 629
 630         # TODO: Extract audio-only formats
 631         for bid in set(traverse_obj(initial_format_data, ('program', 'video', ..., 'bid'), expected_type=str_or_none)):
 632             dash_path = dash_paths.get(bid)
 633             if not dash_path:
 634                 self.report_warning(f'Unknown format id: {bid}. It is currently not being extracted')
 635                 continue
 636             format_data = traverse_obj(self._download_json(
 637                 urljoin('https://cache-video.iq.com', dash_path), video_id,
 638                 note=f'Downloading format data for {self._BID_TAGS[bid]}', errnote='Unable to download format data',
 639                 fatal=False), 'data', expected_type=dict)
 640
 641             video_format = traverse_obj(format_data, ('program', 'video', lambda _, v: str(v['bid']) == bid),
 642                                         expected_type=dict, get_all=False) or {}
 643             extracted_formats = []
 644             if video_format.get('m3u8Url'):
 645                 extracted_formats.extend(self._extract_m3u8_formats(
 646                     urljoin(format_data.get('dm3u8', 'https://cache-m.iq.com/dc/dt/'), video_format['m3u8Url']),
 647                     'mp4', m3u8_id=bid, fatal=False))
 648             if video_format.get('mpdUrl'):
 649                 # TODO: Properly extract mpd hostname
 650                 extracted_formats.extend(self._extract_mpd_formats(
 651                     urljoin(format_data.get('dm3u8', 'https://cache-m.iq.com/dc/dt/'), video_format['mpdUrl']),
 652                     mpd_id=bid, fatal=False))
 653             if video_format.get('m3u8'):
 654                 ff = video_format.get('ff', 'ts')
 655                 if ff == 'ts':
 656                     m3u8_formats, _ = self._parse_m3u8_formats_and_subtitles(
 657                         video_format['m3u8'], ext='mp4', m3u8_id=bid, fatal=False)
 658                     extracted_formats.extend(m3u8_formats)
 659                 elif ff == 'm4s':
 660                     mpd_data = traverse_obj(
 661                         self._parse_json(video_format['m3u8'], video_id, fatal=False), ('payload', ..., 'data'), expected_type=str)
 662                     if not mpd_data:
 663                         continue
 664                     mpd_formats, _ = self._parse_mpd_formats_and_subtitles(
 665                         mpd_data, bid, format_data.get('dm3u8', 'https://cache-m.iq.com/dc/dt/'))
 666                     extracted_formats.extend(mpd_formats)
 667                 else:
 668                     self.report_warning(f'{ff} formats are currently not supported')
 669
 670             if not extracted_formats:
 671                 if video_format.get('s'):
 672                     self.report_warning(f'{self._BID_TAGS[bid]} format is restricted')
 673                 else:
 674                     self.report_warning(f'Unable to extract {self._BID_TAGS[bid]} format')
 675             for f in extracted_formats:
 676                 f.update({
 677                     'quality': qualities(list(self._BID_TAGS.keys()))(bid),
 678                     'format_note': self._BID_TAGS[bid],
 679                     **parse_resolution(video_format.get('scrsz'))
 680                 })
 681             formats.extend(extracted_formats)
 682
 683         for sub_format in traverse_obj(initial_format_data, ('program', 'stl', ...), expected_type=dict):
 684             lang = self._LID_TAGS.get(str_or_none(sub_format.get('lid')), sub_format.get('_name'))
 685             subtitles.setdefault(lang, []).extend([{
 686                 'ext': format_ext,
 687                 'url': urljoin(initial_format_data.get('dstl', 'http://meta.video.iqiyi.com'), sub_format[format_key])
 688             } for format_key, format_ext in [('srt', 'srt'), ('webvtt', 'vtt')] if sub_format.get(format_key)])
 689
 690         extra_metadata = page_data.get('albumInfo') if video_info.get('albumId') and page_data.get('albumInfo') else video_info
 691         return {
 692             'id': video_id,
 693             'title': video_info['name'],
 694             'formats': formats,
 695             'subtitles': subtitles,
 696             'description': video_info.get('mergeDesc'),
 697             'duration': parse_duration(video_info.get('len')),
 698             'age_limit': parse_age_limit(video_info.get('rating')),
 699             'average_rating': traverse_obj(page_data, ('playScoreInfo', 'score'), expected_type=float_or_none),
 700             'timestamp': parse_iso8601(video_info.get('isoUploadDate')),
 701             'categories': traverse_obj(extra_metadata, ('videoTagMap', ..., ..., 'name'), expected_type=str),
 702             'cast': traverse_obj(extra_metadata, ('actorArr', ..., 'name'), expected_type=str),
 703             'episode_number': int_or_none(video_info.get('order')) or None,
 704             'series': video_info.get('albumName'),
 705         }
 706
 707
 708 class IqAlbumIE(InfoExtractor):
 709     IE_NAME = 'iq.com:album'
 710     _VALID_URL = r'https?://(?:www\.)?iq\.com/album/(?:[\w%-]*-)?(?P<id>\w+)'
 711     _TESTS = [{
 712         'url': 'https://www.iq.com/album/one-piece-1999-1bk9icvr331',
 713         'info_dict': {
 714             'id': '1bk9icvr331',
 715             'title': 'One Piece',
 716             'description': 'Subtitle available on Sunday 4PM（GMT+8）.'
 717         },
 718         'playlist_mincount': 238
 719     }, {
 720         # Movie/single video
 721         'url': 'https://www.iq.com/album/九龙城寨-2021-22yjnij099k',
 722         'info_dict': {
 723             'ext': 'mp4',
 724             'id': '22yjnij099k',
 725             'title': '九龙城寨',
 726             'description': 'md5:8a09f50b8ba0db4dc69bc7c844228044',
 727             'duration': 5000,
 728             'timestamp': 1641911371,
 729             'upload_date': '20220111',
 730             'series': '九龙城寨',
 731             'cast': ['Shi Yan Neng', 'Yu Lang', 'Peter  lv', 'Sun Zi Jun', 'Yang Xiao Bo'],
 732             'age_limit': 13,
 733             'average_rating': float,
 734         },
 735         'expected_warnings': ['format is restricted']
 736     }]
 737
 738     def _entries(self, album_id_num, page_ranges, album_id=None, mode_code='intl', lang_code='en_us'):
 739         for page_range in page_ranges:
 740             page = self._download_json(
 741                 f'https://pcw-api.iq.com/api/episodeListSource/{album_id_num}', album_id,
 742                 note=f'Downloading video list episodes {page_range.get("msg", "")}',
 743                 errnote='Unable to download video list', query={
 744                     'platformId': 3,
 745                     'modeCode': mode_code,
 746                     'langCode': lang_code,
 747                     'endOrder': page_range['to'],
 748                     'startOrder': page_range['from']
 749                 })
 750             for video in page['data']['epg']:
 751                 yield self.url_result('https://www.iq.com/play/%s' % (video.get('playLocSuffix') or video['qipuIdStr']),
 752                                       IqIE.ie_key(), video.get('qipuIdStr'), video.get('name'))
 753
 754     def _real_extract(self, url):
 755         album_id = self._match_id(url)
 756         webpage = self._download_webpage(url, album_id)
 757         next_data = self._search_nextjs_data(webpage, album_id)
 758         album_data = next_data['props']['initialState']['album']['videoAlbumInfo']
 759
 760         if album_data.get('videoType') == 'singleVideo':
 761             return self.url_result('https://www.iq.com/play/%s' % album_id, IqIE.ie_key())
 762         return self.playlist_result(
 763             self._entries(album_data['albumId'], album_data['totalPageRange'], album_id,
 764                           traverse_obj(next_data, ('props', 'initialProps', 'pageProps', 'modeCode')),
 765                           traverse_obj(next_data, ('props', 'initialProps', 'pageProps', 'langCode'))),
 766             album_id, album_data.get('name'), album_data.get('desc'))