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