You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

982 lines
39 KiB

10 years ago
11 years ago
11 years ago
11 years ago
  1. # encoding: utf-8
  2. from __future__ import unicode_literals
  3. import os
  4. import re
  5. from .common import InfoExtractor
  6. from .youtube import YoutubeIE
  7. from ..utils import (
  8. compat_urllib_parse,
  9. compat_urlparse,
  10. compat_xml_parse_error,
  11. determine_ext,
  12. ExtractorError,
  13. float_or_none,
  14. HEADRequest,
  15. orderedSet,
  16. parse_xml,
  17. smuggle_url,
  18. unescapeHTML,
  19. unified_strdate,
  20. unsmuggle_url,
  21. url_basename,
  22. )
  23. from .brightcove import BrightcoveIE
  24. from .ooyala import OoyalaIE
  25. from .rutv import RUTVIE
  26. from .smotri import SmotriIE
  27. from .condenast import CondeNastIE
  28. class GenericIE(InfoExtractor):
  29. IE_DESC = 'Generic downloader that works on some sites'
  30. _VALID_URL = r'.*'
  31. IE_NAME = 'generic'
  32. _TESTS = [
  33. {
  34. 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  35. 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  36. 'info_dict': {
  37. 'id': '13601338388002',
  38. 'ext': 'mp4',
  39. 'uploader': 'www.hodiho.fr',
  40. 'title': 'R\u00e9gis plante sa Jeep',
  41. }
  42. },
  43. # bandcamp page with custom domain
  44. {
  45. 'add_ie': ['Bandcamp'],
  46. 'url': 'http://bronyrock.com/track/the-pony-mash',
  47. 'info_dict': {
  48. 'id': '3235767654',
  49. 'ext': 'mp3',
  50. 'title': 'The Pony Mash',
  51. 'uploader': 'M_Pallante',
  52. },
  53. 'skip': 'There is a limit of 200 free downloads / month for the test song',
  54. },
  55. # embedded brightcove video
  56. # it also tests brightcove videos that need to set the 'Referer' in the
  57. # http requests
  58. {
  59. 'add_ie': ['Brightcove'],
  60. 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  61. 'info_dict': {
  62. 'id': '2765128793001',
  63. 'ext': 'mp4',
  64. 'title': 'Le cours de bourse : l’analyse technique',
  65. 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  66. 'uploader': 'BFM BUSINESS',
  67. },
  68. 'params': {
  69. 'skip_download': True,
  70. },
  71. },
  72. {
  73. # https://github.com/rg3/youtube-dl/issues/2253
  74. 'url': 'http://bcove.me/i6nfkrc3',
  75. 'md5': '0ba9446db037002366bab3b3eb30c88c',
  76. 'info_dict': {
  77. 'id': '3101154703001',
  78. 'ext': 'mp4',
  79. 'title': 'Still no power',
  80. 'uploader': 'thestar.com',
  81. 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
  82. },
  83. 'add_ie': ['Brightcove'],
  84. },
  85. {
  86. 'url': 'http://www.championat.com/video/football/v/87/87499.html',
  87. 'md5': 'fb973ecf6e4a78a67453647444222983',
  88. 'info_dict': {
  89. 'id': '3414141473001',
  90. 'ext': 'mp4',
  91. 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
  92. 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
  93. 'uploader': 'Championat',
  94. },
  95. },
  96. # Direct link to a video
  97. {
  98. 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  99. 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  100. 'info_dict': {
  101. 'id': 'trailer',
  102. 'ext': 'mp4',
  103. 'title': 'trailer',
  104. 'upload_date': '20100513',
  105. }
  106. },
  107. # ooyala video
  108. {
  109. 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
  110. 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
  111. 'info_dict': {
  112. 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
  113. 'ext': 'mp4',
  114. 'title': '2cc213299525360.mov', # that's what we get
  115. },
  116. },
  117. # google redirect
  118. {
  119. 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
  120. 'info_dict': {
  121. 'id': 'cmQHVoWB5FY',
  122. 'ext': 'mp4',
  123. 'upload_date': '20130224',
  124. 'uploader_id': 'TheVerge',
  125. 'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.',
  126. 'uploader': 'The Verge',
  127. 'title': 'First Firefox OS phones side-by-side',
  128. },
  129. 'params': {
  130. 'skip_download': False,
  131. }
  132. },
  133. # embed.ly video
  134. {
  135. 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
  136. 'info_dict': {
  137. 'id': '9ODmcdjQcHQ',
  138. 'ext': 'mp4',
  139. 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
  140. 'upload_date': '20140225',
  141. 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
  142. 'uploader': 'Tested',
  143. 'uploader_id': 'testedcom',
  144. },
  145. # No need to test YoutubeIE here
  146. 'params': {
  147. 'skip_download': True,
  148. },
  149. },
  150. # funnyordie embed
  151. {
  152. 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
  153. 'info_dict': {
  154. 'id': '18e820ec3f',
  155. 'ext': 'mp4',
  156. 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
  157. 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
  158. },
  159. },
  160. # RUTV embed
  161. {
  162. 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
  163. 'info_dict': {
  164. 'id': '776940',
  165. 'ext': 'mp4',
  166. 'title': 'Охотское море стало целиком российским',
  167. 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
  168. },
  169. 'params': {
  170. # m3u8 download
  171. 'skip_download': True,
  172. },
  173. },
  174. # Embedded TED video
  175. {
  176. 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
  177. 'md5': '65fdff94098e4a607385a60c5177c638',
  178. 'info_dict': {
  179. 'id': '1969',
  180. 'ext': 'mp4',
  181. 'title': 'Hidden miracles of the natural world',
  182. 'uploader': 'Louie Schwartzberg',
  183. 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
  184. }
  185. },
  186. # Embeded Ustream video
  187. {
  188. 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
  189. 'md5': '27b99cdb639c9b12a79bca876a073417',
  190. 'info_dict': {
  191. 'id': '45734260',
  192. 'ext': 'flv',
  193. 'uploader': 'AU SPA: The NSA and Privacy',
  194. 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
  195. }
  196. },
  197. # nowvideo embed hidden behind percent encoding
  198. {
  199. 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
  200. 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
  201. 'info_dict': {
  202. 'id': '06e53103ca9aa',
  203. 'ext': 'flv',
  204. 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
  205. 'description': 'No description',
  206. },
  207. },
  208. # arte embed
  209. {
  210. 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
  211. 'md5': '7653032cbb25bf6c80d80f217055fa43',
  212. 'info_dict': {
  213. 'id': '048195-004_PLUS7-F',
  214. 'ext': 'flv',
  215. 'title': 'X:enius',
  216. 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
  217. 'upload_date': '20140320',
  218. },
  219. 'params': {
  220. 'skip_download': 'Requires rtmpdump'
  221. }
  222. },
  223. # Condé Nast embed
  224. {
  225. 'url': 'http://www.wired.com/2014/04/honda-asimo/',
  226. 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
  227. 'info_dict': {
  228. 'id': '53501be369702d3275860000',
  229. 'ext': 'mp4',
  230. 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
  231. }
  232. },
  233. # Dailymotion embed
  234. {
  235. 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
  236. 'md5': '441aeeb82eb72c422c7f14ec533999cd',
  237. 'info_dict': {
  238. 'id': 'k2mm4bCdJ6CQ2i7c8o2',
  239. 'ext': 'mp4',
  240. 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
  241. 'uploader': 'Spi0n',
  242. },
  243. 'add_ie': ['Dailymotion'],
  244. },
  245. # YouTube embed
  246. {
  247. 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
  248. 'info_dict': {
  249. 'id': 'FXRb4ykk4S0',
  250. 'ext': 'mp4',
  251. 'title': 'The NBL Auction 2014',
  252. 'uploader': 'BADMINTON England',
  253. 'uploader_id': 'BADMINTONEvents',
  254. 'upload_date': '20140603',
  255. 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
  256. },
  257. 'add_ie': ['Youtube'],
  258. 'params': {
  259. 'skip_download': True,
  260. }
  261. },
  262. # MTVSercices embed
  263. {
  264. 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
  265. 'md5': '35727f82f58c76d996fc188f9755b0d5',
  266. 'info_dict': {
  267. 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
  268. 'ext': 'mp4',
  269. 'title': 'Review',
  270. 'description': 'Mario\'s life in the fast lane has never looked so good.',
  271. },
  272. },
  273. # YouTube embed via <data-embed-url="">
  274. {
  275. 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
  276. 'info_dict': {
  277. 'id': '4vAffPZIT44',
  278. 'ext': 'mp4',
  279. 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
  280. 'uploader': 'Gameloft',
  281. 'uploader_id': 'gameloft',
  282. 'upload_date': '20140828',
  283. 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
  284. },
  285. 'params': {
  286. 'skip_download': True,
  287. }
  288. },
  289. # Camtasia studio
  290. {
  291. 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
  292. 'playlist': [{
  293. 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
  294. 'info_dict': {
  295. 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
  296. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
  297. 'ext': 'flv',
  298. 'duration': 2235.90,
  299. }
  300. }, {
  301. 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
  302. 'info_dict': {
  303. 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
  304. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
  305. 'ext': 'flv',
  306. 'duration': 2235.93,
  307. }
  308. }],
  309. 'info_dict': {
  310. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
  311. }
  312. },
  313. # Flowplayer
  314. {
  315. 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
  316. 'md5': '9d65602bf31c6e20014319c7d07fba27',
  317. 'info_dict': {
  318. 'id': '5123ea6d5e5a7',
  319. 'ext': 'mp4',
  320. 'age_limit': 18,
  321. 'uploader': 'www.handjobhub.com',
  322. 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
  323. }
  324. },
  325. # RSS feed
  326. {
  327. 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
  328. 'info_dict': {
  329. 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
  330. 'title': 'Zero Punctuation',
  331. 'description': 're:'
  332. },
  333. 'playlist_mincount': 11,
  334. },
  335. # Multiple brightcove videos
  336. # https://github.com/rg3/youtube-dl/issues/2283
  337. {
  338. 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
  339. 'info_dict': {
  340. 'id': 'always-never',
  341. 'title': 'Always / Never - The New Yorker',
  342. },
  343. 'playlist_count': 3,
  344. 'params': {
  345. 'extract_flat': False,
  346. 'skip_download': True,
  347. }
  348. },
  349. # MLB embed
  350. {
  351. 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
  352. 'md5': '96f09a37e44da40dd083e12d9a683327',
  353. 'info_dict': {
  354. 'id': '33322633',
  355. 'ext': 'mp4',
  356. 'title': 'Ump changes call to ball',
  357. 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
  358. 'duration': 48,
  359. 'timestamp': 1401537900,
  360. 'upload_date': '20140531',
  361. 'thumbnail': 're:^https?://.*\.jpg$',
  362. },
  363. },
  364. # Wistia embed
  365. {
  366. 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
  367. 'md5': '8788b683c777a5cf25621eaf286d0c23',
  368. 'info_dict': {
  369. 'id': '1cfaf6b7ea',
  370. 'ext': 'mov',
  371. 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
  372. 'duration': 643.0,
  373. 'filesize': 182808282,
  374. 'uploader': 'education-portal.com',
  375. },
  376. },
  377. {
  378. 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
  379. 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
  380. 'info_dict': {
  381. 'id': 'uxjb0lwrcz',
  382. 'ext': 'mp4',
  383. 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
  384. 'duration': 1715.0,
  385. 'uploader': 'thoughtworks.wistia.com',
  386. },
  387. },
  388. # Direct download with broken HEAD
  389. {
  390. 'url': 'http://ai-radio.org:8000/radio.opus',
  391. 'info_dict': {
  392. 'id': 'radio',
  393. 'ext': 'opus',
  394. 'title': 'radio',
  395. },
  396. 'params': {
  397. 'skip_download': True, # infinite live stream
  398. },
  399. 'expected_warnings': [
  400. r'501.*Not Implemented'
  401. ],
  402. }
  403. ]
  404. def report_following_redirect(self, new_url):
  405. """Report information extraction."""
  406. self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
  407. def _extract_rss(self, url, video_id, doc):
  408. playlist_title = doc.find('./channel/title').text
  409. playlist_desc_el = doc.find('./channel/description')
  410. playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
  411. entries = [{
  412. '_type': 'url',
  413. 'url': e.find('link').text,
  414. 'title': e.find('title').text,
  415. } for e in doc.findall('./channel/item')]
  416. return {
  417. '_type': 'playlist',
  418. 'id': url,
  419. 'title': playlist_title,
  420. 'description': playlist_desc,
  421. 'entries': entries,
  422. }
  423. def _extract_camtasia(self, url, video_id, webpage):
  424. """ Returns None if no camtasia video can be found. """
  425. camtasia_cfg = self._search_regex(
  426. r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
  427. webpage, 'camtasia configuration file', default=None)
  428. if camtasia_cfg is None:
  429. return None
  430. title = self._html_search_meta('DC.title', webpage, fatal=True)
  431. camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
  432. camtasia_cfg = self._download_xml(
  433. camtasia_url, video_id,
  434. note='Downloading camtasia configuration',
  435. errnote='Failed to download camtasia configuration')
  436. fileset_node = camtasia_cfg.find('./playlist/array/fileset')
  437. entries = []
  438. for n in fileset_node.getchildren():
  439. url_n = n.find('./uri')
  440. if url_n is None:
  441. continue
  442. entries.append({
  443. 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
  444. 'title': '%s - %s' % (title, n.tag),
  445. 'url': compat_urlparse.urljoin(url, url_n.text),
  446. 'duration': float_or_none(n.find('./duration').text),
  447. })
  448. return {
  449. '_type': 'playlist',
  450. 'entries': entries,
  451. 'title': title,
  452. }
  453. def _real_extract(self, url):
  454. if url.startswith('//'):
  455. return {
  456. '_type': 'url',
  457. 'url': self.http_scheme() + url,
  458. }
  459. parsed_url = compat_urlparse.urlparse(url)
  460. if not parsed_url.scheme:
  461. default_search = self._downloader.params.get('default_search')
  462. if default_search is None:
  463. default_search = 'fixup_error'
  464. if default_search in ('auto', 'auto_warning', 'fixup_error'):
  465. if '/' in url:
  466. self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
  467. return self.url_result('http://' + url)
  468. elif default_search != 'fixup_error':
  469. if default_search == 'auto_warning':
  470. if re.match(r'^(?:url|URL)$', url):
  471. raise ExtractorError(
  472. 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
  473. expected=True)
  474. else:
  475. self._downloader.report_warning(
  476. 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
  477. return self.url_result('ytsearch:' + url)
  478. if default_search in ('error', 'fixup_error'):
  479. raise ExtractorError(
  480. ('%r is not a valid URL. '
  481. 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
  482. ) % (url, url), expected=True)
  483. else:
  484. if ':' not in default_search:
  485. default_search += ':'
  486. return self.url_result(default_search + url)
  487. url, smuggled_data = unsmuggle_url(url)
  488. force_videoid = None
  489. is_intentional = smuggled_data and smuggled_data.get('to_generic')
  490. if smuggled_data and 'force_videoid' in smuggled_data:
  491. force_videoid = smuggled_data['force_videoid']
  492. video_id = force_videoid
  493. else:
  494. video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
  495. self.to_screen('%s: Requesting header' % video_id)
  496. head_req = HEADRequest(url)
  497. head_response = self._request_webpage(
  498. head_req, video_id,
  499. note=False, errnote='Could not send HEAD request to %s' % url,
  500. fatal=False)
  501. if head_response is not False:
  502. # Check for redirect
  503. new_url = head_response.geturl()
  504. if url != new_url:
  505. self.report_following_redirect(new_url)
  506. if force_videoid:
  507. new_url = smuggle_url(
  508. new_url, {'force_videoid': force_videoid})
  509. return self.url_result(new_url)
  510. full_response = None
  511. if head_response is False:
  512. full_response = self._request_webpage(url, video_id)
  513. head_response = full_response
  514. # Check for direct link to a video
  515. content_type = head_response.headers.get('Content-Type', '')
  516. m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
  517. if m:
  518. upload_date = unified_strdate(
  519. head_response.headers.get('Last-Modified'))
  520. return {
  521. 'id': video_id,
  522. 'title': os.path.splitext(url_basename(url))[0],
  523. 'formats': [{
  524. 'format_id': m.group('format_id'),
  525. 'url': url,
  526. 'vcodec': 'none' if m.group('type') == 'audio' else None
  527. }],
  528. 'upload_date': upload_date,
  529. }
  530. if not self._downloader.params.get('test', False) and not is_intentional:
  531. self._downloader.report_warning('Falling back on generic information extractor.')
  532. if full_response:
  533. webpage = self._webpage_read_content(full_response, url, video_id)
  534. else:
  535. webpage = self._download_webpage(url, video_id)
  536. self.report_extraction(video_id)
  537. # Is it an RSS feed?
  538. try:
  539. doc = parse_xml(webpage)
  540. if doc.tag == 'rss':
  541. return self._extract_rss(url, video_id, doc)
  542. except compat_xml_parse_error:
  543. pass
  544. # Is it a Camtasia project?
  545. camtasia_res = self._extract_camtasia(url, video_id, webpage)
  546. if camtasia_res is not None:
  547. return camtasia_res
  548. # Sometimes embedded video player is hidden behind percent encoding
  549. # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
  550. # Unescaping the whole page allows to handle those cases in a generic way
  551. webpage = compat_urllib_parse.unquote(webpage)
  552. # it's tempting to parse this further, but you would
  553. # have to take into account all the variations like
  554. # Video Title - Site Name
  555. # Site Name | Video Title
  556. # Video Title - Tagline | Site Name
  557. # and so on and so forth; it's just not practical
  558. video_title = self._html_search_regex(
  559. r'(?s)<title>(.*?)</title>', webpage, 'video title',
  560. default='video')
  561. # Try to detect age limit automatically
  562. age_limit = self._rta_search(webpage)
  563. # And then there are the jokers who advertise that they use RTA,
  564. # but actually don't.
  565. AGE_LIMIT_MARKERS = [
  566. r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
  567. ]
  568. if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
  569. age_limit = 18
  570. # video uploader is domain name
  571. video_uploader = self._search_regex(
  572. r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
  573. # Helper method
  574. def _playlist_from_matches(matches, getter, ie=None):
  575. urlrs = orderedSet(
  576. self.url_result(self._proto_relative_url(getter(m)), ie)
  577. for m in matches)
  578. return self.playlist_result(
  579. urlrs, playlist_id=video_id, playlist_title=video_title)
  580. # Look for BrightCove:
  581. bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
  582. if bc_urls:
  583. self.to_screen('Brightcove video detected.')
  584. entries = [{
  585. '_type': 'url',
  586. 'url': smuggle_url(bc_url, {'Referer': url}),
  587. 'ie_key': 'Brightcove'
  588. } for bc_url in bc_urls]
  589. return {
  590. '_type': 'playlist',
  591. 'title': video_title,
  592. 'id': video_id,
  593. 'entries': entries,
  594. }
  595. # Look for embedded (iframe) Vimeo player
  596. mobj = re.search(
  597. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
  598. if mobj:
  599. player_url = unescapeHTML(mobj.group('url'))
  600. surl = smuggle_url(player_url, {'Referer': url})
  601. return self.url_result(surl)
  602. # Look for embedded (swf embed) Vimeo player
  603. mobj = re.search(
  604. r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
  605. if mobj:
  606. return self.url_result(mobj.group(1))
  607. # Look for embedded YouTube player
  608. matches = re.findall(r'''(?x)
  609. (?:
  610. <iframe[^>]+?src=|
  611. data-video-url=|
  612. <embed[^>]+?src=|
  613. embedSWF\(?:\s*|
  614. new\s+SWFObject\(
  615. )
  616. (["\'])
  617. (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
  618. (?:embed|v|p)/.+?)
  619. \1''', webpage)
  620. if matches:
  621. return _playlist_from_matches(
  622. matches, lambda m: unescapeHTML(m[1]))
  623. # Look for embedded Dailymotion player
  624. matches = re.findall(
  625. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
  626. if matches:
  627. return _playlist_from_matches(
  628. matches, lambda m: unescapeHTML(m[1]))
  629. # Look for embedded Dailymotion playlist player (#3822)
  630. m = re.search(
  631. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
  632. if m:
  633. playlists = re.findall(
  634. r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
  635. if playlists:
  636. return _playlist_from_matches(
  637. playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
  638. # Look for embedded Wistia player
  639. match = re.search(
  640. r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
  641. if match:
  642. embed_url = self._proto_relative_url(
  643. unescapeHTML(match.group('url')))
  644. return {
  645. '_type': 'url_transparent',
  646. 'url': embed_url,
  647. 'ie_key': 'Wistia',
  648. 'uploader': video_uploader,
  649. 'title': video_title,
  650. 'id': video_id,
  651. }
  652. match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
  653. if match:
  654. return {
  655. '_type': 'url_transparent',
  656. 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
  657. 'ie_key': 'Wistia',
  658. 'uploader': video_uploader,
  659. 'title': video_title,
  660. 'id': match.group('id')
  661. }
  662. # Look for embedded blip.tv player
  663. mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
  664. if mobj:
  665. return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
  666. mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
  667. if mobj:
  668. return self.url_result(mobj.group(1), 'BlipTV')
  669. # Look for embedded condenast player
  670. matches = re.findall(
  671. r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
  672. webpage)
  673. if matches:
  674. return {
  675. '_type': 'playlist',
  676. 'entries': [{
  677. '_type': 'url',
  678. 'ie_key': 'CondeNast',
  679. 'url': ma,
  680. } for ma in matches],
  681. 'title': video_title,
  682. 'id': video_id,
  683. }
  684. # Look for Bandcamp pages with custom domain
  685. mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
  686. if mobj is not None:
  687. burl = unescapeHTML(mobj.group(1))
  688. # Don't set the extractor because it can be a track url or an album
  689. return self.url_result(burl)
  690. # Look for embedded Vevo player
  691. mobj = re.search(
  692. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
  693. if mobj is not None:
  694. return self.url_result(mobj.group('url'))
  695. # Look for Ooyala videos
  696. mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
  697. re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
  698. if mobj is not None:
  699. return OoyalaIE._build_url_result(mobj.group('ec'))
  700. # Look for Aparat videos
  701. mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
  702. if mobj is not None:
  703. return self.url_result(mobj.group(1), 'Aparat')
  704. # Look for MPORA videos
  705. mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
  706. if mobj is not None:
  707. return self.url_result(mobj.group(1), 'Mpora')
  708. # Look for embedded NovaMov-based player
  709. mobj = re.search(
  710. r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
  711. (?P<url>http://(?:(?:embed|www)\.)?
  712. (?:novamov\.com|
  713. nowvideo\.(?:ch|sx|eu|at|ag|co)|
  714. videoweed\.(?:es|com)|
  715. movshare\.(?:net|sx|ag)|
  716. divxstage\.(?:eu|net|ch|co|at|ag))
  717. /embed\.php.+?)\1''', webpage)
  718. if mobj is not None:
  719. return self.url_result(mobj.group('url'))
  720. # Look for embedded Facebook player
  721. mobj = re.search(
  722. r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
  723. if mobj is not None:
  724. return self.url_result(mobj.group('url'), 'Facebook')
  725. # Look for embedded VK player
  726. mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
  727. if mobj is not None:
  728. return self.url_result(mobj.group('url'), 'VK')
  729. # Look for embedded ivi player
  730. mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
  731. if mobj is not None:
  732. return self.url_result(mobj.group('url'), 'Ivi')
  733. # Look for embedded Huffington Post player
  734. mobj = re.search(
  735. r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
  736. if mobj is not None:
  737. return self.url_result(mobj.group('url'), 'HuffPost')
  738. # Look for embed.ly
  739. mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
  740. if mobj is not None:
  741. return self.url_result(mobj.group('url'))
  742. mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
  743. if mobj is not None:
  744. return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
  745. # Look for funnyordie embed
  746. matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
  747. if matches:
  748. return _playlist_from_matches(
  749. matches, getter=unescapeHTML, ie='FunnyOrDie')
  750. # Look for embedded RUTV player
  751. rutv_url = RUTVIE._extract_url(webpage)
  752. if rutv_url:
  753. return self.url_result(rutv_url, 'RUTV')
  754. # Look for embedded TED player
  755. mobj = re.search(
  756. r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
  757. if mobj is not None:
  758. return self.url_result(mobj.group('url'), 'TED')
  759. # Look for embedded Ustream videos
  760. mobj = re.search(
  761. r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
  762. if mobj is not None:
  763. return self.url_result(mobj.group('url'), 'Ustream')
  764. # Look for embedded arte.tv player
  765. mobj = re.search(
  766. r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
  767. webpage)
  768. if mobj is not None:
  769. return self.url_result(mobj.group('url'), 'ArteTVEmbed')
  770. # Look for embedded smotri.com player
  771. smotri_url = SmotriIE._extract_url(webpage)
  772. if smotri_url:
  773. return self.url_result(smotri_url, 'Smotri')
  774. # Look for embeded soundcloud player
  775. mobj = re.search(
  776. r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
  777. webpage)
  778. if mobj is not None:
  779. url = unescapeHTML(mobj.group('url'))
  780. return self.url_result(url)
  781. # Look for embedded vulture.com player
  782. mobj = re.search(
  783. r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
  784. webpage)
  785. if mobj is not None:
  786. url = unescapeHTML(mobj.group('url'))
  787. return self.url_result(url, ie='Vulture')
  788. # Look for embedded mtvservices player
  789. mobj = re.search(
  790. r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
  791. webpage)
  792. if mobj is not None:
  793. url = unescapeHTML(mobj.group('url'))
  794. return self.url_result(url, ie='MTVServicesEmbedded')
  795. # Look for embedded yahoo player
  796. mobj = re.search(
  797. r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
  798. webpage)
  799. if mobj is not None:
  800. return self.url_result(mobj.group('url'), 'Yahoo')
  801. # Look for embedded sbs.com.au player
  802. mobj = re.search(
  803. r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
  804. webpage)
  805. if mobj is not None:
  806. return self.url_result(mobj.group('url'), 'SBS')
  807. mobj = re.search(
  808. r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
  809. webpage)
  810. if mobj is not None:
  811. return self.url_result(mobj.group('url'), 'MLB')
  812. mobj = re.search(
  813. r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
  814. webpage)
  815. if mobj is not None:
  816. return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
  817. def check_video(vurl):
  818. vpath = compat_urlparse.urlparse(vurl).path
  819. vext = determine_ext(vpath)
  820. return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
  821. def filter_video(urls):
  822. return list(filter(check_video, urls))
  823. # Start with something easy: JW Player in SWFObject
  824. found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
  825. if not found:
  826. # Look for gorilla-vid style embedding
  827. found = filter_video(re.findall(r'''(?sx)
  828. (?:
  829. jw_plugins|
  830. JWPlayerOptions|
  831. jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
  832. )
  833. .*?file\s*:\s*["\'](.*?)["\']''', webpage))
  834. if not found:
  835. # Broaden the search a little bit
  836. found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
  837. if not found:
  838. # Broaden the findall a little bit: JWPlayer JS loader
  839. found = filter_video(re.findall(
  840. r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
  841. if not found:
  842. # Flow player
  843. found = filter_video(re.findall(r'''(?xs)
  844. flowplayer\("[^"]+",\s*
  845. \{[^}]+?\}\s*,
  846. \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
  847. ["']?url["']?\s*:\s*["']([^"']+)["']
  848. ''', webpage))
  849. if not found:
  850. # Try to find twitter cards info
  851. found = filter_video(re.findall(
  852. r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
  853. if not found:
  854. # We look for Open Graph info:
  855. # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
  856. m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
  857. # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
  858. if m_video_type is not None:
  859. found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
  860. if not found:
  861. # HTML5 video
  862. found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
  863. if not found:
  864. found = re.search(
  865. r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
  866. r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
  867. webpage)
  868. if found:
  869. new_url = found.group(1)
  870. self.report_following_redirect(new_url)
  871. return {
  872. '_type': 'url',
  873. 'url': new_url,
  874. }
  875. if not found:
  876. raise ExtractorError('Unsupported URL: %s' % url)
  877. entries = []
  878. for video_url in found:
  879. video_url = compat_urlparse.urljoin(url, video_url)
  880. video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
  881. # Sometimes, jwplayer extraction will result in a YouTube URL
  882. if YoutubeIE.suitable(video_url):
  883. entries.append(self.url_result(video_url, 'Youtube'))
  884. continue
  885. # here's a fun little line of code for you:
  886. video_id = os.path.splitext(video_id)[0]
  887. entries.append({
  888. 'id': video_id,
  889. 'url': video_url,
  890. 'uploader': video_uploader,
  891. 'title': video_title,
  892. 'age_limit': age_limit,
  893. })
  894. if len(entries) == 1:
  895. return entries[0]
  896. else:
  897. for num, e in enumerate(entries, start=1):
  898. e['title'] = '%s (%d)' % (e['title'], num)
  899. return {
  900. '_type': 'playlist',
  901. 'entries': entries,
  902. }