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.

1028 lines
40 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 ..compat import (
  8. compat_urllib_parse,
  9. compat_urlparse,
  10. compat_xml_parse_error,
  11. )
  12. from ..utils import (
  13. determine_ext,
  14. ExtractorError,
  15. float_or_none,
  16. HEADRequest,
  17. orderedSet,
  18. parse_xml,
  19. smuggle_url,
  20. unescapeHTML,
  21. unified_strdate,
  22. unsmuggle_url,
  23. url_basename,
  24. )
  25. from .brightcove import BrightcoveIE
  26. from .ooyala import OoyalaIE
  27. from .rutv import RUTVIE
  28. from .smotri import SmotriIE
  29. from .condenast import CondeNastIE
  30. class GenericIE(InfoExtractor):
  31. IE_DESC = 'Generic downloader that works on some sites'
  32. _VALID_URL = r'.*'
  33. IE_NAME = 'generic'
  34. _TESTS = [
  35. {
  36. 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  37. 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  38. 'info_dict': {
  39. 'id': '13601338388002',
  40. 'ext': 'mp4',
  41. 'uploader': 'www.hodiho.fr',
  42. 'title': 'R\u00e9gis plante sa Jeep',
  43. }
  44. },
  45. # bandcamp page with custom domain
  46. {
  47. 'add_ie': ['Bandcamp'],
  48. 'url': 'http://bronyrock.com/track/the-pony-mash',
  49. 'info_dict': {
  50. 'id': '3235767654',
  51. 'ext': 'mp3',
  52. 'title': 'The Pony Mash',
  53. 'uploader': 'M_Pallante',
  54. },
  55. 'skip': 'There is a limit of 200 free downloads / month for the test song',
  56. },
  57. # embedded brightcove video
  58. # it also tests brightcove videos that need to set the 'Referer' in the
  59. # http requests
  60. {
  61. 'add_ie': ['Brightcove'],
  62. 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  63. 'info_dict': {
  64. 'id': '2765128793001',
  65. 'ext': 'mp4',
  66. 'title': 'Le cours de bourse : l’analyse technique',
  67. 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  68. 'uploader': 'BFM BUSINESS',
  69. },
  70. 'params': {
  71. 'skip_download': True,
  72. },
  73. },
  74. {
  75. # https://github.com/rg3/youtube-dl/issues/2253
  76. 'url': 'http://bcove.me/i6nfkrc3',
  77. 'md5': '0ba9446db037002366bab3b3eb30c88c',
  78. 'info_dict': {
  79. 'id': '3101154703001',
  80. 'ext': 'mp4',
  81. 'title': 'Still no power',
  82. 'uploader': 'thestar.com',
  83. '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.',
  84. },
  85. 'add_ie': ['Brightcove'],
  86. },
  87. {
  88. 'url': 'http://www.championat.com/video/football/v/87/87499.html',
  89. 'md5': 'fb973ecf6e4a78a67453647444222983',
  90. 'info_dict': {
  91. 'id': '3414141473001',
  92. 'ext': 'mp4',
  93. 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
  94. 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
  95. 'uploader': 'Championat',
  96. },
  97. },
  98. {
  99. # https://github.com/rg3/youtube-dl/issues/3541
  100. 'add_ie': ['Brightcove'],
  101. 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1',
  102. 'info_dict': {
  103. 'id': '3866516442001',
  104. 'ext': 'mp4',
  105. 'title': 'Leer mij vrouwen kennen: Aflevering 1',
  106. 'description': 'Leer mij vrouwen kennen: Aflevering 1',
  107. 'uploader': 'SBS Broadcasting',
  108. },
  109. 'skip': 'Restricted to Netherlands',
  110. 'params': {
  111. 'skip_download': True, # m3u8 download
  112. },
  113. },
  114. # Direct link to a video
  115. {
  116. 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  117. 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  118. 'info_dict': {
  119. 'id': 'trailer',
  120. 'ext': 'mp4',
  121. 'title': 'trailer',
  122. 'upload_date': '20100513',
  123. }
  124. },
  125. # ooyala video
  126. {
  127. 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
  128. 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
  129. 'info_dict': {
  130. 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
  131. 'ext': 'mp4',
  132. 'title': '2cc213299525360.mov', # that's what we get
  133. },
  134. },
  135. # google redirect
  136. {
  137. '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',
  138. 'info_dict': {
  139. 'id': 'cmQHVoWB5FY',
  140. 'ext': 'mp4',
  141. 'upload_date': '20130224',
  142. 'uploader_id': 'TheVerge',
  143. '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.',
  144. 'uploader': 'The Verge',
  145. 'title': 'First Firefox OS phones side-by-side',
  146. },
  147. 'params': {
  148. 'skip_download': False,
  149. }
  150. },
  151. # embed.ly video
  152. {
  153. 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
  154. 'info_dict': {
  155. 'id': '9ODmcdjQcHQ',
  156. 'ext': 'mp4',
  157. 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
  158. 'upload_date': '20140225',
  159. 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
  160. 'uploader': 'Tested',
  161. 'uploader_id': 'testedcom',
  162. },
  163. # No need to test YoutubeIE here
  164. 'params': {
  165. 'skip_download': True,
  166. },
  167. },
  168. # funnyordie embed
  169. {
  170. 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
  171. 'info_dict': {
  172. 'id': '18e820ec3f',
  173. 'ext': 'mp4',
  174. 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
  175. 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
  176. },
  177. },
  178. # RUTV embed
  179. {
  180. 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
  181. 'info_dict': {
  182. 'id': '776940',
  183. 'ext': 'mp4',
  184. 'title': 'Охотское море стало целиком российским',
  185. 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
  186. },
  187. 'params': {
  188. # m3u8 download
  189. 'skip_download': True,
  190. },
  191. },
  192. # Embedded TED video
  193. {
  194. 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
  195. 'md5': '65fdff94098e4a607385a60c5177c638',
  196. 'info_dict': {
  197. 'id': '1969',
  198. 'ext': 'mp4',
  199. 'title': 'Hidden miracles of the natural world',
  200. 'uploader': 'Louie Schwartzberg',
  201. 'description': 'md5:8145d19d320ff3e52f28401f4c4283b9',
  202. }
  203. },
  204. # Embeded Ustream video
  205. {
  206. 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
  207. 'md5': '27b99cdb639c9b12a79bca876a073417',
  208. 'info_dict': {
  209. 'id': '45734260',
  210. 'ext': 'flv',
  211. 'uploader': 'AU SPA: The NSA and Privacy',
  212. 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
  213. }
  214. },
  215. # nowvideo embed hidden behind percent encoding
  216. {
  217. 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
  218. 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
  219. 'info_dict': {
  220. 'id': '06e53103ca9aa',
  221. 'ext': 'flv',
  222. 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
  223. 'description': 'No description',
  224. },
  225. },
  226. # arte embed
  227. {
  228. 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
  229. 'md5': '7653032cbb25bf6c80d80f217055fa43',
  230. 'info_dict': {
  231. 'id': '048195-004_PLUS7-F',
  232. 'ext': 'flv',
  233. 'title': 'X:enius',
  234. 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
  235. 'upload_date': '20140320',
  236. },
  237. 'params': {
  238. 'skip_download': 'Requires rtmpdump'
  239. }
  240. },
  241. # Condé Nast embed
  242. {
  243. 'url': 'http://www.wired.com/2014/04/honda-asimo/',
  244. 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
  245. 'info_dict': {
  246. 'id': '53501be369702d3275860000',
  247. 'ext': 'mp4',
  248. 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
  249. }
  250. },
  251. # Dailymotion embed
  252. {
  253. 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
  254. 'md5': '441aeeb82eb72c422c7f14ec533999cd',
  255. 'info_dict': {
  256. 'id': 'k2mm4bCdJ6CQ2i7c8o2',
  257. 'ext': 'mp4',
  258. 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
  259. 'uploader': 'Spi0n',
  260. },
  261. 'add_ie': ['Dailymotion'],
  262. },
  263. # YouTube embed
  264. {
  265. 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
  266. 'info_dict': {
  267. 'id': 'FXRb4ykk4S0',
  268. 'ext': 'mp4',
  269. 'title': 'The NBL Auction 2014',
  270. 'uploader': 'BADMINTON England',
  271. 'uploader_id': 'BADMINTONEvents',
  272. 'upload_date': '20140603',
  273. 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
  274. },
  275. 'add_ie': ['Youtube'],
  276. 'params': {
  277. 'skip_download': True,
  278. }
  279. },
  280. # MTVSercices embed
  281. {
  282. 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
  283. 'md5': '35727f82f58c76d996fc188f9755b0d5',
  284. 'info_dict': {
  285. 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
  286. 'ext': 'mp4',
  287. 'title': 'Review',
  288. 'description': 'Mario\'s life in the fast lane has never looked so good.',
  289. },
  290. },
  291. # YouTube embed via <data-embed-url="">
  292. {
  293. 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM',
  294. 'info_dict': {
  295. 'id': '4vAffPZIT44',
  296. 'ext': 'mp4',
  297. 'title': 'Asphalt 8: Airborne - Update - Welcome to Dubai!',
  298. 'uploader': 'Gameloft',
  299. 'uploader_id': 'gameloft',
  300. 'upload_date': '20140828',
  301. 'description': 'md5:c80da9ed3d83ae6d1876c834de03e1c4',
  302. },
  303. 'params': {
  304. 'skip_download': True,
  305. }
  306. },
  307. # Camtasia studio
  308. {
  309. 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/',
  310. 'playlist': [{
  311. 'md5': '0c5e352edabf715d762b0ad4e6d9ee67',
  312. 'info_dict': {
  313. 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
  314. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1',
  315. 'ext': 'flv',
  316. 'duration': 2235.90,
  317. }
  318. }, {
  319. 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63',
  320. 'info_dict': {
  321. 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP',
  322. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip',
  323. 'ext': 'flv',
  324. 'duration': 2235.93,
  325. }
  326. }],
  327. 'info_dict': {
  328. 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final',
  329. }
  330. },
  331. # Flowplayer
  332. {
  333. 'url': 'http://www.handjobhub.com/video/busty-blonde-siri-tit-fuck-while-wank-6313.html',
  334. 'md5': '9d65602bf31c6e20014319c7d07fba27',
  335. 'info_dict': {
  336. 'id': '5123ea6d5e5a7',
  337. 'ext': 'mp4',
  338. 'age_limit': 18,
  339. 'uploader': 'www.handjobhub.com',
  340. 'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
  341. }
  342. },
  343. # RSS feed
  344. {
  345. 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
  346. 'info_dict': {
  347. 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
  348. 'title': 'Zero Punctuation',
  349. 'description': 're:'
  350. },
  351. 'playlist_mincount': 11,
  352. },
  353. # Multiple brightcove videos
  354. # https://github.com/rg3/youtube-dl/issues/2283
  355. {
  356. 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
  357. 'info_dict': {
  358. 'id': 'always-never',
  359. 'title': 'Always / Never - The New Yorker',
  360. },
  361. 'playlist_count': 3,
  362. 'params': {
  363. 'extract_flat': False,
  364. 'skip_download': True,
  365. }
  366. },
  367. # MLB embed
  368. {
  369. 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
  370. 'md5': '96f09a37e44da40dd083e12d9a683327',
  371. 'info_dict': {
  372. 'id': '33322633',
  373. 'ext': 'mp4',
  374. 'title': 'Ump changes call to ball',
  375. 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
  376. 'duration': 48,
  377. 'timestamp': 1401537900,
  378. 'upload_date': '20140531',
  379. 'thumbnail': 're:^https?://.*\.jpg$',
  380. },
  381. },
  382. # Wistia embed
  383. {
  384. 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
  385. 'md5': '8788b683c777a5cf25621eaf286d0c23',
  386. 'info_dict': {
  387. 'id': '1cfaf6b7ea',
  388. 'ext': 'mov',
  389. 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
  390. 'duration': 643.0,
  391. 'filesize': 182808282,
  392. 'uploader': 'education-portal.com',
  393. },
  394. },
  395. {
  396. 'url': 'http://thoughtworks.wistia.com/medias/uxjb0lwrcz',
  397. 'md5': 'baf49c2baa8a7de5f3fc145a8506dcd4',
  398. 'info_dict': {
  399. 'id': 'uxjb0lwrcz',
  400. 'ext': 'mp4',
  401. 'title': 'Conversation about Hexagonal Rails Part 1 - ThoughtWorks',
  402. 'duration': 1715.0,
  403. 'uploader': 'thoughtworks.wistia.com',
  404. },
  405. },
  406. # Direct download with broken HEAD
  407. {
  408. 'url': 'http://ai-radio.org:8000/radio.opus',
  409. 'info_dict': {
  410. 'id': 'radio',
  411. 'ext': 'opus',
  412. 'title': 'radio',
  413. },
  414. 'params': {
  415. 'skip_download': True, # infinite live stream
  416. },
  417. 'expected_warnings': [
  418. r'501.*Not Implemented'
  419. ],
  420. },
  421. # Soundcloud embed
  422. {
  423. 'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
  424. 'info_dict': {
  425. 'id': '174391317',
  426. 'ext': 'mp3',
  427. 'description': 'md5:ff867d6b555488ad3c52572bb33d432c',
  428. 'uploader': 'Sophos Security',
  429. 'title': 'Chet Chat 171 - Oct 29, 2014',
  430. 'upload_date': '20141029',
  431. }
  432. },
  433. # Livestream embed
  434. {
  435. 'url': 'http://www.esa.int/Our_Activities/Space_Science/Rosetta/Philae_comet_touch-down_webcast',
  436. 'info_dict': {
  437. 'id': '67864563',
  438. 'ext': 'flv',
  439. 'upload_date': '20141112',
  440. 'title': 'Rosetta #CometLanding webcast HL 10',
  441. }
  442. },
  443. ]
  444. def report_following_redirect(self, new_url):
  445. """Report information extraction."""
  446. self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
  447. def _extract_rss(self, url, video_id, doc):
  448. playlist_title = doc.find('./channel/title').text
  449. playlist_desc_el = doc.find('./channel/description')
  450. playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
  451. entries = [{
  452. '_type': 'url',
  453. 'url': e.find('link').text,
  454. 'title': e.find('title').text,
  455. } for e in doc.findall('./channel/item')]
  456. return {
  457. '_type': 'playlist',
  458. 'id': url,
  459. 'title': playlist_title,
  460. 'description': playlist_desc,
  461. 'entries': entries,
  462. }
  463. def _extract_camtasia(self, url, video_id, webpage):
  464. """ Returns None if no camtasia video can be found. """
  465. camtasia_cfg = self._search_regex(
  466. r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);',
  467. webpage, 'camtasia configuration file', default=None)
  468. if camtasia_cfg is None:
  469. return None
  470. title = self._html_search_meta('DC.title', webpage, fatal=True)
  471. camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg)
  472. camtasia_cfg = self._download_xml(
  473. camtasia_url, video_id,
  474. note='Downloading camtasia configuration',
  475. errnote='Failed to download camtasia configuration')
  476. fileset_node = camtasia_cfg.find('./playlist/array/fileset')
  477. entries = []
  478. for n in fileset_node.getchildren():
  479. url_n = n.find('./uri')
  480. if url_n is None:
  481. continue
  482. entries.append({
  483. 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0],
  484. 'title': '%s - %s' % (title, n.tag),
  485. 'url': compat_urlparse.urljoin(url, url_n.text),
  486. 'duration': float_or_none(n.find('./duration').text),
  487. })
  488. return {
  489. '_type': 'playlist',
  490. 'entries': entries,
  491. 'title': title,
  492. }
  493. def _real_extract(self, url):
  494. if url.startswith('//'):
  495. return {
  496. '_type': 'url',
  497. 'url': self.http_scheme() + url,
  498. }
  499. parsed_url = compat_urlparse.urlparse(url)
  500. if not parsed_url.scheme:
  501. default_search = self._downloader.params.get('default_search')
  502. if default_search is None:
  503. default_search = 'fixup_error'
  504. if default_search in ('auto', 'auto_warning', 'fixup_error'):
  505. if '/' in url:
  506. self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
  507. return self.url_result('http://' + url)
  508. elif default_search != 'fixup_error':
  509. if default_search == 'auto_warning':
  510. if re.match(r'^(?:url|URL)$', url):
  511. raise ExtractorError(
  512. 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
  513. expected=True)
  514. else:
  515. self._downloader.report_warning(
  516. 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
  517. return self.url_result('ytsearch:' + url)
  518. if default_search in ('error', 'fixup_error'):
  519. raise ExtractorError(
  520. ('%r is not a valid URL. '
  521. 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
  522. ) % (url, url), expected=True)
  523. else:
  524. if ':' not in default_search:
  525. default_search += ':'
  526. return self.url_result(default_search + url)
  527. url, smuggled_data = unsmuggle_url(url)
  528. force_videoid = None
  529. is_intentional = smuggled_data and smuggled_data.get('to_generic')
  530. if smuggled_data and 'force_videoid' in smuggled_data:
  531. force_videoid = smuggled_data['force_videoid']
  532. video_id = force_videoid
  533. else:
  534. video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
  535. self.to_screen('%s: Requesting header' % video_id)
  536. head_req = HEADRequest(url)
  537. head_response = self._request_webpage(
  538. head_req, video_id,
  539. note=False, errnote='Could not send HEAD request to %s' % url,
  540. fatal=False)
  541. if head_response is not False:
  542. # Check for redirect
  543. new_url = head_response.geturl()
  544. if url != new_url:
  545. self.report_following_redirect(new_url)
  546. if force_videoid:
  547. new_url = smuggle_url(
  548. new_url, {'force_videoid': force_videoid})
  549. return self.url_result(new_url)
  550. full_response = None
  551. if head_response is False:
  552. full_response = self._request_webpage(url, video_id)
  553. head_response = full_response
  554. # Check for direct link to a video
  555. content_type = head_response.headers.get('Content-Type', '')
  556. m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
  557. if m:
  558. upload_date = unified_strdate(
  559. head_response.headers.get('Last-Modified'))
  560. return {
  561. 'id': video_id,
  562. 'title': os.path.splitext(url_basename(url))[0],
  563. 'direct': True,
  564. 'formats': [{
  565. 'format_id': m.group('format_id'),
  566. 'url': url,
  567. 'vcodec': 'none' if m.group('type') == 'audio' else None
  568. }],
  569. 'upload_date': upload_date,
  570. }
  571. if not self._downloader.params.get('test', False) and not is_intentional:
  572. self._downloader.report_warning('Falling back on generic information extractor.')
  573. if full_response:
  574. webpage = self._webpage_read_content(full_response, url, video_id)
  575. else:
  576. webpage = self._download_webpage(url, video_id)
  577. self.report_extraction(video_id)
  578. # Is it an RSS feed?
  579. try:
  580. doc = parse_xml(webpage)
  581. if doc.tag == 'rss':
  582. return self._extract_rss(url, video_id, doc)
  583. except compat_xml_parse_error:
  584. pass
  585. # Is it a Camtasia project?
  586. camtasia_res = self._extract_camtasia(url, video_id, webpage)
  587. if camtasia_res is not None:
  588. return camtasia_res
  589. # Sometimes embedded video player is hidden behind percent encoding
  590. # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
  591. # Unescaping the whole page allows to handle those cases in a generic way
  592. webpage = compat_urllib_parse.unquote(webpage)
  593. # it's tempting to parse this further, but you would
  594. # have to take into account all the variations like
  595. # Video Title - Site Name
  596. # Site Name | Video Title
  597. # Video Title - Tagline | Site Name
  598. # and so on and so forth; it's just not practical
  599. video_title = self._html_search_regex(
  600. r'(?s)<title>(.*?)</title>', webpage, 'video title',
  601. default='video')
  602. # Try to detect age limit automatically
  603. age_limit = self._rta_search(webpage)
  604. # And then there are the jokers who advertise that they use RTA,
  605. # but actually don't.
  606. AGE_LIMIT_MARKERS = [
  607. r'Proudly Labeled <a href="http://www.rtalabel.org/" title="Restricted to Adults">RTA</a>',
  608. ]
  609. if any(re.search(marker, webpage) for marker in AGE_LIMIT_MARKERS):
  610. age_limit = 18
  611. # video uploader is domain name
  612. video_uploader = self._search_regex(
  613. r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
  614. # Helper method
  615. def _playlist_from_matches(matches, getter, ie=None):
  616. urlrs = orderedSet(
  617. self.url_result(self._proto_relative_url(getter(m)), ie)
  618. for m in matches)
  619. return self.playlist_result(
  620. urlrs, playlist_id=video_id, playlist_title=video_title)
  621. # Look for BrightCove:
  622. bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
  623. if bc_urls:
  624. self.to_screen('Brightcove video detected.')
  625. entries = [{
  626. '_type': 'url',
  627. 'url': smuggle_url(bc_url, {'Referer': url}),
  628. 'ie_key': 'Brightcove'
  629. } for bc_url in bc_urls]
  630. return {
  631. '_type': 'playlist',
  632. 'title': video_title,
  633. 'id': video_id,
  634. 'entries': entries,
  635. }
  636. # Look for embedded (iframe) Vimeo player
  637. mobj = re.search(
  638. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
  639. if mobj:
  640. player_url = unescapeHTML(mobj.group('url'))
  641. surl = smuggle_url(player_url, {'Referer': url})
  642. return self.url_result(surl)
  643. # Look for embedded (swf embed) Vimeo player
  644. mobj = re.search(
  645. r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
  646. if mobj:
  647. return self.url_result(mobj.group(1))
  648. # Look for embedded YouTube player
  649. matches = re.findall(r'''(?x)
  650. (?:
  651. <iframe[^>]+?src=|
  652. data-video-url=|
  653. <embed[^>]+?src=|
  654. embedSWF\(?:\s*|
  655. new\s+SWFObject\(
  656. )
  657. (["\'])
  658. (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
  659. (?:embed|v|p)/.+?)
  660. \1''', webpage)
  661. if matches:
  662. return _playlist_from_matches(
  663. matches, lambda m: unescapeHTML(m[1]))
  664. # Look for embedded Dailymotion player
  665. matches = re.findall(
  666. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
  667. if matches:
  668. return _playlist_from_matches(
  669. matches, lambda m: unescapeHTML(m[1]))
  670. # Look for embedded Dailymotion playlist player (#3822)
  671. m = re.search(
  672. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1', webpage)
  673. if m:
  674. playlists = re.findall(
  675. r'list\[\]=/playlist/([^/]+)/', unescapeHTML(m.group('url')))
  676. if playlists:
  677. return _playlist_from_matches(
  678. playlists, lambda p: '//dailymotion.com/playlist/%s' % p)
  679. # Look for embedded Wistia player
  680. match = re.search(
  681. r'<(?:meta[^>]+?content|iframe[^>]+?src)=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
  682. if match:
  683. embed_url = self._proto_relative_url(
  684. unescapeHTML(match.group('url')))
  685. return {
  686. '_type': 'url_transparent',
  687. 'url': embed_url,
  688. 'ie_key': 'Wistia',
  689. 'uploader': video_uploader,
  690. 'title': video_title,
  691. 'id': video_id,
  692. }
  693. match = re.search(r'(?:id=["\']wistia_|data-wistia-?id=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
  694. if match:
  695. return {
  696. '_type': 'url_transparent',
  697. 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
  698. 'ie_key': 'Wistia',
  699. 'uploader': video_uploader,
  700. 'title': video_title,
  701. 'id': match.group('id')
  702. }
  703. # Look for embedded blip.tv player
  704. mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
  705. if mobj:
  706. return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
  707. mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
  708. if mobj:
  709. return self.url_result(mobj.group(1), 'BlipTV')
  710. # Look for embedded condenast player
  711. matches = re.findall(
  712. r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
  713. webpage)
  714. if matches:
  715. return {
  716. '_type': 'playlist',
  717. 'entries': [{
  718. '_type': 'url',
  719. 'ie_key': 'CondeNast',
  720. 'url': ma,
  721. } for ma in matches],
  722. 'title': video_title,
  723. 'id': video_id,
  724. }
  725. # Look for Bandcamp pages with custom domain
  726. mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
  727. if mobj is not None:
  728. burl = unescapeHTML(mobj.group(1))
  729. # Don't set the extractor because it can be a track url or an album
  730. return self.url_result(burl)
  731. # Look for embedded Vevo player
  732. mobj = re.search(
  733. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
  734. if mobj is not None:
  735. return self.url_result(mobj.group('url'))
  736. # Look for Ooyala videos
  737. mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
  738. re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
  739. if mobj is not None:
  740. return OoyalaIE._build_url_result(mobj.group('ec'))
  741. # Look for Aparat videos
  742. mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
  743. if mobj is not None:
  744. return self.url_result(mobj.group(1), 'Aparat')
  745. # Look for MPORA videos
  746. mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
  747. if mobj is not None:
  748. return self.url_result(mobj.group(1), 'Mpora')
  749. # Look for embedded NovaMov-based player
  750. mobj = re.search(
  751. r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
  752. (?P<url>http://(?:(?:embed|www)\.)?
  753. (?:novamov\.com|
  754. nowvideo\.(?:ch|sx|eu|at|ag|co)|
  755. videoweed\.(?:es|com)|
  756. movshare\.(?:net|sx|ag)|
  757. divxstage\.(?:eu|net|ch|co|at|ag))
  758. /embed\.php.+?)\1''', webpage)
  759. if mobj is not None:
  760. return self.url_result(mobj.group('url'))
  761. # Look for embedded Facebook player
  762. mobj = re.search(
  763. r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
  764. if mobj is not None:
  765. return self.url_result(mobj.group('url'), 'Facebook')
  766. # Look for embedded VK player
  767. mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
  768. if mobj is not None:
  769. return self.url_result(mobj.group('url'), 'VK')
  770. # Look for embedded ivi player
  771. mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
  772. if mobj is not None:
  773. return self.url_result(mobj.group('url'), 'Ivi')
  774. # Look for embedded Huffington Post player
  775. mobj = re.search(
  776. r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
  777. if mobj is not None:
  778. return self.url_result(mobj.group('url'), 'HuffPost')
  779. # Look for embed.ly
  780. mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
  781. if mobj is not None:
  782. return self.url_result(mobj.group('url'))
  783. mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
  784. if mobj is not None:
  785. return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
  786. # Look for funnyordie embed
  787. matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
  788. if matches:
  789. return _playlist_from_matches(
  790. matches, getter=unescapeHTML, ie='FunnyOrDie')
  791. # Look for embedded RUTV player
  792. rutv_url = RUTVIE._extract_url(webpage)
  793. if rutv_url:
  794. return self.url_result(rutv_url, 'RUTV')
  795. # Look for embedded TED player
  796. mobj = re.search(
  797. r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
  798. if mobj is not None:
  799. return self.url_result(mobj.group('url'), 'TED')
  800. # Look for embedded Ustream videos
  801. mobj = re.search(
  802. r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
  803. if mobj is not None:
  804. return self.url_result(mobj.group('url'), 'Ustream')
  805. # Look for embedded arte.tv player
  806. mobj = re.search(
  807. r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
  808. webpage)
  809. if mobj is not None:
  810. return self.url_result(mobj.group('url'), 'ArteTVEmbed')
  811. # Look for embedded smotri.com player
  812. smotri_url = SmotriIE._extract_url(webpage)
  813. if smotri_url:
  814. return self.url_result(smotri_url, 'Smotri')
  815. # Look for embeded soundcloud player
  816. mobj = re.search(
  817. r'<iframe\s+(?:[a-zA-Z0-9_-]+="[^"]+"\s+)*src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
  818. webpage)
  819. if mobj is not None:
  820. url = unescapeHTML(mobj.group('url'))
  821. return self.url_result(url)
  822. # Look for embedded vulture.com player
  823. mobj = re.search(
  824. r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
  825. webpage)
  826. if mobj is not None:
  827. url = unescapeHTML(mobj.group('url'))
  828. return self.url_result(url, ie='Vulture')
  829. # Look for embedded mtvservices player
  830. mobj = re.search(
  831. r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
  832. webpage)
  833. if mobj is not None:
  834. url = unescapeHTML(mobj.group('url'))
  835. return self.url_result(url, ie='MTVServicesEmbedded')
  836. # Look for embedded yahoo player
  837. mobj = re.search(
  838. r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1',
  839. webpage)
  840. if mobj is not None:
  841. return self.url_result(mobj.group('url'), 'Yahoo')
  842. # Look for embedded sbs.com.au player
  843. mobj = re.search(
  844. r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
  845. webpage)
  846. if mobj is not None:
  847. return self.url_result(mobj.group('url'), 'SBS')
  848. mobj = re.search(
  849. r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
  850. webpage)
  851. if mobj is not None:
  852. return self.url_result(mobj.group('url'), 'MLB')
  853. mobj = re.search(
  854. r'<iframe[^>]+?src=(["\'])(?P<url>%s)\1' % CondeNastIE.EMBED_URL,
  855. webpage)
  856. if mobj is not None:
  857. return self.url_result(self._proto_relative_url(mobj.group('url'), scheme='http:'), 'CondeNast')
  858. mobj = re.search(
  859. r'<iframe[^>]+src="(?P<url>https?://new\.livestream\.com/[^"]+/player[^"]+)"',
  860. webpage)
  861. if mobj is not None:
  862. return self.url_result(mobj.group('url'), 'Livestream')
  863. def check_video(vurl):
  864. vpath = compat_urlparse.urlparse(vurl).path
  865. vext = determine_ext(vpath)
  866. return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
  867. def filter_video(urls):
  868. return list(filter(check_video, urls))
  869. # Start with something easy: JW Player in SWFObject
  870. found = filter_video(re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage))
  871. if not found:
  872. # Look for gorilla-vid style embedding
  873. found = filter_video(re.findall(r'''(?sx)
  874. (?:
  875. jw_plugins|
  876. JWPlayerOptions|
  877. jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
  878. )
  879. .*?file\s*:\s*["\'](.*?)["\']''', webpage))
  880. if not found:
  881. # Broaden the search a little bit
  882. found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))
  883. if not found:
  884. # Broaden the findall a little bit: JWPlayer JS loader
  885. found = filter_video(re.findall(
  886. r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage))
  887. if not found:
  888. # Flow player
  889. found = filter_video(re.findall(r'''(?xs)
  890. flowplayer\("[^"]+",\s*
  891. \{[^}]+?\}\s*,
  892. \s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
  893. ["']?url["']?\s*:\s*["']([^"']+)["']
  894. ''', webpage))
  895. if not found:
  896. # Try to find twitter cards info
  897. found = filter_video(re.findall(
  898. r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage))
  899. if not found:
  900. # We look for Open Graph info:
  901. # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
  902. m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
  903. # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
  904. if m_video_type is not None:
  905. found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
  906. if not found:
  907. # HTML5 video
  908. found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
  909. if not found:
  910. found = re.search(
  911. r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
  912. r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
  913. webpage)
  914. if found:
  915. new_url = found.group(1)
  916. self.report_following_redirect(new_url)
  917. return {
  918. '_type': 'url',
  919. 'url': new_url,
  920. }
  921. if not found:
  922. raise ExtractorError('Unsupported URL: %s' % url)
  923. entries = []
  924. for video_url in found:
  925. video_url = compat_urlparse.urljoin(url, video_url)
  926. video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
  927. # Sometimes, jwplayer extraction will result in a YouTube URL
  928. if YoutubeIE.suitable(video_url):
  929. entries.append(self.url_result(video_url, 'Youtube'))
  930. continue
  931. # here's a fun little line of code for you:
  932. video_id = os.path.splitext(video_id)[0]
  933. entries.append({
  934. 'id': video_id,
  935. 'url': video_url,
  936. 'uploader': video_uploader,
  937. 'title': video_title,
  938. 'age_limit': age_limit,
  939. })
  940. if len(entries) == 1:
  941. return entries[0]
  942. else:
  943. for num, e in enumerate(entries, start=1):
  944. e['title'] = '%s (%d)' % (e['title'], num)
  945. return {
  946. '_type': 'playlist',
  947. 'entries': entries,
  948. }