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.

783 lines
32 KiB

  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_error,
  9. compat_urllib_parse,
  10. compat_urllib_request,
  11. compat_urlparse,
  12. compat_xml_parse_error,
  13. ExtractorError,
  14. HEADRequest,
  15. parse_xml,
  16. smuggle_url,
  17. unescapeHTML,
  18. unified_strdate,
  19. url_basename,
  20. )
  21. from .brightcove import BrightcoveIE
  22. from .ooyala import OoyalaIE
  23. from .rutv import RUTVIE
  24. from .smotri import SmotriIE
  25. class GenericIE(InfoExtractor):
  26. IE_DESC = 'Generic downloader that works on some sites'
  27. _VALID_URL = r'.*'
  28. IE_NAME = 'generic'
  29. _TESTS = [
  30. {
  31. 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
  32. 'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
  33. 'info_dict': {
  34. 'id': '13601338388002',
  35. 'ext': 'mp4',
  36. 'uploader': 'www.hodiho.fr',
  37. 'title': 'R\u00e9gis plante sa Jeep',
  38. }
  39. },
  40. # bandcamp page with custom domain
  41. {
  42. 'add_ie': ['Bandcamp'],
  43. 'url': 'http://bronyrock.com/track/the-pony-mash',
  44. 'info_dict': {
  45. 'id': '3235767654',
  46. 'ext': 'mp3',
  47. 'title': 'The Pony Mash',
  48. 'uploader': 'M_Pallante',
  49. },
  50. 'skip': 'There is a limit of 200 free downloads / month for the test song',
  51. },
  52. # embedded brightcove video
  53. # it also tests brightcove videos that need to set the 'Referer' in the
  54. # http requests
  55. {
  56. 'add_ie': ['Brightcove'],
  57. 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
  58. 'info_dict': {
  59. 'id': '2765128793001',
  60. 'ext': 'mp4',
  61. 'title': 'Le cours de bourse : l’analyse technique',
  62. 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
  63. 'uploader': 'BFM BUSINESS',
  64. },
  65. 'params': {
  66. 'skip_download': True,
  67. },
  68. },
  69. {
  70. # https://github.com/rg3/youtube-dl/issues/2253
  71. 'url': 'http://bcove.me/i6nfkrc3',
  72. 'md5': '0ba9446db037002366bab3b3eb30c88c',
  73. 'info_dict': {
  74. 'id': '3101154703001',
  75. 'ext': 'mp4',
  76. 'title': 'Still no power',
  77. 'uploader': 'thestar.com',
  78. '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.',
  79. },
  80. 'add_ie': ['Brightcove'],
  81. },
  82. {
  83. 'url': 'http://www.championat.com/video/football/v/87/87499.html',
  84. 'md5': 'fb973ecf6e4a78a67453647444222983',
  85. 'info_dict': {
  86. 'id': '3414141473001',
  87. 'ext': 'mp4',
  88. 'title': 'Видео. Удаление Дзагоева (ЦСКА)',
  89. 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"',
  90. 'uploader': 'Championat',
  91. },
  92. },
  93. # Direct link to a video
  94. {
  95. 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
  96. 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
  97. 'info_dict': {
  98. 'id': 'trailer',
  99. 'ext': 'mp4',
  100. 'title': 'trailer',
  101. 'upload_date': '20100513',
  102. }
  103. },
  104. # ooyala video
  105. {
  106. 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
  107. 'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
  108. 'info_dict': {
  109. 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
  110. 'ext': 'mp4',
  111. 'title': '2cc213299525360.mov', # that's what we get
  112. },
  113. },
  114. # google redirect
  115. {
  116. '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',
  117. 'info_dict': {
  118. 'id': 'cmQHVoWB5FY',
  119. 'ext': 'mp4',
  120. 'upload_date': '20130224',
  121. 'uploader_id': 'TheVerge',
  122. '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.',
  123. 'uploader': 'The Verge',
  124. 'title': 'First Firefox OS phones side-by-side',
  125. },
  126. 'params': {
  127. 'skip_download': False,
  128. }
  129. },
  130. # embed.ly video
  131. {
  132. 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
  133. 'info_dict': {
  134. 'id': '9ODmcdjQcHQ',
  135. 'ext': 'mp4',
  136. 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second',
  137. 'upload_date': '20140225',
  138. 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff',
  139. 'uploader': 'Tested',
  140. 'uploader_id': 'testedcom',
  141. },
  142. # No need to test YoutubeIE here
  143. 'params': {
  144. 'skip_download': True,
  145. },
  146. },
  147. # funnyordie embed
  148. {
  149. 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns',
  150. 'md5': '7cf780be104d40fea7bae52eed4a470e',
  151. 'info_dict': {
  152. 'id': '18e820ec3f',
  153. 'ext': 'mp4',
  154. 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama',
  155. 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.',
  156. },
  157. },
  158. # RUTV embed
  159. {
  160. 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html',
  161. 'info_dict': {
  162. 'id': '776940',
  163. 'ext': 'mp4',
  164. 'title': 'Охотское море стало целиком российским',
  165. 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43',
  166. },
  167. 'params': {
  168. # m3u8 download
  169. 'skip_download': True,
  170. },
  171. },
  172. # Embedded TED video
  173. {
  174. 'url': 'http://en.support.wordpress.com/videos/ted-talks/',
  175. 'md5': 'deeeabcc1085eb2ba205474e7235a3d5',
  176. 'info_dict': {
  177. 'id': '981',
  178. 'ext': 'mp4',
  179. 'title': 'My web playroom',
  180. 'uploader': 'Ze Frank',
  181. 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b',
  182. }
  183. },
  184. # Embeded Ustream video
  185. {
  186. 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm',
  187. 'md5': '27b99cdb639c9b12a79bca876a073417',
  188. 'info_dict': {
  189. 'id': '45734260',
  190. 'ext': 'flv',
  191. 'uploader': 'AU SPA: The NSA and Privacy',
  192. 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman'
  193. }
  194. },
  195. # nowvideo embed hidden behind percent encoding
  196. {
  197. 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/',
  198. 'md5': '2baf4ddd70f697d94b1c18cf796d5107',
  199. 'info_dict': {
  200. 'id': '06e53103ca9aa',
  201. 'ext': 'flv',
  202. 'title': 'Macross Episode 001 Watch Macross Episode 001 onl',
  203. 'description': 'No description',
  204. },
  205. },
  206. # arte embed
  207. {
  208. 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html',
  209. 'md5': '7653032cbb25bf6c80d80f217055fa43',
  210. 'info_dict': {
  211. 'id': '048195-004_PLUS7-F',
  212. 'ext': 'flv',
  213. 'title': 'X:enius',
  214. 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168',
  215. 'upload_date': '20140320',
  216. },
  217. 'params': {
  218. 'skip_download': 'Requires rtmpdump'
  219. }
  220. },
  221. # smotri embed
  222. {
  223. 'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml',
  224. 'md5': 'ec40048448e9284c9a1de77bb188108b',
  225. 'info_dict': {
  226. 'id': 'v27008541fad',
  227. 'ext': 'mp4',
  228. 'title': 'Крым и Севастополь вошли в состав России',
  229. 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175',
  230. 'duration': 900,
  231. 'upload_date': '20140318',
  232. 'uploader': 'rbctv_2012_4',
  233. 'uploader_id': 'rbctv_2012_4',
  234. },
  235. },
  236. # Condé Nast embed
  237. {
  238. 'url': 'http://www.wired.com/2014/04/honda-asimo/',
  239. 'md5': 'ba0dfe966fa007657bd1443ee672db0f',
  240. 'info_dict': {
  241. 'id': '53501be369702d3275860000',
  242. 'ext': 'mp4',
  243. 'title': 'Honda’s New Asimo Robot Is More Human Than Ever',
  244. }
  245. },
  246. # Dailymotion embed
  247. {
  248. 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/',
  249. 'md5': '441aeeb82eb72c422c7f14ec533999cd',
  250. 'info_dict': {
  251. 'id': 'k2mm4bCdJ6CQ2i7c8o2',
  252. 'ext': 'mp4',
  253. 'title': 'Le Zap de Spi0n n°216 - Zapping du Web',
  254. 'uploader': 'Spi0n',
  255. },
  256. 'add_ie': ['Dailymotion'],
  257. },
  258. # YouTube embed
  259. {
  260. 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html',
  261. 'info_dict': {
  262. 'id': 'FXRb4ykk4S0',
  263. 'ext': 'mp4',
  264. 'title': 'The NBL Auction 2014',
  265. 'uploader': 'BADMINTON England',
  266. 'uploader_id': 'BADMINTONEvents',
  267. 'upload_date': '20140603',
  268. 'description': 'md5:9ef128a69f1e262a700ed83edb163a73',
  269. },
  270. 'add_ie': ['Youtube'],
  271. 'params': {
  272. 'skip_download': True,
  273. }
  274. },
  275. # MTVSercices embed
  276. {
  277. 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
  278. 'md5': '35727f82f58c76d996fc188f9755b0d5',
  279. 'info_dict': {
  280. 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
  281. 'ext': 'mp4',
  282. 'title': 'Review',
  283. 'description': 'Mario\'s life in the fast lane has never looked so good.',
  284. },
  285. },
  286. ]
  287. def report_download_webpage(self, video_id):
  288. """Report webpage download."""
  289. if not self._downloader.params.get('test', False):
  290. self._downloader.report_warning('Falling back on generic information extractor.')
  291. super(GenericIE, self).report_download_webpage(video_id)
  292. def report_following_redirect(self, new_url):
  293. """Report information extraction."""
  294. self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
  295. def _send_head(self, url):
  296. """Check if it is a redirect, like url shorteners, in case return the new url."""
  297. class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
  298. """
  299. Subclass the HTTPRedirectHandler to make it use our
  300. HEADRequest also on the redirected URL
  301. """
  302. def redirect_request(self, req, fp, code, msg, headers, newurl):
  303. if code in (301, 302, 303, 307):
  304. newurl = newurl.replace(' ', '%20')
  305. newheaders = dict((k,v) for k,v in req.headers.items()
  306. if k.lower() not in ("content-length", "content-type"))
  307. try:
  308. # This function was deprecated in python 3.3 and removed in 3.4
  309. origin_req_host = req.get_origin_req_host()
  310. except AttributeError:
  311. origin_req_host = req.origin_req_host
  312. return HEADRequest(newurl,
  313. headers=newheaders,
  314. origin_req_host=origin_req_host,
  315. unverifiable=True)
  316. else:
  317. raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp)
  318. class HTTPMethodFallback(compat_urllib_request.BaseHandler):
  319. """
  320. Fallback to GET if HEAD is not allowed (405 HTTP error)
  321. """
  322. def http_error_405(self, req, fp, code, msg, headers):
  323. fp.read()
  324. fp.close()
  325. newheaders = dict((k,v) for k,v in req.headers.items()
  326. if k.lower() not in ("content-length", "content-type"))
  327. return self.parent.open(compat_urllib_request.Request(req.get_full_url(),
  328. headers=newheaders,
  329. origin_req_host=req.get_origin_req_host(),
  330. unverifiable=True))
  331. # Build our opener
  332. opener = compat_urllib_request.OpenerDirector()
  333. for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler,
  334. HTTPMethodFallback, HEADRedirectHandler,
  335. compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
  336. opener.add_handler(handler())
  337. response = opener.open(HEADRequest(url))
  338. if response is None:
  339. raise ExtractorError('Invalid URL protocol')
  340. return response
  341. def _extract_rss(self, url, video_id, doc):
  342. playlist_title = doc.find('./channel/title').text
  343. playlist_desc_el = doc.find('./channel/description')
  344. playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text
  345. entries = [{
  346. '_type': 'url',
  347. 'url': e.find('link').text,
  348. 'title': e.find('title').text,
  349. } for e in doc.findall('./channel/item')]
  350. return {
  351. '_type': 'playlist',
  352. 'id': url,
  353. 'title': playlist_title,
  354. 'description': playlist_desc,
  355. 'entries': entries,
  356. }
  357. def _real_extract(self, url):
  358. if url.startswith('//'):
  359. return {
  360. '_type': 'url',
  361. 'url': self.http_scheme() + url,
  362. }
  363. parsed_url = compat_urlparse.urlparse(url)
  364. if not parsed_url.scheme:
  365. default_search = self._downloader.params.get('default_search')
  366. if default_search is None:
  367. default_search = 'fixup_error'
  368. if default_search in ('auto', 'auto_warning', 'fixup_error'):
  369. if '/' in url:
  370. self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
  371. return self.url_result('http://' + url)
  372. elif default_search != 'fixup_error':
  373. if default_search == 'auto_warning':
  374. if re.match(r'^(?:url|URL)$', url):
  375. raise ExtractorError(
  376. 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url,
  377. expected=True)
  378. else:
  379. self._downloader.report_warning(
  380. 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url)
  381. return self.url_result('ytsearch:' + url)
  382. if default_search in ('error', 'fixup_error'):
  383. raise ExtractorError(
  384. ('%r is not a valid URL. '
  385. 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube'
  386. ) % (url, url), expected=True)
  387. else:
  388. assert ':' in default_search
  389. return self.url_result(default_search + url)
  390. video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
  391. self.to_screen('%s: Requesting header' % video_id)
  392. try:
  393. response = self._send_head(url)
  394. # Check for redirect
  395. new_url = response.geturl()
  396. if url != new_url:
  397. self.report_following_redirect(new_url)
  398. return self.url_result(new_url)
  399. # Check for direct link to a video
  400. content_type = response.headers.get('Content-Type', '')
  401. m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
  402. if m:
  403. upload_date = response.headers.get('Last-Modified')
  404. if upload_date:
  405. upload_date = unified_strdate(upload_date)
  406. return {
  407. 'id': video_id,
  408. 'title': os.path.splitext(url_basename(url))[0],
  409. 'formats': [{
  410. 'format_id': m.group('format_id'),
  411. 'url': url,
  412. 'vcodec': 'none' if m.group('type') == 'audio' else None
  413. }],
  414. 'upload_date': upload_date,
  415. }
  416. except compat_urllib_error.HTTPError:
  417. # This may be a stupid server that doesn't like HEAD, our UA, or so
  418. pass
  419. try:
  420. webpage = self._download_webpage(url, video_id)
  421. except ValueError:
  422. # since this is the last-resort InfoExtractor, if
  423. # this error is thrown, it'll be thrown here
  424. raise ExtractorError('Failed to download URL: %s' % url)
  425. self.report_extraction(video_id)
  426. # Is it an RSS feed?
  427. try:
  428. doc = parse_xml(webpage)
  429. if doc.tag == 'rss':
  430. return self._extract_rss(url, video_id, doc)
  431. except compat_xml_parse_error:
  432. pass
  433. # Sometimes embedded video player is hidden behind percent encoding
  434. # (e.g. https://github.com/rg3/youtube-dl/issues/2448)
  435. # Unescaping the whole page allows to handle those cases in a generic way
  436. webpage = compat_urllib_parse.unquote(webpage)
  437. # it's tempting to parse this further, but you would
  438. # have to take into account all the variations like
  439. # Video Title - Site Name
  440. # Site Name | Video Title
  441. # Video Title - Tagline | Site Name
  442. # and so on and so forth; it's just not practical
  443. video_title = self._html_search_regex(
  444. r'(?s)<title>(.*?)</title>', webpage, 'video title',
  445. default='video')
  446. # video uploader is domain name
  447. video_uploader = self._search_regex(
  448. r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
  449. # Look for BrightCove:
  450. bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
  451. if bc_urls:
  452. self.to_screen('Brightcove video detected.')
  453. entries = [{
  454. '_type': 'url',
  455. 'url': smuggle_url(bc_url, {'Referer': url}),
  456. 'ie_key': 'Brightcove'
  457. } for bc_url in bc_urls]
  458. return {
  459. '_type': 'playlist',
  460. 'title': video_title,
  461. 'id': video_id,
  462. 'entries': entries,
  463. }
  464. # Look for embedded (iframe) Vimeo player
  465. mobj = re.search(
  466. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
  467. if mobj:
  468. player_url = unescapeHTML(mobj.group('url'))
  469. surl = smuggle_url(player_url, {'Referer': url})
  470. return self.url_result(surl, 'Vimeo')
  471. # Look for embedded (swf embed) Vimeo player
  472. mobj = re.search(
  473. r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
  474. if mobj:
  475. return self.url_result(mobj.group(1), 'Vimeo')
  476. # Look for embedded YouTube player
  477. matches = re.findall(r'''(?x)
  478. (?:
  479. <iframe[^>]+?src=|
  480. <embed[^>]+?src=|
  481. embedSWF\(?:\s*
  482. )
  483. (["\'])
  484. (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
  485. (?:embed|v)/.+?)
  486. \1''', webpage)
  487. if matches:
  488. urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
  489. for tuppl in matches]
  490. return self.playlist_result(
  491. urlrs, playlist_id=video_id, playlist_title=video_title)
  492. # Look for embedded Dailymotion player
  493. matches = re.findall(
  494. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
  495. if matches:
  496. urlrs = [self.url_result(unescapeHTML(tuppl[1]))
  497. for tuppl in matches]
  498. return self.playlist_result(
  499. urlrs, playlist_id=video_id, playlist_title=video_title)
  500. # Look for embedded Wistia player
  501. match = re.search(
  502. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
  503. if match:
  504. return {
  505. '_type': 'url_transparent',
  506. 'url': unescapeHTML(match.group('url')),
  507. 'ie_key': 'Wistia',
  508. 'uploader': video_uploader,
  509. 'title': video_title,
  510. 'id': video_id,
  511. }
  512. # Look for embedded blip.tv player
  513. mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
  514. if mobj:
  515. return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
  516. mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage)
  517. if mobj:
  518. return self.url_result(mobj.group(1), 'BlipTV')
  519. # Look for embedded condenast player
  520. matches = re.findall(
  521. r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")',
  522. webpage)
  523. if matches:
  524. return {
  525. '_type': 'playlist',
  526. 'entries': [{
  527. '_type': 'url',
  528. 'ie_key': 'CondeNast',
  529. 'url': ma,
  530. } for ma in matches],
  531. 'title': video_title,
  532. 'id': video_id,
  533. }
  534. # Look for Bandcamp pages with custom domain
  535. mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
  536. if mobj is not None:
  537. burl = unescapeHTML(mobj.group(1))
  538. # Don't set the extractor because it can be a track url or an album
  539. return self.url_result(burl)
  540. # Look for embedded Vevo player
  541. mobj = re.search(
  542. r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
  543. if mobj is not None:
  544. return self.url_result(mobj.group('url'))
  545. # Look for Ooyala videos
  546. mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or
  547. re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage))
  548. if mobj is not None:
  549. return OoyalaIE._build_url_result(mobj.group('ec'))
  550. # Look for Aparat videos
  551. mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
  552. if mobj is not None:
  553. return self.url_result(mobj.group(1), 'Aparat')
  554. # Look for MPORA videos
  555. mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
  556. if mobj is not None:
  557. return self.url_result(mobj.group(1), 'Mpora')
  558. # Look for embedded NovaMov-based player
  559. mobj = re.search(
  560. r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\'])
  561. (?P<url>http://(?:(?:embed|www)\.)?
  562. (?:novamov\.com|
  563. nowvideo\.(?:ch|sx|eu|at|ag|co)|
  564. videoweed\.(?:es|com)|
  565. movshare\.(?:net|sx|ag)|
  566. divxstage\.(?:eu|net|ch|co|at|ag))
  567. /embed\.php.+?)\1''', webpage)
  568. if mobj is not None:
  569. return self.url_result(mobj.group('url'))
  570. # Look for embedded Facebook player
  571. mobj = re.search(
  572. r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
  573. if mobj is not None:
  574. return self.url_result(mobj.group('url'), 'Facebook')
  575. # Look for embedded VK player
  576. mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
  577. if mobj is not None:
  578. return self.url_result(mobj.group('url'), 'VK')
  579. # Look for embedded ivi player
  580. mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage)
  581. if mobj is not None:
  582. return self.url_result(mobj.group('url'), 'Ivi')
  583. # Look for embedded Huffington Post player
  584. mobj = re.search(
  585. r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
  586. if mobj is not None:
  587. return self.url_result(mobj.group('url'), 'HuffPost')
  588. # Look for embed.ly
  589. mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage)
  590. if mobj is not None:
  591. return self.url_result(mobj.group('url'))
  592. mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage)
  593. if mobj is not None:
  594. return self.url_result(compat_urllib_parse.unquote(mobj.group('url')))
  595. # Look for funnyordie embed
  596. matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage)
  597. if matches:
  598. urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie')
  599. for eurl in matches]
  600. return self.playlist_result(
  601. urlrs, playlist_id=video_id, playlist_title=video_title)
  602. # Look for embedded RUTV player
  603. rutv_url = RUTVIE._extract_url(webpage)
  604. if rutv_url:
  605. return self.url_result(rutv_url, 'RUTV')
  606. # Look for embedded TED player
  607. mobj = re.search(
  608. r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage)
  609. if mobj is not None:
  610. return self.url_result(mobj.group('url'), 'TED')
  611. # Look for embedded Ustream videos
  612. mobj = re.search(
  613. r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage)
  614. if mobj is not None:
  615. return self.url_result(mobj.group('url'), 'Ustream')
  616. # Look for embedded arte.tv player
  617. mobj = re.search(
  618. r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"',
  619. webpage)
  620. if mobj is not None:
  621. return self.url_result(mobj.group('url'), 'ArteTVEmbed')
  622. # Look for embedded smotri.com player
  623. smotri_url = SmotriIE._extract_url(webpage)
  624. if smotri_url:
  625. return self.url_result(smotri_url, 'Smotri')
  626. # Look for embeded soundcloud player
  627. mobj = re.search(
  628. r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"',
  629. webpage)
  630. if mobj is not None:
  631. url = unescapeHTML(mobj.group('url'))
  632. return self.url_result(url)
  633. # Look for embedded vulture.com player
  634. mobj = re.search(
  635. r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
  636. webpage)
  637. if mobj is not None:
  638. url = unescapeHTML(mobj.group('url'))
  639. return self.url_result(url, ie='Vulture')
  640. # Look for embedded mtvservices player
  641. mobj = re.search(
  642. r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"',
  643. webpage)
  644. if mobj is not None:
  645. url = unescapeHTML(mobj.group('url'))
  646. return self.url_result(url, ie='MTVServicesEmbedded')
  647. # Start with something easy: JW Player in SWFObject
  648. found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
  649. if not found:
  650. # Look for gorilla-vid style embedding
  651. found = re.findall(r'''(?sx)
  652. (?:
  653. jw_plugins|
  654. JWPlayerOptions|
  655. jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
  656. )
  657. .*?file\s*:\s*["\'](.*?)["\']''', webpage)
  658. if not found:
  659. # Broaden the search a little bit
  660. found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
  661. if not found:
  662. # Broaden the findall a little bit: JWPlayer JS loader
  663. found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
  664. if not found:
  665. # Try to find twitter cards info
  666. found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
  667. if not found:
  668. # We look for Open Graph info:
  669. # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am)
  670. m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage)
  671. # We only look in og:video if the MIME type is a video, don't try if it's a Flash player:
  672. if m_video_type is not None:
  673. found = re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)
  674. if not found:
  675. # HTML5 video
  676. found = re.findall(r'(?s)<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage)
  677. if not found:
  678. found = re.search(
  679. r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
  680. r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"',
  681. webpage)
  682. if found:
  683. new_url = found.group(1)
  684. self.report_following_redirect(new_url)
  685. return {
  686. '_type': 'url',
  687. 'url': new_url,
  688. }
  689. if not found:
  690. raise ExtractorError('Unsupported URL: %s' % url)
  691. entries = []
  692. for video_url in found:
  693. video_url = compat_urlparse.urljoin(url, video_url)
  694. video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
  695. # Sometimes, jwplayer extraction will result in a YouTube URL
  696. if YoutubeIE.suitable(video_url):
  697. entries.append(self.url_result(video_url, 'Youtube'))
  698. continue
  699. # here's a fun little line of code for you:
  700. video_id = os.path.splitext(video_id)[0]
  701. entries.append({
  702. 'id': video_id,
  703. 'url': video_url,
  704. 'uploader': video_uploader,
  705. 'title': video_title,
  706. })
  707. if len(entries) == 1:
  708. return entries[0]
  709. else:
  710. for num, e in enumerate(entries, start=1):
  711. e['title'] = '%s (%d)' % (e['title'], num)
  712. return {
  713. '_type': 'playlist',
  714. 'entries': entries,
  715. }