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.

1170 lines
51 KiB

12 years ago
12 years ago
12 years ago
12 years ago
  1. # coding: utf-8
  2. import json
  3. import netrc
  4. import re
  5. import socket
  6. import itertools
  7. from .common import InfoExtractor, SearchInfoExtractor
  8. from ..utils import (
  9. compat_http_client,
  10. compat_parse_qs,
  11. compat_urllib_error,
  12. compat_urllib_parse,
  13. compat_urllib_request,
  14. compat_str,
  15. clean_html,
  16. get_element_by_id,
  17. ExtractorError,
  18. unescapeHTML,
  19. unified_strdate,
  20. orderedSet,
  21. )
  22. class YoutubeBaseInfoExtractor(InfoExtractor):
  23. """Provide base functions for Youtube extractors"""
  24. _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
  25. _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
  26. _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
  27. _NETRC_MACHINE = 'youtube'
  28. # If True it will raise an error if no login info is provided
  29. _LOGIN_REQUIRED = False
  30. def report_lang(self):
  31. """Report attempt to set language."""
  32. self.to_screen(u'Setting language')
  33. def _set_language(self):
  34. request = compat_urllib_request.Request(self._LANG_URL)
  35. try:
  36. self.report_lang()
  37. compat_urllib_request.urlopen(request).read()
  38. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  39. self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
  40. return False
  41. return True
  42. def _login(self):
  43. (username, password) = self._get_login_info()
  44. # No authentication to be performed
  45. if username is None:
  46. if self._LOGIN_REQUIRED:
  47. raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  48. return False
  49. request = compat_urllib_request.Request(self._LOGIN_URL)
  50. try:
  51. login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
  52. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  53. self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
  54. return False
  55. galx = None
  56. dsh = None
  57. match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
  58. if match:
  59. galx = match.group(1)
  60. match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
  61. if match:
  62. dsh = match.group(1)
  63. # Log in
  64. login_form_strs = {
  65. u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
  66. u'Email': username,
  67. u'GALX': galx,
  68. u'Passwd': password,
  69. u'PersistentCookie': u'yes',
  70. u'_utf8': u'',
  71. u'bgresponse': u'js_disabled',
  72. u'checkConnection': u'',
  73. u'checkedDomains': u'youtube',
  74. u'dnConn': u'',
  75. u'dsh': dsh,
  76. u'pstMsg': u'0',
  77. u'rmShown': u'1',
  78. u'secTok': u'',
  79. u'signIn': u'Sign in',
  80. u'timeStmp': u'',
  81. u'service': u'youtube',
  82. u'uilel': u'3',
  83. u'hl': u'en_US',
  84. }
  85. # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
  86. # chokes on unicode
  87. login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
  88. login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
  89. request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
  90. try:
  91. self.report_login()
  92. login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
  93. if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
  94. self._downloader.report_warning(u'unable to log in: bad username or password')
  95. return False
  96. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  97. self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
  98. return False
  99. return True
  100. def _confirm_age(self):
  101. age_form = {
  102. 'next_url': '/',
  103. 'action_confirm': 'Confirm',
  104. }
  105. request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
  106. try:
  107. self.report_age_confirmation()
  108. compat_urllib_request.urlopen(request).read().decode('utf-8')
  109. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  110. raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
  111. return True
  112. def _real_initialize(self):
  113. if self._downloader is None:
  114. return
  115. if not self._set_language():
  116. return
  117. if not self._login():
  118. return
  119. self._confirm_age()
  120. class YoutubeIE(YoutubeBaseInfoExtractor):
  121. IE_DESC = u'YouTube.com'
  122. _VALID_URL = r"""^
  123. (
  124. (?:https?://)? # http(s):// (optional)
  125. (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
  126. tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
  127. (?:.*?\#/)? # handle anchor (#/) redirect urls
  128. (?: # the various things that can precede the ID:
  129. (?:(?:v|embed|e)/) # v/ or embed/ or e/
  130. |(?: # or the v= param in all its forms
  131. (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
  132. (?:\?|\#!?) # the params delimiter ? or # or #!
  133. (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
  134. v=
  135. )
  136. )? # optional -> youtube.com/xxxx is OK
  137. )? # all until now is optional -> you can pass the naked ID
  138. ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
  139. (?(1).+)? # if we found the ID, everything can follow
  140. $"""
  141. _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
  142. # Listed in order of quality
  143. _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13',
  144. '95', '94', '93', '92', '132', '151',
  145. # 3D
  146. '85', '84', '102', '83', '101', '82', '100',
  147. # Dash video
  148. '138', '137', '248', '136', '247', '135', '246',
  149. '245', '244', '134', '243', '133', '242', '160',
  150. # Dash audio
  151. '141', '172', '140', '171', '139',
  152. ]
  153. _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
  154. '95', '94', '93', '92', '132', '151',
  155. '85', '102', '84', '101', '83', '100', '82',
  156. # Dash video
  157. '138', '248', '137', '247', '136', '246', '245',
  158. '244', '135', '243', '134', '242', '133', '160',
  159. # Dash audio
  160. '172', '141', '171', '140', '139',
  161. ]
  162. _video_extensions = {
  163. '13': '3gp',
  164. '17': 'mp4',
  165. '18': 'mp4',
  166. '22': 'mp4',
  167. '37': 'mp4',
  168. '38': 'mp4',
  169. '43': 'webm',
  170. '44': 'webm',
  171. '45': 'webm',
  172. '46': 'webm',
  173. # 3d videos
  174. '82': 'mp4',
  175. '83': 'mp4',
  176. '84': 'mp4',
  177. '85': 'mp4',
  178. '100': 'webm',
  179. '101': 'webm',
  180. '102': 'webm',
  181. # videos that use m3u8
  182. '92': 'mp4',
  183. '93': 'mp4',
  184. '94': 'mp4',
  185. '95': 'mp4',
  186. '96': 'mp4',
  187. '132': 'mp4',
  188. '151': 'mp4',
  189. # Dash mp4
  190. '133': 'mp4',
  191. '134': 'mp4',
  192. '135': 'mp4',
  193. '136': 'mp4',
  194. '137': 'mp4',
  195. '138': 'mp4',
  196. '139': 'mp4',
  197. '140': 'mp4',
  198. '141': 'mp4',
  199. '160': 'mp4',
  200. # Dash webm
  201. '171': 'webm',
  202. '172': 'webm',
  203. '242': 'webm',
  204. '243': 'webm',
  205. '244': 'webm',
  206. '245': 'webm',
  207. '246': 'webm',
  208. '247': 'webm',
  209. '248': 'webm',
  210. }
  211. _video_dimensions = {
  212. '5': '240x400',
  213. '6': '???',
  214. '13': '???',
  215. '17': '144x176',
  216. '18': '360x640',
  217. '22': '720x1280',
  218. '34': '360x640',
  219. '35': '480x854',
  220. '37': '1080x1920',
  221. '38': '3072x4096',
  222. '43': '360x640',
  223. '44': '480x854',
  224. '45': '720x1280',
  225. '46': '1080x1920',
  226. '82': '360p',
  227. '83': '480p',
  228. '84': '720p',
  229. '85': '1080p',
  230. '92': '240p',
  231. '93': '360p',
  232. '94': '480p',
  233. '95': '720p',
  234. '96': '1080p',
  235. '100': '360p',
  236. '101': '480p',
  237. '102': '720p',
  238. '132': '240p',
  239. '151': '72p',
  240. '133': '240p',
  241. '134': '360p',
  242. '135': '480p',
  243. '136': '720p',
  244. '137': '1080p',
  245. '138': '>1080p',
  246. '139': '48k',
  247. '140': '128k',
  248. '141': '256k',
  249. '160': '192p',
  250. '171': '128k',
  251. '172': '256k',
  252. '242': '240p',
  253. '243': '360p',
  254. '244': '480p',
  255. '245': '480p',
  256. '246': '480p',
  257. '247': '720p',
  258. '248': '1080p',
  259. }
  260. _special_itags = {
  261. '82': '3D',
  262. '83': '3D',
  263. '84': '3D',
  264. '85': '3D',
  265. '100': '3D',
  266. '101': '3D',
  267. '102': '3D',
  268. '133': 'DASH Video',
  269. '134': 'DASH Video',
  270. '135': 'DASH Video',
  271. '136': 'DASH Video',
  272. '137': 'DASH Video',
  273. '138': 'DASH Video',
  274. '139': 'DASH Audio',
  275. '140': 'DASH Audio',
  276. '141': 'DASH Audio',
  277. '160': 'DASH Video',
  278. '171': 'DASH Audio',
  279. '172': 'DASH Audio',
  280. '242': 'DASH Video',
  281. '243': 'DASH Video',
  282. '244': 'DASH Video',
  283. '245': 'DASH Video',
  284. '246': 'DASH Video',
  285. '247': 'DASH Video',
  286. '248': 'DASH Video',
  287. }
  288. IE_NAME = u'youtube'
  289. _TESTS = [
  290. {
  291. u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
  292. u"file": u"BaW_jenozKc.mp4",
  293. u"info_dict": {
  294. u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
  295. u"uploader": u"Philipp Hagemeister",
  296. u"uploader_id": u"phihag",
  297. u"upload_date": u"20121002",
  298. u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
  299. }
  300. },
  301. {
  302. u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
  303. u"file": u"1ltcDfZMA3U.flv",
  304. u"note": u"Test VEVO video (#897)",
  305. u"info_dict": {
  306. u"upload_date": u"20070518",
  307. u"title": u"Maps - It Will Find You",
  308. u"description": u"Music video by Maps performing It Will Find You.",
  309. u"uploader": u"MuteUSA",
  310. u"uploader_id": u"MuteUSA"
  311. }
  312. },
  313. {
  314. u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
  315. u"file": u"UxxajLWwzqY.mp4",
  316. u"note": u"Test generic use_cipher_signature video (#897)",
  317. u"info_dict": {
  318. u"upload_date": u"20120506",
  319. u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
  320. u"description": u"md5:3e2666e0a55044490499ea45fe9037b7",
  321. u"uploader": u"Icona Pop",
  322. u"uploader_id": u"IconaPop"
  323. }
  324. },
  325. {
  326. u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
  327. u"file": u"07FYdnEawAQ.mp4",
  328. u"note": u"Test VEVO video with age protection (#956)",
  329. u"info_dict": {
  330. u"upload_date": u"20130703",
  331. u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
  332. u"description": u"md5:64249768eec3bc4276236606ea996373",
  333. u"uploader": u"justintimberlakeVEVO",
  334. u"uploader_id": u"justintimberlakeVEVO"
  335. }
  336. },
  337. {
  338. u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE',
  339. u'file': u'TGi3HqYrWHE.mp4',
  340. u'note': u'm3u8 video',
  341. u'info_dict': {
  342. u'title': u'Triathlon - Men - London 2012 Olympic Games',
  343. u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
  344. u'uploader': u'olympic',
  345. u'upload_date': u'20120807',
  346. u'uploader_id': u'olympic',
  347. },
  348. u'params': {
  349. u'skip_download': True,
  350. },
  351. },
  352. ]
  353. @classmethod
  354. def suitable(cls, url):
  355. """Receives a URL and returns True if suitable for this IE."""
  356. if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
  357. return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  358. def report_video_webpage_download(self, video_id):
  359. """Report attempt to download video webpage."""
  360. self.to_screen(u'%s: Downloading video webpage' % video_id)
  361. def report_video_info_webpage_download(self, video_id):
  362. """Report attempt to download video info webpage."""
  363. self.to_screen(u'%s: Downloading video info webpage' % video_id)
  364. def report_video_subtitles_download(self, video_id):
  365. """Report attempt to download video info webpage."""
  366. self.to_screen(u'%s: Checking available subtitles' % video_id)
  367. def report_video_subtitles_request(self, video_id, sub_lang, format):
  368. """Report attempt to download video info webpage."""
  369. self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
  370. def report_video_subtitles_available(self, video_id, sub_lang_list):
  371. """Report available subtitles."""
  372. sub_lang = ",".join(list(sub_lang_list.keys()))
  373. self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
  374. def report_information_extraction(self, video_id):
  375. """Report attempt to extract video information."""
  376. self.to_screen(u'%s: Extracting video information' % video_id)
  377. def report_unavailable_format(self, video_id, format):
  378. """Report extracted video URL."""
  379. self.to_screen(u'%s: Format %s not available' % (video_id, format))
  380. def report_rtmp_download(self):
  381. """Indicate the download will use the RTMP protocol."""
  382. self.to_screen(u'RTMP download detected')
  383. def _decrypt_signature(self, s):
  384. """Turn the encrypted s field into a working signature"""
  385. if len(s) == 92:
  386. return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
  387. elif len(s) == 90:
  388. return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
  389. elif len(s) == 89:
  390. return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1]
  391. elif len(s) == 88:
  392. return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28]
  393. elif len(s) == 87:
  394. return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
  395. elif len(s) == 86:
  396. return s[5:40] + s[3] + s[41:48] + s[0] + s[49:86]
  397. elif len(s) == 85:
  398. return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27]
  399. elif len(s) == 84:
  400. return s[5:40] + s[3] + s[41:48] + s[0] + s[49:84]
  401. elif len(s) == 83:
  402. return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0]
  403. elif len(s) == 82:
  404. return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82]
  405. elif len(s) == 81:
  406. return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9]
  407. elif len(s) == 80:
  408. return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80]
  409. elif len(s) == 79:
  410. return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9]
  411. else:
  412. raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
  413. def _decrypt_signature_age_gate(self, s):
  414. # The videos with age protection use another player, so the algorithms
  415. # can be different.
  416. if len(s) == 86:
  417. return s[2:63] + s[82] + s[64:82] + s[63]
  418. else:
  419. # Fallback to the other algortihms
  420. return self._decrypt_signature(s)
  421. def _get_available_subtitles(self, video_id):
  422. self.report_video_subtitles_download(video_id)
  423. request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
  424. try:
  425. sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
  426. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  427. self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
  428. return {}
  429. sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
  430. sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
  431. if not sub_lang_list:
  432. self._downloader.report_warning(u'video doesn\'t have subtitles')
  433. return {}
  434. return sub_lang_list
  435. def _list_available_subtitles(self, video_id):
  436. sub_lang_list = self._get_available_subtitles(video_id)
  437. self.report_video_subtitles_available(video_id, sub_lang_list)
  438. def _request_subtitle(self, sub_lang, sub_name, video_id, format):
  439. """
  440. Return the subtitle as a string or None if they are not found
  441. """
  442. self.report_video_subtitles_request(video_id, sub_lang, format)
  443. params = compat_urllib_parse.urlencode({
  444. 'lang': sub_lang,
  445. 'name': sub_name,
  446. 'v': video_id,
  447. 'fmt': format,
  448. })
  449. url = 'http://www.youtube.com/api/timedtext?' + params
  450. try:
  451. sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
  452. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  453. self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
  454. return
  455. if not sub:
  456. self._downloader.report_warning(u'Did not fetch video subtitles')
  457. return
  458. return sub
  459. def _request_automatic_caption(self, video_id, webpage):
  460. """We need the webpage for getting the captions url, pass it as an
  461. argument to speed up the process."""
  462. sub_lang = (self._downloader.params.get('subtitleslangs') or ['en'])[0]
  463. sub_format = self._downloader.params.get('subtitlesformat')
  464. self.to_screen(u'%s: Looking for automatic captions' % video_id)
  465. mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
  466. err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
  467. if mobj is None:
  468. self._downloader.report_warning(err_msg)
  469. return {}
  470. player_config = json.loads(mobj.group(1))
  471. try:
  472. args = player_config[u'args']
  473. caption_url = args[u'ttsurl']
  474. timestamp = args[u'timestamp']
  475. params = compat_urllib_parse.urlencode({
  476. 'lang': 'en',
  477. 'tlang': sub_lang,
  478. 'fmt': sub_format,
  479. 'ts': timestamp,
  480. 'kind': 'asr',
  481. })
  482. subtitles_url = caption_url + '&' + params
  483. sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
  484. return {sub_lang: sub}
  485. # An extractor error can be raise by the download process if there are
  486. # no automatic captions but there are subtitles
  487. except (KeyError, ExtractorError):
  488. self._downloader.report_warning(err_msg)
  489. return {}
  490. def _extract_subtitles(self, video_id):
  491. """
  492. Return a dictionary: {language: subtitles} or {} if the subtitles
  493. couldn't be found
  494. """
  495. available_subs_list = self._get_available_subtitles(video_id)
  496. sub_format = self._downloader.params.get('subtitlesformat')
  497. if not available_subs_list: #There was some error, it didn't get the available subtitles
  498. return {}
  499. if self._downloader.params.get('allsubtitles', False):
  500. sub_lang_list = available_subs_list
  501. else:
  502. if self._downloader.params.get('subtitleslangs', False):
  503. reqested_langs = self._downloader.params.get('subtitleslangs')
  504. elif 'en' in available_subs_list:
  505. reqested_langs = ['en']
  506. else:
  507. reqested_langs = [list(available_subs_list.keys())[0]]
  508. sub_lang_list = {}
  509. for sub_lang in reqested_langs:
  510. if not sub_lang in available_subs_list:
  511. self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
  512. continue
  513. sub_lang_list[sub_lang] = available_subs_list[sub_lang]
  514. subtitles = {}
  515. for sub_lang in sub_lang_list:
  516. subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
  517. if subtitle:
  518. subtitles[sub_lang] = subtitle
  519. return subtitles
  520. def _print_formats(self, formats):
  521. print('Available formats:')
  522. for x in formats:
  523. print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
  524. self._video_dimensions.get(x, '???'),
  525. ' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
  526. def _extract_id(self, url):
  527. mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  528. if mobj is None:
  529. raise ExtractorError(u'Invalid URL: %s' % url)
  530. video_id = mobj.group(2)
  531. return video_id
  532. def _get_video_url_list(self, url_map):
  533. """
  534. Transform a dictionary in the format {itag:url} to a list of (itag, url)
  535. with the requested formats.
  536. """
  537. req_format = self._downloader.params.get('format', None)
  538. format_limit = self._downloader.params.get('format_limit', None)
  539. available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
  540. if format_limit is not None and format_limit in available_formats:
  541. format_list = available_formats[available_formats.index(format_limit):]
  542. else:
  543. format_list = available_formats
  544. existing_formats = [x for x in format_list if x in url_map]
  545. if len(existing_formats) == 0:
  546. raise ExtractorError(u'no known formats available for video')
  547. if self._downloader.params.get('listformats', None):
  548. self._print_formats(existing_formats)
  549. return
  550. if req_format is None or req_format == 'best':
  551. video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
  552. elif req_format == 'worst':
  553. video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
  554. elif req_format in ('-1', 'all'):
  555. video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
  556. else:
  557. # Specific formats. We pick the first in a slash-delimeted sequence.
  558. # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
  559. req_formats = req_format.split('/')
  560. video_url_list = None
  561. for rf in req_formats:
  562. if rf in url_map:
  563. video_url_list = [(rf, url_map[rf])]
  564. break
  565. if video_url_list is None:
  566. raise ExtractorError(u'requested format not available')
  567. return video_url_list
  568. def _extract_from_m3u8(self, manifest_url, video_id):
  569. url_map = {}
  570. def _get_urls(_manifest):
  571. lines = _manifest.split('\n')
  572. urls = filter(lambda l: l and not l.startswith('#'),
  573. lines)
  574. return urls
  575. manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
  576. formats_urls = _get_urls(manifest)
  577. for format_url in formats_urls:
  578. itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
  579. url_map[itag] = format_url
  580. return url_map
  581. def _real_extract(self, url):
  582. if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
  583. self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).')
  584. # Extract original video URL from URL with redirection, like age verification, using next_url parameter
  585. mobj = re.search(self._NEXT_URL_RE, url)
  586. if mobj:
  587. url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
  588. video_id = self._extract_id(url)
  589. # Get video webpage
  590. self.report_video_webpage_download(video_id)
  591. url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
  592. request = compat_urllib_request.Request(url)
  593. try:
  594. video_webpage_bytes = compat_urllib_request.urlopen(request).read()
  595. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  596. raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
  597. video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
  598. # Attempt to extract SWF player URL
  599. mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
  600. if mobj is not None:
  601. player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
  602. else:
  603. player_url = None
  604. # Get video info
  605. self.report_video_info_webpage_download(video_id)
  606. if re.search(r'player-age-gate-content">', video_webpage) is not None:
  607. self.report_age_confirmation()
  608. age_gate = True
  609. # We simulate the access to the video from www.youtube.com/v/{video_id}
  610. # this can be viewed without login into Youtube
  611. data = compat_urllib_parse.urlencode({'video_id': video_id,
  612. 'el': 'embedded',
  613. 'gl': 'US',
  614. 'hl': 'en',
  615. 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
  616. 'asv': 3,
  617. 'sts':'1588',
  618. })
  619. video_info_url = 'https://www.youtube.com/get_video_info?' + data
  620. video_info_webpage = self._download_webpage(video_info_url, video_id,
  621. note=False,
  622. errnote='unable to download video info webpage')
  623. video_info = compat_parse_qs(video_info_webpage)
  624. else:
  625. age_gate = False
  626. for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
  627. video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
  628. % (video_id, el_type))
  629. video_info_webpage = self._download_webpage(video_info_url, video_id,
  630. note=False,
  631. errnote='unable to download video info webpage')
  632. video_info = compat_parse_qs(video_info_webpage)
  633. if 'token' in video_info:
  634. break
  635. if 'token' not in video_info:
  636. if 'reason' in video_info:
  637. raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
  638. else:
  639. raise ExtractorError(u'"token" parameter not in video info for unknown reason')
  640. # Check for "rental" videos
  641. if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
  642. raise ExtractorError(u'"rental" videos not supported')
  643. # Start extracting information
  644. self.report_information_extraction(video_id)
  645. # uploader
  646. if 'author' not in video_info:
  647. raise ExtractorError(u'Unable to extract uploader name')
  648. video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
  649. # uploader_id
  650. video_uploader_id = None
  651. mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
  652. if mobj is not None:
  653. video_uploader_id = mobj.group(1)
  654. else:
  655. self._downloader.report_warning(u'unable to extract uploader nickname')
  656. # title
  657. if 'title' not in video_info:
  658. raise ExtractorError(u'Unable to extract video title')
  659. video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
  660. # thumbnail image
  661. # We try first to get a high quality image:
  662. m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
  663. video_webpage, re.DOTALL)
  664. if m_thumb is not None:
  665. video_thumbnail = m_thumb.group(1)
  666. elif 'thumbnail_url' not in video_info:
  667. self._downloader.report_warning(u'unable to extract video thumbnail')
  668. video_thumbnail = ''
  669. else: # don't panic if we can't find it
  670. video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
  671. # upload date
  672. upload_date = None
  673. mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
  674. if mobj is not None:
  675. upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
  676. upload_date = unified_strdate(upload_date)
  677. # description
  678. video_description = get_element_by_id("eow-description", video_webpage)
  679. if video_description:
  680. video_description = clean_html(video_description)
  681. else:
  682. fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
  683. if fd_mobj:
  684. video_description = unescapeHTML(fd_mobj.group(1))
  685. else:
  686. video_description = u''
  687. # subtitles
  688. video_subtitles = None
  689. if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
  690. video_subtitles = self._extract_subtitles(video_id)
  691. elif self._downloader.params.get('writeautomaticsub', False):
  692. video_subtitles = self._request_automatic_caption(video_id, video_webpage)
  693. if self._downloader.params.get('listsubtitles', False):
  694. self._list_available_subtitles(video_id)
  695. return
  696. if 'length_seconds' not in video_info:
  697. self._downloader.report_warning(u'unable to extract video duration')
  698. video_duration = ''
  699. else:
  700. video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
  701. # Decide which formats to download
  702. try:
  703. mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
  704. if not mobj:
  705. raise ValueError('Could not find vevo ID')
  706. info = json.loads(mobj.group(1))
  707. args = info['args']
  708. # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
  709. # this signatures are encrypted
  710. m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
  711. if m_s is not None:
  712. self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
  713. video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
  714. m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u''))
  715. if m_s is not None:
  716. if 'url_encoded_fmt_stream_map' in video_info:
  717. video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
  718. else:
  719. video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
  720. elif 'adaptive_fmts' in video_info:
  721. if 'url_encoded_fmt_stream_map' in video_info:
  722. video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
  723. else:
  724. video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
  725. except ValueError:
  726. pass
  727. if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
  728. self.report_rtmp_download()
  729. video_url_list = [(None, video_info['conn'][0])]
  730. elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
  731. if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
  732. raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
  733. url_map = {}
  734. for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
  735. url_data = compat_parse_qs(url_data_str)
  736. if 'itag' in url_data and 'url' in url_data:
  737. url = url_data['url'][0]
  738. if 'sig' in url_data:
  739. url += '&signature=' + url_data['sig'][0]
  740. elif 's' in url_data:
  741. if self._downloader.params.get('verbose'):
  742. s = url_data['s'][0]
  743. if age_gate:
  744. player_version = self._search_regex(r'ad3-(.+?)\.swf',
  745. video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND',
  746. 'flash player', fatal=False)
  747. player = 'flash player %s' % player_version
  748. else:
  749. player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
  750. 'html5 player', fatal=False)
  751. parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
  752. self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
  753. (len(s), parts_sizes, url_data['itag'][0], player))
  754. encrypted_sig = url_data['s'][0]
  755. if age_gate:
  756. signature = self._decrypt_signature_age_gate(encrypted_sig)
  757. else:
  758. signature = self._decrypt_signature(encrypted_sig)
  759. url += '&signature=' + signature
  760. if 'ratebypass' not in url:
  761. url += '&ratebypass=yes'
  762. url_map[url_data['itag'][0]] = url
  763. video_url_list = self._get_video_url_list(url_map)
  764. if not video_url_list:
  765. return
  766. elif video_info.get('hlsvp'):
  767. manifest_url = video_info['hlsvp'][0]
  768. url_map = self._extract_from_m3u8(manifest_url, video_id)
  769. video_url_list = self._get_video_url_list(url_map)
  770. if not video_url_list:
  771. return
  772. else:
  773. raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
  774. results = []
  775. for format_param, video_real_url in video_url_list:
  776. # Extension
  777. video_extension = self._video_extensions.get(format_param, 'flv')
  778. video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
  779. self._video_dimensions.get(format_param, '???'),
  780. ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
  781. results.append({
  782. 'id': video_id,
  783. 'url': video_real_url,
  784. 'uploader': video_uploader,
  785. 'uploader_id': video_uploader_id,
  786. 'upload_date': upload_date,
  787. 'title': video_title,
  788. 'ext': video_extension,
  789. 'format': video_format,
  790. 'thumbnail': video_thumbnail,
  791. 'description': video_description,
  792. 'player_url': player_url,
  793. 'subtitles': video_subtitles,
  794. 'duration': video_duration
  795. })
  796. return results
  797. class YoutubePlaylistIE(InfoExtractor):
  798. IE_DESC = u'YouTube.com playlists'
  799. _VALID_URL = r"""(?:
  800. (?:https?://)?
  801. (?:\w+\.)?
  802. youtube\.com/
  803. (?:
  804. (?:course|view_play_list|my_playlists|artist|playlist|watch)
  805. \? (?:.*?&)*? (?:p|a|list)=
  806. | p/
  807. )
  808. ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
  809. .*
  810. |
  811. ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
  812. )"""
  813. _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
  814. _MAX_RESULTS = 50
  815. IE_NAME = u'youtube:playlist'
  816. @classmethod
  817. def suitable(cls, url):
  818. """Receives a URL and returns True if suitable for this IE."""
  819. return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  820. def _real_extract(self, url):
  821. # Extract playlist id
  822. mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  823. if mobj is None:
  824. raise ExtractorError(u'Invalid URL: %s' % url)
  825. # Download playlist videos from API
  826. playlist_id = mobj.group(1) or mobj.group(2)
  827. videos = []
  828. for page_num in itertools.count(1):
  829. start_index = self._MAX_RESULTS * (page_num - 1) + 1
  830. if start_index >= 1000:
  831. self._downloader.report_warning(u'Max number of results reached')
  832. break
  833. url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
  834. page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
  835. try:
  836. response = json.loads(page)
  837. except ValueError as err:
  838. raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
  839. if 'feed' not in response:
  840. raise ExtractorError(u'Got a malformed response from YouTube API')
  841. playlist_title = response['feed']['title']['$t']
  842. if 'entry' not in response['feed']:
  843. # Number of videos is a multiple of self._MAX_RESULTS
  844. break
  845. for entry in response['feed']['entry']:
  846. index = entry['yt$position']['$t']
  847. if 'media$group' in entry and 'media$player' in entry['media$group']:
  848. videos.append((index, entry['media$group']['media$player']['url']))
  849. videos = [v[1] for v in sorted(videos)]
  850. url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
  851. return [self.playlist_result(url_results, playlist_id, playlist_title)]
  852. class YoutubeChannelIE(InfoExtractor):
  853. IE_DESC = u'YouTube.com channels'
  854. _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
  855. _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
  856. _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
  857. _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
  858. IE_NAME = u'youtube:channel'
  859. def extract_videos_from_page(self, page):
  860. ids_in_page = []
  861. for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
  862. if mobj.group(1) not in ids_in_page:
  863. ids_in_page.append(mobj.group(1))
  864. return ids_in_page
  865. def _real_extract(self, url):
  866. # Extract channel id
  867. mobj = re.match(self._VALID_URL, url)
  868. if mobj is None:
  869. raise ExtractorError(u'Invalid URL: %s' % url)
  870. # Download channel page
  871. channel_id = mobj.group(1)
  872. video_ids = []
  873. pagenum = 1
  874. url = self._TEMPLATE_URL % (channel_id, pagenum)
  875. page = self._download_webpage(url, channel_id,
  876. u'Downloading page #%s' % pagenum)
  877. # Extract video identifiers
  878. ids_in_page = self.extract_videos_from_page(page)
  879. video_ids.extend(ids_in_page)
  880. # Download any subsequent channel pages using the json-based channel_ajax query
  881. if self._MORE_PAGES_INDICATOR in page:
  882. for pagenum in itertools.count(1):
  883. url = self._MORE_PAGES_URL % (pagenum, channel_id)
  884. page = self._download_webpage(url, channel_id,
  885. u'Downloading page #%s' % pagenum)
  886. page = json.loads(page)
  887. ids_in_page = self.extract_videos_from_page(page['content_html'])
  888. video_ids.extend(ids_in_page)
  889. if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
  890. break
  891. self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
  892. urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
  893. url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
  894. return [self.playlist_result(url_entries, channel_id)]
  895. class YoutubeUserIE(InfoExtractor):
  896. IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
  897. _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
  898. _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
  899. _GDATA_PAGE_SIZE = 50
  900. _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
  901. _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
  902. IE_NAME = u'youtube:user'
  903. def _real_extract(self, url):
  904. # Extract username
  905. mobj = re.match(self._VALID_URL, url)
  906. if mobj is None:
  907. raise ExtractorError(u'Invalid URL: %s' % url)
  908. username = mobj.group(1)
  909. # Download video ids using YouTube Data API. Result size per
  910. # query is limited (currently to 50 videos) so we need to query
  911. # page by page until there are no video ids - it means we got
  912. # all of them.
  913. video_ids = []
  914. for pagenum in itertools.count(0):
  915. start_index = pagenum * self._GDATA_PAGE_SIZE + 1
  916. gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
  917. page = self._download_webpage(gdata_url, username,
  918. u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
  919. # Extract video identifiers
  920. ids_in_page = []
  921. for mobj in re.finditer(self._VIDEO_INDICATOR, page):
  922. if mobj.group(1) not in ids_in_page:
  923. ids_in_page.append(mobj.group(1))
  924. video_ids.extend(ids_in_page)
  925. # A little optimization - if current page is not
  926. # "full", ie. does not contain PAGE_SIZE video ids then
  927. # we can assume that this page is the last one - there
  928. # are no more ids on further pages - no need to query
  929. # again.
  930. if len(ids_in_page) < self._GDATA_PAGE_SIZE:
  931. break
  932. urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
  933. url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
  934. return [self.playlist_result(url_results, playlist_title = username)]
  935. class YoutubeSearchIE(SearchInfoExtractor):
  936. IE_DESC = u'YouTube.com searches'
  937. _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
  938. _MAX_RESULTS = 1000
  939. IE_NAME = u'youtube:search'
  940. _SEARCH_KEY = 'ytsearch'
  941. def report_download_page(self, query, pagenum):
  942. """Report attempt to download search page with given number."""
  943. self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
  944. def _get_n_results(self, query, n):
  945. """Get a specified number of results for a query"""
  946. video_ids = []
  947. pagenum = 0
  948. limit = n
  949. while (50 * pagenum) < limit:
  950. self.report_download_page(query, pagenum+1)
  951. result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
  952. request = compat_urllib_request.Request(result_url)
  953. try:
  954. data = compat_urllib_request.urlopen(request).read().decode('utf-8')
  955. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  956. raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
  957. api_response = json.loads(data)['data']
  958. if not 'items' in api_response:
  959. raise ExtractorError(u'[youtube] No video results')
  960. new_ids = list(video['id'] for video in api_response['items'])
  961. video_ids += new_ids
  962. limit = min(n, api_response['totalItems'])
  963. pagenum += 1
  964. if len(video_ids) > n:
  965. video_ids = video_ids[:n]
  966. videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
  967. return self.playlist_result(videos, query)
  968. class YoutubeShowIE(InfoExtractor):
  969. IE_DESC = u'YouTube.com (multi-season) shows'
  970. _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
  971. IE_NAME = u'youtube:show'
  972. def _real_extract(self, url):
  973. mobj = re.match(self._VALID_URL, url)
  974. show_name = mobj.group(1)
  975. webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
  976. # There's one playlist for each season of the show
  977. m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
  978. self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
  979. return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
  980. class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
  981. """
  982. Base class for extractors that fetch info from
  983. http://www.youtube.com/feed_ajax
  984. Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
  985. """
  986. _LOGIN_REQUIRED = True
  987. _PAGING_STEP = 30
  988. # use action_load_personal_feed instead of action_load_system_feed
  989. _PERSONAL_FEED = False
  990. @property
  991. def _FEED_TEMPLATE(self):
  992. action = 'action_load_system_feed'
  993. if self._PERSONAL_FEED:
  994. action = 'action_load_personal_feed'
  995. return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
  996. @property
  997. def IE_NAME(self):
  998. return u'youtube:%s' % self._FEED_NAME
  999. def _real_initialize(self):
  1000. self._login()
  1001. def _real_extract(self, url):
  1002. feed_entries = []
  1003. # The step argument is available only in 2.7 or higher
  1004. for i in itertools.count(0):
  1005. paging = i*self._PAGING_STEP
  1006. info = self._download_webpage(self._FEED_TEMPLATE % paging,
  1007. u'%s feed' % self._FEED_NAME,
  1008. u'Downloading page %s' % i)
  1009. info = json.loads(info)
  1010. feed_html = info['feed_html']
  1011. m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
  1012. ids = orderedSet(m.group(1) for m in m_ids)
  1013. feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
  1014. if info['paging'] is None:
  1015. break
  1016. return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
  1017. class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
  1018. IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
  1019. _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
  1020. _FEED_NAME = 'subscriptions'
  1021. _PLAYLIST_TITLE = u'Youtube Subscriptions'
  1022. class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
  1023. IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
  1024. _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
  1025. _FEED_NAME = 'recommended'
  1026. _PLAYLIST_TITLE = u'Youtube Recommended videos'
  1027. class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
  1028. IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
  1029. _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
  1030. _FEED_NAME = 'watch_later'
  1031. _PLAYLIST_TITLE = u'Youtube Watch Later'
  1032. _PAGING_STEP = 100
  1033. _PERSONAL_FEED = True
  1034. class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
  1035. IE_NAME = u'youtube:favorites'
  1036. IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
  1037. _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
  1038. _LOGIN_REQUIRED = True
  1039. def _real_extract(self, url):
  1040. webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
  1041. playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
  1042. return self.url_result(playlist_id, 'YoutubePlaylist')