3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from .subtitles import SubtitlesInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
21 compat_urllib_request,
27 get_element_by_attribute,
38 class YoutubeBaseInfoExtractor(InfoExtractor):
39 """Provide base functions for Youtube extractors"""
40 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
41 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
42 _NETRC_MACHINE = 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED = False
46 def _set_language(self):
48 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
49 # YouTube sets the expire time to about two months
50 expire_time=time.time() + 2 * 30 * 24 * 3600)
54 Attempt to log in to YouTube.
55 True is returned if successful or skipped.
56 False is returned if login failed.
58 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
60 (username, password) = self._get_login_info()
61 # No authentication to be performed
63 if self._LOGIN_REQUIRED:
64 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
67 login_page = self._download_webpage(
68 self._LOGIN_URL, None,
69 note='Downloading login page',
70 errnote='unable to fetch login page', fatal=False)
71 if login_page is False:
74 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
75 login_page, 'Login GALX parameter')
79 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
84 'PersistentCookie': 'yes',
86 'bgresponse': 'js_disabled',
87 'checkConnection': '',
88 'checkedDomains': 'youtube',
100 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
102 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
103 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
105 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
106 login_results = self._download_webpage(
108 note='Logging in', errnote='unable to log in', fatal=False)
109 if login_results is False:
112 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
113 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
116 # TODO add SMS and phone call support - these require making a request and then prompting the user
118 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
119 tfa_code = self._get_tfa_info()
122 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
123 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
126 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
128 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
130 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
131 secTok = match.group(1)
132 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
134 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
135 timeStmp = match.group(1)
138 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
140 'smsUserPin': tfa_code,
141 'smsVerifyPin': 'Verify',
143 'PersistentCookie': 'yes',
144 'checkConnection': '',
145 'checkedDomains': 'youtube',
148 'timeStmp': timeStmp,
149 'service': 'youtube',
152 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
153 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
155 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
156 tfa_results = self._download_webpage(
158 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
160 if tfa_results is False:
163 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
164 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
166 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
167 self._downloader.report_warning('unable to log in - did the page structure change?')
169 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
170 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
173 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
174 self._downloader.report_warning('unable to log in: bad username or password')
178 def _real_initialize(self):
179 if self._downloader is None:
182 if not self._login():
186 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
187 IE_DESC = 'YouTube.com'
188 _VALID_URL = r"""(?x)^
190 (?:https?://|//) # http(s):// or protocol-independent URL
191 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
192 (?:www\.)?deturl\.com/www\.youtube\.com/|
193 (?:www\.)?pwnyoutube\.com/|
194 (?:www\.)?yourepeat\.com/|
195 tube\.majestyc\.net/|
196 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
197 (?:.*?\#/)? # handle anchor (#/) redirect urls
198 (?: # the various things that can precede the ID:
199 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
200 |(?: # or the v= param in all its forms
201 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
202 (?:\?|\#!?) # the params delimiter ? or # or #!
203 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
207 |youtu\.be/ # just youtu.be/xxxx
208 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
210 )? # all until now is optional -> you can pass the naked ID
211 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
212 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
213 (?(1).+)? # if we found the ID, everything can follow
215 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
217 '5': {'ext': 'flv', 'width': 400, 'height': 240},
218 '6': {'ext': 'flv', 'width': 450, 'height': 270},
219 '13': {'ext': '3gp'},
220 '17': {'ext': '3gp', 'width': 176, 'height': 144},
221 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
222 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
223 '34': {'ext': 'flv', 'width': 640, 'height': 360},
224 '35': {'ext': 'flv', 'width': 854, 'height': 480},
225 '36': {'ext': '3gp', 'width': 320, 'height': 240},
226 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
227 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
228 '43': {'ext': 'webm', 'width': 640, 'height': 360},
229 '44': {'ext': 'webm', 'width': 854, 'height': 480},
230 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
231 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
235 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
236 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
237 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
238 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
239 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
240 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
241 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
243 # Apple HTTP Live Streaming
244 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
245 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
246 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
247 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
248 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
249 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
250 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
253 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
254 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
262 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
266 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
267 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
268 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
271 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
272 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
278 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
279 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
288 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
292 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
293 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
295 # Dash webm audio with opus inside
296 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
297 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
298 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
301 '_rtmp': {'protocol': 'rtmp'},
307 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
311 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
312 'uploader': 'Philipp Hagemeister',
313 'uploader_id': 'phihag',
314 'upload_date': '20121002',
315 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
316 'categories': ['Science & Technology'],
318 'dislike_count': int,
322 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
323 'note': 'Test generic use_cipher_signature video (#897)',
327 'upload_date': '20120506',
328 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
329 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
330 'uploader': 'Icona Pop',
331 'uploader_id': 'IconaPop',
335 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
336 'note': 'Test VEVO video with age protection (#956)',
340 'upload_date': '20130703',
341 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
342 'description': 'md5:64249768eec3bc4276236606ea996373',
343 'uploader': 'justintimberlakeVEVO',
344 'uploader_id': 'justintimberlakeVEVO',
348 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
349 'note': 'Embed-only video (#1746)',
353 'upload_date': '20120608',
354 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
355 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
356 'uploader': 'SET India',
357 'uploader_id': 'setindia'
361 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
362 'note': '256k DASH audio (format 141) via DASH manifest',
366 'upload_date': '20121002',
367 'uploader_id': '8KVIDEO',
369 'uploader': '8KVIDEO',
370 'title': 'UHDTV TEST 8K VIDEO.mp4'
373 'youtube_include_dash_manifest': True,
377 # DASH manifest with encrypted signature
379 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
383 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
384 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
385 'uploader': 'AfrojackVEVO',
386 'uploader_id': 'AfrojackVEVO',
387 'upload_date': '20131011',
390 'youtube_include_dash_manifest': True,
396 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
400 'upload_date': '20100909',
401 'uploader': 'The Amazing Atheist',
402 'uploader_id': 'TheAmazingAtheist',
403 'title': 'Burning Everyone\'s Koran',
404 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
407 # Normal age-gate video (No vevo, embed allowed)
409 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
413 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
414 'description': 'md5:eca57043abae25130f58f655ad9a7771',
415 'uploader': 'The Witcher',
416 'uploader_id': 'WitcherGame',
417 'upload_date': '20140605',
422 def __init__(self, *args, **kwargs):
423 super(YoutubeIE, self).__init__(*args, **kwargs)
424 self._player_cache = {}
426 def report_video_info_webpage_download(self, video_id):
427 """Report attempt to download video info webpage."""
428 self.to_screen('%s: Downloading video info webpage' % video_id)
430 def report_information_extraction(self, video_id):
431 """Report attempt to extract video information."""
432 self.to_screen('%s: Extracting video information' % video_id)
434 def report_unavailable_format(self, video_id, format):
435 """Report extracted video URL."""
436 self.to_screen('%s: Format %s not available' % (video_id, format))
438 def report_rtmp_download(self):
439 """Indicate the download will use the RTMP protocol."""
440 self.to_screen('RTMP download detected')
442 def _signature_cache_id(self, example_sig):
443 """ Return a string representation of a signature """
444 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
446 def _extract_signature_function(self, video_id, player_url, example_sig):
448 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
451 raise ExtractorError('Cannot identify player %r' % player_url)
452 player_type = id_m.group('ext')
453 player_id = id_m.group('id')
455 # Read from filesystem cache
456 func_id = '%s_%s_%s' % (
457 player_type, player_id, self._signature_cache_id(example_sig))
458 assert os.path.basename(func_id) == func_id
460 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
461 if cache_spec is not None:
462 return lambda s: ''.join(s[i] for i in cache_spec)
464 if player_type == 'js':
465 code = self._download_webpage(
466 player_url, video_id,
467 note='Downloading %s player %s' % (player_type, player_id),
468 errnote='Download of %s failed' % player_url)
469 res = self._parse_sig_js(code)
470 elif player_type == 'swf':
471 urlh = self._request_webpage(
472 player_url, video_id,
473 note='Downloading %s player %s' % (player_type, player_id),
474 errnote='Download of %s failed' % player_url)
476 res = self._parse_sig_swf(code)
478 assert False, 'Invalid player type %r' % player_type
480 if cache_spec is None:
481 test_string = ''.join(map(compat_chr, range(len(example_sig))))
482 cache_res = res(test_string)
483 cache_spec = [ord(c) for c in cache_res]
485 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
488 def _print_sig_code(self, func, example_sig):
489 def gen_sig_code(idxs):
490 def _genslice(start, end, step):
491 starts = '' if start == 0 else str(start)
492 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
493 steps = '' if step == 1 else (':%d' % step)
494 return 's[%s%s%s]' % (starts, ends, steps)
497 start = '(Never used)' # Quelch pyflakes warnings - start will be
498 # set as soon as step is set
499 for i, prev in zip(idxs[1:], idxs[:-1]):
503 yield _genslice(start, prev, step)
506 if i - prev in [-1, 1]:
515 yield _genslice(start, i, step)
517 test_string = ''.join(map(compat_chr, range(len(example_sig))))
518 cache_res = func(test_string)
519 cache_spec = [ord(c) for c in cache_res]
520 expr_code = ' + '.join(gen_sig_code(cache_spec))
521 signature_id_tuple = '(%s)' % (
522 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
523 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
524 ' return %s\n') % (signature_id_tuple, expr_code)
525 self.to_screen('Extracted signature function:\n' + code)
527 def _parse_sig_js(self, jscode):
528 funcname = self._search_regex(
529 r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
530 'Initial JS player signature function name')
532 jsi = JSInterpreter(jscode)
533 initial_function = jsi.extract_function(funcname)
534 return lambda s: initial_function([s])
536 def _parse_sig_swf(self, file_contents):
537 swfi = SWFInterpreter(file_contents)
538 TARGET_CLASSNAME = 'SignatureDecipher'
539 searched_class = swfi.extract_class(TARGET_CLASSNAME)
540 initial_function = swfi.extract_function(searched_class, 'decipher')
541 return lambda s: initial_function([s])
543 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
544 """Turn the encrypted s field into a working signature"""
546 if player_url is None:
547 raise ExtractorError('Cannot decrypt signature without player_url')
549 if player_url.startswith('//'):
550 player_url = 'https:' + player_url
552 player_id = (player_url, self._signature_cache_id(s))
553 if player_id not in self._player_cache:
554 func = self._extract_signature_function(
555 video_id, player_url, s
557 self._player_cache[player_id] = func
558 func = self._player_cache[player_id]
559 if self._downloader.params.get('youtube_print_sig_code'):
560 self._print_sig_code(func, s)
562 except Exception as e:
563 tb = traceback.format_exc()
564 raise ExtractorError(
565 'Signature extraction failed: ' + tb, cause=e)
567 def _get_available_subtitles(self, video_id, webpage):
569 sub_list = self._download_webpage(
570 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
571 video_id, note=False)
572 except ExtractorError as err:
573 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
575 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
580 if lang in sub_lang_list:
582 params = compat_urllib_parse.urlencode({
585 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
586 'name': unescapeHTML(l[0]).encode('utf-8'),
588 url = 'https://www.youtube.com/api/timedtext?' + params
589 sub_lang_list[lang] = url
590 if not sub_lang_list:
591 self._downloader.report_warning('video doesn\'t have subtitles')
595 def _get_available_automatic_caption(self, video_id, webpage):
596 """We need the webpage for getting the captions url, pass it as an
597 argument to speed up the process."""
598 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
599 self.to_screen('%s: Looking for automatic captions' % video_id)
600 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
601 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
603 self._downloader.report_warning(err_msg)
605 player_config = json.loads(mobj.group(1))
607 args = player_config['args']
608 caption_url = args['ttsurl']
609 timestamp = args['timestamp']
610 # We get the available subtitles
611 list_params = compat_urllib_parse.urlencode({
616 list_url = caption_url + '&' + list_params
617 caption_list = self._download_xml(list_url, video_id)
618 original_lang_node = caption_list.find('track')
619 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr':
620 self._downloader.report_warning('Video doesn\'t have automatic captions')
622 original_lang = original_lang_node.attrib['lang_code']
625 for lang_node in caption_list.findall('target'):
626 sub_lang = lang_node.attrib['lang_code']
627 params = compat_urllib_parse.urlencode({
628 'lang': original_lang,
634 sub_lang_list[sub_lang] = caption_url + '&' + params
636 # An extractor error can be raise by the download process if there are
637 # no automatic captions but there are subtitles
638 except (KeyError, ExtractorError):
639 self._downloader.report_warning(err_msg)
643 def extract_id(cls, url):
644 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
646 raise ExtractorError('Invalid URL: %s' % url)
647 video_id = mobj.group(2)
650 def _extract_from_m3u8(self, manifest_url, video_id):
653 def _get_urls(_manifest):
654 lines = _manifest.split('\n')
655 urls = filter(lambda l: l and not l.startswith('#'),
658 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
659 formats_urls = _get_urls(manifest)
660 for format_url in formats_urls:
661 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
662 url_map[itag] = format_url
665 def _extract_annotations(self, video_id):
666 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
667 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
669 def _real_extract(self, url):
671 'http' if self._downloader.params.get('prefer_insecure', False)
674 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
675 mobj = re.search(self._NEXT_URL_RE, url)
677 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
678 video_id = self.extract_id(url)
681 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
682 video_webpage = self._download_webpage(url, video_id)
684 # Attempt to extract SWF player URL
685 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
687 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
692 if re.search(r'player-age-gate-content">', video_webpage) is not None:
694 # We simulate the access to the video from www.youtube.com/v/{video_id}
695 # this can be viewed without login into Youtube
696 data = compat_urllib_parse.urlencode({
697 'video_id': video_id,
698 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
699 'sts': self._search_regex(
700 r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
702 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
703 video_info_webpage = self._download_webpage(
704 video_info_url, video_id,
705 note='Refetching age-gated info webpage',
706 errnote='unable to download video info webpage')
707 video_info = compat_parse_qs(video_info_webpage)
711 # Try looking directly into the video webpage
712 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
714 raise ValueError('Could not find ytplayer.config') # caught below
715 json_code = uppercase_escape(mobj.group(1))
716 ytplayer_config = json.loads(json_code)
717 args = ytplayer_config['args']
718 # Convert to the same format returned by compat_parse_qs
719 video_info = dict((k, [v]) for k, v in args.items())
720 if 'url_encoded_fmt_stream_map' not in args:
721 raise ValueError('No stream_map present') # caught below
723 # We fallback to the get_video_info pages (used by the embed page)
724 self.report_video_info_webpage_download(video_id)
725 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
727 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
728 % (proto, video_id, el_type))
729 video_info_webpage = self._download_webpage(
731 video_id, note=False,
732 errnote='unable to download video info webpage')
733 video_info = compat_parse_qs(video_info_webpage)
734 if 'token' in video_info:
736 if 'token' not in video_info:
737 if 'reason' in video_info:
738 raise ExtractorError(
739 'YouTube said: %s' % video_info['reason'][0],
740 expected=True, video_id=video_id)
742 raise ExtractorError(
743 '"token" parameter not in video info for unknown reason',
746 if 'view_count' in video_info:
747 view_count = int(video_info['view_count'][0])
751 # Check for "rental" videos
752 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
753 raise ExtractorError('"rental" videos not supported')
755 # Start extracting information
756 self.report_information_extraction(video_id)
759 if 'author' not in video_info:
760 raise ExtractorError('Unable to extract uploader name')
761 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
764 video_uploader_id = None
765 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
767 video_uploader_id = mobj.group(1)
769 self._downloader.report_warning('unable to extract uploader nickname')
772 if 'title' in video_info:
773 video_title = video_info['title'][0]
775 self._downloader.report_warning('Unable to extract video title')
779 # We try first to get a high quality image:
780 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
781 video_webpage, re.DOTALL)
782 if m_thumb is not None:
783 video_thumbnail = m_thumb.group(1)
784 elif 'thumbnail_url' not in video_info:
785 self._downloader.report_warning('unable to extract video thumbnail')
786 video_thumbnail = None
787 else: # don't panic if we can't find it
788 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
792 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
795 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
798 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
799 upload_date = unified_strdate(upload_date)
801 m_cat_container = self._search_regex(
802 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
803 video_webpage, 'categories', fatal=False)
805 category = self._html_search_regex(
806 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
808 video_categories = None if category is None else [category]
810 video_categories = None
813 video_description = get_element_by_id("eow-description", video_webpage)
814 if video_description:
815 video_description = re.sub(r'''(?x)
817 (?:[a-zA-Z-]+="[^"]+"\s+)*?
819 (?:[a-zA-Z-]+="[^"]+"\s+)*?
820 class="yt-uix-redirect-link"\s*>
823 ''', r'\1', video_description)
824 video_description = clean_html(video_description)
826 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
828 video_description = unescapeHTML(fd_mobj.group(1))
830 video_description = ''
832 def _extract_count(count_name):
833 count = self._search_regex(
834 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
835 video_webpage, count_name, default=None)
836 if count is not None:
837 return int(count.replace(',', ''))
839 like_count = _extract_count('like')
840 dislike_count = _extract_count('dislike')
843 video_subtitles = self.extract_subtitles(video_id, video_webpage)
845 if self._downloader.params.get('listsubtitles', False):
846 self._list_available_subtitles(video_id, video_webpage)
849 if 'length_seconds' not in video_info:
850 self._downloader.report_warning('unable to extract video duration')
851 video_duration = None
853 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
856 video_annotations = None
857 if self._downloader.params.get('writeannotations', False):
858 video_annotations = self._extract_annotations(video_id)
860 def _map_to_format_list(urlmap):
862 for itag, video_real_url in urlmap.items():
865 'url': video_real_url,
866 'player_url': player_url,
868 if itag in self._formats:
869 dct.update(self._formats[itag])
873 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
874 self.report_rtmp_download()
876 'format_id': '_rtmp',
878 'url': video_info['conn'][0],
879 'player_url': player_url,
881 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
882 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
883 if 'rtmpe%3Dyes' in encoded_url_map:
884 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
886 for url_data_str in encoded_url_map.split(','):
887 url_data = compat_parse_qs(url_data_str)
888 if 'itag' not in url_data or 'url' not in url_data:
890 format_id = url_data['itag'][0]
891 url = url_data['url'][0]
893 if 'sig' in url_data:
894 url += '&signature=' + url_data['sig'][0]
895 elif 's' in url_data:
896 encrypted_sig = url_data['s'][0]
899 jsplayer_url_json = self._search_regex(
900 r'"assets":.+?"js":\s*("[^"]+")',
901 video_webpage, 'JS player URL')
902 player_url = json.loads(jsplayer_url_json)
903 if player_url is None:
904 player_url_json = self._search_regex(
905 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
906 video_webpage, 'age gate player URL')
907 player_url = json.loads(player_url_json)
909 if self._downloader.params.get('verbose'):
910 if player_url is None:
911 player_version = 'unknown'
912 player_desc = 'unknown'
914 if player_url.endswith('swf'):
915 player_version = self._search_regex(
916 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
917 'flash player', fatal=False)
918 player_desc = 'flash player %s' % player_version
920 player_version = self._search_regex(
921 r'html5player-([^/]+?)(?:/html5player)?\.js',
923 'html5 player', fatal=False)
924 player_desc = 'html5 player %s' % player_version
926 parts_sizes = self._signature_cache_id(encrypted_sig)
927 self.to_screen('{%s} signature length %s, %s' %
928 (format_id, parts_sizes, player_desc))
930 signature = self._decrypt_signature(
931 encrypted_sig, video_id, player_url, age_gate)
932 url += '&signature=' + signature
933 if 'ratebypass' not in url:
934 url += '&ratebypass=yes'
935 url_map[format_id] = url
936 formats = _map_to_format_list(url_map)
937 elif video_info.get('hlsvp'):
938 manifest_url = video_info['hlsvp'][0]
939 url_map = self._extract_from_m3u8(manifest_url, video_id)
940 formats = _map_to_format_list(url_map)
942 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
944 # Look for the DASH manifest
945 if self._downloader.params.get('youtube_include_dash_manifest', True):
947 # The DASH manifest used needs to be the one from the original video_webpage.
948 # The one found in get_video_info seems to be using different signatures.
949 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
950 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
951 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
952 dash_manifest_url = video_info.get('dashmpd')[0]
954 def decrypt_sig(mobj):
956 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
957 return '/signature/%s' % dec_s
958 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
959 dash_doc = self._download_xml(
960 dash_manifest_url, video_id,
961 note='Downloading DASH manifest',
962 errnote='Could not download DASH manifest')
963 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
964 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
967 format_id = r.attrib['id']
968 video_url = url_el.text
969 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
971 'format_id': format_id,
973 'width': int_or_none(r.attrib.get('width')),
974 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
975 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
976 'filesize': filesize,
977 'fps': int_or_none(r.attrib.get('frameRate')),
980 existing_format = next(
982 if fo['format_id'] == format_id)
983 except StopIteration:
984 f.update(self._formats.get(format_id, {}))
987 existing_format.update(f)
989 except (ExtractorError, KeyError) as e:
990 self.report_warning('Skipping DASH manifest: %r' % e, video_id)
992 self._sort_formats(formats)
996 'uploader': video_uploader,
997 'uploader_id': video_uploader_id,
998 'upload_date': upload_date,
999 'title': video_title,
1000 'thumbnail': video_thumbnail,
1001 'description': video_description,
1002 'categories': video_categories,
1003 'subtitles': video_subtitles,
1004 'duration': video_duration,
1005 'age_limit': 18 if age_gate else 0,
1006 'annotations': video_annotations,
1007 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1008 'view_count': view_count,
1009 'like_count': like_count,
1010 'dislike_count': dislike_count,
1015 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1016 IE_DESC = 'YouTube.com playlists'
1017 _VALID_URL = r"""(?x)(?:
1022 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1023 \? (?:.*?&)*? (?:p|a|list)=
1027 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1028 # Top tracks, they can also include dots
1033 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1035 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1036 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
1037 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1038 IE_NAME = 'youtube:playlist'
1040 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1042 'title': 'ytdl test PL',
1043 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1045 'playlist_count': 3,
1047 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1049 'title': 'YDL_Empty_List',
1051 'playlist_count': 0,
1053 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1054 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1056 'title': '29C3: Not my department',
1058 'playlist_count': 95,
1060 'note': 'issue #673',
1061 'url': 'PLBB231211A4F62143',
1063 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1065 'playlist_mincount': 26,
1067 'note': 'Large playlist',
1068 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1070 'title': 'Uploads from Cauchemar',
1072 'playlist_mincount': 799,
1074 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1076 'title': 'YDL_safe_search',
1078 'playlist_count': 2,
1081 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1082 'playlist_count': 4,
1087 'note': 'Embedded SWF player',
1088 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1089 'playlist_count': 4,
1095 def _real_initialize(self):
1098 def _ids_to_results(self, ids):
1100 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1103 def _extract_mix(self, playlist_id):
1104 # The mixes are generated from a a single video
1105 # the id of the playlist is just 'RD' + video_id
1106 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1107 webpage = self._download_webpage(
1108 url, playlist_id, 'Downloading Youtube mix')
1109 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1111 search_title('playlist-title') or
1112 search_title('title long-title') or
1113 search_title('title'))
1114 title = clean_html(title_span)
1115 ids = orderedSet(re.findall(
1116 r'''(?xs)data-video-username=".*?".*?
1117 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1119 url_results = self._ids_to_results(ids)
1121 return self.playlist_result(url_results, playlist_id, title)
1123 def _real_extract(self, url):
1124 # Extract playlist id
1125 mobj = re.match(self._VALID_URL, url)
1127 raise ExtractorError('Invalid URL: %s' % url)
1128 playlist_id = mobj.group(1) or mobj.group(2)
1130 # Check if it's a video-specific URL
1131 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1132 if 'v' in query_dict:
1133 video_id = query_dict['v'][0]
1134 if self._downloader.params.get('noplaylist'):
1135 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1136 return self.url_result(video_id, 'Youtube', video_id=video_id)
1138 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1140 if playlist_id.startswith('RD'):
1141 # Mixes require a custom extraction process
1142 return self._extract_mix(playlist_id)
1143 if playlist_id.startswith('TL'):
1144 raise ExtractorError('For downloading YouTube.com top lists, use '
1145 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1147 url = self._TEMPLATE_URL % playlist_id
1148 page = self._download_webpage(url, playlist_id)
1149 more_widget_html = content_html = page
1151 # Check if the playlist exists or is private
1152 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1153 raise ExtractorError(
1154 'The playlist doesn\'t exist or is private, use --username or '
1155 '--netrc to access it.',
1158 # Extract the video ids from the playlist pages
1161 for page_num in itertools.count(1):
1162 matches = re.finditer(self._VIDEO_RE, content_html)
1163 # We remove the duplicates and the link with index 0
1164 # (it's not the first video of the playlist)
1165 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1168 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1172 more = self._download_json(
1173 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1174 'Downloading page #%s' % page_num,
1175 transform_source=uppercase_escape)
1176 content_html = more['content_html']
1177 more_widget_html = more['load_more_widget_html']
1179 playlist_title = self._html_search_regex(
1180 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1183 url_results = self._ids_to_results(ids)
1184 return self.playlist_result(url_results, playlist_id, playlist_title)
1187 class YoutubeTopListIE(YoutubePlaylistIE):
1188 IE_NAME = 'youtube:toplist'
1189 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1190 ' (Example: "yttoplist:music:Top Tracks")')
1191 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1193 'url': 'yttoplist:music:Trending',
1194 'playlist_mincount': 5,
1195 'skip': 'Only works for logged-in users',
1198 def _real_extract(self, url):
1199 mobj = re.match(self._VALID_URL, url)
1200 channel = mobj.group('chann')
1201 title = mobj.group('title')
1202 query = compat_urllib_parse.urlencode({'title': title})
1203 channel_page = self._download_webpage(
1204 'https://www.youtube.com/%s' % channel, title)
1205 link = self._html_search_regex(
1207 <a\s+href="([^"]+)".*?>\s*
1208 <span\s+class="branded-page-module-title-text">\s*
1209 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1210 channel_page, 'list')
1211 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1213 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1215 # sometimes the webpage doesn't contain the videos
1216 # retry until we get them
1217 for i in itertools.count(0):
1218 msg = 'Downloading Youtube mix'
1220 msg += ', retry #%d' % i
1222 webpage = self._download_webpage(url, title, msg)
1223 ids = orderedSet(re.findall(video_re, webpage))
1226 url_results = self._ids_to_results(ids)
1227 return self.playlist_result(url_results, playlist_title=title)
1230 class YoutubeChannelIE(InfoExtractor):
1231 IE_DESC = 'YouTube.com channels'
1232 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1233 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1234 _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
1235 IE_NAME = 'youtube:channel'
1237 'note': 'paginated channel',
1238 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1239 'playlist_mincount': 91,
1242 def extract_videos_from_page(self, page):
1244 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1245 if mobj.group(1) not in ids_in_page:
1246 ids_in_page.append(mobj.group(1))
1249 def _real_extract(self, url):
1250 # Extract channel id
1251 mobj = re.match(self._VALID_URL, url)
1253 raise ExtractorError('Invalid URL: %s' % url)
1255 # Download channel page
1256 channel_id = mobj.group(1)
1258 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1259 channel_page = self._download_webpage(url, channel_id)
1260 autogenerated = re.search(r'''(?x)
1262 channel-header-autogenerated-label|
1263 yt-channel-title-autogenerated
1264 )[^"]*"''', channel_page) is not None
1267 # The videos are contained in a single page
1268 # the ajax pages can't be used, they are empty
1269 video_ids = self.extract_videos_from_page(channel_page)
1271 # Download all channel pages using the json-based channel_ajax query
1272 for pagenum in itertools.count(1):
1273 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1274 page = self._download_json(
1275 url, channel_id, note='Downloading page #%s' % pagenum,
1276 transform_source=uppercase_escape)
1278 ids_in_page = self.extract_videos_from_page(page['content_html'])
1279 video_ids.extend(ids_in_page)
1281 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1284 self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1286 url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
1287 for video_id in video_ids]
1288 return self.playlist_result(url_entries, channel_id)
1291 class YoutubeUserIE(InfoExtractor):
1292 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1293 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1294 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1295 _GDATA_PAGE_SIZE = 50
1296 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1297 IE_NAME = 'youtube:user'
1300 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1301 'playlist_mincount': 320,
1303 'title': 'TheLinuxFoundation',
1306 'url': 'ytuser:phihag',
1307 'only_matching': True,
1311 def suitable(cls, url):
1312 # Don't return True if the url can be extracted with other youtube
1313 # extractor, the regex would is too permissive and it would match.
1314 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1315 if any(ie.suitable(url) for ie in other_ies):
1318 return super(YoutubeUserIE, cls).suitable(url)
1320 def _real_extract(self, url):
1322 mobj = re.match(self._VALID_URL, url)
1324 raise ExtractorError('Invalid URL: %s' % url)
1326 username = mobj.group(1)
1328 # Download video ids using YouTube Data API. Result size per
1329 # query is limited (currently to 50 videos) so we need to query
1330 # page by page until there are no video ids - it means we got
1333 def download_page(pagenum):
1334 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1336 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1337 page = self._download_webpage(
1338 gdata_url, username,
1339 'Downloading video ids from %d to %d' % (
1340 start_index, start_index + self._GDATA_PAGE_SIZE))
1343 response = json.loads(page)
1344 except ValueError as err:
1345 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1346 if 'entry' not in response['feed']:
1349 # Extract video identifiers
1350 entries = response['feed']['entry']
1351 for entry in entries:
1352 title = entry['title']['$t']
1353 video_id = entry['id']['$t'].split('/')[-1]
1357 'ie_key': 'Youtube',
1361 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1363 return self.playlist_result(url_results, playlist_title=username)
1366 class YoutubeSearchIE(SearchInfoExtractor):
1367 IE_DESC = 'YouTube.com searches'
1368 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1370 IE_NAME = 'youtube:search'
1371 _SEARCH_KEY = 'ytsearch'
1373 def _get_n_results(self, query, n):
1374 """Get a specified number of results for a query"""
1381 while (PAGE_SIZE * pagenum) < limit:
1382 result_url = self._API_URL % (
1383 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1384 (PAGE_SIZE * pagenum) + 1)
1385 data_json = self._download_webpage(
1386 result_url, video_id='query "%s"' % query,
1387 note='Downloading page %s' % (pagenum + 1),
1388 errnote='Unable to download API page')
1389 data = json.loads(data_json)
1390 api_response = data['data']
1392 if 'items' not in api_response:
1393 raise ExtractorError(
1394 '[youtube] No video results', expected=True)
1396 new_ids = list(video['id'] for video in api_response['items'])
1397 video_ids += new_ids
1399 limit = min(n, api_response['totalItems'])
1402 if len(video_ids) > n:
1403 video_ids = video_ids[:n]
1404 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1405 for video_id in video_ids]
1406 return self.playlist_result(videos, query)
1409 class YoutubeSearchDateIE(YoutubeSearchIE):
1410 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1411 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1412 _SEARCH_KEY = 'ytsearchdate'
1413 IE_DESC = 'YouTube.com searches, newest videos first'
1416 class YoutubeSearchURLIE(InfoExtractor):
1417 IE_DESC = 'YouTube.com search URLs'
1418 IE_NAME = 'youtube:search_url'
1419 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1421 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1422 'playlist_mincount': 5,
1424 'title': 'youtube-dl test video',
1428 def _real_extract(self, url):
1429 mobj = re.match(self._VALID_URL, url)
1430 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1432 webpage = self._download_webpage(url, query)
1433 result_code = self._search_regex(
1434 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1436 part_codes = re.findall(
1437 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1439 for part_code in part_codes:
1440 part_title = self._html_search_regex(
1441 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1442 part_url_snippet = self._html_search_regex(
1443 r'(?s)href="([^"]+)"', part_code, 'item URL')
1444 part_url = compat_urlparse.urljoin(
1445 'https://www.youtube.com/', part_url_snippet)
1449 'title': part_title,
1453 '_type': 'playlist',
1459 class YoutubeShowIE(InfoExtractor):
1460 IE_DESC = 'YouTube.com (multi-season) shows'
1461 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1462 IE_NAME = 'youtube:show'
1464 'url': 'http://www.youtube.com/show/airdisasters',
1465 'playlist_mincount': 3,
1467 'id': 'airdisasters',
1468 'title': 'Air Disasters',
1472 def _real_extract(self, url):
1473 mobj = re.match(self._VALID_URL, url)
1474 playlist_id = mobj.group('id')
1475 webpage = self._download_webpage(
1476 url, playlist_id, 'Downloading show webpage')
1477 # There's one playlist for each season of the show
1478 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1479 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1482 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1483 for season in m_seasons
1485 title = self._og_search_title(webpage, fatal=False)
1488 '_type': 'playlist',
1495 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1497 Base class for extractors that fetch info from
1498 http://www.youtube.com/feed_ajax
1499 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1501 _LOGIN_REQUIRED = True
1502 # use action_load_personal_feed instead of action_load_system_feed
1503 _PERSONAL_FEED = False
1506 def _FEED_TEMPLATE(self):
1507 action = 'action_load_system_feed'
1508 if self._PERSONAL_FEED:
1509 action = 'action_load_personal_feed'
1510 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1514 return 'youtube:%s' % self._FEED_NAME
1516 def _real_initialize(self):
1519 def _real_extract(self, url):
1522 for i in itertools.count(1):
1523 info = self._download_json(self._FEED_TEMPLATE % paging,
1524 '%s feed' % self._FEED_NAME,
1525 'Downloading page %s' % i)
1526 feed_html = info.get('feed_html') or info.get('content_html')
1527 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1528 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1529 ids = orderedSet(m.group(1) for m in m_ids)
1530 feed_entries.extend(
1531 self.url_result(video_id, 'Youtube', video_id=video_id)
1532 for video_id in ids)
1534 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1535 load_more_widget_html)
1538 paging = mobj.group('paging')
1539 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1542 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1543 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1544 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1545 _FEED_NAME = 'recommended'
1546 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1549 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1550 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1551 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1552 _FEED_NAME = 'watch_later'
1553 _PLAYLIST_TITLE = 'Youtube Watch Later'
1554 _PERSONAL_FEED = True
1557 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1558 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1559 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1560 _FEED_NAME = 'history'
1561 _PERSONAL_FEED = True
1562 _PLAYLIST_TITLE = 'Youtube Watch History'
1565 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1566 IE_NAME = 'youtube:favorites'
1567 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1568 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1569 _LOGIN_REQUIRED = True
1571 def _real_extract(self, url):
1572 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1573 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1574 return self.url_result(playlist_id, 'YoutubePlaylist')
1577 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1578 IE_NAME = 'youtube:subscriptions'
1579 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1580 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1583 def _real_extract(self, url):
1584 title = 'Youtube Subscriptions'
1585 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1587 # The extraction process is the same as for playlists, but the regex
1588 # for the video ids doesn't contain an index
1590 more_widget_html = content_html = page
1592 for page_num in itertools.count(1):
1593 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1594 new_ids = orderedSet(matches)
1597 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1601 more = self._download_json(
1602 'https://youtube.com/%s' % mobj.group('more'), title,
1603 'Downloading page #%s' % page_num,
1604 transform_source=uppercase_escape)
1605 content_html = more['content_html']
1606 more_widget_html = more['load_more_widget_html']
1609 '_type': 'playlist',
1611 'entries': self._ids_to_results(ids),
1615 class YoutubeTruncatedURLIE(InfoExtractor):
1616 IE_NAME = 'youtube:truncated_url'
1617 IE_DESC = False # Do not list
1618 _VALID_URL = r'''(?x)
1619 (?:https?://)?[^/]+/watch\?(?:
1621 annotation_id=annotation_[^&]+
1623 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1627 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1628 'only_matching': True,
1630 'url': 'http://www.youtube.com/watch?',
1631 'only_matching': True,
1634 def _real_extract(self, url):
1635 raise ExtractorError(
1636 'Did you forget to quote the URL? Remember that & is a meta '
1637 'character in most shells, so you want to put the URL in quotes, '
1639 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1640 ' or simply youtube-dl BaW_jenozKc .',