3 from __future__ import unicode_literals
12 from .common import InfoExtractor, SearchInfoExtractor
13 from .subtitles import SubtitlesInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
20 compat_urllib_request,
26 get_element_by_attribute,
36 class YoutubeBaseInfoExtractor(InfoExtractor):
37 """Provide base functions for Youtube extractors"""
38 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
39 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
40 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
41 _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
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):
47 return bool(self._download_webpage(
49 note='Setting language', errnote='unable to set language',
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 _confirm_age(self):
181 'action_confirm': 'Confirm',
183 req = compat_urllib_request.Request(self._AGE_URL,
184 compat_urllib_parse.urlencode(age_form).encode('ascii'))
186 self._download_webpage(
188 note='Confirming age', errnote='Unable to confirm age',
191 def _real_initialize(self):
192 if self._downloader is None:
194 if self._get_login_info()[0] is not None:
195 if not self._set_language():
197 if not self._login():
202 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
203 IE_DESC = 'YouTube.com'
204 _VALID_URL = r"""(?x)^
206 (?:https?://|//) # http(s):// or protocol-independent URL
207 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
208 (?:www\.)?deturl\.com/www\.youtube\.com/|
209 (?:www\.)?pwnyoutube\.com/|
210 (?:www\.)?yourepeat\.com/|
211 tube\.majestyc\.net/|
212 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
213 (?:.*?\#/)? # handle anchor (#/) redirect urls
214 (?: # the various things that can precede the ID:
215 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
216 |(?: # or the v= param in all its forms
217 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
218 (?:\?|\#!?) # the params delimiter ? or # or #!
219 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
223 |youtu\.be/ # just youtu.be/xxxx
224 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
226 )? # all until now is optional -> you can pass the naked ID
227 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
228 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
229 (?(1).+)? # if we found the ID, everything can follow
231 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
233 '5': {'ext': 'flv', 'width': 400, 'height': 240},
234 '6': {'ext': 'flv', 'width': 450, 'height': 270},
235 '13': {'ext': '3gp'},
236 '17': {'ext': '3gp', 'width': 176, 'height': 144},
237 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
238 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
239 '34': {'ext': 'flv', 'width': 640, 'height': 360},
240 '35': {'ext': 'flv', 'width': 854, 'height': 480},
241 '36': {'ext': '3gp', 'width': 320, 'height': 240},
242 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
243 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
244 '43': {'ext': 'webm', 'width': 640, 'height': 360},
245 '44': {'ext': 'webm', 'width': 854, 'height': 480},
246 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
247 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
251 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
252 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
253 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
254 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
255 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
256 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
257 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
259 # Apple HTTP Live Streaming
260 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
261 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
262 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
263 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
264 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
265 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
266 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
269 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
276 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
279 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
280 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
281 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
284 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
291 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
301 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
304 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
305 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
308 '298': {'ext': 'mov', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
309 '299': {'ext': 'mov', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
312 '_rtmp': {'protocol': 'rtmp'},
318 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
322 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
323 'uploader': 'Philipp Hagemeister',
324 'uploader_id': 'phihag',
325 'upload_date': '20121002',
326 '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 .',
327 'categories': ['Science & Technology'],
329 'dislike_count': int,
333 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
334 'note': 'Test generic use_cipher_signature video (#897)',
338 'upload_date': '20120506',
339 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
340 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
341 'uploader': 'Icona Pop',
342 'uploader_id': 'IconaPop',
346 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
347 'note': 'Test VEVO video with age protection (#956)',
351 'upload_date': '20130703',
352 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
353 'description': 'md5:64249768eec3bc4276236606ea996373',
354 'uploader': 'justintimberlakeVEVO',
355 'uploader_id': 'justintimberlakeVEVO',
359 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
360 'note': 'Embed-only video (#1746)',
364 'upload_date': '20120608',
365 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
366 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
367 'uploader': 'SET India',
368 'uploader_id': 'setindia'
372 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
373 'note': '256k DASH audio (format 141) via DASH manifest',
377 'upload_date': '20121002',
378 'uploader_id': '8KVIDEO',
380 'uploader': '8KVIDEO',
381 'title': 'UHDTV TEST 8K VIDEO.mp4'
384 'youtube_include_dash_manifest': True,
388 # DASH manifest with encrypted signature
390 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
394 'title': 'Afrojack - The Spark ft. Spree Wilson',
395 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8',
396 'uploader': 'AfrojackVEVO',
397 'uploader_id': 'AfrojackVEVO',
398 'upload_date': '20131011',
401 'youtube_include_dash_manifest': True,
407 def __init__(self, *args, **kwargs):
408 super(YoutubeIE, self).__init__(*args, **kwargs)
409 self._player_cache = {}
411 def report_video_info_webpage_download(self, video_id):
412 """Report attempt to download video info webpage."""
413 self.to_screen('%s: Downloading video info webpage' % video_id)
415 def report_information_extraction(self, video_id):
416 """Report attempt to extract video information."""
417 self.to_screen('%s: Extracting video information' % video_id)
419 def report_unavailable_format(self, video_id, format):
420 """Report extracted video URL."""
421 self.to_screen('%s: Format %s not available' % (video_id, format))
423 def report_rtmp_download(self):
424 """Indicate the download will use the RTMP protocol."""
425 self.to_screen('RTMP download detected')
427 def _signature_cache_id(self, example_sig):
428 """ Return a string representation of a signature """
429 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
431 def _extract_signature_function(self, video_id, player_url, example_sig):
433 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
436 raise ExtractorError('Cannot identify player %r' % player_url)
437 player_type = id_m.group('ext')
438 player_id = id_m.group('id')
440 # Read from filesystem cache
441 func_id = '%s_%s_%s' % (
442 player_type, player_id, self._signature_cache_id(example_sig))
443 assert os.path.basename(func_id) == func_id
445 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
446 if cache_spec is not None:
447 return lambda s: ''.join(s[i] for i in cache_spec)
449 if player_type == 'js':
450 code = self._download_webpage(
451 player_url, video_id,
452 note='Downloading %s player %s' % (player_type, player_id),
453 errnote='Download of %s failed' % player_url)
454 res = self._parse_sig_js(code)
455 elif player_type == 'swf':
456 urlh = self._request_webpage(
457 player_url, video_id,
458 note='Downloading %s player %s' % (player_type, player_id),
459 errnote='Download of %s failed' % player_url)
461 res = self._parse_sig_swf(code)
463 assert False, 'Invalid player type %r' % player_type
465 if cache_spec is None:
466 test_string = ''.join(map(compat_chr, range(len(example_sig))))
467 cache_res = res(test_string)
468 cache_spec = [ord(c) for c in cache_res]
470 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
473 def _print_sig_code(self, func, example_sig):
474 def gen_sig_code(idxs):
475 def _genslice(start, end, step):
476 starts = '' if start == 0 else str(start)
477 ends = (':%d' % (end+step)) if end + step >= 0 else ':'
478 steps = '' if step == 1 else (':%d' % step)
479 return 's[%s%s%s]' % (starts, ends, steps)
482 start = '(Never used)' # Quelch pyflakes warnings - start will be
483 # set as soon as step is set
484 for i, prev in zip(idxs[1:], idxs[:-1]):
488 yield _genslice(start, prev, step)
491 if i - prev in [-1, 1]:
500 yield _genslice(start, i, step)
502 test_string = ''.join(map(compat_chr, range(len(example_sig))))
503 cache_res = func(test_string)
504 cache_spec = [ord(c) for c in cache_res]
505 expr_code = ' + '.join(gen_sig_code(cache_spec))
506 signature_id_tuple = '(%s)' % (
507 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
508 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
509 ' return %s\n') % (signature_id_tuple, expr_code)
510 self.to_screen('Extracted signature function:\n' + code)
512 def _parse_sig_js(self, jscode):
513 funcname = self._search_regex(
514 r'signature=([$a-zA-Z]+)', jscode,
515 'Initial JS player signature function name')
517 jsi = JSInterpreter(jscode)
518 initial_function = jsi.extract_function(funcname)
519 return lambda s: initial_function([s])
521 def _parse_sig_swf(self, file_contents):
522 swfi = SWFInterpreter(file_contents)
523 TARGET_CLASSNAME = 'SignatureDecipher'
524 searched_class = swfi.extract_class(TARGET_CLASSNAME)
525 initial_function = swfi.extract_function(searched_class, 'decipher')
526 return lambda s: initial_function([s])
528 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
529 """Turn the encrypted s field into a working signature"""
531 if player_url is None:
532 raise ExtractorError('Cannot decrypt signature without player_url')
534 if player_url.startswith('//'):
535 player_url = 'https:' + player_url
537 player_id = (player_url, self._signature_cache_id(s))
538 if player_id not in self._player_cache:
539 func = self._extract_signature_function(
540 video_id, player_url, s
542 self._player_cache[player_id] = func
543 func = self._player_cache[player_id]
544 if self._downloader.params.get('youtube_print_sig_code'):
545 self._print_sig_code(func, s)
547 except Exception as e:
548 tb = traceback.format_exc()
549 raise ExtractorError(
550 'Signature extraction failed: ' + tb, cause=e)
552 def _get_available_subtitles(self, video_id, webpage):
554 sub_list = self._download_webpage(
555 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
556 video_id, note=False)
557 except ExtractorError as err:
558 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
560 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
565 if lang in sub_lang_list:
567 params = compat_urllib_parse.urlencode({
570 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
571 'name': unescapeHTML(l[0]).encode('utf-8'),
573 url = 'https://www.youtube.com/api/timedtext?' + params
574 sub_lang_list[lang] = url
575 if not sub_lang_list:
576 self._downloader.report_warning('video doesn\'t have subtitles')
580 def _get_available_automatic_caption(self, video_id, webpage):
581 """We need the webpage for getting the captions url, pass it as an
582 argument to speed up the process."""
583 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
584 self.to_screen('%s: Looking for automatic captions' % video_id)
585 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
586 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
588 self._downloader.report_warning(err_msg)
590 player_config = json.loads(mobj.group(1))
592 args = player_config[u'args']
593 caption_url = args[u'ttsurl']
594 timestamp = args[u'timestamp']
595 # We get the available subtitles
596 list_params = compat_urllib_parse.urlencode({
601 list_url = caption_url + '&' + list_params
602 caption_list = self._download_xml(list_url, video_id)
603 original_lang_node = caption_list.find('track')
604 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' :
605 self._downloader.report_warning('Video doesn\'t have automatic captions')
607 original_lang = original_lang_node.attrib['lang_code']
610 for lang_node in caption_list.findall('target'):
611 sub_lang = lang_node.attrib['lang_code']
612 params = compat_urllib_parse.urlencode({
613 'lang': original_lang,
619 sub_lang_list[sub_lang] = caption_url + '&' + params
621 # An extractor error can be raise by the download process if there are
622 # no automatic captions but there are subtitles
623 except (KeyError, ExtractorError):
624 self._downloader.report_warning(err_msg)
628 def extract_id(cls, url):
629 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
631 raise ExtractorError('Invalid URL: %s' % url)
632 video_id = mobj.group(2)
635 def _extract_from_m3u8(self, manifest_url, video_id):
637 def _get_urls(_manifest):
638 lines = _manifest.split('\n')
639 urls = filter(lambda l: l and not l.startswith('#'),
642 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
643 formats_urls = _get_urls(manifest)
644 for format_url in formats_urls:
645 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
646 url_map[itag] = format_url
649 def _extract_annotations(self, video_id):
650 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
651 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
653 def _real_extract(self, url):
655 'http' if self._downloader.params.get('prefer_insecure', False)
658 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
659 mobj = re.search(self._NEXT_URL_RE, url)
661 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
662 video_id = self.extract_id(url)
665 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
667 c for c in self._downloader.cookiejar
668 if c.domain == '.youtube.com' and c.name == 'PREF']
669 for pc in pref_cookies:
670 if 'hl=' in pc.value:
671 pc.value = re.sub(r'hl=[^&]+', 'hl=en', pc.value)
676 video_webpage = self._download_webpage(url, video_id)
678 # Attempt to extract SWF player URL
679 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
681 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
686 self.report_video_info_webpage_download(video_id)
687 if re.search(r'player-age-gate-content">', video_webpage) is not None:
688 self.report_age_confirmation()
690 # We simulate the access to the video from www.youtube.com/v/{video_id}
691 # this can be viewed without login into Youtube
692 data = compat_urllib_parse.urlencode({
693 'video_id': video_id,
694 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
695 'sts': self._search_regex(
696 r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'),
698 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
699 video_info_webpage = self._download_webpage(video_info_url, video_id,
701 errnote='unable to download video info webpage')
702 video_info = compat_parse_qs(video_info_webpage)
705 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
706 video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
707 % (video_id, el_type))
708 video_info_webpage = self._download_webpage(video_info_url, video_id,
710 errnote='unable to download video info webpage')
711 video_info = compat_parse_qs(video_info_webpage)
712 if 'token' in video_info:
714 if 'token' not in video_info:
715 if 'reason' in video_info:
716 raise ExtractorError(
717 'YouTube said: %s' % video_info['reason'][0],
718 expected=True, video_id=video_id)
720 raise ExtractorError(
721 '"token" parameter not in video info for unknown reason',
724 if 'view_count' in video_info:
725 view_count = int(video_info['view_count'][0])
729 # Check for "rental" videos
730 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
731 raise ExtractorError('"rental" videos not supported')
733 # Start extracting information
734 self.report_information_extraction(video_id)
737 if 'author' not in video_info:
738 raise ExtractorError('Unable to extract uploader name')
739 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
742 video_uploader_id = None
743 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
745 video_uploader_id = mobj.group(1)
747 self._downloader.report_warning('unable to extract uploader nickname')
750 if 'title' in video_info:
751 video_title = video_info['title'][0]
753 self._downloader.report_warning('Unable to extract video title')
757 # We try first to get a high quality image:
758 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
759 video_webpage, re.DOTALL)
760 if m_thumb is not None:
761 video_thumbnail = m_thumb.group(1)
762 elif 'thumbnail_url' not in video_info:
763 self._downloader.report_warning('unable to extract video thumbnail')
764 video_thumbnail = None
765 else: # don't panic if we can't find it
766 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
770 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
773 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
776 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
777 upload_date = unified_strdate(upload_date)
779 m_cat_container = self._search_regex(
780 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
781 video_webpage, 'categories', fatal=False)
783 category = self._html_search_regex(
784 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
786 video_categories = None if category is None else [category]
788 video_categories = None
791 video_description = get_element_by_id("eow-description", video_webpage)
792 if video_description:
793 video_description = re.sub(r'''(?x)
795 (?:[a-zA-Z-]+="[^"]+"\s+)*?
797 (?:[a-zA-Z-]+="[^"]+"\s+)*?
798 class="yt-uix-redirect-link"\s*>
801 ''', r'\1', video_description)
802 video_description = clean_html(video_description)
804 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
806 video_description = unescapeHTML(fd_mobj.group(1))
808 video_description = ''
810 def _extract_count(count_name):
811 count = self._search_regex(
812 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
813 video_webpage, count_name, default=None)
814 if count is not None:
815 return int(count.replace(',', ''))
817 like_count = _extract_count('like')
818 dislike_count = _extract_count('dislike')
821 video_subtitles = self.extract_subtitles(video_id, video_webpage)
823 if self._downloader.params.get('listsubtitles', False):
824 self._list_available_subtitles(video_id, video_webpage)
827 if 'length_seconds' not in video_info:
828 self._downloader.report_warning('unable to extract video duration')
829 video_duration = None
831 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
834 video_annotations = None
835 if self._downloader.params.get('writeannotations', False):
836 video_annotations = self._extract_annotations(video_id)
838 # Decide which formats to download
840 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
842 raise ValueError('Could not find vevo ID')
843 json_code = uppercase_escape(mobj.group(1))
844 ytplayer_config = json.loads(json_code)
845 args = ytplayer_config['args']
846 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
847 # this signatures are encrypted
848 if 'url_encoded_fmt_stream_map' not in args:
849 raise ValueError('No stream_map present') # caught below
850 re_signature = re.compile(r'[&,]s=')
851 m_s = re_signature.search(args['url_encoded_fmt_stream_map'])
853 self.to_screen('%s: Encrypted signatures detected.' % video_id)
854 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
855 m_s = re_signature.search(args.get('adaptive_fmts', ''))
857 if 'adaptive_fmts' in video_info:
858 video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
860 video_info['adaptive_fmts'] = [args['adaptive_fmts']]
864 def _map_to_format_list(urlmap):
866 for itag, video_real_url in urlmap.items():
869 'url': video_real_url,
870 'player_url': player_url,
872 if itag in self._formats:
873 dct.update(self._formats[itag])
877 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
878 self.report_rtmp_download()
880 'format_id': '_rtmp',
882 'url': video_info['conn'][0],
883 'player_url': player_url,
885 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
886 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0]
887 if 'rtmpe%3Dyes' in encoded_url_map:
888 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
890 for url_data_str in encoded_url_map.split(','):
891 url_data = compat_parse_qs(url_data_str)
892 if 'itag' not in url_data or 'url' not in url_data:
894 format_id = url_data['itag'][0]
895 url = url_data['url'][0]
897 if 'sig' in url_data:
898 url += '&signature=' + url_data['sig'][0]
899 elif 's' in url_data:
900 encrypted_sig = url_data['s'][0]
903 jsplayer_url_json = self._search_regex(
904 r'"assets":.+?"js":\s*("[^"]+")',
905 video_webpage, 'JS player URL')
906 player_url = json.loads(jsplayer_url_json)
907 if player_url is None:
908 player_url_json = self._search_regex(
909 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
910 video_webpage, 'age gate player URL')
911 player_url = json.loads(player_url_json)
913 if self._downloader.params.get('verbose'):
914 if player_url is None:
915 player_version = 'unknown'
916 player_desc = 'unknown'
918 if player_url.endswith('swf'):
919 player_version = self._search_regex(
920 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
921 'flash player', fatal=False)
922 player_desc = 'flash player %s' % player_version
924 player_version = self._search_regex(
925 r'html5player-([^/]+?)(?:/html5player)?\.js',
927 'html5 player', fatal=False)
928 player_desc = 'html5 player %s' % player_version
930 parts_sizes = self._signature_cache_id(encrypted_sig)
931 self.to_screen('{%s} signature length %s, %s' %
932 (format_id, parts_sizes, player_desc))
934 signature = self._decrypt_signature(
935 encrypted_sig, video_id, player_url, age_gate)
936 url += '&signature=' + signature
937 if 'ratebypass' not in url:
938 url += '&ratebypass=yes'
939 url_map[format_id] = url
940 formats = _map_to_format_list(url_map)
941 elif video_info.get('hlsvp'):
942 manifest_url = video_info['hlsvp'][0]
943 url_map = self._extract_from_m3u8(manifest_url, video_id)
944 formats = _map_to_format_list(url_map)
946 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
948 # Look for the DASH manifest
949 if self._downloader.params.get('youtube_include_dash_manifest', True):
951 # The DASH manifest used needs to be the one from the original video_webpage.
952 # The one found in get_video_info seems to be using different signatures.
953 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
954 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
955 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
957 dash_manifest_url = video_info.get('dashmpd')[0]
959 dash_manifest_url = ytplayer_config['args']['dashmpd']
960 def decrypt_sig(mobj):
962 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
963 return '/signature/%s' % dec_s
964 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
965 dash_doc = self._download_xml(
966 dash_manifest_url, video_id,
967 note='Downloading DASH manifest',
968 errnote='Could not download DASH manifest')
969 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
970 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
973 format_id = r.attrib['id']
974 video_url = url_el.text
975 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
977 'format_id': format_id,
979 'width': int_or_none(r.attrib.get('width')),
980 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
981 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
982 'filesize': filesize,
985 existing_format = next(
987 if fo['format_id'] == format_id)
988 except StopIteration:
989 f.update(self._formats.get(format_id, {}))
992 existing_format.update(f)
994 except (ExtractorError, KeyError) as e:
995 self.report_warning('Skipping DASH manifest: %s' % e, video_id)
997 self._sort_formats(formats)
1001 'uploader': video_uploader,
1002 'uploader_id': video_uploader_id,
1003 'upload_date': upload_date,
1004 'title': video_title,
1005 'thumbnail': video_thumbnail,
1006 'description': video_description,
1007 'categories': video_categories,
1008 'subtitles': video_subtitles,
1009 'duration': video_duration,
1010 'age_limit': 18 if age_gate else 0,
1011 'annotations': video_annotations,
1012 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1013 'view_count': view_count,
1014 'like_count': like_count,
1015 'dislike_count': dislike_count,
1019 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1020 IE_DESC = 'YouTube.com playlists'
1021 _VALID_URL = r"""(?x)(?:
1026 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1027 \? (?:.*?&)*? (?:p|a|list)=
1031 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1032 # Top tracks, they can also include dots
1037 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1039 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1040 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
1041 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1042 IE_NAME = 'youtube:playlist'
1044 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1046 'title': 'ytdl test PL',
1048 'playlist_count': 3,
1050 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1052 'title': 'YDL_Empty_List',
1054 'playlist_count': 0,
1056 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1057 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1059 'title': '29C3: Not my department',
1061 'playlist_count': 95,
1063 'note': 'issue #673',
1064 'url': 'PLBB231211A4F62143',
1066 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1068 'playlist_mincount': 26,
1070 'note': 'Large playlist',
1071 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1073 'title': 'Uploads from Cauchemar',
1075 'playlist_mincount': 799,
1077 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1079 'title': 'YDL_safe_search',
1081 'playlist_count': 2,
1084 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1085 'playlist_count': 4,
1090 'note': 'Embedded SWF player',
1091 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1092 'playlist_count': 4,
1098 def _real_initialize(self):
1101 def _ids_to_results(self, ids):
1103 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1106 def _extract_mix(self, playlist_id):
1107 # The mixes are generated from a a single video
1108 # the id of the playlist is just 'RD' + video_id
1109 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1110 webpage = self._download_webpage(
1111 url, playlist_id, 'Downloading Youtube mix')
1112 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1114 search_title('playlist-title') or
1115 search_title('title long-title') or
1116 search_title('title'))
1117 title = clean_html(title_span)
1118 ids = orderedSet(re.findall(
1119 r'''(?xs)data-video-username=".*?".*?
1120 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1122 url_results = self._ids_to_results(ids)
1124 return self.playlist_result(url_results, playlist_id, title)
1126 def _real_extract(self, url):
1127 # Extract playlist id
1128 mobj = re.match(self._VALID_URL, url)
1130 raise ExtractorError('Invalid URL: %s' % url)
1131 playlist_id = mobj.group(1) or mobj.group(2)
1133 # Check if it's a video-specific URL
1134 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1135 if 'v' in query_dict:
1136 video_id = query_dict['v'][0]
1137 if self._downloader.params.get('noplaylist'):
1138 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1139 return self.url_result(video_id, 'Youtube', video_id=video_id)
1141 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1143 if playlist_id.startswith('RD'):
1144 # Mixes require a custom extraction process
1145 return self._extract_mix(playlist_id)
1146 if playlist_id.startswith('TL'):
1147 raise ExtractorError('For downloading YouTube.com top lists, use '
1148 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1150 url = self._TEMPLATE_URL % playlist_id
1151 page = self._download_webpage(url, playlist_id)
1152 more_widget_html = content_html = page
1154 # Check if the playlist exists or is private
1155 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1156 raise ExtractorError(
1157 'The playlist doesn\'t exist or is private, use --username or '
1158 '--netrc to access it.',
1161 # Extract the video ids from the playlist pages
1164 for page_num in itertools.count(1):
1165 matches = re.finditer(self._VIDEO_RE, content_html)
1166 # We remove the duplicates and the link with index 0
1167 # (it's not the first video of the playlist)
1168 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1171 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1175 more = self._download_json(
1176 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1177 'Downloading page #%s' % page_num,
1178 transform_source=uppercase_escape)
1179 content_html = more['content_html']
1180 more_widget_html = more['load_more_widget_html']
1182 playlist_title = self._html_search_regex(
1183 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1186 url_results = self._ids_to_results(ids)
1187 return self.playlist_result(url_results, playlist_id, playlist_title)
1190 class YoutubeTopListIE(YoutubePlaylistIE):
1191 IE_NAME = 'youtube:toplist'
1192 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1193 ' (Example: "yttoplist:music:Top Tracks")')
1194 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1196 'url': 'yttoplist:music:Trending',
1197 'playlist_mincount': 5,
1198 'skip': 'Only works for logged-in users',
1201 def _real_extract(self, url):
1202 mobj = re.match(self._VALID_URL, url)
1203 channel = mobj.group('chann')
1204 title = mobj.group('title')
1205 query = compat_urllib_parse.urlencode({'title': title})
1206 channel_page = self._download_webpage(
1207 'https://www.youtube.com/%s' % channel, title)
1208 link = self._html_search_regex(
1210 <a\s+href="([^"]+)".*?>\s*
1211 <span\s+class="branded-page-module-title-text">\s*
1212 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1213 channel_page, 'list')
1214 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1216 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1218 # sometimes the webpage doesn't contain the videos
1219 # retry until we get them
1220 for i in itertools.count(0):
1221 msg = 'Downloading Youtube mix'
1223 msg += ', retry #%d' % i
1225 webpage = self._download_webpage(url, title, msg)
1226 ids = orderedSet(re.findall(video_re, webpage))
1229 url_results = self._ids_to_results(ids)
1230 return self.playlist_result(url_results, playlist_title=title)
1233 class YoutubeChannelIE(InfoExtractor):
1234 IE_DESC = 'YouTube.com channels'
1235 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1236 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1237 _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'
1238 IE_NAME = 'youtube:channel'
1240 'note': 'paginated channel',
1241 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1242 'playlist_mincount': 91,
1245 def extract_videos_from_page(self, page):
1247 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1248 if mobj.group(1) not in ids_in_page:
1249 ids_in_page.append(mobj.group(1))
1252 def _real_extract(self, url):
1253 # Extract channel id
1254 mobj = re.match(self._VALID_URL, url)
1256 raise ExtractorError('Invalid URL: %s' % url)
1258 # Download channel page
1259 channel_id = mobj.group(1)
1261 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1262 channel_page = self._download_webpage(url, channel_id)
1263 autogenerated = re.search(r'''(?x)
1265 channel-header-autogenerated-label|
1266 yt-channel-title-autogenerated
1267 )[^"]*"''', channel_page) is not None
1270 # The videos are contained in a single page
1271 # the ajax pages can't be used, they are empty
1272 video_ids = self.extract_videos_from_page(channel_page)
1274 # Download all channel pages using the json-based channel_ajax query
1275 for pagenum in itertools.count(1):
1276 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1277 page = self._download_json(
1278 url, channel_id, note='Downloading page #%s' % pagenum,
1279 transform_source=uppercase_escape)
1281 ids_in_page = self.extract_videos_from_page(page['content_html'])
1282 video_ids.extend(ids_in_page)
1284 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1287 self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1289 url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
1290 for video_id in video_ids]
1291 return self.playlist_result(url_entries, channel_id)
1294 class YoutubeUserIE(InfoExtractor):
1295 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1296 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1297 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1298 _GDATA_PAGE_SIZE = 50
1299 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1300 IE_NAME = 'youtube:user'
1303 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1304 'playlist_mincount': 320,
1306 'title': 'TheLinuxFoundation',
1309 'url': 'ytuser:phihag',
1310 'only_matching': True,
1314 def suitable(cls, url):
1315 # Don't return True if the url can be extracted with other youtube
1316 # extractor, the regex would is too permissive and it would match.
1317 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1318 if any(ie.suitable(url) for ie in other_ies): return False
1319 else: return super(YoutubeUserIE, cls).suitable(url)
1321 def _real_extract(self, url):
1323 mobj = re.match(self._VALID_URL, url)
1325 raise ExtractorError('Invalid URL: %s' % url)
1327 username = mobj.group(1)
1329 # Download video ids using YouTube Data API. Result size per
1330 # query is limited (currently to 50 videos) so we need to query
1331 # page by page until there are no video ids - it means we got
1334 def download_page(pagenum):
1335 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1337 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1338 page = self._download_webpage(
1339 gdata_url, username,
1340 'Downloading video ids from %d to %d' % (
1341 start_index, start_index + self._GDATA_PAGE_SIZE))
1344 response = json.loads(page)
1345 except ValueError as err:
1346 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1347 if 'entry' not in response['feed']:
1350 # Extract video identifiers
1351 entries = response['feed']['entry']
1352 for entry in entries:
1353 title = entry['title']['$t']
1354 video_id = entry['id']['$t'].split('/')[-1]
1358 'ie_key': 'Youtube',
1362 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1364 return self.playlist_result(url_results, playlist_title=username)
1367 class YoutubeSearchIE(SearchInfoExtractor):
1368 IE_DESC = 'YouTube.com searches'
1369 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1371 IE_NAME = 'youtube:search'
1372 _SEARCH_KEY = 'ytsearch'
1374 def _get_n_results(self, query, n):
1375 """Get a specified number of results for a query"""
1382 while (PAGE_SIZE * pagenum) < limit:
1383 result_url = self._API_URL % (
1384 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1385 (PAGE_SIZE * pagenum) + 1)
1386 data_json = self._download_webpage(
1387 result_url, video_id='query "%s"' % query,
1388 note='Downloading page %s' % (pagenum + 1),
1389 errnote='Unable to download API page')
1390 data = json.loads(data_json)
1391 api_response = data['data']
1393 if 'items' not in api_response:
1394 raise ExtractorError(
1395 '[youtube] No video results', expected=True)
1397 new_ids = list(video['id'] for video in api_response['items'])
1398 video_ids += new_ids
1400 limit = min(n, api_response['totalItems'])
1403 if len(video_ids) > n:
1404 video_ids = video_ids[:n]
1405 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1406 for video_id in video_ids]
1407 return self.playlist_result(videos, query)
1410 class YoutubeSearchDateIE(YoutubeSearchIE):
1411 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1412 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1413 _SEARCH_KEY = 'ytsearchdate'
1414 IE_DESC = 'YouTube.com searches, newest videos first'
1417 class YoutubeSearchURLIE(InfoExtractor):
1418 IE_DESC = 'YouTube.com search URLs'
1419 IE_NAME = 'youtube:search_url'
1420 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1422 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1423 'playlist_mincount': 5,
1425 'title': 'youtube-dl test video',
1429 def _real_extract(self, url):
1430 mobj = re.match(self._VALID_URL, url)
1431 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1433 webpage = self._download_webpage(url, query)
1434 result_code = self._search_regex(
1435 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1437 part_codes = re.findall(
1438 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1440 for part_code in part_codes:
1441 part_title = self._html_search_regex(
1442 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1443 part_url_snippet = self._html_search_regex(
1444 r'(?s)href="([^"]+)"', part_code, 'item URL')
1445 part_url = compat_urlparse.urljoin(
1446 'https://www.youtube.com/', part_url_snippet)
1450 'title': part_title,
1454 '_type': 'playlist',
1460 class YoutubeShowIE(InfoExtractor):
1461 IE_DESC = 'YouTube.com (multi-season) shows'
1462 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1463 IE_NAME = 'youtube:show'
1465 'url': 'http://www.youtube.com/show/airdisasters',
1466 'playlist_mincount': 3,
1468 'id': 'airdisasters',
1469 'title': 'Air Disasters',
1473 def _real_extract(self, url):
1474 mobj = re.match(self._VALID_URL, url)
1475 playlist_id = mobj.group('id')
1476 webpage = self._download_webpage(
1477 url, playlist_id, 'Downloading show webpage')
1478 # There's one playlist for each season of the show
1479 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1480 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1483 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1484 for season in m_seasons
1486 title = self._og_search_title(webpage, fatal=False)
1489 '_type': 'playlist',
1496 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1498 Base class for extractors that fetch info from
1499 http://www.youtube.com/feed_ajax
1500 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1502 _LOGIN_REQUIRED = True
1503 # use action_load_personal_feed instead of action_load_system_feed
1504 _PERSONAL_FEED = False
1507 def _FEED_TEMPLATE(self):
1508 action = 'action_load_system_feed'
1509 if self._PERSONAL_FEED:
1510 action = 'action_load_personal_feed'
1511 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1515 return 'youtube:%s' % self._FEED_NAME
1517 def _real_initialize(self):
1520 def _real_extract(self, url):
1523 for i in itertools.count(1):
1524 info = self._download_json(self._FEED_TEMPLATE % paging,
1525 '%s feed' % self._FEED_NAME,
1526 'Downloading page %s' % i)
1527 feed_html = info.get('feed_html') or info.get('content_html')
1528 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1529 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1530 ids = orderedSet(m.group(1) for m in m_ids)
1531 feed_entries.extend(
1532 self.url_result(video_id, 'Youtube', video_id=video_id)
1533 for video_id in ids)
1535 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1536 load_more_widget_html)
1539 paging = mobj.group('paging')
1540 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1542 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1543 IE_DESC = 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1544 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1545 _FEED_NAME = 'recommended'
1546 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1548 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1549 IE_DESC = 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1550 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1551 _FEED_NAME = 'watch_later'
1552 _PLAYLIST_TITLE = 'Youtube Watch Later'
1553 _PERSONAL_FEED = True
1555 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1556 IE_DESC = 'Youtube watch history, "ythistory" keyword (requires authentication)'
1557 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1558 _FEED_NAME = 'history'
1559 _PERSONAL_FEED = True
1560 _PLAYLIST_TITLE = 'Youtube Watch History'
1562 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1563 IE_NAME = 'youtube:favorites'
1564 IE_DESC = 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1565 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1566 _LOGIN_REQUIRED = True
1568 def _real_extract(self, url):
1569 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1570 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1571 return self.url_result(playlist_id, 'YoutubePlaylist')
1574 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1575 IE_NAME = 'youtube:subscriptions'
1576 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1577 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1580 def _real_extract(self, url):
1581 title = 'Youtube Subscriptions'
1582 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1584 # The extraction process is the same as for playlists, but the regex
1585 # for the video ids doesn't contain an index
1587 more_widget_html = content_html = page
1589 for page_num in itertools.count(1):
1590 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1591 new_ids = orderedSet(matches)
1594 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1598 more = self._download_json(
1599 'https://youtube.com/%s' % mobj.group('more'), title,
1600 'Downloading page #%s' % page_num,
1601 transform_source=uppercase_escape)
1602 content_html = more['content_html']
1603 more_widget_html = more['load_more_widget_html']
1606 '_type': 'playlist',
1608 'entries': self._ids_to_results(ids),
1612 class YoutubeTruncatedURLIE(InfoExtractor):
1613 IE_NAME = 'youtube:truncated_url'
1614 IE_DESC = False # Do not list
1615 _VALID_URL = r'''(?x)
1616 (?:https?://)?[^/]+/watch\?(?:
1618 annotation_id=annotation_[^&]+
1620 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1624 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1625 'only_matching': True,
1627 'url': 'http://www.youtube.com/watch?',
1628 'only_matching': True,
1631 def _real_extract(self, url):
1632 raise ExtractorError(
1633 'Did you forget to quote the URL? Remember that & is a meta '
1634 'character in most shells, so you want to put the URL in quotes, '
1636 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1637 ' or simply youtube-dl BaW_jenozKc .',