3 from __future__
import unicode_literals
13 from .common
import InfoExtractor
, SearchInfoExtractor
14 from ..jsinterp
import JSInterpreter
15 from ..swfinterp
import SWFInterpreter
16 from ..compat
import (
20 compat_urllib_parse_unquote
,
21 compat_urllib_parse_unquote_plus
,
22 compat_urllib_parse_urlparse
,
32 get_element_by_attribute
,
50 class YoutubeBaseInfoExtractor(InfoExtractor
):
51 """Provide base functions for Youtube extractors"""
52 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
53 _TWOFACTOR_URL
= 'https://accounts.google.com/signin/challenge'
54 _NETRC_MACHINE
= 'youtube'
55 # If True it will raise an error if no login info is provided
56 _LOGIN_REQUIRED
= False
58 def _set_language(self
):
60 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
61 # YouTube sets the expire time to about two months
62 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
64 def _ids_to_results(self
, ids
):
66 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
71 Attempt to log in to YouTube.
72 True is returned if successful or skipped.
73 False is returned if login failed.
75 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
77 (username
, password
) = self
._get
_login
_info
()
78 # No authentication to be performed
80 if self
._LOGIN
_REQUIRED
:
81 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
84 login_page
= self
._download
_webpage
(
85 self
._LOGIN
_URL
, None,
86 note
='Downloading login page',
87 errnote
='unable to fetch login page', fatal
=False)
88 if login_page
is False:
91 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
92 login_page
, 'Login GALX parameter')
96 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
101 'PersistentCookie': 'yes',
103 'bgresponse': 'js_disabled',
104 'checkConnection': '',
105 'checkedDomains': 'youtube',
112 'service': 'youtube',
117 login_data
= compat_urllib_parse
.urlencode(encode_dict(login_form_strs
)).encode('ascii')
119 req
= sanitized_Request(self
._LOGIN
_URL
, login_data
)
120 login_results
= self
._download
_webpage
(
122 note
='Logging in', errnote
='unable to log in', fatal
=False)
123 if login_results
is False:
126 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
127 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
130 # TODO add SMS and phone call support - these require making a request and then prompting the user
132 if re
.search(r
'(?i)<form[^>]* id="challenge"', login_results
) is not None:
133 tfa_code
= self
._get
_tfa
_info
('2-step verification code')
136 self
._downloader
.report_warning(
137 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
138 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
141 tfa_code
= remove_start(tfa_code
, 'G-')
143 tfa_form_strs
= self
._form
_hidden
_inputs
('challenge', login_results
)
145 tfa_form_strs
.update({
150 tfa_data
= compat_urllib_parse
.urlencode(encode_dict(tfa_form_strs
)).encode('ascii')
152 tfa_req
= sanitized_Request(self
._TWOFACTOR
_URL
, tfa_data
)
153 tfa_results
= self
._download
_webpage
(
155 note
='Submitting TFA code', errnote
='unable to submit tfa', fatal
=False)
157 if tfa_results
is False:
160 if re
.search(r
'(?i)<form[^>]* id="challenge"', tfa_results
) is not None:
161 self
._downloader
.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
163 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', tfa_results
) is not None:
164 self
._downloader
.report_warning('unable to log in - did the page structure change?')
166 if re
.search(r
'smsauth-interstitial-reviewsettings', tfa_results
) is not None:
167 self
._downloader
.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
170 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', login_results
) is not None:
171 self
._downloader
.report_warning('unable to log in: bad username or password')
175 def _real_initialize(self
):
176 if self
._downloader
is None:
179 if not self
._login
():
183 class YoutubeEntryListBaseInfoExtractor(InfoExtractor
):
184 # Extract entries from page with "Load more" button
185 def _entries(self
, page
, playlist_id
):
186 more_widget_html
= content_html
= page
187 for page_num
in itertools
.count(1):
188 for entry
in self
._process
_page
(content_html
):
191 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
195 more
= self
._download
_json
(
196 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
197 'Downloading page #%s' % page_num
,
198 transform_source
=uppercase_escape
)
199 content_html
= more
['content_html']
200 if not content_html
.strip():
201 # Some webpages show a "Load more" button but they don't
204 more_widget_html
= more
['load_more_widget_html']
207 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor
):
208 def _process_page(self
, content
):
209 for video_id
, video_title
in self
.extract_videos_from_page(content
):
210 yield self
.url_result(video_id
, 'Youtube', video_id
, video_title
)
212 def extract_videos_from_page(self
, page
):
215 for mobj
in re
.finditer(self
._VIDEO
_RE
, page
):
216 # The link with index 0 is not the first video of the playlist (not sure if still actual)
217 if 'index' in mobj
.groupdict() and mobj
.group('id') == '0':
219 video_id
= mobj
.group('id')
220 video_title
= unescapeHTML(mobj
.group('title'))
222 video_title
= video_title
.strip()
224 idx
= ids_in_page
.index(video_id
)
225 if video_title
and not titles_in_page
[idx
]:
226 titles_in_page
[idx
] = video_title
228 ids_in_page
.append(video_id
)
229 titles_in_page
.append(video_title
)
230 return zip(ids_in_page
, titles_in_page
)
233 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor
):
234 def _process_page(self
, content
):
235 for playlist_id
in re
.findall(r
'href="/?playlist\?list=(.+?)"', content
):
236 yield self
.url_result(
237 'https://www.youtube.com/playlist?list=%s' % playlist_id
, 'YoutubePlaylist')
239 def _real_extract(self
, url
):
240 playlist_id
= self
._match
_id
(url
)
241 webpage
= self
._download
_webpage
(url
, playlist_id
)
242 title
= self
._og
_search
_title
(webpage
, fatal
=False)
243 return self
.playlist_result(self
._entries
(webpage
, playlist_id
), playlist_id
, title
)
246 class YoutubeIE(YoutubeBaseInfoExtractor
):
247 IE_DESC
= 'YouTube.com'
248 _VALID_URL
= r
"""(?x)^
250 (?:https?://|//) # http(s):// or protocol-independent URL
251 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
252 (?:www\.)?deturl\.com/www\.youtube\.com/|
253 (?:www\.)?pwnyoutube\.com/|
254 (?:www\.)?yourepeat\.com/|
255 tube\.majestyc\.net/|
256 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
257 (?:.*?\#/)? # handle anchor (#/) redirect urls
258 (?: # the various things that can precede the ID:
259 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
260 |(?: # or the v= param in all its forms
261 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
262 (?:\?|\#!?) # the params delimiter ? or # or #!
263 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
268 youtu\.be| # just youtu.be/xxxx
269 vid\.plus # or vid.plus/xxxx
271 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
273 )? # all until now is optional -> you can pass the naked ID
274 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
275 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
276 (?(1).+)? # if we found the ID, everything can follow
278 _NEXT_URL_RE
= r
'[\?&]next_url=([^&]+)'
280 '5': {'ext': 'flv', 'width': 400, 'height': 240}
,
281 '6': {'ext': 'flv', 'width': 450, 'height': 270}
,
282 '13': {'ext': '3gp'}
,
283 '17': {'ext': '3gp', 'width': 176, 'height': 144}
,
284 '18': {'ext': 'mp4', 'width': 640, 'height': 360}
,
285 '22': {'ext': 'mp4', 'width': 1280, 'height': 720}
,
286 '34': {'ext': 'flv', 'width': 640, 'height': 360}
,
287 '35': {'ext': 'flv', 'width': 854, 'height': 480}
,
288 '36': {'ext': '3gp', 'width': 320, 'height': 240}
,
289 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}
,
290 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}
,
291 '43': {'ext': 'webm', 'width': 640, 'height': 360}
,
292 '44': {'ext': 'webm', 'width': 854, 'height': 480}
,
293 '45': {'ext': 'webm', 'width': 1280, 'height': 720}
,
294 '46': {'ext': 'webm', 'width': 1920, 'height': 1080}
,
295 '59': {'ext': 'mp4', 'width': 854, 'height': 480}
,
296 '78': {'ext': 'mp4', 'width': 854, 'height': 480}
,
300 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}
,
301 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}
,
302 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}
,
303 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}
,
304 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}
,
305 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}
,
306 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}
,
308 # Apple HTTP Live Streaming
309 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}
,
310 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}
,
311 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}
,
312 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}
,
313 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}
,
314 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}
,
315 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}
,
318 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
319 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
320 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
321 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
322 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
323 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
324 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
325 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
326 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}
,
327 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}
,
328 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}
,
331 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}
,
332 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}
,
333 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}
,
336 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
337 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
338 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
339 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
340 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
341 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
342 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'}
,
343 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
344 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
345 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
346 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
347 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
348 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
349 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
350 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
351 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
352 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
353 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
354 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
355 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
356 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'}
,
357 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
360 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}
,
361 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}
,
363 # Dash webm audio with opus inside
364 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}
,
365 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}
,
366 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}
,
369 '_rtmp': {'protocol': 'rtmp'}
,
375 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
379 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
380 'uploader': 'Philipp Hagemeister',
381 'uploader_id': 'phihag',
382 'upload_date': '20121002',
383 '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 .',
384 'categories': ['Science & Technology'],
385 'tags': ['youtube-dl'],
387 'dislike_count': int,
393 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
394 'note': 'Test generic use_cipher_signature video (#897)',
398 'upload_date': '20120506',
399 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
400 'alt_title': 'I Love It (feat. Charli XCX)',
401 'description': 'md5:782e8651347686cba06e58f71ab51773',
402 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
403 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
404 'iconic ep', 'iconic', 'love', 'it'],
405 'uploader': 'Icona Pop',
406 'uploader_id': 'IconaPop',
407 'creator': 'Icona Pop',
411 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
412 'note': 'Test VEVO video with age protection (#956)',
416 'upload_date': '20130703',
417 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
418 'alt_title': 'Tunnel Vision',
419 'description': 'md5:64249768eec3bc4276236606ea996373',
420 'uploader': 'justintimberlakeVEVO',
421 'uploader_id': 'justintimberlakeVEVO',
422 'creator': 'Justin Timberlake',
427 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
428 'note': 'Embed-only video (#1746)',
432 'upload_date': '20120608',
433 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
434 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
435 'uploader': 'SET India',
436 'uploader_id': 'setindia',
441 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
442 'note': 'Use the first video ID in the URL',
446 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
447 'uploader': 'Philipp Hagemeister',
448 'uploader_id': 'phihag',
449 'upload_date': '20121002',
450 '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 .',
451 'categories': ['Science & Technology'],
452 'tags': ['youtube-dl'],
454 'dislike_count': int,
457 'skip_download': True,
461 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
462 'note': '256k DASH audio (format 141) via DASH manifest',
466 'upload_date': '20121002',
467 'uploader_id': '8KVIDEO',
469 'uploader': '8KVIDEO',
470 'title': 'UHDTV TEST 8K VIDEO.mp4'
473 'youtube_include_dash_manifest': True,
477 # DASH manifest with encrypted signature
479 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
483 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
484 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
485 'uploader': 'AfrojackVEVO',
486 'uploader_id': 'AfrojackVEVO',
487 'upload_date': '20131011',
490 'youtube_include_dash_manifest': True,
494 # JS player signature function name containing $
496 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
500 'title': 'Taylor Swift - Shake It Off',
501 'alt_title': 'Shake It Off',
502 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
503 'uploader': 'TaylorSwiftVEVO',
504 'uploader_id': 'TaylorSwiftVEVO',
505 'upload_date': '20140818',
506 'creator': 'Taylor Swift',
509 'youtube_include_dash_manifest': True,
515 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
519 'upload_date': '20100909',
520 'uploader': 'The Amazing Atheist',
521 'uploader_id': 'TheAmazingAtheist',
522 'title': 'Burning Everyone\'s Koran',
523 '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',
526 # Normal age-gate video (No vevo, embed allowed)
528 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
532 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
533 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
534 'uploader': 'The Witcher',
535 'uploader_id': 'WitcherGame',
536 'upload_date': '20140605',
540 # Age-gate video with encrypted signature
542 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
546 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
547 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
548 'uploader': 'LloydVEVO',
549 'uploader_id': 'LloydVEVO',
550 'upload_date': '20110629',
554 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
556 'url': '__2ABJjxzNo',
560 'upload_date': '20100430',
561 'uploader_id': 'deadmau5',
562 'creator': 'deadmau5',
563 'description': 'md5:12c56784b8032162bb936a5f76d55360',
564 'uploader': 'deadmau5',
565 'title': 'Deadmau5 - Some Chords (HD)',
566 'alt_title': 'Some Chords',
568 'expected_warnings': [
569 'DASH manifest missing',
572 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
574 'url': 'lqQg6PlCWgI',
578 'upload_date': '20150827',
579 'uploader_id': 'olympic',
580 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
581 'uploader': 'Olympics',
582 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
585 'skip_download': 'requires avconv',
590 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
594 'stretched_ratio': 16 / 9.,
595 'upload_date': '20110310',
596 'uploader_id': 'AllenMeow',
597 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
599 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
602 # url_encoded_fmt_stream_map is empty string
604 'url': 'qEJwOuvDf7I',
608 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
610 'upload_date': '20150404',
611 'uploader_id': 'spbelect',
612 'uploader': 'Наблюдатели Петербурга',
615 'skip_download': 'requires avconv',
618 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
620 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
624 'title': 'md5:7b81415841e02ecd4313668cde88737a',
625 'description': 'md5:116377fd2963b81ec4ce64b542173306',
626 'upload_date': '20150625',
627 'uploader_id': 'dorappi2000',
628 'uploader': 'dorappi2000',
629 'formats': 'mincount:33',
632 # DASH manifest with segment_list
634 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
635 'md5': '8ce563a1d667b599d21064e982ab9e31',
639 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
640 'uploader': 'Airtek',
641 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
642 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
643 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
646 'youtube_include_dash_manifest': True,
647 'format': '135', # bestvideo
651 # Multifeed videos (multiple cameras), URL is for Main Camera
652 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
655 'title': 'teamPGP: Rocket League Noob Stream',
656 'description': 'md5:dc7872fb300e143831327f1bae3af010',
662 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
663 'description': 'md5:dc7872fb300e143831327f1bae3af010',
664 'upload_date': '20150721',
665 'uploader': 'Beer Games Beer',
666 'uploader_id': 'beergamesbeer',
672 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
673 'description': 'md5:dc7872fb300e143831327f1bae3af010',
674 'upload_date': '20150721',
675 'uploader': 'Beer Games Beer',
676 'uploader_id': 'beergamesbeer',
682 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
683 'description': 'md5:dc7872fb300e143831327f1bae3af010',
684 'upload_date': '20150721',
685 'uploader': 'Beer Games Beer',
686 'uploader_id': 'beergamesbeer',
692 'title': 'teamPGP: Rocket League Noob Stream (zim)',
693 'description': 'md5:dc7872fb300e143831327f1bae3af010',
694 'upload_date': '20150721',
695 'uploader': 'Beer Games Beer',
696 'uploader_id': 'beergamesbeer',
700 'skip_download': True,
704 'url': 'http://vid.plus/FlRa-iH7PGw',
705 'only_matching': True,
708 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
709 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
713 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
714 'alt_title': 'Dark Walk',
715 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
716 'upload_date': '20151119',
717 'uploader_id': 'IronSoulElf',
718 'uploader': 'IronSoulElf',
719 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
722 'skip_download': True,
726 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
727 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
728 'only_matching': True,
731 # Video with yt:stretch=17:0
732 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
736 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
737 'description': 'md5:ee18a25c350637c8faff806845bddee9',
738 'upload_date': '20151107',
739 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
740 'uploader': 'CH GAMER DROID',
743 'skip_download': True,
747 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
748 'only_matching': True,
752 def __init__(self
, *args
, **kwargs
):
753 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
754 self
._player
_cache
= {}
756 def report_video_info_webpage_download(self
, video_id
):
757 """Report attempt to download video info webpage."""
758 self
.to_screen('%s: Downloading video info webpage' % video_id
)
760 def report_information_extraction(self
, video_id
):
761 """Report attempt to extract video information."""
762 self
.to_screen('%s: Extracting video information' % video_id
)
764 def report_unavailable_format(self
, video_id
, format
):
765 """Report extracted video URL."""
766 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
768 def report_rtmp_download(self
):
769 """Indicate the download will use the RTMP protocol."""
770 self
.to_screen('RTMP download detected')
772 def _signature_cache_id(self
, example_sig
):
773 """ Return a string representation of a signature """
774 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
776 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
778 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
781 raise ExtractorError('Cannot identify player %r' % player_url
)
782 player_type
= id_m
.group('ext')
783 player_id
= id_m
.group('id')
785 # Read from filesystem cache
786 func_id
= '%s_%s_%s' % (
787 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
788 assert os
.path
.basename(func_id
) == func_id
790 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
791 if cache_spec
is not None:
792 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
795 'Downloading player %s' % player_url
796 if self
._downloader
.params
.get('verbose') else
797 'Downloading %s player %s' % (player_type
, player_id
)
799 if player_type
== 'js':
800 code
= self
._download
_webpage
(
801 player_url
, video_id
,
803 errnote
='Download of %s failed' % player_url
)
804 res
= self
._parse
_sig
_js
(code
)
805 elif player_type
== 'swf':
806 urlh
= self
._request
_webpage
(
807 player_url
, video_id
,
809 errnote
='Download of %s failed' % player_url
)
811 res
= self
._parse
_sig
_swf
(code
)
813 assert False, 'Invalid player type %r' % player_type
815 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
816 cache_res
= res(test_string
)
817 cache_spec
= [ord(c
) for c
in cache_res
]
819 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
822 def _print_sig_code(self
, func
, example_sig
):
823 def gen_sig_code(idxs
):
824 def _genslice(start
, end
, step
):
825 starts
= '' if start
== 0 else str(start
)
826 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
827 steps
= '' if step
== 1 else (':%d' % step
)
828 return 's[%s%s%s]' % (starts
, ends
, steps
)
831 # Quelch pyflakes warnings - start will be set when step is set
832 start
= '(Never used)'
833 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
837 yield _genslice(start
, prev
, step
)
840 if i
- prev
in [-1, 1]:
849 yield _genslice(start
, i
, step
)
851 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
852 cache_res
= func(test_string
)
853 cache_spec
= [ord(c
) for c
in cache_res
]
854 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
855 signature_id_tuple
= '(%s)' % (
856 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
857 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
858 ' return %s\n') % (signature_id_tuple
, expr_code
)
859 self
.to_screen('Extracted signature function:\n' + code
)
861 def _parse_sig_js(self
, jscode
):
862 funcname
= self
._search
_regex
(
863 r
'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode
,
864 'Initial JS player signature function name')
866 jsi
= JSInterpreter(jscode
)
867 initial_function
= jsi
.extract_function(funcname
)
868 return lambda s
: initial_function([s
])
870 def _parse_sig_swf(self
, file_contents
):
871 swfi
= SWFInterpreter(file_contents
)
872 TARGET_CLASSNAME
= 'SignatureDecipher'
873 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
874 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
875 return lambda s
: initial_function([s
])
877 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
878 """Turn the encrypted s field into a working signature"""
880 if player_url
is None:
881 raise ExtractorError('Cannot decrypt signature without player_url')
883 if player_url
.startswith('//'):
884 player_url
= 'https:' + player_url
886 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
887 if player_id
not in self
._player
_cache
:
888 func
= self
._extract
_signature
_function
(
889 video_id
, player_url
, s
891 self
._player
_cache
[player_id
] = func
892 func
= self
._player
_cache
[player_id
]
893 if self
._downloader
.params
.get('youtube_print_sig_code'):
894 self
._print
_sig
_code
(func
, s
)
896 except Exception as e
:
897 tb
= traceback
.format_exc()
898 raise ExtractorError(
899 'Signature extraction failed: ' + tb
, cause
=e
)
901 def _get_subtitles(self
, video_id
, webpage
):
903 subs_doc
= self
._download
_xml
(
904 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
905 video_id
, note
=False)
906 except ExtractorError
as err
:
907 self
._downloader
.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err
))
911 for track
in subs_doc
.findall('track'):
912 lang
= track
.attrib
['lang_code']
913 if lang
in sub_lang_list
:
916 for ext
in ['sbv', 'vtt', 'srt']:
917 params
= compat_urllib_parse
.urlencode({
921 'name': track
.attrib
['name'].encode('utf-8'),
924 'url': 'https://www.youtube.com/api/timedtext?' + params
,
927 sub_lang_list
[lang
] = sub_formats
928 if not sub_lang_list
:
929 self
._downloader
.report_warning('video doesn\'t have subtitles')
933 def _get_ytplayer_config(self
, video_id
, webpage
):
935 # User data may contain arbitrary character sequences that may affect
936 # JSON extraction with regex, e.g. when '};' is contained the second
937 # regex won't capture the whole JSON. Yet working around by trying more
938 # concrete regex first keeping in mind proper quoted string handling
939 # to be implemented in future that will replace this workaround (see
940 # https://github.com/rg3/youtube-dl/issues/7468,
941 # https://github.com/rg3/youtube-dl/pull/7599)
942 r
';ytplayer\.config\s*=\s*({.+?});ytplayer',
943 r
';ytplayer\.config\s*=\s*({.+?});',
945 config
= self
._search
_regex
(
946 patterns
, webpage
, 'ytplayer.config', default
=None)
948 return self
._parse
_json
(
949 uppercase_escape(config
), video_id
, fatal
=False)
951 def _get_automatic_captions(self
, video_id
, webpage
):
952 """We need the webpage for getting the captions url, pass it as an
953 argument to speed up the process."""
954 self
.to_screen('%s: Looking for automatic captions' % video_id
)
955 player_config
= self
._get
_ytplayer
_config
(video_id
, webpage
)
956 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
957 if not player_config
:
958 self
._downloader
.report_warning(err_msg
)
961 args
= player_config
['args']
962 caption_url
= args
['ttsurl']
963 timestamp
= args
['timestamp']
964 # We get the available subtitles
965 list_params
= compat_urllib_parse
.urlencode({
970 list_url
= caption_url
+ '&' + list_params
971 caption_list
= self
._download
_xml
(list_url
, video_id
)
972 original_lang_node
= caption_list
.find('track')
973 if original_lang_node
is None:
974 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
976 original_lang
= original_lang_node
.attrib
['lang_code']
977 caption_kind
= original_lang_node
.attrib
.get('kind', '')
980 for lang_node
in caption_list
.findall('target'):
981 sub_lang
= lang_node
.attrib
['lang_code']
983 for ext
in ['sbv', 'vtt', 'srt']:
984 params
= compat_urllib_parse
.urlencode({
985 'lang': original_lang
,
989 'kind': caption_kind
,
992 'url': caption_url
+ '&' + params
,
995 sub_lang_list
[sub_lang
] = sub_formats
997 # An extractor error can be raise by the download process if there are
998 # no automatic captions but there are subtitles
999 except (KeyError, ExtractorError
):
1000 self
._downloader
.report_warning(err_msg
)
1004 def extract_id(cls
, url
):
1005 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
1007 raise ExtractorError('Invalid URL: %s' % url
)
1008 video_id
= mobj
.group(2)
1011 def _extract_from_m3u8(self
, manifest_url
, video_id
):
1014 def _get_urls(_manifest
):
1015 lines
= _manifest
.split('\n')
1016 urls
= filter(lambda l
: l
and not l
.startswith('#'),
1019 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
1020 formats_urls
= _get_urls(manifest
)
1021 for format_url
in formats_urls
:
1022 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
1023 url_map
[itag
] = format_url
1026 def _extract_annotations(self
, video_id
):
1027 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1028 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
1030 def _parse_dash_manifest(
1031 self
, video_id
, dash_manifest_url
, player_url
, age_gate
, fatal
=True):
1032 def decrypt_sig(mobj
):
1034 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
1035 return '/signature/%s' % dec_s
1036 dash_manifest_url
= re
.sub(r
'/s/([a-fA-F0-9\.]+)', decrypt_sig
, dash_manifest_url
)
1037 dash_doc
= self
._download
_xml
(
1038 dash_manifest_url
, video_id
,
1039 note
='Downloading DASH manifest',
1040 errnote
='Could not download DASH manifest',
1043 if dash_doc
is False:
1047 for a
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1048 mime_type
= a
.attrib
.get('mimeType')
1049 for r
in a
.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1050 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1053 if mime_type
== 'text/vtt':
1054 # TODO implement WebVTT downloading
1056 elif mime_type
.startswith('audio/') or mime_type
.startswith('video/'):
1057 segment_list
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1058 format_id
= r
.attrib
['id']
1059 video_url
= url_el
.text
1060 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1062 'format_id': format_id
,
1064 'width': int_or_none(r
.attrib
.get('width')),
1065 'height': int_or_none(r
.attrib
.get('height')),
1066 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
1067 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
1068 'filesize': filesize
,
1069 'fps': int_or_none(r
.attrib
.get('frameRate')),
1071 if segment_list
is not None:
1073 'initialization_url': segment_list
.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib
['sourceURL'],
1074 'segment_urls': [segment
.attrib
.get('media') for segment
in segment_list
.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1075 'protocol': 'http_dash_segments',
1078 existing_format
= next(
1079 fo
for fo
in formats
1080 if fo
['format_id'] == format_id
)
1081 except StopIteration:
1082 full_info
= self
._formats
.get(format_id
, {}).copy()
1084 codecs
= r
.attrib
.get('codecs')
1086 if full_info
.get('acodec') == 'none' and 'vcodec' not in full_info
:
1087 full_info
['vcodec'] = codecs
1088 elif full_info
.get('vcodec') == 'none' and 'acodec' not in full_info
:
1089 full_info
['acodec'] = codecs
1090 formats
.append(full_info
)
1092 existing_format
.update(f
)
1094 self
.report_warning('Unknown MIME type %s in DASH manifest' % mime_type
)
1097 def _real_extract(self
, url
):
1098 url
, smuggled_data
= unsmuggle_url(url
, {})
1101 'http' if self
._downloader
.params
.get('prefer_insecure', False)
1106 parsed_url
= compat_urllib_parse_urlparse(url
)
1107 for component
in [parsed_url
.fragment
, parsed_url
.query
]:
1108 query
= compat_parse_qs(component
)
1109 if start_time
is None and 't' in query
:
1110 start_time
= parse_duration(query
['t'][0])
1111 if start_time
is None and 'start' in query
:
1112 start_time
= parse_duration(query
['start'][0])
1113 if end_time
is None and 'end' in query
:
1114 end_time
= parse_duration(query
['end'][0])
1116 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1117 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
1119 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse_unquote(mobj
.group(1)).lstrip('/')
1120 video_id
= self
.extract_id(url
)
1123 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1124 video_webpage
= self
._download
_webpage
(url
, video_id
)
1126 # Attempt to extract SWF player URL
1127 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
1128 if mobj
is not None:
1129 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
1135 def add_dash_mpd(video_info
):
1136 dash_mpd
= video_info
.get('dashmpd')
1137 if dash_mpd
and dash_mpd
[0] not in dash_mpds
:
1138 dash_mpds
.append(dash_mpd
[0])
1141 embed_webpage
= None
1143 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
1145 # We simulate the access to the video from www.youtube.com/v/{video_id}
1146 # this can be viewed without login into Youtube
1147 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1148 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
1149 data
= compat_urllib_parse
.urlencode({
1150 'video_id': video_id
,
1151 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
1152 'sts': self
._search
_regex
(
1153 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
1155 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
1156 video_info_webpage
= self
._download
_webpage
(
1157 video_info_url
, video_id
,
1158 note
='Refetching age-gated info webpage',
1159 errnote
='unable to download video info webpage')
1160 video_info
= compat_parse_qs(video_info_webpage
)
1161 add_dash_mpd(video_info
)
1165 # Try looking directly into the video webpage
1166 ytplayer_config
= self
._get
_ytplayer
_config
(video_id
, video_webpage
)
1168 args
= ytplayer_config
['args']
1169 if args
.get('url_encoded_fmt_stream_map'):
1170 # Convert to the same format returned by compat_parse_qs
1171 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
1172 add_dash_mpd(video_info
)
1173 if args
.get('livestream') == '1' or args
.get('live_playback') == 1:
1175 if not video_info
or self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1176 # We also try looking in get_video_info since it may contain different dashmpd
1177 # URL that points to a DASH manifest with possibly different itag set (some itags
1178 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1179 # manifest pointed by get_video_info's dashmpd).
1180 # The general idea is to take a union of itags of both DASH manifests (for example
1181 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1182 self
.report_video_info_webpage_download(video_id
)
1183 for el_type
in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1185 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1186 % (proto
, video_id
, el_type
))
1187 video_info_webpage
= self
._download
_webpage
(
1189 video_id
, note
=False,
1190 errnote
='unable to download video info webpage')
1191 get_video_info
= compat_parse_qs(video_info_webpage
)
1192 if get_video_info
.get('use_cipher_signature') != ['True']:
1193 add_dash_mpd(get_video_info
)
1195 video_info
= get_video_info
1196 if 'token' in get_video_info
:
1197 # Different get_video_info requests may report different results, e.g.
1198 # some may report video unavailability, but some may serve it without
1199 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1200 # the original webpage as well as el=info and el=embedded get_video_info
1201 # requests report video unavailability due to geo restriction while
1202 # el=detailpage succeeds and returns valid data). This is probably
1203 # due to YouTube measures against IP ranges of hosting providers.
1204 # Working around by preferring the first succeeded video_info containing
1205 # the token if no such video_info yet was found.
1206 if 'token' not in video_info
:
1207 video_info
= get_video_info
1209 if 'token' not in video_info
:
1210 if 'reason' in video_info
:
1211 if 'The uploader has not made this video available in your country.' in video_info
['reason']:
1212 regions_allowed
= self
._html
_search
_meta
('regionsAllowed', video_webpage
, default
=None)
1214 raise ExtractorError('YouTube said: This video is available in %s only' % (
1215 ', '.join(map(ISO3166Utils
.short2full
, regions_allowed
.split(',')))),
1217 raise ExtractorError(
1218 'YouTube said: %s' % video_info
['reason'][0],
1219 expected
=True, video_id
=video_id
)
1221 raise ExtractorError(
1222 '"token" parameter not in video info for unknown reason',
1226 if 'title' in video_info
:
1227 video_title
= video_info
['title'][0]
1229 self
._downloader
.report_warning('Unable to extract video title')
1233 video_description
= get_element_by_id("eow-description", video_webpage
)
1234 if video_description
:
1235 video_description
= re
.sub(r
'''(?x)
1237 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1239 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1240 class="yt-uix-redirect-link"\s*>
1243 ''', r
'\1', video_description
)
1244 video_description
= clean_html(video_description
)
1246 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
1248 video_description
= unescapeHTML(fd_mobj
.group(1))
1250 video_description
= ''
1252 if 'multifeed_metadata_list' in video_info
and not smuggled_data
.get('force_singlefeed', False):
1253 if not self
._downloader
.params
.get('noplaylist'):
1256 multifeed_metadata_list
= compat_urllib_parse_unquote_plus(video_info
['multifeed_metadata_list'][0])
1257 for feed
in multifeed_metadata_list
.split(','):
1258 feed_data
= compat_parse_qs(feed
)
1260 '_type': 'url_transparent',
1261 'ie_key': 'Youtube',
1263 '%s://www.youtube.com/watch?v=%s' % (proto
, feed_data
['id'][0]),
1264 {'force_singlefeed': True}
),
1265 'title': '%s (%s)' % (video_title
, feed_data
['title'][0]),
1267 feed_ids
.append(feed_data
['id'][0])
1269 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1270 % (', '.join(feed_ids
), video_id
))
1271 return self
.playlist_result(entries
, video_id
, video_title
, video_description
)
1272 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1274 if 'view_count' in video_info
:
1275 view_count
= int(video_info
['view_count'][0])
1279 # Check for "rental" videos
1280 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
1281 raise ExtractorError('"rental" videos not supported')
1283 # Start extracting information
1284 self
.report_information_extraction(video_id
)
1287 if 'author' not in video_info
:
1288 raise ExtractorError('Unable to extract uploader name')
1289 video_uploader
= compat_urllib_parse_unquote_plus(video_info
['author'][0])
1292 video_uploader_id
= None
1293 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
1294 if mobj
is not None:
1295 video_uploader_id
= mobj
.group(1)
1297 self
._downloader
.report_warning('unable to extract uploader nickname')
1300 # We try first to get a high quality image:
1301 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
1302 video_webpage
, re
.DOTALL
)
1303 if m_thumb
is not None:
1304 video_thumbnail
= m_thumb
.group(1)
1305 elif 'thumbnail_url' not in video_info
:
1306 self
._downloader
.report_warning('unable to extract video thumbnail')
1307 video_thumbnail
= None
1308 else: # don't panic if we can't find it
1309 video_thumbnail
= compat_urllib_parse_unquote_plus(video_info
['thumbnail_url'][0])
1312 upload_date
= self
._html
_search
_meta
(
1313 'datePublished', video_webpage
, 'upload date', default
=None)
1315 upload_date
= self
._search
_regex
(
1316 [r
'(?s)id="eow-date.*?>(.*?)</span>',
1317 r
'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1318 video_webpage
, 'upload date', default
=None)
1320 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
1321 upload_date
= unified_strdate(upload_date
)
1323 m_music
= re
.search(
1324 r
'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1327 video_alt_title
= remove_quotes(unescapeHTML(m_music
.group('title')))
1328 video_creator
= clean_html(m_music
.group('creator'))
1330 video_alt_title
= video_creator
= None
1332 m_cat_container
= self
._search
_regex
(
1333 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1334 video_webpage
, 'categories', default
=None)
1336 category
= self
._html
_search
_regex
(
1337 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
1339 video_categories
= None if category
is None else [category
]
1341 video_categories
= None
1344 unescapeHTML(m
.group('content'))
1345 for m
in re
.finditer(self
._meta
_regex
('og:video:tag'), video_webpage
)]
1347 def _extract_count(count_name
):
1348 return str_to_int(self
._search
_regex
(
1349 r
'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1350 % re
.escape(count_name
),
1351 video_webpage
, count_name
, default
=None))
1353 like_count
= _extract_count('like')
1354 dislike_count
= _extract_count('dislike')
1357 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
1358 automatic_captions
= self
.extract_automatic_captions(video_id
, video_webpage
)
1360 if 'length_seconds' not in video_info
:
1361 self
._downloader
.report_warning('unable to extract video duration')
1362 video_duration
= None
1364 video_duration
= int(compat_urllib_parse_unquote_plus(video_info
['length_seconds'][0]))
1367 video_annotations
= None
1368 if self
._downloader
.params
.get('writeannotations', False):
1369 video_annotations
= self
._extract
_annotations
(video_id
)
1371 def _map_to_format_list(urlmap
):
1373 for itag
, video_real_url
in urlmap
.items():
1376 'url': video_real_url
,
1377 'player_url': player_url
,
1379 if itag
in self
._formats
:
1380 dct
.update(self
._formats
[itag
])
1384 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
1385 self
.report_rtmp_download()
1387 'format_id': '_rtmp',
1389 'url': video_info
['conn'][0],
1390 'player_url': player_url
,
1392 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
1393 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
1394 if 'rtmpe%3Dyes' in encoded_url_map
:
1395 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
1397 for url_data_str
in encoded_url_map
.split(','):
1398 url_data
= compat_parse_qs(url_data_str
)
1399 if 'itag' not in url_data
or 'url' not in url_data
:
1401 format_id
= url_data
['itag'][0]
1402 url
= url_data
['url'][0]
1404 if 'sig' in url_data
:
1405 url
+= '&signature=' + url_data
['sig'][0]
1406 elif 's' in url_data
:
1407 encrypted_sig
= url_data
['s'][0]
1408 ASSETS_RE
= r
'"assets":.+?"js":\s*("[^"]+")'
1410 jsplayer_url_json
= self
._search
_regex
(
1412 embed_webpage
if age_gate
else video_webpage
,
1413 'JS player URL (1)', default
=None)
1414 if not jsplayer_url_json
and not age_gate
:
1415 # We need the embed website after all
1416 if embed_webpage
is None:
1417 embed_url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1418 embed_webpage
= self
._download
_webpage
(
1419 embed_url
, video_id
, 'Downloading embed webpage')
1420 jsplayer_url_json
= self
._search
_regex
(
1421 ASSETS_RE
, embed_webpage
, 'JS player URL')
1423 player_url
= json
.loads(jsplayer_url_json
)
1424 if player_url
is None:
1425 player_url_json
= self
._search
_regex
(
1426 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1427 video_webpage
, 'age gate player URL')
1428 player_url
= json
.loads(player_url_json
)
1430 if self
._downloader
.params
.get('verbose'):
1431 if player_url
is None:
1432 player_version
= 'unknown'
1433 player_desc
= 'unknown'
1435 if player_url
.endswith('swf'):
1436 player_version
= self
._search
_regex
(
1437 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1438 'flash player', fatal
=False)
1439 player_desc
= 'flash player %s' % player_version
1441 player_version
= self
._search
_regex
(
1442 [r
'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r
'(?:www|player)-([^/]+)/base\.js'],
1444 'html5 player', fatal
=False)
1445 player_desc
= 'html5 player %s' % player_version
1447 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1448 self
.to_screen('{%s} signature length %s, %s' %
1449 (format_id
, parts_sizes
, player_desc
))
1451 signature
= self
._decrypt
_signature
(
1452 encrypted_sig
, video_id
, player_url
, age_gate
)
1453 url
+= '&signature=' + signature
1454 if 'ratebypass' not in url
:
1455 url
+= '&ratebypass=yes'
1457 # Some itags are not included in DASH manifest thus corresponding formats will
1458 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1459 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1460 mobj
= re
.search(r
'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data
.get('size', [''])[0])
1461 width
, height
= (int(mobj
.group('width')), int(mobj
.group('height'))) if mobj
else (None, None)
1463 'format_id': format_id
,
1465 'player_url': player_url
,
1466 'filesize': int_or_none(url_data
.get('clen', [None])[0]),
1467 'tbr': float_or_none(url_data
.get('bitrate', [None])[0], 1000),
1470 'fps': int_or_none(url_data
.get('fps', [None])[0]),
1471 'format_note': url_data
.get('quality_label', [None])[0] or url_data
.get('quality', [None])[0],
1473 type_
= url_data
.get('type', [None])[0]
1475 type_split
= type_
.split(';')
1476 kind_ext
= type_split
[0].split('/')
1477 if len(kind_ext
) == 2:
1478 kind
, ext
= kind_ext
1480 if kind
in ('audio', 'video'):
1482 for mobj
in re
.finditer(
1483 r
'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?
)(?P
<val
>.+?
)(?P
=quote
)(?
:;|$
)', type_):
1484 if mobj.group('key
') == 'codecs
':
1485 codecs = mobj.group('val
')
1488 codecs = codecs.split(',')
1489 if len(codecs) == 2:
1490 acodec, vcodec = codecs[0], codecs[1]
1492 acodec, vcodec = (codecs[0], 'none
') if kind == 'audio
' else ('none
', codecs[0])
1497 if format_id in self._formats:
1498 dct.update(self._formats[format_id])
1500 elif video_info.get('hlsvp
'):
1501 manifest_url = video_info['hlsvp
'][0]
1502 url_map = self._extract_from_m3u8(manifest_url, video_id)
1503 formats = _map_to_format_list(url_map)
1504 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1505 for a_format in formats:
1506 a_format.setdefault('http_headers
', {})['Youtubedl
-no
-compression
'] = 'True'
1508 raise ExtractorError('no conn
, hlsvp
or url_encoded_fmt_stream_map information found
in video info
')
1510 # Look for the DASH manifest
1511 if self._downloader.params.get('youtube_include_dash_manifest
', True):
1512 dash_mpd_fatal = True
1513 for dash_manifest_url in dash_mpds:
1516 for df in self._parse_dash_manifest(
1517 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1518 # Do not overwrite DASH format found in some previous DASH manifest
1519 if df['format_id
'] not in dash_formats:
1520 dash_formats[df['format_id
']] = df
1521 # Additional DASH manifests may end up in HTTP Error 403 therefore
1522 # allow them to fail without bug report message if we already have
1523 # some DASH manifest succeeded. This is temporary workaround to reduce
1524 # burst of bug reports until we figure out the reason and whether it
1525 # can be fixed at all.
1526 dash_mpd_fatal = False
1527 except (ExtractorError, KeyError) as e:
1528 self.report_warning(
1529 'Skipping DASH manifest
: %r' % e, video_id)
1531 # Remove the formats we found through non-DASH, they
1532 # contain less info and it can be wrong, because we use
1533 # fixed values (for example the resolution). See
1534 # https://github.com/rg3/youtube-dl/issues/5774 for an
1536 formats = [f for f in formats if f['format_id
'] not in dash_formats.keys()]
1537 formats.extend(dash_formats.values())
1539 # Check for malformed aspect ratio
1540 stretched_m = re.search(
1541 r'<meta\s
+property="og:video:tag".*?content
="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1544 w = float(stretched_m.group('w
'))
1545 h = float(stretched_m.group('h
'))
1546 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1547 # We will only process correct ratios.
1551 if f.get('vcodec
') != 'none
':
1552 f['stretched_ratio
'] = ratio
1554 self._sort_formats(formats)
1558 'uploader
': video_uploader,
1559 'uploader_id
': video_uploader_id,
1560 'upload_date
': upload_date,
1561 'creator
': video_creator,
1562 'title
': video_title,
1563 'alt_title
': video_alt_title,
1564 'thumbnail
': video_thumbnail,
1565 'description
': video_description,
1566 'categories
': video_categories,
1568 'subtitles
': video_subtitles,
1569 'automatic_captions
': automatic_captions,
1570 'duration
': video_duration,
1571 'age_limit
': 18 if age_gate else 0,
1572 'annotations
': video_annotations,
1573 'webpage_url
': proto + '://www
.youtube
.com
/watch?v
=%s' % video_id,
1574 'view_count
': view_count,
1575 'like_count
': like_count,
1576 'dislike_count
': dislike_count,
1577 'average_rating
': float_or_none(video_info.get('avg_rating
', [None])[0]),
1580 'start_time
': start_time,
1581 'end_time
': end_time,
1585 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1586 IE_DESC = 'YouTube
.com playlists
'
1587 _VALID_URL = r"""(?x)(?:
1592 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1593 \? (?:.*?[&;])*? (?:p|a|list)=
1597 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1598 # Top tracks, they can also include dots
1603 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1605 _TEMPLATE_URL = 'https
://www
.youtube
.com
/playlist?
list=%s'
1606 _VIDEO_RE = r'href
="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index
=(?P
<index
>\d
+)(?
:[^
>]+>(?P
<title
>[^
<]+))?
'
1607 IE_NAME = 'youtube
:playlist
'
1609 'url
': 'https
://www
.youtube
.com
/playlist?
list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re
',
1611 'title
': 'ytdl test PL
',
1612 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re
',
1614 'playlist_count
': 3,
1616 'url
': 'https
://www
.youtube
.com
/playlist?
list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx
',
1618 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx
',
1619 'title
': 'YDL_Empty_List
',
1621 'playlist_count
': 0,
1623 'note
': 'Playlist
with deleted
videos (#651). As a bonus, the video #51 is also twice in this list.',
1624 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1626 'title': '29C3: Not my department',
1627 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1629 'playlist_count': 95,
1631 'note': 'issue #673',
1632 'url': 'PLBB231211A4F62143',
1634 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1635 'id': 'PLBB231211A4F62143',
1637 'playlist_mincount': 26,
1639 'note': 'Large playlist',
1640 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1642 'title': 'Uploads from Cauchemar',
1643 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1645 'playlist_mincount': 799,
1647 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1649 'title': 'YDL_safe_search',
1650 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1652 'playlist_count': 2,
1655 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1656 'playlist_count': 4,
1659 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1662 'note': 'Embedded SWF player',
1663 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1664 'playlist_count': 4,
1667 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1670 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1671 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1673 'title': 'Uploads from Interstellar Movie',
1674 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1676 'playlist_mincout': 21,
1679 def _real_initialize(self
):
1682 def _extract_mix(self
, playlist_id
):
1683 # The mixes are generated from a single video
1684 # the id of the playlist is just 'RD' + video_id
1685 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1686 webpage
= self
._download
_webpage
(
1687 url
, playlist_id
, 'Downloading Youtube mix')
1688 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1690 search_title('playlist-title') or
1691 search_title('title long-title') or
1692 search_title('title'))
1693 title
= clean_html(title_span
)
1694 ids
= orderedSet(re
.findall(
1695 r
'''(?xs)data-video-username=".*?".*?
1696 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1698 url_results
= self
._ids
_to
_results
(ids
)
1700 return self
.playlist_result(url_results
, playlist_id
, title
)
1702 def _extract_playlist(self
, playlist_id
):
1703 url
= self
._TEMPLATE
_URL
% playlist_id
1704 page
= self
._download
_webpage
(url
, playlist_id
)
1706 for match
in re
.findall(r
'<div class="yt-alert-message">([^<]+)</div>', page
):
1707 match
= match
.strip()
1708 # Check if the playlist exists or is private
1709 if re
.match(r
'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match
):
1710 raise ExtractorError(
1711 'The playlist doesn\'t exist or is private, use --username or '
1712 '--netrc to access it.',
1714 elif re
.match(r
'[^<]*Invalid parameters[^<]*', match
):
1715 raise ExtractorError(
1716 'Invalid parameters. Maybe URL is incorrect.',
1718 elif re
.match(r
'[^<]*Choose your language[^<]*', match
):
1721 self
.report_warning('Youtube gives an alert message: ' + match
)
1723 playlist_title
= self
._html
_search
_regex
(
1724 r
'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1727 return self
.playlist_result(self
._entries
(page
, playlist_id
), playlist_id
, playlist_title
)
1729 def _real_extract(self
, url
):
1730 # Extract playlist id
1731 mobj
= re
.match(self
._VALID
_URL
, url
)
1733 raise ExtractorError('Invalid URL: %s' % url
)
1734 playlist_id
= mobj
.group(1) or mobj
.group(2)
1736 # Check if it's a video-specific URL
1737 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1738 if 'v' in query_dict
:
1739 video_id
= query_dict
['v'][0]
1740 if self
._downloader
.params
.get('noplaylist'):
1741 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1742 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1744 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1746 if playlist_id
.startswith('RD') or playlist_id
.startswith('UL'):
1747 # Mixes require a custom extraction process
1748 return self
._extract
_mix
(playlist_id
)
1750 return self
._extract
_playlist
(playlist_id
)
1753 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor
):
1754 IE_DESC
= 'YouTube.com channels'
1755 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1756 _TEMPLATE_URL
= 'https://www.youtube.com/channel/%s/videos'
1757 _VIDEO_RE
= r
'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1758 IE_NAME
= 'youtube:channel'
1760 'note': 'paginated channel',
1761 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1762 'playlist_mincount': 91,
1764 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1765 'title': 'Uploads from lex will',
1768 'note': 'Age restricted channel',
1769 # from https://www.youtube.com/user/DeusExOfficial
1770 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1771 'playlist_mincount': 64,
1773 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1774 'title': 'Uploads from Deus Ex',
1779 def suitable(cls
, url
):
1780 return False if YoutubePlaylistsIE
.suitable(url
) else super(YoutubeChannelIE
, cls
).suitable(url
)
1782 def _real_extract(self
, url
):
1783 channel_id
= self
._match
_id
(url
)
1785 url
= self
._TEMPLATE
_URL
% channel_id
1787 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1788 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1789 # otherwise fallback on channel by page extraction
1790 channel_page
= self
._download
_webpage
(
1791 url
+ '?view=57', channel_id
,
1792 'Downloading channel page', fatal
=False)
1793 if channel_page
is False:
1794 channel_playlist_id
= False
1796 channel_playlist_id
= self
._html
_search
_meta
(
1797 'channelId', channel_page
, 'channel id', default
=None)
1798 if not channel_playlist_id
:
1799 channel_playlist_id
= self
._search
_regex
(
1800 r
'data-(?:channel-external-|yt)id="([^"]+)"',
1801 channel_page
, 'channel id', default
=None)
1802 if channel_playlist_id
and channel_playlist_id
.startswith('UC'):
1803 playlist_id
= 'UU' + channel_playlist_id
[2:]
1804 return self
.url_result(
1805 compat_urlparse
.urljoin(url
, '/playlist?list=%s' % playlist_id
), 'YoutubePlaylist')
1807 channel_page
= self
._download
_webpage
(url
, channel_id
, 'Downloading page #1')
1808 autogenerated
= re
.search(r
'''(?x)
1810 channel-header-autogenerated-label|
1811 yt-channel-title-autogenerated
1812 )[^"]*"''', channel_page
) is not None
1815 # The videos are contained in a single page
1816 # the ajax pages can't be used, they are empty
1819 video_id
, 'Youtube', video_id
=video_id
,
1820 video_title
=video_title
)
1821 for video_id
, video_title
in self
.extract_videos_from_page(channel_page
)]
1822 return self
.playlist_result(entries
, channel_id
)
1824 return self
.playlist_result(self
._entries
(channel_page
, channel_id
), channel_id
)
1827 class YoutubeUserIE(YoutubeChannelIE
):
1828 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1829 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1830 _TEMPLATE_URL
= 'https://www.youtube.com/user/%s/videos'
1831 IE_NAME
= 'youtube:user'
1834 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1835 'playlist_mincount': 320,
1837 'title': 'TheLinuxFoundation',
1840 'url': 'ytuser:phihag',
1841 'only_matching': True,
1845 def suitable(cls
, url
):
1846 # Don't return True if the url can be extracted with other youtube
1847 # extractor, the regex would is too permissive and it would match.
1848 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1849 if any(ie
.suitable(url
) for ie
in other_ies
):
1852 return super(YoutubeUserIE
, cls
).suitable(url
)
1855 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor
):
1856 IE_DESC
= 'YouTube.com user/channel playlists'
1857 _VALID_URL
= r
'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1858 IE_NAME
= 'youtube:playlists'
1861 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1862 'playlist_mincount': 4,
1864 'id': 'ThirstForScience',
1865 'title': 'Thirst for Science',
1868 # with "Load more" button
1869 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1870 'playlist_mincount': 70,
1873 'title': 'Игорь Клейнер',
1876 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1877 'playlist_mincount': 17,
1879 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1880 'title': 'Chem Player',
1885 class YoutubeSearchIE(SearchInfoExtractor
, YoutubePlaylistIE
):
1886 IE_DESC
= 'YouTube.com searches'
1887 # there doesn't appear to be a real limit, for example if you search for
1888 # 'python' you get more than 8.000.000 results
1889 _MAX_RESULTS
= float('inf')
1890 IE_NAME
= 'youtube:search'
1891 _SEARCH_KEY
= 'ytsearch'
1892 _EXTRA_QUERY_ARGS
= {}
1895 def _get_n_results(self
, query
, n
):
1896 """Get a specified number of results for a query"""
1901 for pagenum
in itertools
.count(1):
1903 'search_query': query
.encode('utf-8'),
1907 url_query
.update(self
._EXTRA
_QUERY
_ARGS
)
1908 result_url
= 'https://www.youtube.com/results?' + compat_urllib_parse
.urlencode(url_query
)
1909 data
= self
._download
_json
(
1910 result_url
, video_id
='query "%s"' % query
,
1911 note
='Downloading page %s' % pagenum
,
1912 errnote
='Unable to download API page')
1913 html_content
= data
[1]['body']['content']
1915 if 'class="search-message' in html_content
:
1916 raise ExtractorError(
1917 '[youtube] No video results', expected
=True)
1919 new_videos
= self
._ids
_to
_results
(orderedSet(re
.findall(
1920 r
'href="/watch\?v=(.{11})', html_content
)))
1921 videos
+= new_videos
1922 if not new_videos
or len(videos
) > limit
:
1927 return self
.playlist_result(videos
, query
)
1930 class YoutubeSearchDateIE(YoutubeSearchIE
):
1931 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1932 _SEARCH_KEY
= 'ytsearchdate'
1933 IE_DESC
= 'YouTube.com searches, newest videos first'
1934 _EXTRA_QUERY_ARGS
= {'search_sort': 'video_date_uploaded'}
1937 class YoutubeSearchURLIE(InfoExtractor
):
1938 IE_DESC
= 'YouTube.com search URLs'
1939 IE_NAME
= 'youtube:search_url'
1940 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1942 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1943 'playlist_mincount': 5,
1945 'title': 'youtube-dl test video',
1949 def _real_extract(self
, url
):
1950 mobj
= re
.match(self
._VALID
_URL
, url
)
1951 query
= compat_urllib_parse_unquote_plus(mobj
.group('query'))
1953 webpage
= self
._download
_webpage
(url
, query
)
1954 result_code
= self
._search
_regex
(
1955 r
'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1957 part_codes
= re
.findall(
1958 r
'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code
)
1960 for part_code
in part_codes
:
1961 part_title
= self
._html
_search
_regex
(
1962 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1963 part_url_snippet
= self
._html
_search
_regex
(
1964 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1965 part_url
= compat_urlparse
.urljoin(
1966 'https://www.youtube.com/', part_url_snippet
)
1970 'title': part_title
,
1974 '_type': 'playlist',
1980 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor
):
1981 IE_DESC
= 'YouTube.com (multi-season) shows'
1982 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1983 IE_NAME
= 'youtube:show'
1985 'url': 'https://www.youtube.com/show/airdisasters',
1986 'playlist_mincount': 5,
1988 'id': 'airdisasters',
1989 'title': 'Air Disasters',
1993 def _real_extract(self
, url
):
1994 playlist_id
= self
._match
_id
(url
)
1995 return super(YoutubeShowIE
, self
)._real
_extract
(
1996 'https://www.youtube.com/show/%s/playlists' % playlist_id
)
1999 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
2001 Base class for feed extractors
2002 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2004 _LOGIN_REQUIRED
= True
2008 return 'youtube:%s' % self
._FEED
_NAME
2010 def _real_initialize(self
):
2013 def _real_extract(self
, url
):
2014 page
= self
._download
_webpage
(
2015 'https://www.youtube.com/feed/%s' % self
._FEED
_NAME
, self
._PLAYLIST
_TITLE
)
2017 # The extraction process is the same as for playlists, but the regex
2018 # for the video ids doesn't contain an index
2020 more_widget_html
= content_html
= page
2021 for page_num
in itertools
.count(1):
2022 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
2024 # 'recommended' feed has infinite 'load more' and each new portion spins
2025 # the same videos in (sometimes) slightly different order, so we'll check
2026 # for unicity and break when portion has no new videos
2027 new_ids
= filter(lambda video_id
: video_id
not in ids
, orderedSet(matches
))
2033 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
2037 more
= self
._download
_json
(
2038 'https://youtube.com/%s' % mobj
.group('more'), self
._PLAYLIST
_TITLE
,
2039 'Downloading page #%s' % page_num
,
2040 transform_source
=uppercase_escape
)
2041 content_html
= more
['content_html']
2042 more_widget_html
= more
['load_more_widget_html']
2044 return self
.playlist_result(
2045 self
._ids
_to
_results
(ids
), playlist_title
=self
._PLAYLIST
_TITLE
)
2048 class YoutubeWatchLaterIE(YoutubePlaylistIE
):
2049 IE_NAME
= 'youtube:watchlater'
2050 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2051 _VALID_URL
= r
'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2053 _TESTS
= [] # override PlaylistIE tests
2055 def _real_extract(self
, url
):
2056 return self
._extract
_playlist
('WL')
2059 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
2060 IE_NAME
= 'youtube:favorites'
2061 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2062 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2063 _LOGIN_REQUIRED
= True
2065 def _real_extract(self
, url
):
2066 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2067 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
2068 return self
.url_result(playlist_id
, 'YoutubePlaylist')
2071 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
2072 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2073 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2074 _FEED_NAME
= 'recommended'
2075 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
2078 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor
):
2079 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2080 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2081 _FEED_NAME
= 'subscriptions'
2082 _PLAYLIST_TITLE
= 'Youtube Subscriptions'
2085 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
2086 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
2087 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
2088 _FEED_NAME
= 'history'
2089 _PLAYLIST_TITLE
= 'Youtube History'
2092 class YoutubeTruncatedURLIE(InfoExtractor
):
2093 IE_NAME
= 'youtube:truncated_url'
2094 IE_DESC
= False # Do not list
2095 _VALID_URL
= r
'''(?x)
2097 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2100 annotation_id=annotation_[^&]+|
2106 attribution_link\?a=[^&]+
2112 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2113 'only_matching': True,
2115 'url': 'http://www.youtube.com/watch?',
2116 'only_matching': True,
2118 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2119 'only_matching': True,
2121 'url': 'https://www.youtube.com/watch?feature=foo',
2122 'only_matching': True,
2124 'url': 'https://www.youtube.com/watch?hl=en-GB',
2125 'only_matching': True,
2127 'url': 'https://www.youtube.com/watch?t=2372',
2128 'only_matching': True,
2131 def _real_extract(self
, url
):
2132 raise ExtractorError(
2133 'Did you forget to quote the URL? Remember that & is a meta '
2134 'character in most shells, so you want to put the URL in quotes, '
2136 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2137 ' or simply youtube-dl BaW_jenozKc .',
2141 class YoutubeTruncatedIDIE(InfoExtractor
):
2142 IE_NAME
= 'youtube:truncated_id'
2143 IE_DESC
= False # Do not list
2144 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2147 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2148 'only_matching': True,
2151 def _real_extract(self
, url
):
2152 video_id
= self
._match
_id
(url
)
2153 raise ExtractorError(
2154 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),