]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[youtube] Add YouTube Red paid video reference test (#10059)
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_parse_qs,
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 compat_str,
26 )
27 from ..utils import (
28 clean_html,
29 error_to_compat_str,
30 ExtractorError,
31 float_or_none,
32 get_element_by_attribute,
33 get_element_by_id,
34 int_or_none,
35 mimetype2ext,
36 orderedSet,
37 parse_duration,
38 remove_quotes,
39 remove_start,
40 sanitized_Request,
41 smuggle_url,
42 str_to_int,
43 unescapeHTML,
44 unified_strdate,
45 unsmuggle_url,
46 uppercase_escape,
47 urlencode_postdata,
48 ISO3166Utils,
49 )
50
51
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
56 _NETRC_MACHINE = 'youtube'
57 # If True it will raise an error if no login info is provided
58 _LOGIN_REQUIRED = False
59
60 def _set_language(self):
61 self._set_cookie(
62 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
63 # YouTube sets the expire time to about two months
64 expire_time=time.time() + 2 * 30 * 24 * 3600)
65
66 def _ids_to_results(self, ids):
67 return [
68 self.url_result(vid_id, 'Youtube', video_id=vid_id)
69 for vid_id in ids]
70
71 def _login(self):
72 """
73 Attempt to log in to YouTube.
74 True is returned if successful or skipped.
75 False is returned if login failed.
76
77 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
78 """
79 (username, password) = self._get_login_info()
80 # No authentication to be performed
81 if username is None:
82 if self._LOGIN_REQUIRED:
83 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
84 return True
85
86 login_page = self._download_webpage(
87 self._LOGIN_URL, None,
88 note='Downloading login page',
89 errnote='unable to fetch login page', fatal=False)
90 if login_page is False:
91 return
92
93 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
94 login_page, 'Login GALX parameter')
95
96 # Log in
97 login_form_strs = {
98 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
99 'Email': username,
100 'GALX': galx,
101 'Passwd': password,
102
103 'PersistentCookie': 'yes',
104 '_utf8': '霱',
105 'bgresponse': 'js_disabled',
106 'checkConnection': '',
107 'checkedDomains': 'youtube',
108 'dnConn': '',
109 'pstMsg': '0',
110 'rmShown': '1',
111 'secTok': '',
112 'signIn': 'Sign in',
113 'timeStmp': '',
114 'service': 'youtube',
115 'uilel': '3',
116 'hl': 'en_US',
117 }
118
119 login_data = urlencode_postdata(login_form_strs)
120
121 req = sanitized_Request(self._LOGIN_URL, login_data)
122 login_results = self._download_webpage(
123 req, None,
124 note='Logging in', errnote='unable to log in', fatal=False)
125 if login_results is False:
126 return False
127
128 error_msg = self._html_search_regex(
129 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
130 login_results, 'error message', default=None)
131 if error_msg:
132 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
133
134 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
135 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
136
137 # Two-Factor
138 # TODO add SMS and phone call support - these require making a request and then prompting the user
139
140 if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
141 tfa_code = self._get_tfa_info('2-step verification code')
142
143 if not tfa_code:
144 self._downloader.report_warning(
145 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
146 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
147 return False
148
149 tfa_code = remove_start(tfa_code, 'G-')
150
151 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
152
153 tfa_form_strs.update({
154 'Pin': tfa_code,
155 'TrustDevice': 'on',
156 })
157
158 tfa_data = urlencode_postdata(tfa_form_strs)
159
160 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
161 tfa_results = self._download_webpage(
162 tfa_req, None,
163 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
164
165 if tfa_results is False:
166 return False
167
168 if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
169 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
170 return False
171 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
172 self._downloader.report_warning('unable to log in - did the page structure change?')
173 return False
174 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
175 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
176 return False
177
178 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
179 self._downloader.report_warning('unable to log in: bad username or password')
180 return False
181 return True
182
183 def _real_initialize(self):
184 if self._downloader is None:
185 return
186 self._set_language()
187 if not self._login():
188 return
189
190
191 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
192 # Extract entries from page with "Load more" button
193 def _entries(self, page, playlist_id):
194 more_widget_html = content_html = page
195 for page_num in itertools.count(1):
196 for entry in self._process_page(content_html):
197 yield entry
198
199 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
200 if not mobj:
201 break
202
203 more = self._download_json(
204 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
205 'Downloading page #%s' % page_num,
206 transform_source=uppercase_escape)
207 content_html = more['content_html']
208 if not content_html.strip():
209 # Some webpages show a "Load more" button but they don't
210 # have more videos
211 break
212 more_widget_html = more['load_more_widget_html']
213
214
215 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
216 def _process_page(self, content):
217 for video_id, video_title in self.extract_videos_from_page(content):
218 yield self.url_result(video_id, 'Youtube', video_id, video_title)
219
220 def extract_videos_from_page(self, page):
221 ids_in_page = []
222 titles_in_page = []
223 for mobj in re.finditer(self._VIDEO_RE, page):
224 # The link with index 0 is not the first video of the playlist (not sure if still actual)
225 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
226 continue
227 video_id = mobj.group('id')
228 video_title = unescapeHTML(mobj.group('title'))
229 if video_title:
230 video_title = video_title.strip()
231 try:
232 idx = ids_in_page.index(video_id)
233 if video_title and not titles_in_page[idx]:
234 titles_in_page[idx] = video_title
235 except ValueError:
236 ids_in_page.append(video_id)
237 titles_in_page.append(video_title)
238 return zip(ids_in_page, titles_in_page)
239
240
241 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
242 def _process_page(self, content):
243 for playlist_id in orderedSet(re.findall(
244 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
245 content)):
246 yield self.url_result(
247 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
248
249 def _real_extract(self, url):
250 playlist_id = self._match_id(url)
251 webpage = self._download_webpage(url, playlist_id)
252 title = self._og_search_title(webpage, fatal=False)
253 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
254
255
256 class YoutubeIE(YoutubeBaseInfoExtractor):
257 IE_DESC = 'YouTube.com'
258 _VALID_URL = r"""(?x)^
259 (
260 (?:https?://|//) # http(s):// or protocol-independent URL
261 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
262 (?:www\.)?deturl\.com/www\.youtube\.com/|
263 (?:www\.)?pwnyoutube\.com/|
264 (?:www\.)?yourepeat\.com/|
265 tube\.majestyc\.net/|
266 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
267 (?:.*?\#/)? # handle anchor (#/) redirect urls
268 (?: # the various things that can precede the ID:
269 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
270 |(?: # or the v= param in all its forms
271 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
272 (?:\?|\#!?) # the params delimiter ? or # or #!
273 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
274 v=
275 )
276 ))
277 |(?:
278 youtu\.be| # just youtu.be/xxxx
279 vid\.plus| # or vid.plus/xxxx
280 zwearz\.com/watch| # or zwearz.com/watch/xxxx
281 )/
282 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
283 )
284 )? # all until now is optional -> you can pass the naked ID
285 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
286 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
287 (?(1).+)? # if we found the ID, everything can follow
288 $"""
289 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
290 _formats = {
291 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
292 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
293 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
294 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
295 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
296 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
297 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
298 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
300 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
301 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
302 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
303 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
304 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
305 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
306 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
307 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
308 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
309
310
311 # 3D videos
312 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
313 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
314 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
315 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
316 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
317 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
318 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
319
320 # Apple HTTP Live Streaming
321 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
322 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
323 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
324 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
325 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
326 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
327 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
328 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
329
330 # DASH mp4 video
331 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
332 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
333 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
334 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
335 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
336 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
337 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
338 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
339 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
340 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
341 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
342
343 # Dash mp4 audio
344 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
345 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
346 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
347 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
348 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
349
350 # Dash webm
351 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
352 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
353 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
354 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
355 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
356 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
357 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
358 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
359 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
361 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
362 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
363 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
364 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
365 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
366 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
367 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
368 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
369 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
370 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
371 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
372 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
373
374 # Dash webm audio
375 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
376 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
377
378 # Dash webm audio with opus inside
379 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
380 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
381 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
382
383 # RTMP (unnamed)
384 '_rtmp': {'protocol': 'rtmp'},
385 }
386 _SUBTITLE_FORMATS = ('ttml', 'vtt')
387
388 IE_NAME = 'youtube'
389 _TESTS = [
390 {
391 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
392 'info_dict': {
393 'id': 'BaW_jenozKc',
394 'ext': 'mp4',
395 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
396 'uploader': 'Philipp Hagemeister',
397 'uploader_id': 'phihag',
398 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
399 'upload_date': '20121002',
400 'license': 'Standard YouTube License',
401 '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 .',
402 'categories': ['Science & Technology'],
403 'tags': ['youtube-dl'],
404 'like_count': int,
405 'dislike_count': int,
406 'start_time': 1,
407 'end_time': 9,
408 }
409 },
410 {
411 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
412 'note': 'Test generic use_cipher_signature video (#897)',
413 'info_dict': {
414 'id': 'UxxajLWwzqY',
415 'ext': 'mp4',
416 'upload_date': '20120506',
417 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
418 'alt_title': 'I Love It (feat. Charli XCX)',
419 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
420 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
421 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
422 'iconic ep', 'iconic', 'love', 'it'],
423 'uploader': 'Icona Pop',
424 'uploader_id': 'IconaPop',
425 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
426 'license': 'Standard YouTube License',
427 'creator': 'Icona Pop',
428 }
429 },
430 {
431 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
432 'note': 'Test VEVO video with age protection (#956)',
433 'info_dict': {
434 'id': '07FYdnEawAQ',
435 'ext': 'mp4',
436 'upload_date': '20130703',
437 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
438 'alt_title': 'Tunnel Vision',
439 'description': 'md5:64249768eec3bc4276236606ea996373',
440 'uploader': 'justintimberlakeVEVO',
441 'uploader_id': 'justintimberlakeVEVO',
442 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
443 'license': 'Standard YouTube License',
444 'creator': 'Justin Timberlake',
445 'age_limit': 18,
446 }
447 },
448 {
449 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
450 'note': 'Embed-only video (#1746)',
451 'info_dict': {
452 'id': 'yZIXLfi8CZQ',
453 'ext': 'mp4',
454 'upload_date': '20120608',
455 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
456 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
457 'uploader': 'SET India',
458 'uploader_id': 'setindia',
459 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
460 'license': 'Standard YouTube License',
461 'age_limit': 18,
462 }
463 },
464 {
465 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
466 'note': 'Use the first video ID in the URL',
467 'info_dict': {
468 'id': 'BaW_jenozKc',
469 'ext': 'mp4',
470 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
471 'uploader': 'Philipp Hagemeister',
472 'uploader_id': 'phihag',
473 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
474 'upload_date': '20121002',
475 'license': 'Standard YouTube License',
476 '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 .',
477 'categories': ['Science & Technology'],
478 'tags': ['youtube-dl'],
479 'like_count': int,
480 'dislike_count': int,
481 },
482 'params': {
483 'skip_download': True,
484 },
485 },
486 {
487 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
488 'note': '256k DASH audio (format 141) via DASH manifest',
489 'info_dict': {
490 'id': 'a9LDPn-MO4I',
491 'ext': 'm4a',
492 'upload_date': '20121002',
493 'uploader_id': '8KVIDEO',
494 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
495 'description': '',
496 'uploader': '8KVIDEO',
497 'license': 'Standard YouTube License',
498 'title': 'UHDTV TEST 8K VIDEO.mp4'
499 },
500 'params': {
501 'youtube_include_dash_manifest': True,
502 'format': '141',
503 },
504 'skip': 'format 141 not served anymore',
505 },
506 # DASH manifest with encrypted signature
507 {
508 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
509 'info_dict': {
510 'id': 'IB3lcPjvWLA',
511 'ext': 'm4a',
512 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
513 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
514 'uploader': 'AfrojackVEVO',
515 'uploader_id': 'AfrojackVEVO',
516 'upload_date': '20131011',
517 'license': 'Standard YouTube License',
518 },
519 'params': {
520 'youtube_include_dash_manifest': True,
521 'format': '141/bestaudio[ext=m4a]',
522 },
523 },
524 # JS player signature function name containing $
525 {
526 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
527 'info_dict': {
528 'id': 'nfWlot6h_JM',
529 'ext': 'm4a',
530 'title': 'Taylor Swift - Shake It Off',
531 'alt_title': 'Shake It Off',
532 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
533 'uploader': 'TaylorSwiftVEVO',
534 'uploader_id': 'TaylorSwiftVEVO',
535 'upload_date': '20140818',
536 'license': 'Standard YouTube License',
537 'creator': 'Taylor Swift',
538 },
539 'params': {
540 'youtube_include_dash_manifest': True,
541 'format': '141/bestaudio[ext=m4a]',
542 },
543 },
544 # Controversy video
545 {
546 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
547 'info_dict': {
548 'id': 'T4XJQO3qol8',
549 'ext': 'mp4',
550 'upload_date': '20100909',
551 'uploader': 'The Amazing Atheist',
552 'uploader_id': 'TheAmazingAtheist',
553 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
554 'license': 'Standard YouTube License',
555 'title': 'Burning Everyone\'s Koran',
556 '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',
557 }
558 },
559 # Normal age-gate video (No vevo, embed allowed)
560 {
561 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
562 'info_dict': {
563 'id': 'HtVdAasjOgU',
564 'ext': 'mp4',
565 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
566 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
567 'uploader': 'The Witcher',
568 'uploader_id': 'WitcherGame',
569 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
570 'upload_date': '20140605',
571 'license': 'Standard YouTube License',
572 'age_limit': 18,
573 },
574 },
575 # Age-gate video with encrypted signature
576 {
577 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
578 'info_dict': {
579 'id': '6kLq3WMV1nU',
580 'ext': 'mp4',
581 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
582 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
583 'uploader': 'LloydVEVO',
584 'uploader_id': 'LloydVEVO',
585 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
586 'upload_date': '20110629',
587 'license': 'Standard YouTube License',
588 'age_limit': 18,
589 },
590 },
591 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
592 {
593 'url': '__2ABJjxzNo',
594 'info_dict': {
595 'id': '__2ABJjxzNo',
596 'ext': 'mp4',
597 'upload_date': '20100430',
598 'uploader_id': 'deadmau5',
599 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
600 'creator': 'deadmau5',
601 'description': 'md5:12c56784b8032162bb936a5f76d55360',
602 'uploader': 'deadmau5',
603 'license': 'Standard YouTube License',
604 'title': 'Deadmau5 - Some Chords (HD)',
605 'alt_title': 'Some Chords',
606 },
607 'expected_warnings': [
608 'DASH manifest missing',
609 ]
610 },
611 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
612 {
613 'url': 'lqQg6PlCWgI',
614 'info_dict': {
615 'id': 'lqQg6PlCWgI',
616 'ext': 'mp4',
617 'upload_date': '20150827',
618 'uploader_id': 'olympic',
619 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
620 'license': 'Standard YouTube License',
621 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
622 'uploader': 'Olympic',
623 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
624 },
625 'params': {
626 'skip_download': 'requires avconv',
627 }
628 },
629 # Non-square pixels
630 {
631 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
632 'info_dict': {
633 'id': '_b-2C3KPAM0',
634 'ext': 'mp4',
635 'stretched_ratio': 16 / 9.,
636 'upload_date': '20110310',
637 'uploader_id': 'AllenMeow',
638 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
639 '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',
640 'uploader': '孫艾倫',
641 'license': 'Standard YouTube License',
642 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
643 },
644 },
645 # url_encoded_fmt_stream_map is empty string
646 {
647 'url': 'qEJwOuvDf7I',
648 'info_dict': {
649 'id': 'qEJwOuvDf7I',
650 'ext': 'webm',
651 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
652 'description': '',
653 'upload_date': '20150404',
654 'uploader_id': 'spbelect',
655 'uploader': 'Наблюдатели Петербурга',
656 },
657 'params': {
658 'skip_download': 'requires avconv',
659 },
660 'skip': 'This live event has ended.',
661 },
662 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
663 {
664 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
665 'info_dict': {
666 'id': 'FIl7x6_3R5Y',
667 'ext': 'mp4',
668 'title': 'md5:7b81415841e02ecd4313668cde88737a',
669 'description': 'md5:116377fd2963b81ec4ce64b542173306',
670 'upload_date': '20150625',
671 'uploader_id': 'dorappi2000',
672 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
673 'uploader': 'dorappi2000',
674 'license': 'Standard YouTube License',
675 'formats': 'mincount:32',
676 },
677 },
678 # DASH manifest with segment_list
679 {
680 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
681 'md5': '8ce563a1d667b599d21064e982ab9e31',
682 'info_dict': {
683 'id': 'CsmdDsKjzN8',
684 'ext': 'mp4',
685 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
686 'uploader': 'Airtek',
687 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
688 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
689 'license': 'Standard YouTube License',
690 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
691 },
692 'params': {
693 'youtube_include_dash_manifest': True,
694 'format': '135', # bestvideo
695 },
696 'skip': 'This live event has ended.',
697 },
698 {
699 # Multifeed videos (multiple cameras), URL is for Main Camera
700 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
701 'info_dict': {
702 'id': 'jqWvoWXjCVs',
703 'title': 'teamPGP: Rocket League Noob Stream',
704 'description': 'md5:dc7872fb300e143831327f1bae3af010',
705 },
706 'playlist': [{
707 'info_dict': {
708 'id': 'jqWvoWXjCVs',
709 'ext': 'mp4',
710 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
711 'description': 'md5:dc7872fb300e143831327f1bae3af010',
712 'upload_date': '20150721',
713 'uploader': 'Beer Games Beer',
714 'uploader_id': 'beergamesbeer',
715 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
716 'license': 'Standard YouTube License',
717 },
718 }, {
719 'info_dict': {
720 'id': '6h8e8xoXJzg',
721 'ext': 'mp4',
722 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
723 'description': 'md5:dc7872fb300e143831327f1bae3af010',
724 'upload_date': '20150721',
725 'uploader': 'Beer Games Beer',
726 'uploader_id': 'beergamesbeer',
727 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
728 'license': 'Standard YouTube License',
729 },
730 }, {
731 'info_dict': {
732 'id': 'PUOgX5z9xZw',
733 'ext': 'mp4',
734 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
735 'description': 'md5:dc7872fb300e143831327f1bae3af010',
736 'upload_date': '20150721',
737 'uploader': 'Beer Games Beer',
738 'uploader_id': 'beergamesbeer',
739 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
740 'license': 'Standard YouTube License',
741 },
742 }, {
743 'info_dict': {
744 'id': 'teuwxikvS5k',
745 'ext': 'mp4',
746 'title': 'teamPGP: Rocket League Noob Stream (zim)',
747 'description': 'md5:dc7872fb300e143831327f1bae3af010',
748 'upload_date': '20150721',
749 'uploader': 'Beer Games Beer',
750 'uploader_id': 'beergamesbeer',
751 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
752 'license': 'Standard YouTube License',
753 },
754 }],
755 'params': {
756 'skip_download': True,
757 },
758 },
759 {
760 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
761 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
762 'info_dict': {
763 'id': 'gVfLd0zydlo',
764 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
765 },
766 'playlist_count': 2,
767 'skip': 'Not multifeed anymore',
768 },
769 {
770 'url': 'http://vid.plus/FlRa-iH7PGw',
771 'only_matching': True,
772 },
773 {
774 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
775 'only_matching': True,
776 },
777 {
778 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
779 # Also tests cut-off URL expansion in video description (see
780 # https://github.com/rg3/youtube-dl/issues/1892,
781 # https://github.com/rg3/youtube-dl/issues/8164)
782 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
783 'info_dict': {
784 'id': 'lsguqyKfVQg',
785 'ext': 'mp4',
786 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
787 'alt_title': 'Dark Walk',
788 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
789 'upload_date': '20151119',
790 'uploader_id': 'IronSoulElf',
791 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
792 'uploader': 'IronSoulElf',
793 'license': 'Standard YouTube License',
794 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
795 },
796 'params': {
797 'skip_download': True,
798 },
799 },
800 {
801 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
802 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
803 'only_matching': True,
804 },
805 {
806 # Video with yt:stretch=17:0
807 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
808 'info_dict': {
809 'id': 'Q39EVAstoRM',
810 'ext': 'mp4',
811 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
812 'description': 'md5:ee18a25c350637c8faff806845bddee9',
813 'upload_date': '20151107',
814 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
815 'uploader': 'CH GAMER DROID',
816 },
817 'params': {
818 'skip_download': True,
819 },
820 'skip': 'This video does not exist.',
821 },
822 {
823 # Video licensed under Creative Commons
824 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
825 'info_dict': {
826 'id': 'M4gD1WSo5mA',
827 'ext': 'mp4',
828 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
829 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
830 'upload_date': '20150127',
831 'uploader_id': 'BerkmanCenter',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
833 'uploader': 'BerkmanCenter',
834 'license': 'Creative Commons Attribution license (reuse allowed)',
835 },
836 'params': {
837 'skip_download': True,
838 },
839 },
840 {
841 # Channel-like uploader_url
842 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
843 'info_dict': {
844 'id': 'eQcmzGIKrzg',
845 'ext': 'mp4',
846 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
847 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
848 'upload_date': '20151119',
849 'uploader': 'Bernie 2016',
850 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
851 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
852 'license': 'Creative Commons Attribution license (reuse allowed)',
853 },
854 'params': {
855 'skip_download': True,
856 },
857 },
858 {
859 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
860 'only_matching': True,
861 },
862 {
863 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
864 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
865 'only_matching': True,
866 }
867 ]
868
869 def __init__(self, *args, **kwargs):
870 super(YoutubeIE, self).__init__(*args, **kwargs)
871 self._player_cache = {}
872
873 def report_video_info_webpage_download(self, video_id):
874 """Report attempt to download video info webpage."""
875 self.to_screen('%s: Downloading video info webpage' % video_id)
876
877 def report_information_extraction(self, video_id):
878 """Report attempt to extract video information."""
879 self.to_screen('%s: Extracting video information' % video_id)
880
881 def report_unavailable_format(self, video_id, format):
882 """Report extracted video URL."""
883 self.to_screen('%s: Format %s not available' % (video_id, format))
884
885 def report_rtmp_download(self):
886 """Indicate the download will use the RTMP protocol."""
887 self.to_screen('RTMP download detected')
888
889 def _signature_cache_id(self, example_sig):
890 """ Return a string representation of a signature """
891 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
892
893 def _extract_signature_function(self, video_id, player_url, example_sig):
894 id_m = re.match(
895 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
896 player_url)
897 if not id_m:
898 raise ExtractorError('Cannot identify player %r' % player_url)
899 player_type = id_m.group('ext')
900 player_id = id_m.group('id')
901
902 # Read from filesystem cache
903 func_id = '%s_%s_%s' % (
904 player_type, player_id, self._signature_cache_id(example_sig))
905 assert os.path.basename(func_id) == func_id
906
907 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
908 if cache_spec is not None:
909 return lambda s: ''.join(s[i] for i in cache_spec)
910
911 download_note = (
912 'Downloading player %s' % player_url
913 if self._downloader.params.get('verbose') else
914 'Downloading %s player %s' % (player_type, player_id)
915 )
916 if player_type == 'js':
917 code = self._download_webpage(
918 player_url, video_id,
919 note=download_note,
920 errnote='Download of %s failed' % player_url)
921 res = self._parse_sig_js(code)
922 elif player_type == 'swf':
923 urlh = self._request_webpage(
924 player_url, video_id,
925 note=download_note,
926 errnote='Download of %s failed' % player_url)
927 code = urlh.read()
928 res = self._parse_sig_swf(code)
929 else:
930 assert False, 'Invalid player type %r' % player_type
931
932 test_string = ''.join(map(compat_chr, range(len(example_sig))))
933 cache_res = res(test_string)
934 cache_spec = [ord(c) for c in cache_res]
935
936 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
937 return res
938
939 def _print_sig_code(self, func, example_sig):
940 def gen_sig_code(idxs):
941 def _genslice(start, end, step):
942 starts = '' if start == 0 else str(start)
943 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
944 steps = '' if step == 1 else (':%d' % step)
945 return 's[%s%s%s]' % (starts, ends, steps)
946
947 step = None
948 # Quelch pyflakes warnings - start will be set when step is set
949 start = '(Never used)'
950 for i, prev in zip(idxs[1:], idxs[:-1]):
951 if step is not None:
952 if i - prev == step:
953 continue
954 yield _genslice(start, prev, step)
955 step = None
956 continue
957 if i - prev in [-1, 1]:
958 step = i - prev
959 start = prev
960 continue
961 else:
962 yield 's[%d]' % prev
963 if step is None:
964 yield 's[%d]' % i
965 else:
966 yield _genslice(start, i, step)
967
968 test_string = ''.join(map(compat_chr, range(len(example_sig))))
969 cache_res = func(test_string)
970 cache_spec = [ord(c) for c in cache_res]
971 expr_code = ' + '.join(gen_sig_code(cache_spec))
972 signature_id_tuple = '(%s)' % (
973 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
974 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
975 ' return %s\n') % (signature_id_tuple, expr_code)
976 self.to_screen('Extracted signature function:\n' + code)
977
978 def _parse_sig_js(self, jscode):
979 funcname = self._search_regex(
980 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
981 'Initial JS player signature function name')
982
983 jsi = JSInterpreter(jscode)
984 initial_function = jsi.extract_function(funcname)
985 return lambda s: initial_function([s])
986
987 def _parse_sig_swf(self, file_contents):
988 swfi = SWFInterpreter(file_contents)
989 TARGET_CLASSNAME = 'SignatureDecipher'
990 searched_class = swfi.extract_class(TARGET_CLASSNAME)
991 initial_function = swfi.extract_function(searched_class, 'decipher')
992 return lambda s: initial_function([s])
993
994 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
995 """Turn the encrypted s field into a working signature"""
996
997 if player_url is None:
998 raise ExtractorError('Cannot decrypt signature without player_url')
999
1000 if player_url.startswith('//'):
1001 player_url = 'https:' + player_url
1002 try:
1003 player_id = (player_url, self._signature_cache_id(s))
1004 if player_id not in self._player_cache:
1005 func = self._extract_signature_function(
1006 video_id, player_url, s
1007 )
1008 self._player_cache[player_id] = func
1009 func = self._player_cache[player_id]
1010 if self._downloader.params.get('youtube_print_sig_code'):
1011 self._print_sig_code(func, s)
1012 return func(s)
1013 except Exception as e:
1014 tb = traceback.format_exc()
1015 raise ExtractorError(
1016 'Signature extraction failed: ' + tb, cause=e)
1017
1018 def _get_subtitles(self, video_id, webpage):
1019 try:
1020 subs_doc = self._download_xml(
1021 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1022 video_id, note=False)
1023 except ExtractorError as err:
1024 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1025 return {}
1026
1027 sub_lang_list = {}
1028 for track in subs_doc.findall('track'):
1029 lang = track.attrib['lang_code']
1030 if lang in sub_lang_list:
1031 continue
1032 sub_formats = []
1033 for ext in self._SUBTITLE_FORMATS:
1034 params = compat_urllib_parse_urlencode({
1035 'lang': lang,
1036 'v': video_id,
1037 'fmt': ext,
1038 'name': track.attrib['name'].encode('utf-8'),
1039 })
1040 sub_formats.append({
1041 'url': 'https://www.youtube.com/api/timedtext?' + params,
1042 'ext': ext,
1043 })
1044 sub_lang_list[lang] = sub_formats
1045 if not sub_lang_list:
1046 self._downloader.report_warning('video doesn\'t have subtitles')
1047 return {}
1048 return sub_lang_list
1049
1050 def _get_ytplayer_config(self, video_id, webpage):
1051 patterns = (
1052 # User data may contain arbitrary character sequences that may affect
1053 # JSON extraction with regex, e.g. when '};' is contained the second
1054 # regex won't capture the whole JSON. Yet working around by trying more
1055 # concrete regex first keeping in mind proper quoted string handling
1056 # to be implemented in future that will replace this workaround (see
1057 # https://github.com/rg3/youtube-dl/issues/7468,
1058 # https://github.com/rg3/youtube-dl/pull/7599)
1059 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1060 r';ytplayer\.config\s*=\s*({.+?});',
1061 )
1062 config = self._search_regex(
1063 patterns, webpage, 'ytplayer.config', default=None)
1064 if config:
1065 return self._parse_json(
1066 uppercase_escape(config), video_id, fatal=False)
1067
1068 def _get_automatic_captions(self, video_id, webpage):
1069 """We need the webpage for getting the captions url, pass it as an
1070 argument to speed up the process."""
1071 self.to_screen('%s: Looking for automatic captions' % video_id)
1072 player_config = self._get_ytplayer_config(video_id, webpage)
1073 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1074 if not player_config:
1075 self._downloader.report_warning(err_msg)
1076 return {}
1077 try:
1078 args = player_config['args']
1079 caption_url = args.get('ttsurl')
1080 if caption_url:
1081 timestamp = args['timestamp']
1082 # We get the available subtitles
1083 list_params = compat_urllib_parse_urlencode({
1084 'type': 'list',
1085 'tlangs': 1,
1086 'asrs': 1,
1087 })
1088 list_url = caption_url + '&' + list_params
1089 caption_list = self._download_xml(list_url, video_id)
1090 original_lang_node = caption_list.find('track')
1091 if original_lang_node is None:
1092 self._downloader.report_warning('Video doesn\'t have automatic captions')
1093 return {}
1094 original_lang = original_lang_node.attrib['lang_code']
1095 caption_kind = original_lang_node.attrib.get('kind', '')
1096
1097 sub_lang_list = {}
1098 for lang_node in caption_list.findall('target'):
1099 sub_lang = lang_node.attrib['lang_code']
1100 sub_formats = []
1101 for ext in self._SUBTITLE_FORMATS:
1102 params = compat_urllib_parse_urlencode({
1103 'lang': original_lang,
1104 'tlang': sub_lang,
1105 'fmt': ext,
1106 'ts': timestamp,
1107 'kind': caption_kind,
1108 })
1109 sub_formats.append({
1110 'url': caption_url + '&' + params,
1111 'ext': ext,
1112 })
1113 sub_lang_list[sub_lang] = sub_formats
1114 return sub_lang_list
1115
1116 # Some videos don't provide ttsurl but rather caption_tracks and
1117 # caption_translation_languages (e.g. 20LmZk1hakA)
1118 caption_tracks = args['caption_tracks']
1119 caption_translation_languages = args['caption_translation_languages']
1120 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1121 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1122 caption_qs = compat_parse_qs(parsed_caption_url.query)
1123
1124 sub_lang_list = {}
1125 for lang in caption_translation_languages.split(','):
1126 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1127 sub_lang = lang_qs.get('lc', [None])[0]
1128 if not sub_lang:
1129 continue
1130 sub_formats = []
1131 for ext in self._SUBTITLE_FORMATS:
1132 caption_qs.update({
1133 'tlang': [sub_lang],
1134 'fmt': [ext],
1135 })
1136 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1137 query=compat_urllib_parse_urlencode(caption_qs, True)))
1138 sub_formats.append({
1139 'url': sub_url,
1140 'ext': ext,
1141 })
1142 sub_lang_list[sub_lang] = sub_formats
1143 return sub_lang_list
1144 # An extractor error can be raise by the download process if there are
1145 # no automatic captions but there are subtitles
1146 except (KeyError, ExtractorError):
1147 self._downloader.report_warning(err_msg)
1148 return {}
1149
1150 def _mark_watched(self, video_id, video_info):
1151 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1152 if not playback_url:
1153 return
1154 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1155 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1156
1157 # cpn generation algorithm is reverse engineered from base.js.
1158 # In fact it works even with dummy cpn.
1159 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1160 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1161
1162 qs.update({
1163 'ver': ['2'],
1164 'cpn': [cpn],
1165 })
1166 playback_url = compat_urlparse.urlunparse(
1167 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1168
1169 self._download_webpage(
1170 playback_url, video_id, 'Marking watched',
1171 'Unable to mark watched', fatal=False)
1172
1173 @classmethod
1174 def extract_id(cls, url):
1175 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1176 if mobj is None:
1177 raise ExtractorError('Invalid URL: %s' % url)
1178 video_id = mobj.group(2)
1179 return video_id
1180
1181 def _extract_from_m3u8(self, manifest_url, video_id):
1182 url_map = {}
1183
1184 def _get_urls(_manifest):
1185 lines = _manifest.split('\n')
1186 urls = filter(lambda l: l and not l.startswith('#'),
1187 lines)
1188 return urls
1189 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1190 formats_urls = _get_urls(manifest)
1191 for format_url in formats_urls:
1192 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1193 url_map[itag] = format_url
1194 return url_map
1195
1196 def _extract_annotations(self, video_id):
1197 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1198 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1199
1200 def _real_extract(self, url):
1201 url, smuggled_data = unsmuggle_url(url, {})
1202
1203 proto = (
1204 'http' if self._downloader.params.get('prefer_insecure', False)
1205 else 'https')
1206
1207 start_time = None
1208 end_time = None
1209 parsed_url = compat_urllib_parse_urlparse(url)
1210 for component in [parsed_url.fragment, parsed_url.query]:
1211 query = compat_parse_qs(component)
1212 if start_time is None and 't' in query:
1213 start_time = parse_duration(query['t'][0])
1214 if start_time is None and 'start' in query:
1215 start_time = parse_duration(query['start'][0])
1216 if end_time is None and 'end' in query:
1217 end_time = parse_duration(query['end'][0])
1218
1219 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1220 mobj = re.search(self._NEXT_URL_RE, url)
1221 if mobj:
1222 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1223 video_id = self.extract_id(url)
1224
1225 # Get video webpage
1226 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1227 video_webpage = self._download_webpage(url, video_id)
1228
1229 # Attempt to extract SWF player URL
1230 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1231 if mobj is not None:
1232 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1233 else:
1234 player_url = None
1235
1236 dash_mpds = []
1237
1238 def add_dash_mpd(video_info):
1239 dash_mpd = video_info.get('dashmpd')
1240 if dash_mpd and dash_mpd[0] not in dash_mpds:
1241 dash_mpds.append(dash_mpd[0])
1242
1243 # Get video info
1244 embed_webpage = None
1245 is_live = None
1246 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1247 age_gate = True
1248 # We simulate the access to the video from www.youtube.com/v/{video_id}
1249 # this can be viewed without login into Youtube
1250 url = proto + '://www.youtube.com/embed/%s' % video_id
1251 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1252 data = compat_urllib_parse_urlencode({
1253 'video_id': video_id,
1254 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1255 'sts': self._search_regex(
1256 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1257 })
1258 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1259 video_info_webpage = self._download_webpage(
1260 video_info_url, video_id,
1261 note='Refetching age-gated info webpage',
1262 errnote='unable to download video info webpage')
1263 video_info = compat_parse_qs(video_info_webpage)
1264 add_dash_mpd(video_info)
1265 else:
1266 age_gate = False
1267 video_info = None
1268 # Try looking directly into the video webpage
1269 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1270 if ytplayer_config:
1271 args = ytplayer_config['args']
1272 if args.get('url_encoded_fmt_stream_map'):
1273 # Convert to the same format returned by compat_parse_qs
1274 video_info = dict((k, [v]) for k, v in args.items())
1275 add_dash_mpd(video_info)
1276 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1277 is_live = True
1278 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1279 # We also try looking in get_video_info since it may contain different dashmpd
1280 # URL that points to a DASH manifest with possibly different itag set (some itags
1281 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1282 # manifest pointed by get_video_info's dashmpd).
1283 # The general idea is to take a union of itags of both DASH manifests (for example
1284 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1285 self.report_video_info_webpage_download(video_id)
1286 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1287 video_info_url = (
1288 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1289 % (proto, video_id, el_type))
1290 video_info_webpage = self._download_webpage(
1291 video_info_url,
1292 video_id, note=False,
1293 errnote='unable to download video info webpage')
1294 get_video_info = compat_parse_qs(video_info_webpage)
1295 if get_video_info.get('use_cipher_signature') != ['True']:
1296 add_dash_mpd(get_video_info)
1297 if not video_info:
1298 video_info = get_video_info
1299 if 'token' in get_video_info:
1300 # Different get_video_info requests may report different results, e.g.
1301 # some may report video unavailability, but some may serve it without
1302 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1303 # the original webpage as well as el=info and el=embedded get_video_info
1304 # requests report video unavailability due to geo restriction while
1305 # el=detailpage succeeds and returns valid data). This is probably
1306 # due to YouTube measures against IP ranges of hosting providers.
1307 # Working around by preferring the first succeeded video_info containing
1308 # the token if no such video_info yet was found.
1309 if 'token' not in video_info:
1310 video_info = get_video_info
1311 break
1312 if 'token' not in video_info:
1313 if 'reason' in video_info:
1314 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1315 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1316 if regions_allowed:
1317 raise ExtractorError('YouTube said: This video is available in %s only' % (
1318 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1319 expected=True)
1320 raise ExtractorError(
1321 'YouTube said: %s' % video_info['reason'][0],
1322 expected=True, video_id=video_id)
1323 else:
1324 raise ExtractorError(
1325 '"token" parameter not in video info for unknown reason',
1326 video_id=video_id)
1327
1328 # title
1329 if 'title' in video_info:
1330 video_title = video_info['title'][0]
1331 else:
1332 self._downloader.report_warning('Unable to extract video title')
1333 video_title = '_'
1334
1335 # description
1336 video_description = get_element_by_id("eow-description", video_webpage)
1337 if video_description:
1338 video_description = re.sub(r'''(?x)
1339 <a\s+
1340 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1341 (?:title|href)="([^"]+)"\s+
1342 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1343 class="[^"]*"[^>]*>
1344 [^<]+\.{3}\s*
1345 </a>
1346 ''', r'\1', video_description)
1347 video_description = clean_html(video_description)
1348 else:
1349 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1350 if fd_mobj:
1351 video_description = unescapeHTML(fd_mobj.group(1))
1352 else:
1353 video_description = ''
1354
1355 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1356 if not self._downloader.params.get('noplaylist'):
1357 entries = []
1358 feed_ids = []
1359 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1360 for feed in multifeed_metadata_list.split(','):
1361 # Unquote should take place before split on comma (,) since textual
1362 # fields may contain comma as well (see
1363 # https://github.com/rg3/youtube-dl/issues/8536)
1364 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1365 entries.append({
1366 '_type': 'url_transparent',
1367 'ie_key': 'Youtube',
1368 'url': smuggle_url(
1369 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1370 {'force_singlefeed': True}),
1371 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1372 })
1373 feed_ids.append(feed_data['id'][0])
1374 self.to_screen(
1375 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1376 % (', '.join(feed_ids), video_id))
1377 return self.playlist_result(entries, video_id, video_title, video_description)
1378 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1379
1380 if 'view_count' in video_info:
1381 view_count = int(video_info['view_count'][0])
1382 else:
1383 view_count = None
1384
1385 # Check for "rental" videos
1386 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1387 raise ExtractorError('"rental" videos not supported')
1388
1389 # Start extracting information
1390 self.report_information_extraction(video_id)
1391
1392 # uploader
1393 if 'author' not in video_info:
1394 raise ExtractorError('Unable to extract uploader name')
1395 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1396
1397 # uploader_id
1398 video_uploader_id = None
1399 video_uploader_url = None
1400 mobj = re.search(
1401 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1402 video_webpage)
1403 if mobj is not None:
1404 video_uploader_id = mobj.group('uploader_id')
1405 video_uploader_url = mobj.group('uploader_url')
1406 else:
1407 self._downloader.report_warning('unable to extract uploader nickname')
1408
1409 # thumbnail image
1410 # We try first to get a high quality image:
1411 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1412 video_webpage, re.DOTALL)
1413 if m_thumb is not None:
1414 video_thumbnail = m_thumb.group(1)
1415 elif 'thumbnail_url' not in video_info:
1416 self._downloader.report_warning('unable to extract video thumbnail')
1417 video_thumbnail = None
1418 else: # don't panic if we can't find it
1419 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1420
1421 # upload date
1422 upload_date = self._html_search_meta(
1423 'datePublished', video_webpage, 'upload date', default=None)
1424 if not upload_date:
1425 upload_date = self._search_regex(
1426 [r'(?s)id="eow-date.*?>(.*?)</span>',
1427 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1428 video_webpage, 'upload date', default=None)
1429 if upload_date:
1430 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1431 upload_date = unified_strdate(upload_date)
1432
1433 video_license = self._html_search_regex(
1434 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1435 video_webpage, 'license', default=None)
1436
1437 m_music = re.search(
1438 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1439 video_webpage)
1440 if m_music:
1441 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1442 video_creator = clean_html(m_music.group('creator'))
1443 else:
1444 video_alt_title = video_creator = None
1445
1446 m_cat_container = self._search_regex(
1447 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1448 video_webpage, 'categories', default=None)
1449 if m_cat_container:
1450 category = self._html_search_regex(
1451 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1452 default=None)
1453 video_categories = None if category is None else [category]
1454 else:
1455 video_categories = None
1456
1457 video_tags = [
1458 unescapeHTML(m.group('content'))
1459 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1460
1461 def _extract_count(count_name):
1462 return str_to_int(self._search_regex(
1463 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1464 % re.escape(count_name),
1465 video_webpage, count_name, default=None))
1466
1467 like_count = _extract_count('like')
1468 dislike_count = _extract_count('dislike')
1469
1470 # subtitles
1471 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1472 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1473
1474 if 'length_seconds' not in video_info:
1475 self._downloader.report_warning('unable to extract video duration')
1476 video_duration = None
1477 else:
1478 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1479
1480 # annotations
1481 video_annotations = None
1482 if self._downloader.params.get('writeannotations', False):
1483 video_annotations = self._extract_annotations(video_id)
1484
1485 def _map_to_format_list(urlmap):
1486 formats = []
1487 for itag, video_real_url in urlmap.items():
1488 dct = {
1489 'format_id': itag,
1490 'url': video_real_url,
1491 'player_url': player_url,
1492 }
1493 if itag in self._formats:
1494 dct.update(self._formats[itag])
1495 formats.append(dct)
1496 return formats
1497
1498 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1499 self.report_rtmp_download()
1500 formats = [{
1501 'format_id': '_rtmp',
1502 'protocol': 'rtmp',
1503 'url': video_info['conn'][0],
1504 'player_url': player_url,
1505 }]
1506 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1507 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1508 if 'rtmpe%3Dyes' in encoded_url_map:
1509 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1510 formats_spec = {}
1511 fmt_list = video_info.get('fmt_list', [''])[0]
1512 if fmt_list:
1513 for fmt in fmt_list.split(','):
1514 spec = fmt.split('/')
1515 if len(spec) > 1:
1516 width_height = spec[1].split('x')
1517 if len(width_height) == 2:
1518 formats_spec[spec[0]] = {
1519 'resolution': spec[1],
1520 'width': int_or_none(width_height[0]),
1521 'height': int_or_none(width_height[1]),
1522 }
1523 formats = []
1524 for url_data_str in encoded_url_map.split(','):
1525 url_data = compat_parse_qs(url_data_str)
1526 if 'itag' not in url_data or 'url' not in url_data:
1527 continue
1528 format_id = url_data['itag'][0]
1529 url = url_data['url'][0]
1530
1531 if 'sig' in url_data:
1532 url += '&signature=' + url_data['sig'][0]
1533 elif 's' in url_data:
1534 encrypted_sig = url_data['s'][0]
1535 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1536
1537 jsplayer_url_json = self._search_regex(
1538 ASSETS_RE,
1539 embed_webpage if age_gate else video_webpage,
1540 'JS player URL (1)', default=None)
1541 if not jsplayer_url_json and not age_gate:
1542 # We need the embed website after all
1543 if embed_webpage is None:
1544 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1545 embed_webpage = self._download_webpage(
1546 embed_url, video_id, 'Downloading embed webpage')
1547 jsplayer_url_json = self._search_regex(
1548 ASSETS_RE, embed_webpage, 'JS player URL')
1549
1550 player_url = json.loads(jsplayer_url_json)
1551 if player_url is None:
1552 player_url_json = self._search_regex(
1553 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1554 video_webpage, 'age gate player URL')
1555 player_url = json.loads(player_url_json)
1556
1557 if self._downloader.params.get('verbose'):
1558 if player_url is None:
1559 player_version = 'unknown'
1560 player_desc = 'unknown'
1561 else:
1562 if player_url.endswith('swf'):
1563 player_version = self._search_regex(
1564 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1565 'flash player', fatal=False)
1566 player_desc = 'flash player %s' % player_version
1567 else:
1568 player_version = self._search_regex(
1569 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1570 player_url,
1571 'html5 player', fatal=False)
1572 player_desc = 'html5 player %s' % player_version
1573
1574 parts_sizes = self._signature_cache_id(encrypted_sig)
1575 self.to_screen('{%s} signature length %s, %s' %
1576 (format_id, parts_sizes, player_desc))
1577
1578 signature = self._decrypt_signature(
1579 encrypted_sig, video_id, player_url, age_gate)
1580 url += '&signature=' + signature
1581 if 'ratebypass' not in url:
1582 url += '&ratebypass=yes'
1583
1584 dct = {
1585 'format_id': format_id,
1586 'url': url,
1587 'player_url': player_url,
1588 }
1589 if format_id in self._formats:
1590 dct.update(self._formats[format_id])
1591 if format_id in formats_spec:
1592 dct.update(formats_spec[format_id])
1593
1594 # Some itags are not included in DASH manifest thus corresponding formats will
1595 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1596 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1597 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1598 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1599
1600 more_fields = {
1601 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1602 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1603 'width': width,
1604 'height': height,
1605 'fps': int_or_none(url_data.get('fps', [None])[0]),
1606 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1607 }
1608 for key, value in more_fields.items():
1609 if value:
1610 dct[key] = value
1611 type_ = url_data.get('type', [None])[0]
1612 if type_:
1613 type_split = type_.split(';')
1614 kind_ext = type_split[0].split('/')
1615 if len(kind_ext) == 2:
1616 kind, _ = kind_ext
1617 dct['ext'] = mimetype2ext(type_split[0])
1618 if kind in ('audio', 'video'):
1619 codecs = None
1620 for mobj in re.finditer(
1621 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1622 if mobj.group('key') == 'codecs':
1623 codecs = mobj.group('val')
1624 break
1625 if codecs:
1626 codecs = codecs.split(',')
1627 if len(codecs) == 2:
1628 acodec, vcodec = codecs[1], codecs[0]
1629 else:
1630 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1631 dct.update({
1632 'acodec': acodec,
1633 'vcodec': vcodec,
1634 })
1635 formats.append(dct)
1636 elif video_info.get('hlsvp'):
1637 manifest_url = video_info['hlsvp'][0]
1638 url_map = self._extract_from_m3u8(manifest_url, video_id)
1639 formats = _map_to_format_list(url_map)
1640 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1641 for a_format in formats:
1642 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1643 else:
1644 unavailable_message = self._html_search_regex(
1645 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1646 video_webpage, 'unavailable message', default=None)
1647 if unavailable_message:
1648 raise ExtractorError(unavailable_message, expected=True)
1649 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1650
1651 # Look for the DASH manifest
1652 if self._downloader.params.get('youtube_include_dash_manifest', True):
1653 dash_mpd_fatal = True
1654 for mpd_url in dash_mpds:
1655 dash_formats = {}
1656 try:
1657 def decrypt_sig(mobj):
1658 s = mobj.group(1)
1659 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1660 return '/signature/%s' % dec_s
1661
1662 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1663
1664 for df in self._extract_mpd_formats(
1665 mpd_url, video_id, fatal=dash_mpd_fatal,
1666 formats_dict=self._formats):
1667 # Do not overwrite DASH format found in some previous DASH manifest
1668 if df['format_id'] not in dash_formats:
1669 dash_formats[df['format_id']] = df
1670 # Additional DASH manifests may end up in HTTP Error 403 therefore
1671 # allow them to fail without bug report message if we already have
1672 # some DASH manifest succeeded. This is temporary workaround to reduce
1673 # burst of bug reports until we figure out the reason and whether it
1674 # can be fixed at all.
1675 dash_mpd_fatal = False
1676 except (ExtractorError, KeyError) as e:
1677 self.report_warning(
1678 'Skipping DASH manifest: %r' % e, video_id)
1679 if dash_formats:
1680 # Remove the formats we found through non-DASH, they
1681 # contain less info and it can be wrong, because we use
1682 # fixed values (for example the resolution). See
1683 # https://github.com/rg3/youtube-dl/issues/5774 for an
1684 # example.
1685 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1686 formats.extend(dash_formats.values())
1687
1688 # Check for malformed aspect ratio
1689 stretched_m = re.search(
1690 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1691 video_webpage)
1692 if stretched_m:
1693 w = float(stretched_m.group('w'))
1694 h = float(stretched_m.group('h'))
1695 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1696 # We will only process correct ratios.
1697 if w > 0 and h > 0:
1698 ratio = w / h
1699 for f in formats:
1700 if f.get('vcodec') != 'none':
1701 f['stretched_ratio'] = ratio
1702
1703 self._sort_formats(formats)
1704
1705 self.mark_watched(video_id, video_info)
1706
1707 return {
1708 'id': video_id,
1709 'uploader': video_uploader,
1710 'uploader_id': video_uploader_id,
1711 'uploader_url': video_uploader_url,
1712 'upload_date': upload_date,
1713 'license': video_license,
1714 'creator': video_creator,
1715 'title': video_title,
1716 'alt_title': video_alt_title,
1717 'thumbnail': video_thumbnail,
1718 'description': video_description,
1719 'categories': video_categories,
1720 'tags': video_tags,
1721 'subtitles': video_subtitles,
1722 'automatic_captions': automatic_captions,
1723 'duration': video_duration,
1724 'age_limit': 18 if age_gate else 0,
1725 'annotations': video_annotations,
1726 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1727 'view_count': view_count,
1728 'like_count': like_count,
1729 'dislike_count': dislike_count,
1730 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1731 'formats': formats,
1732 'is_live': is_live,
1733 'start_time': start_time,
1734 'end_time': end_time,
1735 }
1736
1737
1738 class YoutubeSharedVideoIE(InfoExtractor):
1739 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?ci=(?P<id>[0-9A-Za-z_-]{11})'
1740 IE_NAME = 'youtube:shared'
1741
1742 _TEST = {
1743 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1744 'info_dict': {
1745 'id': 'uPDB5I9wfp8',
1746 'ext': 'webm',
1747 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1748 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1749 'upload_date': '20160219',
1750 'uploader': 'Pocoyo - Português (BR)',
1751 'uploader_id': 'PocoyoBrazil',
1752 },
1753 'add_ie': ['Youtube'],
1754 'params': {
1755 # There are already too many Youtube downloads
1756 'skip_download': True,
1757 },
1758 }
1759
1760 def _real_extract(self, url):
1761 video_id = self._match_id(url)
1762
1763 webpage = self._download_webpage(url, video_id)
1764
1765 real_video_id = self._html_search_meta(
1766 'videoId', webpage, 'YouTube video id', fatal=True)
1767
1768 return self.url_result(real_video_id, YoutubeIE.ie_key())
1769
1770
1771 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1772 IE_DESC = 'YouTube.com playlists'
1773 _VALID_URL = r"""(?x)(?:
1774 (?:https?://)?
1775 (?:\w+\.)?
1776 youtube\.com/
1777 (?:
1778 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1779 \? (?:.*?[&;])*? (?:p|a|list)=
1780 | p/
1781 )
1782 (
1783 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1784 # Top tracks, they can also include dots
1785 |(?:MC)[\w\.]*
1786 )
1787 .*
1788 |
1789 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1790 )"""
1791 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1792 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1793 IE_NAME = 'youtube:playlist'
1794 _TESTS = [{
1795 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1796 'info_dict': {
1797 'title': 'ytdl test PL',
1798 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1799 },
1800 'playlist_count': 3,
1801 }, {
1802 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1803 'info_dict': {
1804 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1805 'title': 'YDL_Empty_List',
1806 },
1807 'playlist_count': 0,
1808 }, {
1809 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1810 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1811 'info_dict': {
1812 'title': '29C3: Not my department',
1813 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1814 },
1815 'playlist_count': 95,
1816 }, {
1817 'note': 'issue #673',
1818 'url': 'PLBB231211A4F62143',
1819 'info_dict': {
1820 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1821 'id': 'PLBB231211A4F62143',
1822 },
1823 'playlist_mincount': 26,
1824 }, {
1825 'note': 'Large playlist',
1826 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1827 'info_dict': {
1828 'title': 'Uploads from Cauchemar',
1829 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1830 },
1831 'playlist_mincount': 799,
1832 }, {
1833 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1834 'info_dict': {
1835 'title': 'YDL_safe_search',
1836 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1837 },
1838 'playlist_count': 2,
1839 }, {
1840 'note': 'embedded',
1841 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1842 'playlist_count': 4,
1843 'info_dict': {
1844 'title': 'JODA15',
1845 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1846 }
1847 }, {
1848 'note': 'Embedded SWF player',
1849 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1850 'playlist_count': 4,
1851 'info_dict': {
1852 'title': 'JODA7',
1853 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1854 }
1855 }, {
1856 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1857 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1858 'info_dict': {
1859 'title': 'Uploads from Interstellar Movie',
1860 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1861 },
1862 'playlist_mincout': 21,
1863 }]
1864
1865 def _real_initialize(self):
1866 self._login()
1867
1868 def _extract_mix(self, playlist_id):
1869 # The mixes are generated from a single video
1870 # the id of the playlist is just 'RD' + video_id
1871 ids = []
1872 last_id = playlist_id[-11:]
1873 for n in itertools.count(1):
1874 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1875 webpage = self._download_webpage(
1876 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1877 new_ids = orderedSet(re.findall(
1878 r'''(?xs)data-video-username=".*?".*?
1879 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
1880 webpage))
1881 # Fetch new pages until all the videos are repeated, it seems that
1882 # there are always 51 unique videos.
1883 new_ids = [_id for _id in new_ids if _id not in ids]
1884 if not new_ids:
1885 break
1886 ids.extend(new_ids)
1887 last_id = ids[-1]
1888
1889 url_results = self._ids_to_results(ids)
1890
1891 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1892 title_span = (
1893 search_title('playlist-title') or
1894 search_title('title long-title') or
1895 search_title('title'))
1896 title = clean_html(title_span)
1897
1898 return self.playlist_result(url_results, playlist_id, title)
1899
1900 def _extract_playlist(self, playlist_id):
1901 url = self._TEMPLATE_URL % playlist_id
1902 page = self._download_webpage(url, playlist_id)
1903
1904 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1905 match = match.strip()
1906 # Check if the playlist exists or is private
1907 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1908 raise ExtractorError(
1909 'The playlist doesn\'t exist or is private, use --username or '
1910 '--netrc to access it.',
1911 expected=True)
1912 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1913 raise ExtractorError(
1914 'Invalid parameters. Maybe URL is incorrect.',
1915 expected=True)
1916 elif re.match(r'[^<]*Choose your language[^<]*', match):
1917 continue
1918 else:
1919 self.report_warning('Youtube gives an alert message: ' + match)
1920
1921 playlist_title = self._html_search_regex(
1922 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1923 page, 'title')
1924
1925 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1926
1927 def _check_download_just_video(self, url, playlist_id):
1928 # Check if it's a video-specific URL
1929 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1930 if 'v' in query_dict:
1931 video_id = query_dict['v'][0]
1932 if self._downloader.params.get('noplaylist'):
1933 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1934 return self.url_result(video_id, 'Youtube', video_id=video_id)
1935 else:
1936 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1937
1938 def _real_extract(self, url):
1939 # Extract playlist id
1940 mobj = re.match(self._VALID_URL, url)
1941 if mobj is None:
1942 raise ExtractorError('Invalid URL: %s' % url)
1943 playlist_id = mobj.group(1) or mobj.group(2)
1944
1945 video = self._check_download_just_video(url, playlist_id)
1946 if video:
1947 return video
1948
1949 if playlist_id.startswith(('RD', 'UL', 'PU')):
1950 # Mixes require a custom extraction process
1951 return self._extract_mix(playlist_id)
1952
1953 return self._extract_playlist(playlist_id)
1954
1955
1956 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1957 IE_DESC = 'YouTube.com channels'
1958 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1959 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1960 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1961 IE_NAME = 'youtube:channel'
1962 _TESTS = [{
1963 'note': 'paginated channel',
1964 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1965 'playlist_mincount': 91,
1966 'info_dict': {
1967 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1968 'title': 'Uploads from lex will',
1969 }
1970 }, {
1971 'note': 'Age restricted channel',
1972 # from https://www.youtube.com/user/DeusExOfficial
1973 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1974 'playlist_mincount': 64,
1975 'info_dict': {
1976 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1977 'title': 'Uploads from Deus Ex',
1978 },
1979 }]
1980
1981 @classmethod
1982 def suitable(cls, url):
1983 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1984 else super(YoutubeChannelIE, cls).suitable(url))
1985
1986 def _build_template_url(self, url, channel_id):
1987 return self._TEMPLATE_URL % channel_id
1988
1989 def _real_extract(self, url):
1990 channel_id = self._match_id(url)
1991
1992 url = self._build_template_url(url, channel_id)
1993
1994 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1995 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1996 # otherwise fallback on channel by page extraction
1997 channel_page = self._download_webpage(
1998 url + '?view=57', channel_id,
1999 'Downloading channel page', fatal=False)
2000 if channel_page is False:
2001 channel_playlist_id = False
2002 else:
2003 channel_playlist_id = self._html_search_meta(
2004 'channelId', channel_page, 'channel id', default=None)
2005 if not channel_playlist_id:
2006 channel_url = self._html_search_meta(
2007 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2008 channel_page, 'channel url', default=None)
2009 if channel_url:
2010 channel_playlist_id = self._search_regex(
2011 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2012 channel_url, 'channel id', default=None)
2013 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2014 playlist_id = 'UU' + channel_playlist_id[2:]
2015 return self.url_result(
2016 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2017
2018 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2019 autogenerated = re.search(r'''(?x)
2020 class="[^"]*?(?:
2021 channel-header-autogenerated-label|
2022 yt-channel-title-autogenerated
2023 )[^"]*"''', channel_page) is not None
2024
2025 if autogenerated:
2026 # The videos are contained in a single page
2027 # the ajax pages can't be used, they are empty
2028 entries = [
2029 self.url_result(
2030 video_id, 'Youtube', video_id=video_id,
2031 video_title=video_title)
2032 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2033 return self.playlist_result(entries, channel_id)
2034
2035 try:
2036 next(self._entries(channel_page, channel_id))
2037 except StopIteration:
2038 alert_message = self._html_search_regex(
2039 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2040 channel_page, 'alert', default=None, group='alert')
2041 if alert_message:
2042 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2043
2044 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2045
2046
2047 class YoutubeUserIE(YoutubeChannelIE):
2048 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2049 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2050 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2051 IE_NAME = 'youtube:user'
2052
2053 _TESTS = [{
2054 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2055 'playlist_mincount': 320,
2056 'info_dict': {
2057 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2058 'title': 'Uploads from The Linux Foundation',
2059 }
2060 }, {
2061 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2062 # but not https://www.youtube.com/user/12minuteathlete/videos
2063 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2064 'playlist_mincount': 249,
2065 'info_dict': {
2066 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2067 'title': 'Uploads from 12 Minute Athlete',
2068 }
2069 }, {
2070 'url': 'ytuser:phihag',
2071 'only_matching': True,
2072 }, {
2073 'url': 'https://www.youtube.com/c/gametrailers',
2074 'only_matching': True,
2075 }, {
2076 'url': 'https://www.youtube.com/gametrailers',
2077 'only_matching': True,
2078 }, {
2079 # This channel is not available.
2080 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2081 'only_matching': True,
2082 }]
2083
2084 @classmethod
2085 def suitable(cls, url):
2086 # Don't return True if the url can be extracted with other youtube
2087 # extractor, the regex would is too permissive and it would match.
2088 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2089 if any(ie.suitable(url) for ie in other_yt_ies):
2090 return False
2091 else:
2092 return super(YoutubeUserIE, cls).suitable(url)
2093
2094 def _build_template_url(self, url, channel_id):
2095 mobj = re.match(self._VALID_URL, url)
2096 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2097
2098
2099 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2100 IE_DESC = 'YouTube.com live streams'
2101 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2102 IE_NAME = 'youtube:live'
2103
2104 _TESTS = [{
2105 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2106 'info_dict': {
2107 'id': 'a48o2S1cPoo',
2108 'ext': 'mp4',
2109 'title': 'The Young Turks - Live Main Show',
2110 'uploader': 'The Young Turks',
2111 'uploader_id': 'TheYoungTurks',
2112 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2113 'upload_date': '20150715',
2114 'license': 'Standard YouTube License',
2115 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2116 'categories': ['News & Politics'],
2117 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2118 'like_count': int,
2119 'dislike_count': int,
2120 },
2121 'params': {
2122 'skip_download': True,
2123 },
2124 }, {
2125 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2126 'only_matching': True,
2127 }]
2128
2129 def _real_extract(self, url):
2130 mobj = re.match(self._VALID_URL, url)
2131 channel_id = mobj.group('id')
2132 base_url = mobj.group('base_url')
2133 webpage = self._download_webpage(url, channel_id, fatal=False)
2134 if webpage:
2135 page_type = self._og_search_property(
2136 'type', webpage, 'page type', default=None)
2137 video_id = self._html_search_meta(
2138 'videoId', webpage, 'video id', default=None)
2139 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2140 return self.url_result(video_id, YoutubeIE.ie_key())
2141 return self.url_result(base_url)
2142
2143
2144 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2145 IE_DESC = 'YouTube.com user/channel playlists'
2146 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2147 IE_NAME = 'youtube:playlists'
2148
2149 _TESTS = [{
2150 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2151 'playlist_mincount': 4,
2152 'info_dict': {
2153 'id': 'ThirstForScience',
2154 'title': 'Thirst for Science',
2155 },
2156 }, {
2157 # with "Load more" button
2158 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2159 'playlist_mincount': 70,
2160 'info_dict': {
2161 'id': 'igorkle1',
2162 'title': 'Игорь Клейнер',
2163 },
2164 }, {
2165 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2166 'playlist_mincount': 17,
2167 'info_dict': {
2168 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2169 'title': 'Chem Player',
2170 },
2171 }]
2172
2173
2174 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2175 IE_DESC = 'YouTube.com searches'
2176 # there doesn't appear to be a real limit, for example if you search for
2177 # 'python' you get more than 8.000.000 results
2178 _MAX_RESULTS = float('inf')
2179 IE_NAME = 'youtube:search'
2180 _SEARCH_KEY = 'ytsearch'
2181 _EXTRA_QUERY_ARGS = {}
2182 _TESTS = []
2183
2184 def _get_n_results(self, query, n):
2185 """Get a specified number of results for a query"""
2186
2187 videos = []
2188 limit = n
2189
2190 for pagenum in itertools.count(1):
2191 url_query = {
2192 'search_query': query.encode('utf-8'),
2193 'page': pagenum,
2194 'spf': 'navigate',
2195 }
2196 url_query.update(self._EXTRA_QUERY_ARGS)
2197 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2198 data = self._download_json(
2199 result_url, video_id='query "%s"' % query,
2200 note='Downloading page %s' % pagenum,
2201 errnote='Unable to download API page')
2202 html_content = data[1]['body']['content']
2203
2204 if 'class="search-message' in html_content:
2205 raise ExtractorError(
2206 '[youtube] No video results', expected=True)
2207
2208 new_videos = self._ids_to_results(orderedSet(re.findall(
2209 r'href="/watch\?v=(.{11})', html_content)))
2210 videos += new_videos
2211 if not new_videos or len(videos) > limit:
2212 break
2213
2214 if len(videos) > n:
2215 videos = videos[:n]
2216 return self.playlist_result(videos, query)
2217
2218
2219 class YoutubeSearchDateIE(YoutubeSearchIE):
2220 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2221 _SEARCH_KEY = 'ytsearchdate'
2222 IE_DESC = 'YouTube.com searches, newest videos first'
2223 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2224
2225
2226 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2227 IE_DESC = 'YouTube.com search URLs'
2228 IE_NAME = 'youtube:search_url'
2229 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2230 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2231 _TESTS = [{
2232 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2233 'playlist_mincount': 5,
2234 'info_dict': {
2235 'title': 'youtube-dl test video',
2236 }
2237 }, {
2238 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2239 'only_matching': True,
2240 }]
2241
2242 def _real_extract(self, url):
2243 mobj = re.match(self._VALID_URL, url)
2244 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2245 webpage = self._download_webpage(url, query)
2246 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2247
2248
2249 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2250 IE_DESC = 'YouTube.com (multi-season) shows'
2251 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2252 IE_NAME = 'youtube:show'
2253 _TESTS = [{
2254 'url': 'https://www.youtube.com/show/airdisasters',
2255 'playlist_mincount': 5,
2256 'info_dict': {
2257 'id': 'airdisasters',
2258 'title': 'Air Disasters',
2259 }
2260 }]
2261
2262 def _real_extract(self, url):
2263 playlist_id = self._match_id(url)
2264 return super(YoutubeShowIE, self)._real_extract(
2265 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2266
2267
2268 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2269 """
2270 Base class for feed extractors
2271 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2272 """
2273 _LOGIN_REQUIRED = True
2274
2275 @property
2276 def IE_NAME(self):
2277 return 'youtube:%s' % self._FEED_NAME
2278
2279 def _real_initialize(self):
2280 self._login()
2281
2282 def _real_extract(self, url):
2283 page = self._download_webpage(
2284 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2285
2286 # The extraction process is the same as for playlists, but the regex
2287 # for the video ids doesn't contain an index
2288 ids = []
2289 more_widget_html = content_html = page
2290 for page_num in itertools.count(1):
2291 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2292
2293 # 'recommended' feed has infinite 'load more' and each new portion spins
2294 # the same videos in (sometimes) slightly different order, so we'll check
2295 # for unicity and break when portion has no new videos
2296 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2297 if not new_ids:
2298 break
2299
2300 ids.extend(new_ids)
2301
2302 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2303 if not mobj:
2304 break
2305
2306 more = self._download_json(
2307 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2308 'Downloading page #%s' % page_num,
2309 transform_source=uppercase_escape)
2310 content_html = more['content_html']
2311 more_widget_html = more['load_more_widget_html']
2312
2313 return self.playlist_result(
2314 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2315
2316
2317 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2318 IE_NAME = 'youtube:watchlater'
2319 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2320 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2321
2322 _TESTS = [{
2323 'url': 'https://www.youtube.com/playlist?list=WL',
2324 'only_matching': True,
2325 }, {
2326 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2327 'only_matching': True,
2328 }]
2329
2330 def _real_extract(self, url):
2331 video = self._check_download_just_video(url, 'WL')
2332 if video:
2333 return video
2334 return self._extract_playlist('WL')
2335
2336
2337 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2338 IE_NAME = 'youtube:favorites'
2339 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2340 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2341 _LOGIN_REQUIRED = True
2342
2343 def _real_extract(self, url):
2344 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2345 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2346 return self.url_result(playlist_id, 'YoutubePlaylist')
2347
2348
2349 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2350 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2351 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2352 _FEED_NAME = 'recommended'
2353 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2354
2355
2356 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2357 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2358 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2359 _FEED_NAME = 'subscriptions'
2360 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2361
2362
2363 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2364 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2365 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2366 _FEED_NAME = 'history'
2367 _PLAYLIST_TITLE = 'Youtube History'
2368
2369
2370 class YoutubeTruncatedURLIE(InfoExtractor):
2371 IE_NAME = 'youtube:truncated_url'
2372 IE_DESC = False # Do not list
2373 _VALID_URL = r'''(?x)
2374 (?:https?://)?
2375 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2376 (?:watch\?(?:
2377 feature=[a-z_]+|
2378 annotation_id=annotation_[^&]+|
2379 x-yt-cl=[0-9]+|
2380 hl=[^&]*|
2381 t=[0-9]+
2382 )?
2383 |
2384 attribution_link\?a=[^&]+
2385 )
2386 $
2387 '''
2388
2389 _TESTS = [{
2390 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2391 'only_matching': True,
2392 }, {
2393 'url': 'http://www.youtube.com/watch?',
2394 'only_matching': True,
2395 }, {
2396 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2397 'only_matching': True,
2398 }, {
2399 'url': 'https://www.youtube.com/watch?feature=foo',
2400 'only_matching': True,
2401 }, {
2402 'url': 'https://www.youtube.com/watch?hl=en-GB',
2403 'only_matching': True,
2404 }, {
2405 'url': 'https://www.youtube.com/watch?t=2372',
2406 'only_matching': True,
2407 }]
2408
2409 def _real_extract(self, url):
2410 raise ExtractorError(
2411 'Did you forget to quote the URL? Remember that & is a meta '
2412 'character in most shells, so you want to put the URL in quotes, '
2413 'like youtube-dl '
2414 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2415 ' or simply youtube-dl BaW_jenozKc .',
2416 expected=True)
2417
2418
2419 class YoutubeTruncatedIDIE(InfoExtractor):
2420 IE_NAME = 'youtube:truncated_id'
2421 IE_DESC = False # Do not list
2422 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2423
2424 _TESTS = [{
2425 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2426 'only_matching': True,
2427 }]
2428
2429 def _real_extract(self, url):
2430 video_id = self._match_id(url)
2431 raise ExtractorError(
2432 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2433 expected=True)