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 _PASSWORD_CHALLENGE_URL = 'https://accounts.google.com/signin/challenge/sl/password'
57 _NETRC_MACHINE = 'youtube'
58 # If True it will raise an error if no login info is provided
59 _LOGIN_REQUIRED = False
60
61 def _set_language(self):
62 self._set_cookie(
63 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
64 # YouTube sets the expire time to about two months
65 expire_time=time.time() + 2 * 30 * 24 * 3600)
66
67 def _ids_to_results(self, ids):
68 return [
69 self.url_result(vid_id, 'Youtube', video_id=vid_id)
70 for vid_id in ids]
71
72 def _login(self):
73 """
74 Attempt to log in to YouTube.
75 True is returned if successful or skipped.
76 False is returned if login failed.
77
78 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
79 """
80 (username, password) = self._get_login_info()
81 # No authentication to be performed
82 if username is None:
83 if self._LOGIN_REQUIRED:
84 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
85 return True
86
87 login_page = self._download_webpage(
88 self._LOGIN_URL, None,
89 note='Downloading login page',
90 errnote='unable to fetch login page', fatal=False)
91 if login_page is False:
92 return
93
94 login_form = self._hidden_inputs(login_page)
95
96 login_form.update({
97 'checkConnection': 'youtube',
98 'Email': username,
99 'Passwd': password,
100 })
101
102 login_results = self._download_webpage(
103 self._PASSWORD_CHALLENGE_URL, None,
104 note='Logging in', errnote='unable to log in', fatal=False,
105 data=urlencode_postdata(login_form))
106 if login_results is False:
107 return False
108
109 error_msg = self._html_search_regex(
110 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
111 login_results, 'error message', default=None)
112 if error_msg:
113 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
114
115 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
116 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
117
118 # Two-Factor
119 # TODO add SMS and phone call support - these require making a request and then prompting the user
120
121 if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
122 tfa_code = self._get_tfa_info('2-step verification code')
123
124 if not tfa_code:
125 self._downloader.report_warning(
126 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
127 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
128 return False
129
130 tfa_code = remove_start(tfa_code, 'G-')
131
132 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
133
134 tfa_form_strs.update({
135 'Pin': tfa_code,
136 'TrustDevice': 'on',
137 })
138
139 tfa_data = urlencode_postdata(tfa_form_strs)
140
141 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
142 tfa_results = self._download_webpage(
143 tfa_req, None,
144 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
145
146 if tfa_results is False:
147 return False
148
149 if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
150 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
151 return False
152 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
153 self._downloader.report_warning('unable to log in - did the page structure change?')
154 return False
155 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
156 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
157 return False
158
159 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
160 self._downloader.report_warning('unable to log in: bad username or password')
161 return False
162 return True
163
164 def _real_initialize(self):
165 if self._downloader is None:
166 return
167 self._set_language()
168 if not self._login():
169 return
170
171
172 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
173 # Extract entries from page with "Load more" button
174 def _entries(self, page, playlist_id):
175 more_widget_html = content_html = page
176 for page_num in itertools.count(1):
177 for entry in self._process_page(content_html):
178 yield entry
179
180 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
181 if not mobj:
182 break
183
184 more = self._download_json(
185 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
186 'Downloading page #%s' % page_num,
187 transform_source=uppercase_escape)
188 content_html = more['content_html']
189 if not content_html.strip():
190 # Some webpages show a "Load more" button but they don't
191 # have more videos
192 break
193 more_widget_html = more['load_more_widget_html']
194
195
196 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
197 def _process_page(self, content):
198 for video_id, video_title in self.extract_videos_from_page(content):
199 yield self.url_result(video_id, 'Youtube', video_id, video_title)
200
201 def extract_videos_from_page(self, page):
202 ids_in_page = []
203 titles_in_page = []
204 for mobj in re.finditer(self._VIDEO_RE, page):
205 # The link with index 0 is not the first video of the playlist (not sure if still actual)
206 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
207 continue
208 video_id = mobj.group('id')
209 video_title = unescapeHTML(mobj.group('title'))
210 if video_title:
211 video_title = video_title.strip()
212 try:
213 idx = ids_in_page.index(video_id)
214 if video_title and not titles_in_page[idx]:
215 titles_in_page[idx] = video_title
216 except ValueError:
217 ids_in_page.append(video_id)
218 titles_in_page.append(video_title)
219 return zip(ids_in_page, titles_in_page)
220
221
222 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
223 def _process_page(self, content):
224 for playlist_id in orderedSet(re.findall(
225 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
226 content)):
227 yield self.url_result(
228 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
229
230 def _real_extract(self, url):
231 playlist_id = self._match_id(url)
232 webpage = self._download_webpage(url, playlist_id)
233 title = self._og_search_title(webpage, fatal=False)
234 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
235
236
237 class YoutubeIE(YoutubeBaseInfoExtractor):
238 IE_DESC = 'YouTube.com'
239 _VALID_URL = r"""(?x)^
240 (
241 (?:https?://|//) # http(s):// or protocol-independent URL
242 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
243 (?:www\.)?deturl\.com/www\.youtube\.com/|
244 (?:www\.)?pwnyoutube\.com/|
245 (?:www\.)?yourepeat\.com/|
246 tube\.majestyc\.net/|
247 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
248 (?:.*?\#/)? # handle anchor (#/) redirect urls
249 (?: # the various things that can precede the ID:
250 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
251 |(?: # or the v= param in all its forms
252 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
253 (?:\?|\#!?) # the params delimiter ? or # or #!
254 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
255 v=
256 )
257 ))
258 |(?:
259 youtu\.be| # just youtu.be/xxxx
260 vid\.plus| # or vid.plus/xxxx
261 zwearz\.com/watch| # or zwearz.com/watch/xxxx
262 )/
263 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
264 )
265 )? # all until now is optional -> you can pass the naked ID
266 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
267 (?!.*?\blist=) # combined list/video URLs are handled by the playlist IE
268 (?(1).+)? # if we found the ID, everything can follow
269 $"""
270 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
271 _formats = {
272 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
273 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
274 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
275 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
276 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
277 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
278 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
279 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
280 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
281 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
282 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
283 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
284 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
285 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
286 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
287 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
288 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
290
291
292 # 3D videos
293 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
294 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
295 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
296 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
297 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
298 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
299 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
300
301 # Apple HTTP Live Streaming
302 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
303 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
304 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
305 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
306 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
307 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
308 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
309 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
310
311 # DASH mp4 video
312 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
313 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
314 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
315 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
316 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
317 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
318 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
319 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
320 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
321 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
322 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
323
324 # Dash mp4 audio
325 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
326 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
327 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
328 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
329 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
330
331 # Dash webm
332 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
333 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
334 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
335 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
336 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
337 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
339 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
340 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
341 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
342 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
343 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
344 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
345 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
346 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
348 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
350 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
351 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
352 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
354
355 # Dash webm audio
356 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
357 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
358
359 # Dash webm audio with opus inside
360 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
361 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
362 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
363
364 # RTMP (unnamed)
365 '_rtmp': {'protocol': 'rtmp'},
366 }
367 _SUBTITLE_FORMATS = ('ttml', 'vtt')
368
369 IE_NAME = 'youtube'
370 _TESTS = [
371 {
372 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
373 'info_dict': {
374 'id': 'BaW_jenozKc',
375 'ext': 'mp4',
376 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
377 'uploader': 'Philipp Hagemeister',
378 'uploader_id': 'phihag',
379 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
380 'upload_date': '20121002',
381 'license': 'Standard YouTube License',
382 '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 .',
383 'categories': ['Science & Technology'],
384 'tags': ['youtube-dl'],
385 'like_count': int,
386 'dislike_count': int,
387 'start_time': 1,
388 'end_time': 9,
389 }
390 },
391 {
392 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
393 'note': 'Test generic use_cipher_signature video (#897)',
394 'info_dict': {
395 'id': 'UxxajLWwzqY',
396 'ext': 'mp4',
397 'upload_date': '20120506',
398 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
399 'alt_title': 'I Love It (feat. Charli XCX)',
400 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
401 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
402 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
403 'iconic ep', 'iconic', 'love', 'it'],
404 'uploader': 'Icona Pop',
405 'uploader_id': 'IconaPop',
406 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
407 'license': 'Standard YouTube License',
408 'creator': 'Icona Pop',
409 }
410 },
411 {
412 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
413 'note': 'Test VEVO video with age protection (#956)',
414 'info_dict': {
415 'id': '07FYdnEawAQ',
416 'ext': 'mp4',
417 'upload_date': '20130703',
418 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
419 'alt_title': 'Tunnel Vision',
420 'description': 'md5:64249768eec3bc4276236606ea996373',
421 'uploader': 'justintimberlakeVEVO',
422 'uploader_id': 'justintimberlakeVEVO',
423 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
424 'license': 'Standard YouTube License',
425 'creator': 'Justin Timberlake',
426 'age_limit': 18,
427 }
428 },
429 {
430 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
431 'note': 'Embed-only video (#1746)',
432 'info_dict': {
433 'id': 'yZIXLfi8CZQ',
434 'ext': 'mp4',
435 'upload_date': '20120608',
436 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
437 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
438 'uploader': 'SET India',
439 'uploader_id': 'setindia',
440 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
441 'license': 'Standard YouTube License',
442 'age_limit': 18,
443 }
444 },
445 {
446 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
447 'note': 'Use the first video ID in the URL',
448 'info_dict': {
449 'id': 'BaW_jenozKc',
450 'ext': 'mp4',
451 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
452 'uploader': 'Philipp Hagemeister',
453 'uploader_id': 'phihag',
454 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
455 'upload_date': '20121002',
456 'license': 'Standard YouTube License',
457 '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 .',
458 'categories': ['Science & Technology'],
459 'tags': ['youtube-dl'],
460 'like_count': int,
461 'dislike_count': int,
462 },
463 'params': {
464 'skip_download': True,
465 },
466 },
467 {
468 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
469 'note': '256k DASH audio (format 141) via DASH manifest',
470 'info_dict': {
471 'id': 'a9LDPn-MO4I',
472 'ext': 'm4a',
473 'upload_date': '20121002',
474 'uploader_id': '8KVIDEO',
475 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
476 'description': '',
477 'uploader': '8KVIDEO',
478 'license': 'Standard YouTube License',
479 'title': 'UHDTV TEST 8K VIDEO.mp4'
480 },
481 'params': {
482 'youtube_include_dash_manifest': True,
483 'format': '141',
484 },
485 'skip': 'format 141 not served anymore',
486 },
487 # DASH manifest with encrypted signature
488 {
489 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
490 'info_dict': {
491 'id': 'IB3lcPjvWLA',
492 'ext': 'm4a',
493 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
494 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
495 'uploader': 'AfrojackVEVO',
496 'uploader_id': 'AfrojackVEVO',
497 'upload_date': '20131011',
498 'license': 'Standard YouTube License',
499 },
500 'params': {
501 'youtube_include_dash_manifest': True,
502 'format': '141/bestaudio[ext=m4a]',
503 },
504 },
505 # JS player signature function name containing $
506 {
507 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
508 'info_dict': {
509 'id': 'nfWlot6h_JM',
510 'ext': 'm4a',
511 'title': 'Taylor Swift - Shake It Off',
512 'alt_title': 'Shake It Off',
513 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
514 'uploader': 'TaylorSwiftVEVO',
515 'uploader_id': 'TaylorSwiftVEVO',
516 'upload_date': '20140818',
517 'license': 'Standard YouTube License',
518 'creator': 'Taylor Swift',
519 },
520 'params': {
521 'youtube_include_dash_manifest': True,
522 'format': '141/bestaudio[ext=m4a]',
523 },
524 },
525 # Controversy video
526 {
527 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
528 'info_dict': {
529 'id': 'T4XJQO3qol8',
530 'ext': 'mp4',
531 'upload_date': '20100909',
532 'uploader': 'The Amazing Atheist',
533 'uploader_id': 'TheAmazingAtheist',
534 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
535 'license': 'Standard YouTube License',
536 'title': 'Burning Everyone\'s Koran',
537 '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',
538 }
539 },
540 # Normal age-gate video (No vevo, embed allowed)
541 {
542 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
543 'info_dict': {
544 'id': 'HtVdAasjOgU',
545 'ext': 'mp4',
546 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
547 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
548 'uploader': 'The Witcher',
549 'uploader_id': 'WitcherGame',
550 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
551 'upload_date': '20140605',
552 'license': 'Standard YouTube License',
553 'age_limit': 18,
554 },
555 },
556 # Age-gate video with encrypted signature
557 {
558 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
559 'info_dict': {
560 'id': '6kLq3WMV1nU',
561 'ext': 'mp4',
562 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
563 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
564 'uploader': 'LloydVEVO',
565 'uploader_id': 'LloydVEVO',
566 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
567 'upload_date': '20110629',
568 'license': 'Standard YouTube License',
569 'age_limit': 18,
570 },
571 },
572 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
573 {
574 'url': '__2ABJjxzNo',
575 'info_dict': {
576 'id': '__2ABJjxzNo',
577 'ext': 'mp4',
578 'upload_date': '20100430',
579 'uploader_id': 'deadmau5',
580 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
581 'creator': 'deadmau5',
582 'description': 'md5:12c56784b8032162bb936a5f76d55360',
583 'uploader': 'deadmau5',
584 'license': 'Standard YouTube License',
585 'title': 'Deadmau5 - Some Chords (HD)',
586 'alt_title': 'Some Chords',
587 },
588 'expected_warnings': [
589 'DASH manifest missing',
590 ]
591 },
592 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
593 {
594 'url': 'lqQg6PlCWgI',
595 'info_dict': {
596 'id': 'lqQg6PlCWgI',
597 'ext': 'mp4',
598 'upload_date': '20150827',
599 'uploader_id': 'olympic',
600 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
601 'license': 'Standard YouTube License',
602 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
603 'uploader': 'Olympic',
604 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
605 },
606 'params': {
607 'skip_download': 'requires avconv',
608 }
609 },
610 # Non-square pixels
611 {
612 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
613 'info_dict': {
614 'id': '_b-2C3KPAM0',
615 'ext': 'mp4',
616 'stretched_ratio': 16 / 9.,
617 'upload_date': '20110310',
618 'uploader_id': 'AllenMeow',
619 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
620 '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',
621 'uploader': '孫艾倫',
622 'license': 'Standard YouTube License',
623 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
624 },
625 },
626 # url_encoded_fmt_stream_map is empty string
627 {
628 'url': 'qEJwOuvDf7I',
629 'info_dict': {
630 'id': 'qEJwOuvDf7I',
631 'ext': 'webm',
632 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
633 'description': '',
634 'upload_date': '20150404',
635 'uploader_id': 'spbelect',
636 'uploader': 'Наблюдатели Петербурга',
637 },
638 'params': {
639 'skip_download': 'requires avconv',
640 },
641 'skip': 'This live event has ended.',
642 },
643 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
644 {
645 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
646 'info_dict': {
647 'id': 'FIl7x6_3R5Y',
648 'ext': 'mp4',
649 'title': 'md5:7b81415841e02ecd4313668cde88737a',
650 'description': 'md5:116377fd2963b81ec4ce64b542173306',
651 'upload_date': '20150625',
652 'uploader_id': 'dorappi2000',
653 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
654 'uploader': 'dorappi2000',
655 'license': 'Standard YouTube License',
656 'formats': 'mincount:32',
657 },
658 },
659 # DASH manifest with segment_list
660 {
661 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
662 'md5': '8ce563a1d667b599d21064e982ab9e31',
663 'info_dict': {
664 'id': 'CsmdDsKjzN8',
665 'ext': 'mp4',
666 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
667 'uploader': 'Airtek',
668 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
669 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
670 'license': 'Standard YouTube License',
671 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
672 },
673 'params': {
674 'youtube_include_dash_manifest': True,
675 'format': '135', # bestvideo
676 },
677 'skip': 'This live event has ended.',
678 },
679 {
680 # Multifeed videos (multiple cameras), URL is for Main Camera
681 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
682 'info_dict': {
683 'id': 'jqWvoWXjCVs',
684 'title': 'teamPGP: Rocket League Noob Stream',
685 'description': 'md5:dc7872fb300e143831327f1bae3af010',
686 },
687 'playlist': [{
688 'info_dict': {
689 'id': 'jqWvoWXjCVs',
690 'ext': 'mp4',
691 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
692 'description': 'md5:dc7872fb300e143831327f1bae3af010',
693 'upload_date': '20150721',
694 'uploader': 'Beer Games Beer',
695 'uploader_id': 'beergamesbeer',
696 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
697 'license': 'Standard YouTube License',
698 },
699 }, {
700 'info_dict': {
701 'id': '6h8e8xoXJzg',
702 'ext': 'mp4',
703 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
704 'description': 'md5:dc7872fb300e143831327f1bae3af010',
705 'upload_date': '20150721',
706 'uploader': 'Beer Games Beer',
707 'uploader_id': 'beergamesbeer',
708 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
709 'license': 'Standard YouTube License',
710 },
711 }, {
712 'info_dict': {
713 'id': 'PUOgX5z9xZw',
714 'ext': 'mp4',
715 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
716 'description': 'md5:dc7872fb300e143831327f1bae3af010',
717 'upload_date': '20150721',
718 'uploader': 'Beer Games Beer',
719 'uploader_id': 'beergamesbeer',
720 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
721 'license': 'Standard YouTube License',
722 },
723 }, {
724 'info_dict': {
725 'id': 'teuwxikvS5k',
726 'ext': 'mp4',
727 'title': 'teamPGP: Rocket League Noob Stream (zim)',
728 'description': 'md5:dc7872fb300e143831327f1bae3af010',
729 'upload_date': '20150721',
730 'uploader': 'Beer Games Beer',
731 'uploader_id': 'beergamesbeer',
732 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
733 'license': 'Standard YouTube License',
734 },
735 }],
736 'params': {
737 'skip_download': True,
738 },
739 },
740 {
741 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
742 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
743 'info_dict': {
744 'id': 'gVfLd0zydlo',
745 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
746 },
747 'playlist_count': 2,
748 'skip': 'Not multifeed anymore',
749 },
750 {
751 'url': 'http://vid.plus/FlRa-iH7PGw',
752 'only_matching': True,
753 },
754 {
755 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
756 'only_matching': True,
757 },
758 {
759 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
760 # Also tests cut-off URL expansion in video description (see
761 # https://github.com/rg3/youtube-dl/issues/1892,
762 # https://github.com/rg3/youtube-dl/issues/8164)
763 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
764 'info_dict': {
765 'id': 'lsguqyKfVQg',
766 'ext': 'mp4',
767 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
768 'alt_title': 'Dark Walk',
769 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
770 'upload_date': '20151119',
771 'uploader_id': 'IronSoulElf',
772 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
773 'uploader': 'IronSoulElf',
774 'license': 'Standard YouTube License',
775 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
776 },
777 'params': {
778 'skip_download': True,
779 },
780 },
781 {
782 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
783 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
784 'only_matching': True,
785 },
786 {
787 # Video with yt:stretch=17:0
788 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
789 'info_dict': {
790 'id': 'Q39EVAstoRM',
791 'ext': 'mp4',
792 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
793 'description': 'md5:ee18a25c350637c8faff806845bddee9',
794 'upload_date': '20151107',
795 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
796 'uploader': 'CH GAMER DROID',
797 },
798 'params': {
799 'skip_download': True,
800 },
801 'skip': 'This video does not exist.',
802 },
803 {
804 # Video licensed under Creative Commons
805 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
806 'info_dict': {
807 'id': 'M4gD1WSo5mA',
808 'ext': 'mp4',
809 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
810 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
811 'upload_date': '20150127',
812 'uploader_id': 'BerkmanCenter',
813 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
814 'uploader': 'BerkmanCenter',
815 'license': 'Creative Commons Attribution license (reuse allowed)',
816 },
817 'params': {
818 'skip_download': True,
819 },
820 },
821 {
822 # Channel-like uploader_url
823 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
824 'info_dict': {
825 'id': 'eQcmzGIKrzg',
826 'ext': 'mp4',
827 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
828 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
829 'upload_date': '20151119',
830 'uploader': 'Bernie 2016',
831 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
833 'license': 'Creative Commons Attribution license (reuse allowed)',
834 },
835 'params': {
836 'skip_download': True,
837 },
838 },
839 {
840 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
841 'only_matching': True,
842 },
843 {
844 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
845 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
846 'only_matching': True,
847 },
848 {
849 # Rental video preview
850 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
851 'info_dict': {
852 'id': 'uGpuVWrhIzE',
853 'ext': 'mp4',
854 'title': 'Piku - Trailer',
855 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
856 'upload_date': '20150811',
857 'uploader': 'FlixMatrix',
858 'uploader_id': 'FlixMatrixKaravan',
859 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
860 'license': 'Standard YouTube License',
861 },
862 'params': {
863 'skip_download': True,
864 },
865 }
866 ]
867
868 def __init__(self, *args, **kwargs):
869 super(YoutubeIE, self).__init__(*args, **kwargs)
870 self._player_cache = {}
871
872 def report_video_info_webpage_download(self, video_id):
873 """Report attempt to download video info webpage."""
874 self.to_screen('%s: Downloading video info webpage' % video_id)
875
876 def report_information_extraction(self, video_id):
877 """Report attempt to extract video information."""
878 self.to_screen('%s: Extracting video information' % video_id)
879
880 def report_unavailable_format(self, video_id, format):
881 """Report extracted video URL."""
882 self.to_screen('%s: Format %s not available' % (video_id, format))
883
884 def report_rtmp_download(self):
885 """Indicate the download will use the RTMP protocol."""
886 self.to_screen('RTMP download detected')
887
888 def _signature_cache_id(self, example_sig):
889 """ Return a string representation of a signature """
890 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
891
892 def _extract_signature_function(self, video_id, player_url, example_sig):
893 id_m = re.match(
894 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
895 player_url)
896 if not id_m:
897 raise ExtractorError('Cannot identify player %r' % player_url)
898 player_type = id_m.group('ext')
899 player_id = id_m.group('id')
900
901 # Read from filesystem cache
902 func_id = '%s_%s_%s' % (
903 player_type, player_id, self._signature_cache_id(example_sig))
904 assert os.path.basename(func_id) == func_id
905
906 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
907 if cache_spec is not None:
908 return lambda s: ''.join(s[i] for i in cache_spec)
909
910 download_note = (
911 'Downloading player %s' % player_url
912 if self._downloader.params.get('verbose') else
913 'Downloading %s player %s' % (player_type, player_id)
914 )
915 if player_type == 'js':
916 code = self._download_webpage(
917 player_url, video_id,
918 note=download_note,
919 errnote='Download of %s failed' % player_url)
920 res = self._parse_sig_js(code)
921 elif player_type == 'swf':
922 urlh = self._request_webpage(
923 player_url, video_id,
924 note=download_note,
925 errnote='Download of %s failed' % player_url)
926 code = urlh.read()
927 res = self._parse_sig_swf(code)
928 else:
929 assert False, 'Invalid player type %r' % player_type
930
931 test_string = ''.join(map(compat_chr, range(len(example_sig))))
932 cache_res = res(test_string)
933 cache_spec = [ord(c) for c in cache_res]
934
935 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
936 return res
937
938 def _print_sig_code(self, func, example_sig):
939 def gen_sig_code(idxs):
940 def _genslice(start, end, step):
941 starts = '' if start == 0 else str(start)
942 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
943 steps = '' if step == 1 else (':%d' % step)
944 return 's[%s%s%s]' % (starts, ends, steps)
945
946 step = None
947 # Quelch pyflakes warnings - start will be set when step is set
948 start = '(Never used)'
949 for i, prev in zip(idxs[1:], idxs[:-1]):
950 if step is not None:
951 if i - prev == step:
952 continue
953 yield _genslice(start, prev, step)
954 step = None
955 continue
956 if i - prev in [-1, 1]:
957 step = i - prev
958 start = prev
959 continue
960 else:
961 yield 's[%d]' % prev
962 if step is None:
963 yield 's[%d]' % i
964 else:
965 yield _genslice(start, i, step)
966
967 test_string = ''.join(map(compat_chr, range(len(example_sig))))
968 cache_res = func(test_string)
969 cache_spec = [ord(c) for c in cache_res]
970 expr_code = ' + '.join(gen_sig_code(cache_spec))
971 signature_id_tuple = '(%s)' % (
972 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
973 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
974 ' return %s\n') % (signature_id_tuple, expr_code)
975 self.to_screen('Extracted signature function:\n' + code)
976
977 def _parse_sig_js(self, jscode):
978 funcname = self._search_regex(
979 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
980 'Initial JS player signature function name')
981
982 jsi = JSInterpreter(jscode)
983 initial_function = jsi.extract_function(funcname)
984 return lambda s: initial_function([s])
985
986 def _parse_sig_swf(self, file_contents):
987 swfi = SWFInterpreter(file_contents)
988 TARGET_CLASSNAME = 'SignatureDecipher'
989 searched_class = swfi.extract_class(TARGET_CLASSNAME)
990 initial_function = swfi.extract_function(searched_class, 'decipher')
991 return lambda s: initial_function([s])
992
993 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
994 """Turn the encrypted s field into a working signature"""
995
996 if player_url is None:
997 raise ExtractorError('Cannot decrypt signature without player_url')
998
999 if player_url.startswith('//'):
1000 player_url = 'https:' + player_url
1001 try:
1002 player_id = (player_url, self._signature_cache_id(s))
1003 if player_id not in self._player_cache:
1004 func = self._extract_signature_function(
1005 video_id, player_url, s
1006 )
1007 self._player_cache[player_id] = func
1008 func = self._player_cache[player_id]
1009 if self._downloader.params.get('youtube_print_sig_code'):
1010 self._print_sig_code(func, s)
1011 return func(s)
1012 except Exception as e:
1013 tb = traceback.format_exc()
1014 raise ExtractorError(
1015 'Signature extraction failed: ' + tb, cause=e)
1016
1017 def _get_subtitles(self, video_id, webpage):
1018 try:
1019 subs_doc = self._download_xml(
1020 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1021 video_id, note=False)
1022 except ExtractorError as err:
1023 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1024 return {}
1025
1026 sub_lang_list = {}
1027 for track in subs_doc.findall('track'):
1028 lang = track.attrib['lang_code']
1029 if lang in sub_lang_list:
1030 continue
1031 sub_formats = []
1032 for ext in self._SUBTITLE_FORMATS:
1033 params = compat_urllib_parse_urlencode({
1034 'lang': lang,
1035 'v': video_id,
1036 'fmt': ext,
1037 'name': track.attrib['name'].encode('utf-8'),
1038 })
1039 sub_formats.append({
1040 'url': 'https://www.youtube.com/api/timedtext?' + params,
1041 'ext': ext,
1042 })
1043 sub_lang_list[lang] = sub_formats
1044 if not sub_lang_list:
1045 self._downloader.report_warning('video doesn\'t have subtitles')
1046 return {}
1047 return sub_lang_list
1048
1049 def _get_ytplayer_config(self, video_id, webpage):
1050 patterns = (
1051 # User data may contain arbitrary character sequences that may affect
1052 # JSON extraction with regex, e.g. when '};' is contained the second
1053 # regex won't capture the whole JSON. Yet working around by trying more
1054 # concrete regex first keeping in mind proper quoted string handling
1055 # to be implemented in future that will replace this workaround (see
1056 # https://github.com/rg3/youtube-dl/issues/7468,
1057 # https://github.com/rg3/youtube-dl/pull/7599)
1058 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1059 r';ytplayer\.config\s*=\s*({.+?});',
1060 )
1061 config = self._search_regex(
1062 patterns, webpage, 'ytplayer.config', default=None)
1063 if config:
1064 return self._parse_json(
1065 uppercase_escape(config), video_id, fatal=False)
1066
1067 def _get_automatic_captions(self, video_id, webpage):
1068 """We need the webpage for getting the captions url, pass it as an
1069 argument to speed up the process."""
1070 self.to_screen('%s: Looking for automatic captions' % video_id)
1071 player_config = self._get_ytplayer_config(video_id, webpage)
1072 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1073 if not player_config:
1074 self._downloader.report_warning(err_msg)
1075 return {}
1076 try:
1077 args = player_config['args']
1078 caption_url = args.get('ttsurl')
1079 if caption_url:
1080 timestamp = args['timestamp']
1081 # We get the available subtitles
1082 list_params = compat_urllib_parse_urlencode({
1083 'type': 'list',
1084 'tlangs': 1,
1085 'asrs': 1,
1086 })
1087 list_url = caption_url + '&' + list_params
1088 caption_list = self._download_xml(list_url, video_id)
1089 original_lang_node = caption_list.find('track')
1090 if original_lang_node is None:
1091 self._downloader.report_warning('Video doesn\'t have automatic captions')
1092 return {}
1093 original_lang = original_lang_node.attrib['lang_code']
1094 caption_kind = original_lang_node.attrib.get('kind', '')
1095
1096 sub_lang_list = {}
1097 for lang_node in caption_list.findall('target'):
1098 sub_lang = lang_node.attrib['lang_code']
1099 sub_formats = []
1100 for ext in self._SUBTITLE_FORMATS:
1101 params = compat_urllib_parse_urlencode({
1102 'lang': original_lang,
1103 'tlang': sub_lang,
1104 'fmt': ext,
1105 'ts': timestamp,
1106 'kind': caption_kind,
1107 })
1108 sub_formats.append({
1109 'url': caption_url + '&' + params,
1110 'ext': ext,
1111 })
1112 sub_lang_list[sub_lang] = sub_formats
1113 return sub_lang_list
1114
1115 # Some videos don't provide ttsurl but rather caption_tracks and
1116 # caption_translation_languages (e.g. 20LmZk1hakA)
1117 caption_tracks = args['caption_tracks']
1118 caption_translation_languages = args['caption_translation_languages']
1119 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1120 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1121 caption_qs = compat_parse_qs(parsed_caption_url.query)
1122
1123 sub_lang_list = {}
1124 for lang in caption_translation_languages.split(','):
1125 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1126 sub_lang = lang_qs.get('lc', [None])[0]
1127 if not sub_lang:
1128 continue
1129 sub_formats = []
1130 for ext in self._SUBTITLE_FORMATS:
1131 caption_qs.update({
1132 'tlang': [sub_lang],
1133 'fmt': [ext],
1134 })
1135 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1136 query=compat_urllib_parse_urlencode(caption_qs, True)))
1137 sub_formats.append({
1138 'url': sub_url,
1139 'ext': ext,
1140 })
1141 sub_lang_list[sub_lang] = sub_formats
1142 return sub_lang_list
1143 # An extractor error can be raise by the download process if there are
1144 # no automatic captions but there are subtitles
1145 except (KeyError, ExtractorError):
1146 self._downloader.report_warning(err_msg)
1147 return {}
1148
1149 def _mark_watched(self, video_id, video_info):
1150 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1151 if not playback_url:
1152 return
1153 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1154 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1155
1156 # cpn generation algorithm is reverse engineered from base.js.
1157 # In fact it works even with dummy cpn.
1158 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1159 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1160
1161 qs.update({
1162 'ver': ['2'],
1163 'cpn': [cpn],
1164 })
1165 playback_url = compat_urlparse.urlunparse(
1166 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1167
1168 self._download_webpage(
1169 playback_url, video_id, 'Marking watched',
1170 'Unable to mark watched', fatal=False)
1171
1172 @classmethod
1173 def extract_id(cls, url):
1174 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1175 if mobj is None:
1176 raise ExtractorError('Invalid URL: %s' % url)
1177 video_id = mobj.group(2)
1178 return video_id
1179
1180 def _extract_from_m3u8(self, manifest_url, video_id):
1181 url_map = {}
1182
1183 def _get_urls(_manifest):
1184 lines = _manifest.split('\n')
1185 urls = filter(lambda l: l and not l.startswith('#'),
1186 lines)
1187 return urls
1188 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1189 formats_urls = _get_urls(manifest)
1190 for format_url in formats_urls:
1191 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1192 url_map[itag] = format_url
1193 return url_map
1194
1195 def _extract_annotations(self, video_id):
1196 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1197 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1198
1199 def _real_extract(self, url):
1200 url, smuggled_data = unsmuggle_url(url, {})
1201
1202 proto = (
1203 'http' if self._downloader.params.get('prefer_insecure', False)
1204 else 'https')
1205
1206 start_time = None
1207 end_time = None
1208 parsed_url = compat_urllib_parse_urlparse(url)
1209 for component in [parsed_url.fragment, parsed_url.query]:
1210 query = compat_parse_qs(component)
1211 if start_time is None and 't' in query:
1212 start_time = parse_duration(query['t'][0])
1213 if start_time is None and 'start' in query:
1214 start_time = parse_duration(query['start'][0])
1215 if end_time is None and 'end' in query:
1216 end_time = parse_duration(query['end'][0])
1217
1218 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1219 mobj = re.search(self._NEXT_URL_RE, url)
1220 if mobj:
1221 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1222 video_id = self.extract_id(url)
1223
1224 # Get video webpage
1225 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1226 video_webpage = self._download_webpage(url, video_id)
1227
1228 # Attempt to extract SWF player URL
1229 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1230 if mobj is not None:
1231 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1232 else:
1233 player_url = None
1234
1235 dash_mpds = []
1236
1237 def add_dash_mpd(video_info):
1238 dash_mpd = video_info.get('dashmpd')
1239 if dash_mpd and dash_mpd[0] not in dash_mpds:
1240 dash_mpds.append(dash_mpd[0])
1241
1242 # Get video info
1243 embed_webpage = None
1244 is_live = None
1245 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1246 age_gate = True
1247 # We simulate the access to the video from www.youtube.com/v/{video_id}
1248 # this can be viewed without login into Youtube
1249 url = proto + '://www.youtube.com/embed/%s' % video_id
1250 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1251 data = compat_urllib_parse_urlencode({
1252 'video_id': video_id,
1253 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1254 'sts': self._search_regex(
1255 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1256 })
1257 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1258 video_info_webpage = self._download_webpage(
1259 video_info_url, video_id,
1260 note='Refetching age-gated info webpage',
1261 errnote='unable to download video info webpage')
1262 video_info = compat_parse_qs(video_info_webpage)
1263 add_dash_mpd(video_info)
1264 else:
1265 age_gate = False
1266 video_info = None
1267 # Try looking directly into the video webpage
1268 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1269 if ytplayer_config:
1270 args = ytplayer_config['args']
1271 if args.get('url_encoded_fmt_stream_map'):
1272 # Convert to the same format returned by compat_parse_qs
1273 video_info = dict((k, [v]) for k, v in args.items())
1274 add_dash_mpd(video_info)
1275 # Rental video is not rented but preview is available (e.g.
1276 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1277 # https://github.com/rg3/youtube-dl/issues/10532)
1278 if not video_info and args.get('ypc_vid'):
1279 return self.url_result(
1280 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1281 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1282 is_live = True
1283 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1284 # We also try looking in get_video_info since it may contain different dashmpd
1285 # URL that points to a DASH manifest with possibly different itag set (some itags
1286 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1287 # manifest pointed by get_video_info's dashmpd).
1288 # The general idea is to take a union of itags of both DASH manifests (for example
1289 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1290 self.report_video_info_webpage_download(video_id)
1291 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1292 video_info_url = (
1293 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1294 % (proto, video_id, el_type))
1295 video_info_webpage = self._download_webpage(
1296 video_info_url,
1297 video_id, note=False,
1298 errnote='unable to download video info webpage')
1299 get_video_info = compat_parse_qs(video_info_webpage)
1300 if get_video_info.get('use_cipher_signature') != ['True']:
1301 add_dash_mpd(get_video_info)
1302 if not video_info:
1303 video_info = get_video_info
1304 if 'token' in get_video_info:
1305 # Different get_video_info requests may report different results, e.g.
1306 # some may report video unavailability, but some may serve it without
1307 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1308 # the original webpage as well as el=info and el=embedded get_video_info
1309 # requests report video unavailability due to geo restriction while
1310 # el=detailpage succeeds and returns valid data). This is probably
1311 # due to YouTube measures against IP ranges of hosting providers.
1312 # Working around by preferring the first succeeded video_info containing
1313 # the token if no such video_info yet was found.
1314 if 'token' not in video_info:
1315 video_info = get_video_info
1316 break
1317 if 'token' not in video_info:
1318 if 'reason' in video_info:
1319 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1320 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1321 if regions_allowed:
1322 raise ExtractorError('YouTube said: This video is available in %s only' % (
1323 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1324 expected=True)
1325 raise ExtractorError(
1326 'YouTube said: %s' % video_info['reason'][0],
1327 expected=True, video_id=video_id)
1328 else:
1329 raise ExtractorError(
1330 '"token" parameter not in video info for unknown reason',
1331 video_id=video_id)
1332
1333 # title
1334 if 'title' in video_info:
1335 video_title = video_info['title'][0]
1336 else:
1337 self._downloader.report_warning('Unable to extract video title')
1338 video_title = '_'
1339
1340 # description
1341 video_description = get_element_by_id("eow-description", video_webpage)
1342 if video_description:
1343 video_description = re.sub(r'''(?x)
1344 <a\s+
1345 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1346 (?:title|href)="([^"]+)"\s+
1347 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1348 class="[^"]*"[^>]*>
1349 [^<]+\.{3}\s*
1350 </a>
1351 ''', r'\1', video_description)
1352 video_description = clean_html(video_description)
1353 else:
1354 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1355 if fd_mobj:
1356 video_description = unescapeHTML(fd_mobj.group(1))
1357 else:
1358 video_description = ''
1359
1360 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1361 if not self._downloader.params.get('noplaylist'):
1362 entries = []
1363 feed_ids = []
1364 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1365 for feed in multifeed_metadata_list.split(','):
1366 # Unquote should take place before split on comma (,) since textual
1367 # fields may contain comma as well (see
1368 # https://github.com/rg3/youtube-dl/issues/8536)
1369 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1370 entries.append({
1371 '_type': 'url_transparent',
1372 'ie_key': 'Youtube',
1373 'url': smuggle_url(
1374 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1375 {'force_singlefeed': True}),
1376 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1377 })
1378 feed_ids.append(feed_data['id'][0])
1379 self.to_screen(
1380 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1381 % (', '.join(feed_ids), video_id))
1382 return self.playlist_result(entries, video_id, video_title, video_description)
1383 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1384
1385 if 'view_count' in video_info:
1386 view_count = int(video_info['view_count'][0])
1387 else:
1388 view_count = None
1389
1390 # Check for "rental" videos
1391 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1392 raise ExtractorError('"rental" videos not supported')
1393
1394 # Start extracting information
1395 self.report_information_extraction(video_id)
1396
1397 # uploader
1398 if 'author' not in video_info:
1399 raise ExtractorError('Unable to extract uploader name')
1400 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1401
1402 # uploader_id
1403 video_uploader_id = None
1404 video_uploader_url = None
1405 mobj = re.search(
1406 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1407 video_webpage)
1408 if mobj is not None:
1409 video_uploader_id = mobj.group('uploader_id')
1410 video_uploader_url = mobj.group('uploader_url')
1411 else:
1412 self._downloader.report_warning('unable to extract uploader nickname')
1413
1414 # thumbnail image
1415 # We try first to get a high quality image:
1416 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1417 video_webpage, re.DOTALL)
1418 if m_thumb is not None:
1419 video_thumbnail = m_thumb.group(1)
1420 elif 'thumbnail_url' not in video_info:
1421 self._downloader.report_warning('unable to extract video thumbnail')
1422 video_thumbnail = None
1423 else: # don't panic if we can't find it
1424 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1425
1426 # upload date
1427 upload_date = self._html_search_meta(
1428 'datePublished', video_webpage, 'upload date', default=None)
1429 if not upload_date:
1430 upload_date = self._search_regex(
1431 [r'(?s)id="eow-date.*?>(.*?)</span>',
1432 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1433 video_webpage, 'upload date', default=None)
1434 if upload_date:
1435 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1436 upload_date = unified_strdate(upload_date)
1437
1438 video_license = self._html_search_regex(
1439 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1440 video_webpage, 'license', default=None)
1441
1442 m_music = re.search(
1443 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1444 video_webpage)
1445 if m_music:
1446 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1447 video_creator = clean_html(m_music.group('creator'))
1448 else:
1449 video_alt_title = video_creator = None
1450
1451 m_cat_container = self._search_regex(
1452 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1453 video_webpage, 'categories', default=None)
1454 if m_cat_container:
1455 category = self._html_search_regex(
1456 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1457 default=None)
1458 video_categories = None if category is None else [category]
1459 else:
1460 video_categories = None
1461
1462 video_tags = [
1463 unescapeHTML(m.group('content'))
1464 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1465
1466 def _extract_count(count_name):
1467 return str_to_int(self._search_regex(
1468 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1469 % re.escape(count_name),
1470 video_webpage, count_name, default=None))
1471
1472 like_count = _extract_count('like')
1473 dislike_count = _extract_count('dislike')
1474
1475 # subtitles
1476 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1477 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1478
1479 if 'length_seconds' not in video_info:
1480 self._downloader.report_warning('unable to extract video duration')
1481 video_duration = None
1482 else:
1483 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1484
1485 # annotations
1486 video_annotations = None
1487 if self._downloader.params.get('writeannotations', False):
1488 video_annotations = self._extract_annotations(video_id)
1489
1490 def _map_to_format_list(urlmap):
1491 formats = []
1492 for itag, video_real_url in urlmap.items():
1493 dct = {
1494 'format_id': itag,
1495 'url': video_real_url,
1496 'player_url': player_url,
1497 }
1498 if itag in self._formats:
1499 dct.update(self._formats[itag])
1500 formats.append(dct)
1501 return formats
1502
1503 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1504 self.report_rtmp_download()
1505 formats = [{
1506 'format_id': '_rtmp',
1507 'protocol': 'rtmp',
1508 'url': video_info['conn'][0],
1509 'player_url': player_url,
1510 }]
1511 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1512 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1513 if 'rtmpe%3Dyes' in encoded_url_map:
1514 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1515 formats_spec = {}
1516 fmt_list = video_info.get('fmt_list', [''])[0]
1517 if fmt_list:
1518 for fmt in fmt_list.split(','):
1519 spec = fmt.split('/')
1520 if len(spec) > 1:
1521 width_height = spec[1].split('x')
1522 if len(width_height) == 2:
1523 formats_spec[spec[0]] = {
1524 'resolution': spec[1],
1525 'width': int_or_none(width_height[0]),
1526 'height': int_or_none(width_height[1]),
1527 }
1528 formats = []
1529 for url_data_str in encoded_url_map.split(','):
1530 url_data = compat_parse_qs(url_data_str)
1531 if 'itag' not in url_data or 'url' not in url_data:
1532 continue
1533 format_id = url_data['itag'][0]
1534 url = url_data['url'][0]
1535
1536 if 'sig' in url_data:
1537 url += '&signature=' + url_data['sig'][0]
1538 elif 's' in url_data:
1539 encrypted_sig = url_data['s'][0]
1540 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1541
1542 jsplayer_url_json = self._search_regex(
1543 ASSETS_RE,
1544 embed_webpage if age_gate else video_webpage,
1545 'JS player URL (1)', default=None)
1546 if not jsplayer_url_json and not age_gate:
1547 # We need the embed website after all
1548 if embed_webpage is None:
1549 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1550 embed_webpage = self._download_webpage(
1551 embed_url, video_id, 'Downloading embed webpage')
1552 jsplayer_url_json = self._search_regex(
1553 ASSETS_RE, embed_webpage, 'JS player URL')
1554
1555 player_url = json.loads(jsplayer_url_json)
1556 if player_url is None:
1557 player_url_json = self._search_regex(
1558 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1559 video_webpage, 'age gate player URL')
1560 player_url = json.loads(player_url_json)
1561
1562 if self._downloader.params.get('verbose'):
1563 if player_url is None:
1564 player_version = 'unknown'
1565 player_desc = 'unknown'
1566 else:
1567 if player_url.endswith('swf'):
1568 player_version = self._search_regex(
1569 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1570 'flash player', fatal=False)
1571 player_desc = 'flash player %s' % player_version
1572 else:
1573 player_version = self._search_regex(
1574 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1575 player_url,
1576 'html5 player', fatal=False)
1577 player_desc = 'html5 player %s' % player_version
1578
1579 parts_sizes = self._signature_cache_id(encrypted_sig)
1580 self.to_screen('{%s} signature length %s, %s' %
1581 (format_id, parts_sizes, player_desc))
1582
1583 signature = self._decrypt_signature(
1584 encrypted_sig, video_id, player_url, age_gate)
1585 url += '&signature=' + signature
1586 if 'ratebypass' not in url:
1587 url += '&ratebypass=yes'
1588
1589 dct = {
1590 'format_id': format_id,
1591 'url': url,
1592 'player_url': player_url,
1593 }
1594 if format_id in self._formats:
1595 dct.update(self._formats[format_id])
1596 if format_id in formats_spec:
1597 dct.update(formats_spec[format_id])
1598
1599 # Some itags are not included in DASH manifest thus corresponding formats will
1600 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1601 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1602 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1603 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1604
1605 more_fields = {
1606 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1607 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1608 'width': width,
1609 'height': height,
1610 'fps': int_or_none(url_data.get('fps', [None])[0]),
1611 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1612 }
1613 for key, value in more_fields.items():
1614 if value:
1615 dct[key] = value
1616 type_ = url_data.get('type', [None])[0]
1617 if type_:
1618 type_split = type_.split(';')
1619 kind_ext = type_split[0].split('/')
1620 if len(kind_ext) == 2:
1621 kind, _ = kind_ext
1622 dct['ext'] = mimetype2ext(type_split[0])
1623 if kind in ('audio', 'video'):
1624 codecs = None
1625 for mobj in re.finditer(
1626 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1627 if mobj.group('key') == 'codecs':
1628 codecs = mobj.group('val')
1629 break
1630 if codecs:
1631 codecs = codecs.split(',')
1632 if len(codecs) == 2:
1633 acodec, vcodec = codecs[1], codecs[0]
1634 else:
1635 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1636 dct.update({
1637 'acodec': acodec,
1638 'vcodec': vcodec,
1639 })
1640 formats.append(dct)
1641 elif video_info.get('hlsvp'):
1642 manifest_url = video_info['hlsvp'][0]
1643 url_map = self._extract_from_m3u8(manifest_url, video_id)
1644 formats = _map_to_format_list(url_map)
1645 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1646 for a_format in formats:
1647 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1648 else:
1649 unavailable_message = self._html_search_regex(
1650 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1651 video_webpage, 'unavailable message', default=None)
1652 if unavailable_message:
1653 raise ExtractorError(unavailable_message, expected=True)
1654 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1655
1656 # Look for the DASH manifest
1657 if self._downloader.params.get('youtube_include_dash_manifest', True):
1658 dash_mpd_fatal = True
1659 for mpd_url in dash_mpds:
1660 dash_formats = {}
1661 try:
1662 def decrypt_sig(mobj):
1663 s = mobj.group(1)
1664 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1665 return '/signature/%s' % dec_s
1666
1667 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1668
1669 for df in self._extract_mpd_formats(
1670 mpd_url, video_id, fatal=dash_mpd_fatal,
1671 formats_dict=self._formats):
1672 # Do not overwrite DASH format found in some previous DASH manifest
1673 if df['format_id'] not in dash_formats:
1674 dash_formats[df['format_id']] = df
1675 # Additional DASH manifests may end up in HTTP Error 403 therefore
1676 # allow them to fail without bug report message if we already have
1677 # some DASH manifest succeeded. This is temporary workaround to reduce
1678 # burst of bug reports until we figure out the reason and whether it
1679 # can be fixed at all.
1680 dash_mpd_fatal = False
1681 except (ExtractorError, KeyError) as e:
1682 self.report_warning(
1683 'Skipping DASH manifest: %r' % e, video_id)
1684 if dash_formats:
1685 # Remove the formats we found through non-DASH, they
1686 # contain less info and it can be wrong, because we use
1687 # fixed values (for example the resolution). See
1688 # https://github.com/rg3/youtube-dl/issues/5774 for an
1689 # example.
1690 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1691 formats.extend(dash_formats.values())
1692
1693 # Check for malformed aspect ratio
1694 stretched_m = re.search(
1695 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1696 video_webpage)
1697 if stretched_m:
1698 w = float(stretched_m.group('w'))
1699 h = float(stretched_m.group('h'))
1700 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1701 # We will only process correct ratios.
1702 if w > 0 and h > 0:
1703 ratio = w / h
1704 for f in formats:
1705 if f.get('vcodec') != 'none':
1706 f['stretched_ratio'] = ratio
1707
1708 self._sort_formats(formats)
1709
1710 self.mark_watched(video_id, video_info)
1711
1712 return {
1713 'id': video_id,
1714 'uploader': video_uploader,
1715 'uploader_id': video_uploader_id,
1716 'uploader_url': video_uploader_url,
1717 'upload_date': upload_date,
1718 'license': video_license,
1719 'creator': video_creator,
1720 'title': video_title,
1721 'alt_title': video_alt_title,
1722 'thumbnail': video_thumbnail,
1723 'description': video_description,
1724 'categories': video_categories,
1725 'tags': video_tags,
1726 'subtitles': video_subtitles,
1727 'automatic_captions': automatic_captions,
1728 'duration': video_duration,
1729 'age_limit': 18 if age_gate else 0,
1730 'annotations': video_annotations,
1731 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1732 'view_count': view_count,
1733 'like_count': like_count,
1734 'dislike_count': dislike_count,
1735 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1736 'formats': formats,
1737 'is_live': is_live,
1738 'start_time': start_time,
1739 'end_time': end_time,
1740 }
1741
1742
1743 class YoutubeSharedVideoIE(InfoExtractor):
1744 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1745 IE_NAME = 'youtube:shared'
1746
1747 _TEST = {
1748 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1749 'info_dict': {
1750 'id': 'uPDB5I9wfp8',
1751 'ext': 'webm',
1752 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1753 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1754 'upload_date': '20160219',
1755 'uploader': 'Pocoyo - Português (BR)',
1756 'uploader_id': 'PocoyoBrazil',
1757 },
1758 'add_ie': ['Youtube'],
1759 'params': {
1760 # There are already too many Youtube downloads
1761 'skip_download': True,
1762 },
1763 }
1764
1765 def _real_extract(self, url):
1766 video_id = self._match_id(url)
1767
1768 webpage = self._download_webpage(url, video_id)
1769
1770 real_video_id = self._html_search_meta(
1771 'videoId', webpage, 'YouTube video id', fatal=True)
1772
1773 return self.url_result(real_video_id, YoutubeIE.ie_key())
1774
1775
1776 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1777 IE_DESC = 'YouTube.com playlists'
1778 _VALID_URL = r"""(?x)(?:
1779 (?:https?://)?
1780 (?:\w+\.)?
1781 (?:
1782 youtube\.com/
1783 (?:
1784 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1785 \? (?:.*?[&;])*? (?:p|a|list)=
1786 | p/
1787 )|
1788 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
1789 )
1790 (
1791 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1792 # Top tracks, they can also include dots
1793 |(?:MC)[\w\.]*
1794 )
1795 .*
1796 |
1797 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1798 )"""
1799 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1800 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1801 IE_NAME = 'youtube:playlist'
1802 _TESTS = [{
1803 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1804 'info_dict': {
1805 'title': 'ytdl test PL',
1806 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1807 },
1808 'playlist_count': 3,
1809 }, {
1810 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1811 'info_dict': {
1812 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1813 'title': 'YDL_Empty_List',
1814 },
1815 'playlist_count': 0,
1816 }, {
1817 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1818 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1819 'info_dict': {
1820 'title': '29C3: Not my department',
1821 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1822 },
1823 'playlist_count': 95,
1824 }, {
1825 'note': 'issue #673',
1826 'url': 'PLBB231211A4F62143',
1827 'info_dict': {
1828 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1829 'id': 'PLBB231211A4F62143',
1830 },
1831 'playlist_mincount': 26,
1832 }, {
1833 'note': 'Large playlist',
1834 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1835 'info_dict': {
1836 'title': 'Uploads from Cauchemar',
1837 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1838 },
1839 'playlist_mincount': 799,
1840 }, {
1841 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1842 'info_dict': {
1843 'title': 'YDL_safe_search',
1844 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1845 },
1846 'playlist_count': 2,
1847 }, {
1848 'note': 'embedded',
1849 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1850 'playlist_count': 4,
1851 'info_dict': {
1852 'title': 'JODA15',
1853 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1854 }
1855 }, {
1856 'note': 'Embedded SWF player',
1857 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1858 'playlist_count': 4,
1859 'info_dict': {
1860 'title': 'JODA7',
1861 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1862 }
1863 }, {
1864 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1865 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1866 'info_dict': {
1867 'title': 'Uploads from Interstellar Movie',
1868 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1869 },
1870 'playlist_mincout': 21,
1871 }, {
1872 # Playlist URL that does not actually serve a playlist
1873 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
1874 'info_dict': {
1875 'id': 'FqZTN594JQw',
1876 'ext': 'webm',
1877 'title': "Smiley's People 01 detective, Adventure Series, Action",
1878 'uploader': 'STREEM',
1879 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
1880 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
1881 'upload_date': '20150526',
1882 'license': 'Standard YouTube License',
1883 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
1884 'categories': ['People & Blogs'],
1885 'tags': list,
1886 'like_count': int,
1887 'dislike_count': int,
1888 },
1889 'params': {
1890 'skip_download': True,
1891 },
1892 'add_ie': [YoutubeIE.ie_key()],
1893 }, {
1894 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
1895 'only_matching': True,
1896 }]
1897
1898 def _real_initialize(self):
1899 self._login()
1900
1901 def _extract_mix(self, playlist_id):
1902 # The mixes are generated from a single video
1903 # the id of the playlist is just 'RD' + video_id
1904 ids = []
1905 last_id = playlist_id[-11:]
1906 for n in itertools.count(1):
1907 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1908 webpage = self._download_webpage(
1909 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1910 new_ids = orderedSet(re.findall(
1911 r'''(?xs)data-video-username=".*?".*?
1912 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1913 webpage))
1914 # Fetch new pages until all the videos are repeated, it seems that
1915 # there are always 51 unique videos.
1916 new_ids = [_id for _id in new_ids if _id not in ids]
1917 if not new_ids:
1918 break
1919 ids.extend(new_ids)
1920 last_id = ids[-1]
1921
1922 url_results = self._ids_to_results(ids)
1923
1924 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1925 title_span = (
1926 search_title('playlist-title') or
1927 search_title('title long-title') or
1928 search_title('title'))
1929 title = clean_html(title_span)
1930
1931 return self.playlist_result(url_results, playlist_id, title)
1932
1933 def _extract_playlist(self, playlist_id):
1934 url = self._TEMPLATE_URL % playlist_id
1935 page = self._download_webpage(url, playlist_id)
1936
1937 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1938 match = match.strip()
1939 # Check if the playlist exists or is private
1940 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1941 raise ExtractorError(
1942 'The playlist doesn\'t exist or is private, use --username or '
1943 '--netrc to access it.',
1944 expected=True)
1945 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1946 raise ExtractorError(
1947 'Invalid parameters. Maybe URL is incorrect.',
1948 expected=True)
1949 elif re.match(r'[^<]*Choose your language[^<]*', match):
1950 continue
1951 else:
1952 self.report_warning('Youtube gives an alert message: ' + match)
1953
1954 playlist_title = self._html_search_regex(
1955 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1956 page, 'title', default=None)
1957
1958 has_videos = True
1959
1960 if not playlist_title:
1961 try:
1962 # Some playlist URLs don't actually serve a playlist (e.g.
1963 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
1964 next(self._entries(page, playlist_id))
1965 except StopIteration:
1966 has_videos = False
1967
1968 return has_videos, self.playlist_result(
1969 self._entries(page, playlist_id), playlist_id, playlist_title)
1970
1971 def _check_download_just_video(self, url, playlist_id):
1972 # Check if it's a video-specific URL
1973 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1974 if 'v' in query_dict:
1975 video_id = query_dict['v'][0]
1976 if self._downloader.params.get('noplaylist'):
1977 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1978 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
1979 else:
1980 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1981 return video_id, None
1982 return None, None
1983
1984 def _real_extract(self, url):
1985 # Extract playlist id
1986 mobj = re.match(self._VALID_URL, url)
1987 if mobj is None:
1988 raise ExtractorError('Invalid URL: %s' % url)
1989 playlist_id = mobj.group(1) or mobj.group(2)
1990
1991 video_id, video = self._check_download_just_video(url, playlist_id)
1992 if video:
1993 return video
1994
1995 if playlist_id.startswith(('RD', 'UL', 'PU')):
1996 # Mixes require a custom extraction process
1997 return self._extract_mix(playlist_id)
1998
1999 has_videos, playlist = self._extract_playlist(playlist_id)
2000 if has_videos or not video_id:
2001 return playlist
2002
2003 # Some playlist URLs don't actually serve a playlist (see
2004 # https://github.com/rg3/youtube-dl/issues/10537).
2005 # Fallback to plain video extraction if there is a video id
2006 # along with playlist id.
2007 return self.url_result(video_id, 'Youtube', video_id=video_id)
2008
2009
2010 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2011 IE_DESC = 'YouTube.com channels'
2012 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2013 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2014 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2015 IE_NAME = 'youtube:channel'
2016 _TESTS = [{
2017 'note': 'paginated channel',
2018 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2019 'playlist_mincount': 91,
2020 'info_dict': {
2021 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2022 'title': 'Uploads from lex will',
2023 }
2024 }, {
2025 'note': 'Age restricted channel',
2026 # from https://www.youtube.com/user/DeusExOfficial
2027 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2028 'playlist_mincount': 64,
2029 'info_dict': {
2030 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2031 'title': 'Uploads from Deus Ex',
2032 },
2033 }]
2034
2035 @classmethod
2036 def suitable(cls, url):
2037 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2038 else super(YoutubeChannelIE, cls).suitable(url))
2039
2040 def _build_template_url(self, url, channel_id):
2041 return self._TEMPLATE_URL % channel_id
2042
2043 def _real_extract(self, url):
2044 channel_id = self._match_id(url)
2045
2046 url = self._build_template_url(url, channel_id)
2047
2048 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2049 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2050 # otherwise fallback on channel by page extraction
2051 channel_page = self._download_webpage(
2052 url + '?view=57', channel_id,
2053 'Downloading channel page', fatal=False)
2054 if channel_page is False:
2055 channel_playlist_id = False
2056 else:
2057 channel_playlist_id = self._html_search_meta(
2058 'channelId', channel_page, 'channel id', default=None)
2059 if not channel_playlist_id:
2060 channel_url = self._html_search_meta(
2061 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2062 channel_page, 'channel url', default=None)
2063 if channel_url:
2064 channel_playlist_id = self._search_regex(
2065 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2066 channel_url, 'channel id', default=None)
2067 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2068 playlist_id = 'UU' + channel_playlist_id[2:]
2069 return self.url_result(
2070 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2071
2072 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2073 autogenerated = re.search(r'''(?x)
2074 class="[^"]*?(?:
2075 channel-header-autogenerated-label|
2076 yt-channel-title-autogenerated
2077 )[^"]*"''', channel_page) is not None
2078
2079 if autogenerated:
2080 # The videos are contained in a single page
2081 # the ajax pages can't be used, they are empty
2082 entries = [
2083 self.url_result(
2084 video_id, 'Youtube', video_id=video_id,
2085 video_title=video_title)
2086 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2087 return self.playlist_result(entries, channel_id)
2088
2089 try:
2090 next(self._entries(channel_page, channel_id))
2091 except StopIteration:
2092 alert_message = self._html_search_regex(
2093 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2094 channel_page, 'alert', default=None, group='alert')
2095 if alert_message:
2096 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2097
2098 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2099
2100
2101 class YoutubeUserIE(YoutubeChannelIE):
2102 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2103 _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_-]+)'
2104 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2105 IE_NAME = 'youtube:user'
2106
2107 _TESTS = [{
2108 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2109 'playlist_mincount': 320,
2110 'info_dict': {
2111 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2112 'title': 'Uploads from The Linux Foundation',
2113 }
2114 }, {
2115 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2116 # but not https://www.youtube.com/user/12minuteathlete/videos
2117 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2118 'playlist_mincount': 249,
2119 'info_dict': {
2120 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2121 'title': 'Uploads from 12 Minute Athlete',
2122 }
2123 }, {
2124 'url': 'ytuser:phihag',
2125 'only_matching': True,
2126 }, {
2127 'url': 'https://www.youtube.com/c/gametrailers',
2128 'only_matching': True,
2129 }, {
2130 'url': 'https://www.youtube.com/gametrailers',
2131 'only_matching': True,
2132 }, {
2133 # This channel is not available.
2134 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2135 'only_matching': True,
2136 }]
2137
2138 @classmethod
2139 def suitable(cls, url):
2140 # Don't return True if the url can be extracted with other youtube
2141 # extractor, the regex would is too permissive and it would match.
2142 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2143 if any(ie.suitable(url) for ie in other_yt_ies):
2144 return False
2145 else:
2146 return super(YoutubeUserIE, cls).suitable(url)
2147
2148 def _build_template_url(self, url, channel_id):
2149 mobj = re.match(self._VALID_URL, url)
2150 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2151
2152
2153 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2154 IE_DESC = 'YouTube.com live streams'
2155 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2156 IE_NAME = 'youtube:live'
2157
2158 _TESTS = [{
2159 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2160 'info_dict': {
2161 'id': 'a48o2S1cPoo',
2162 'ext': 'mp4',
2163 'title': 'The Young Turks - Live Main Show',
2164 'uploader': 'The Young Turks',
2165 'uploader_id': 'TheYoungTurks',
2166 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2167 'upload_date': '20150715',
2168 'license': 'Standard YouTube License',
2169 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2170 'categories': ['News & Politics'],
2171 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2172 'like_count': int,
2173 'dislike_count': int,
2174 },
2175 'params': {
2176 'skip_download': True,
2177 },
2178 }, {
2179 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2180 'only_matching': True,
2181 }]
2182
2183 def _real_extract(self, url):
2184 mobj = re.match(self._VALID_URL, url)
2185 channel_id = mobj.group('id')
2186 base_url = mobj.group('base_url')
2187 webpage = self._download_webpage(url, channel_id, fatal=False)
2188 if webpage:
2189 page_type = self._og_search_property(
2190 'type', webpage, 'page type', default=None)
2191 video_id = self._html_search_meta(
2192 'videoId', webpage, 'video id', default=None)
2193 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2194 return self.url_result(video_id, YoutubeIE.ie_key())
2195 return self.url_result(base_url)
2196
2197
2198 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2199 IE_DESC = 'YouTube.com user/channel playlists'
2200 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2201 IE_NAME = 'youtube:playlists'
2202
2203 _TESTS = [{
2204 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2205 'playlist_mincount': 4,
2206 'info_dict': {
2207 'id': 'ThirstForScience',
2208 'title': 'Thirst for Science',
2209 },
2210 }, {
2211 # with "Load more" button
2212 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2213 'playlist_mincount': 70,
2214 'info_dict': {
2215 'id': 'igorkle1',
2216 'title': 'Игорь Клейнер',
2217 },
2218 }, {
2219 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2220 'playlist_mincount': 17,
2221 'info_dict': {
2222 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2223 'title': 'Chem Player',
2224 },
2225 }]
2226
2227
2228 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2229 IE_DESC = 'YouTube.com searches'
2230 # there doesn't appear to be a real limit, for example if you search for
2231 # 'python' you get more than 8.000.000 results
2232 _MAX_RESULTS = float('inf')
2233 IE_NAME = 'youtube:search'
2234 _SEARCH_KEY = 'ytsearch'
2235 _EXTRA_QUERY_ARGS = {}
2236 _TESTS = []
2237
2238 def _get_n_results(self, query, n):
2239 """Get a specified number of results for a query"""
2240
2241 videos = []
2242 limit = n
2243
2244 for pagenum in itertools.count(1):
2245 url_query = {
2246 'search_query': query.encode('utf-8'),
2247 'page': pagenum,
2248 'spf': 'navigate',
2249 }
2250 url_query.update(self._EXTRA_QUERY_ARGS)
2251 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2252 data = self._download_json(
2253 result_url, video_id='query "%s"' % query,
2254 note='Downloading page %s' % pagenum,
2255 errnote='Unable to download API page')
2256 html_content = data[1]['body']['content']
2257
2258 if 'class="search-message' in html_content:
2259 raise ExtractorError(
2260 '[youtube] No video results', expected=True)
2261
2262 new_videos = self._ids_to_results(orderedSet(re.findall(
2263 r'href="/watch\?v=(.{11})', html_content)))
2264 videos += new_videos
2265 if not new_videos or len(videos) > limit:
2266 break
2267
2268 if len(videos) > n:
2269 videos = videos[:n]
2270 return self.playlist_result(videos, query)
2271
2272
2273 class YoutubeSearchDateIE(YoutubeSearchIE):
2274 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2275 _SEARCH_KEY = 'ytsearchdate'
2276 IE_DESC = 'YouTube.com searches, newest videos first'
2277 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2278
2279
2280 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2281 IE_DESC = 'YouTube.com search URLs'
2282 IE_NAME = 'youtube:search_url'
2283 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2284 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2285 _TESTS = [{
2286 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2287 'playlist_mincount': 5,
2288 'info_dict': {
2289 'title': 'youtube-dl test video',
2290 }
2291 }, {
2292 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2293 'only_matching': True,
2294 }]
2295
2296 def _real_extract(self, url):
2297 mobj = re.match(self._VALID_URL, url)
2298 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2299 webpage = self._download_webpage(url, query)
2300 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2301
2302
2303 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2304 IE_DESC = 'YouTube.com (multi-season) shows'
2305 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2306 IE_NAME = 'youtube:show'
2307 _TESTS = [{
2308 'url': 'https://www.youtube.com/show/airdisasters',
2309 'playlist_mincount': 5,
2310 'info_dict': {
2311 'id': 'airdisasters',
2312 'title': 'Air Disasters',
2313 }
2314 }]
2315
2316 def _real_extract(self, url):
2317 playlist_id = self._match_id(url)
2318 return super(YoutubeShowIE, self)._real_extract(
2319 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2320
2321
2322 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2323 """
2324 Base class for feed extractors
2325 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2326 """
2327 _LOGIN_REQUIRED = True
2328
2329 @property
2330 def IE_NAME(self):
2331 return 'youtube:%s' % self._FEED_NAME
2332
2333 def _real_initialize(self):
2334 self._login()
2335
2336 def _real_extract(self, url):
2337 page = self._download_webpage(
2338 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2339
2340 # The extraction process is the same as for playlists, but the regex
2341 # for the video ids doesn't contain an index
2342 ids = []
2343 more_widget_html = content_html = page
2344 for page_num in itertools.count(1):
2345 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2346
2347 # 'recommended' feed has infinite 'load more' and each new portion spins
2348 # the same videos in (sometimes) slightly different order, so we'll check
2349 # for unicity and break when portion has no new videos
2350 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2351 if not new_ids:
2352 break
2353
2354 ids.extend(new_ids)
2355
2356 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2357 if not mobj:
2358 break
2359
2360 more = self._download_json(
2361 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2362 'Downloading page #%s' % page_num,
2363 transform_source=uppercase_escape)
2364 content_html = more['content_html']
2365 more_widget_html = more['load_more_widget_html']
2366
2367 return self.playlist_result(
2368 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2369
2370
2371 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2372 IE_NAME = 'youtube:watchlater'
2373 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2374 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2375
2376 _TESTS = [{
2377 'url': 'https://www.youtube.com/playlist?list=WL',
2378 'only_matching': True,
2379 }, {
2380 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2381 'only_matching': True,
2382 }]
2383
2384 def _real_extract(self, url):
2385 _, video = self._check_download_just_video(url, 'WL')
2386 if video:
2387 return video
2388 _, playlist = self._extract_playlist('WL')
2389 return playlist
2390
2391
2392 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2393 IE_NAME = 'youtube:favorites'
2394 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2395 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2396 _LOGIN_REQUIRED = True
2397
2398 def _real_extract(self, url):
2399 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2400 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2401 return self.url_result(playlist_id, 'YoutubePlaylist')
2402
2403
2404 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2405 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2406 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2407 _FEED_NAME = 'recommended'
2408 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2409
2410
2411 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2412 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2413 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2414 _FEED_NAME = 'subscriptions'
2415 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2416
2417
2418 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2419 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2420 _VALID_URL = r'https?://www\.youtube\.com/feed/history|:ythistory'
2421 _FEED_NAME = 'history'
2422 _PLAYLIST_TITLE = 'Youtube History'
2423
2424
2425 class YoutubeTruncatedURLIE(InfoExtractor):
2426 IE_NAME = 'youtube:truncated_url'
2427 IE_DESC = False # Do not list
2428 _VALID_URL = r'''(?x)
2429 (?:https?://)?
2430 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2431 (?:watch\?(?:
2432 feature=[a-z_]+|
2433 annotation_id=annotation_[^&]+|
2434 x-yt-cl=[0-9]+|
2435 hl=[^&]*|
2436 t=[0-9]+
2437 )?
2438 |
2439 attribution_link\?a=[^&]+
2440 )
2441 $
2442 '''
2443
2444 _TESTS = [{
2445 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2446 'only_matching': True,
2447 }, {
2448 'url': 'http://www.youtube.com/watch?',
2449 'only_matching': True,
2450 }, {
2451 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2452 'only_matching': True,
2453 }, {
2454 'url': 'https://www.youtube.com/watch?feature=foo',
2455 'only_matching': True,
2456 }, {
2457 'url': 'https://www.youtube.com/watch?hl=en-GB',
2458 'only_matching': True,
2459 }, {
2460 'url': 'https://www.youtube.com/watch?t=2372',
2461 'only_matching': True,
2462 }]
2463
2464 def _real_extract(self, url):
2465 raise ExtractorError(
2466 'Did you forget to quote the URL? Remember that & is a meta '
2467 'character in most shells, so you want to put the URL in quotes, '
2468 'like youtube-dl '
2469 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2470 ' or simply youtube-dl BaW_jenozKc .',
2471 expected=True)
2472
2473
2474 class YoutubeTruncatedIDIE(InfoExtractor):
2475 IE_NAME = 'youtube:truncated_id'
2476 IE_DESC = False # Do not list
2477 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2478
2479 _TESTS = [{
2480 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2481 'only_matching': True,
2482 }]
2483
2484 def _real_extract(self, url):
2485 video_id = self._match_id(url)
2486 raise ExtractorError(
2487 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2488 expected=True)
|
