]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
Imported Upstream version 2016.06.25
[youtubedl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_parse_qs,
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 compat_str,
26 )
27 from ..utils import (
28 clean_html,
29 error_to_compat_str,
30 ExtractorError,
31 float_or_none,
32 get_element_by_attribute,
33 get_element_by_id,
34 int_or_none,
35 mimetype2ext,
36 orderedSet,
37 parse_duration,
38 remove_quotes,
39 remove_start,
40 sanitized_Request,
41 smuggle_url,
42 str_to_int,
43 unescapeHTML,
44 unified_strdate,
45 unsmuggle_url,
46 uppercase_escape,
47 urlencode_postdata,
48 ISO3166Utils,
49 )
50
51
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
56 _NETRC_MACHINE = 'youtube'
57 # If True it will raise an error if no login info is provided
58 _LOGIN_REQUIRED = False
59
60 def _set_language(self):
61 self._set_cookie(
62 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
63 # YouTube sets the expire time to about two months
64 expire_time=time.time() + 2 * 30 * 24 * 3600)
65
66 def _ids_to_results(self, ids):
67 return [
68 self.url_result(vid_id, 'Youtube', video_id=vid_id)
69 for vid_id in ids]
70
71 def _login(self):
72 """
73 Attempt to log in to YouTube.
74 True is returned if successful or skipped.
75 False is returned if login failed.
76
77 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
78 """
79 (username, password) = self._get_login_info()
80 # No authentication to be performed
81 if username is None:
82 if self._LOGIN_REQUIRED:
83 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
84 return True
85
86 login_page = self._download_webpage(
87 self._LOGIN_URL, None,
88 note='Downloading login page',
89 errnote='unable to fetch login page', fatal=False)
90 if login_page is False:
91 return
92
93 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
94 login_page, 'Login GALX parameter')
95
96 # Log in
97 login_form_strs = {
98 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
99 'Email': username,
100 'GALX': galx,
101 'Passwd': password,
102
103 'PersistentCookie': 'yes',
104 '_utf8': '霱',
105 'bgresponse': 'js_disabled',
106 'checkConnection': '',
107 'checkedDomains': 'youtube',
108 'dnConn': '',
109 'pstMsg': '0',
110 'rmShown': '1',
111 'secTok': '',
112 'signIn': 'Sign in',
113 'timeStmp': '',
114 'service': 'youtube',
115 'uilel': '3',
116 'hl': 'en_US',
117 }
118
119 login_data = urlencode_postdata(login_form_strs)
120
121 req = sanitized_Request(self._LOGIN_URL, login_data)
122 login_results = self._download_webpage(
123 req, None,
124 note='Logging in', errnote='unable to log in', fatal=False)
125 if login_results is False:
126 return False
127
128 error_msg = self._html_search_regex(
129 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
130 login_results, 'error message', default=None)
131 if error_msg:
132 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
133
134 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
135 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
136
137 # Two-Factor
138 # TODO add SMS and phone call support - these require making a request and then prompting the user
139
140 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
141 tfa_code = self._get_tfa_info('2-step verification code')
142
143 if not tfa_code:
144 self._downloader.report_warning(
145 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
146 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
147 return False
148
149 tfa_code = remove_start(tfa_code, 'G-')
150
151 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
152
153 tfa_form_strs.update({
154 'Pin': tfa_code,
155 'TrustDevice': 'on',
156 })
157
158 tfa_data = urlencode_postdata(tfa_form_strs)
159
160 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
161 tfa_results = self._download_webpage(
162 tfa_req, None,
163 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
164
165 if tfa_results is False:
166 return False
167
168 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
169 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
170 return False
171 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
172 self._downloader.report_warning('unable to log in - did the page structure change?')
173 return False
174 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
175 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
176 return False
177
178 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
179 self._downloader.report_warning('unable to log in: bad username or password')
180 return False
181 return True
182
183 def _real_initialize(self):
184 if self._downloader is None:
185 return
186 self._set_language()
187 if not self._login():
188 return
189
190
191 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
192 # Extract entries from page with "Load more" button
193 def _entries(self, page, playlist_id):
194 more_widget_html = content_html = page
195 for page_num in itertools.count(1):
196 for entry in self._process_page(content_html):
197 yield entry
198
199 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
200 if not mobj:
201 break
202
203 more = self._download_json(
204 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
205 'Downloading page #%s' % page_num,
206 transform_source=uppercase_escape)
207 content_html = more['content_html']
208 if not content_html.strip():
209 # Some webpages show a "Load more" button but they don't
210 # have more videos
211 break
212 more_widget_html = more['load_more_widget_html']
213
214
215 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
216 def _process_page(self, content):
217 for video_id, video_title in self.extract_videos_from_page(content):
218 yield self.url_result(video_id, 'Youtube', video_id, video_title)
219
220 def extract_videos_from_page(self, page):
221 ids_in_page = []
222 titles_in_page = []
223 for mobj in re.finditer(self._VIDEO_RE, page):
224 # The link with index 0 is not the first video of the playlist (not sure if still actual)
225 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
226 continue
227 video_id = mobj.group('id')
228 video_title = unescapeHTML(mobj.group('title'))
229 if video_title:
230 video_title = video_title.strip()
231 try:
232 idx = ids_in_page.index(video_id)
233 if video_title and not titles_in_page[idx]:
234 titles_in_page[idx] = video_title
235 except ValueError:
236 ids_in_page.append(video_id)
237 titles_in_page.append(video_title)
238 return zip(ids_in_page, titles_in_page)
239
240
241 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
242 def _process_page(self, content):
243 for playlist_id in orderedSet(re.findall(
244 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
245 content)):
246 yield self.url_result(
247 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
248
249 def _real_extract(self, url):
250 playlist_id = self._match_id(url)
251 webpage = self._download_webpage(url, playlist_id)
252 title = self._og_search_title(webpage, fatal=False)
253 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
254
255
256 class YoutubeIE(YoutubeBaseInfoExtractor):
257 IE_DESC = 'YouTube.com'
258 _VALID_URL = r"""(?x)^
259 (
260 (?:https?://|//) # http(s):// or protocol-independent URL
261 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
262 (?:www\.)?deturl\.com/www\.youtube\.com/|
263 (?:www\.)?pwnyoutube\.com/|
264 (?:www\.)?yourepeat\.com/|
265 tube\.majestyc\.net/|
266 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
267 (?:.*?\#/)? # handle anchor (#/) redirect urls
268 (?: # the various things that can precede the ID:
269 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
270 |(?: # or the v= param in all its forms
271 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
272 (?:\?|\#!?) # the params delimiter ? or # or #!
273 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
274 v=
275 )
276 ))
277 |(?:
278 youtu\.be| # just youtu.be/xxxx
279 vid\.plus| # or vid.plus/xxxx
280 zwearz\.com/watch| # or zwearz.com/watch/xxxx
281 )/
282 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
283 )
284 )? # all until now is optional -> you can pass the naked ID
285 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
286 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
287 (?(1).+)? # if we found the ID, everything can follow
288 $"""
289 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
290 _formats = {
291 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
292 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
293 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
294 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
295 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
296 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
297 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
298 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
300 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
301 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
302 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
303 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
304 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
305 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
306 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
307 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
308 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
309
310
311 # 3D videos
312 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
313 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
314 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
315 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
316 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
317 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
318 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
319
320 # Apple HTTP Live Streaming
321 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
322 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
323 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
324 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
325 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
326 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
327 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
328 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
329
330 # DASH mp4 video
331 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
332 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
333 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
334 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
335 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
336 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
337 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
338 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
339 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
340 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
341 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
342
343 # Dash mp4 audio
344 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
345 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
346 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
347 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
348 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
349
350 # Dash webm
351 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
352 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
353 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
354 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
355 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
356 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
357 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
358 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
359 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
361 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
362 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
363 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
364 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
365 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
366 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
367 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
368 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
369 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
370 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
371 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
372 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
373
374 # Dash webm audio
375 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
376 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
377
378 # Dash webm audio with opus inside
379 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
380 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
381 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
382
383 # RTMP (unnamed)
384 '_rtmp': {'protocol': 'rtmp'},
385 }
386 _SUBTITLE_FORMATS = ('ttml', 'vtt')
387
388 IE_NAME = 'youtube'
389 _TESTS = [
390 {
391 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
392 'info_dict': {
393 'id': 'BaW_jenozKc',
394 'ext': 'mp4',
395 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
396 'uploader': 'Philipp Hagemeister',
397 'uploader_id': 'phihag',
398 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
399 'upload_date': '20121002',
400 'license': 'Standard YouTube License',
401 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
402 'categories': ['Science & Technology'],
403 'tags': ['youtube-dl'],
404 'like_count': int,
405 'dislike_count': int,
406 'start_time': 1,
407 'end_time': 9,
408 }
409 },
410 {
411 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
412 'note': 'Test generic use_cipher_signature video (#897)',
413 'info_dict': {
414 'id': 'UxxajLWwzqY',
415 'ext': 'mp4',
416 'upload_date': '20120506',
417 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
418 'alt_title': 'I Love It (feat. Charli XCX)',
419 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
420 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
421 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
422 'iconic ep', 'iconic', 'love', 'it'],
423 'uploader': 'Icona Pop',
424 'uploader_id': 'IconaPop',
425 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
426 'license': 'Standard YouTube License',
427 'creator': 'Icona Pop',
428 }
429 },
430 {
431 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
432 'note': 'Test VEVO video with age protection (#956)',
433 'info_dict': {
434 'id': '07FYdnEawAQ',
435 'ext': 'mp4',
436 'upload_date': '20130703',
437 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
438 'alt_title': 'Tunnel Vision',
439 'description': 'md5:64249768eec3bc4276236606ea996373',
440 'uploader': 'justintimberlakeVEVO',
441 'uploader_id': 'justintimberlakeVEVO',
442 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
443 'license': 'Standard YouTube License',
444 'creator': 'Justin Timberlake',
445 'age_limit': 18,
446 }
447 },
448 {
449 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
450 'note': 'Embed-only video (#1746)',
451 'info_dict': {
452 'id': 'yZIXLfi8CZQ',
453 'ext': 'mp4',
454 'upload_date': '20120608',
455 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
456 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
457 'uploader': 'SET India',
458 'uploader_id': 'setindia',
459 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
460 'license': 'Standard YouTube License',
461 'age_limit': 18,
462 }
463 },
464 {
465 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
466 'note': 'Use the first video ID in the URL',
467 'info_dict': {
468 'id': 'BaW_jenozKc',
469 'ext': 'mp4',
470 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
471 'uploader': 'Philipp Hagemeister',
472 'uploader_id': 'phihag',
473 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
474 'upload_date': '20121002',
475 'license': 'Standard YouTube License',
476 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
477 'categories': ['Science & Technology'],
478 'tags': ['youtube-dl'],
479 'like_count': int,
480 'dislike_count': int,
481 },
482 'params': {
483 'skip_download': True,
484 },
485 },
486 {
487 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
488 'note': '256k DASH audio (format 141) via DASH manifest',
489 'info_dict': {
490 'id': 'a9LDPn-MO4I',
491 'ext': 'm4a',
492 'upload_date': '20121002',
493 'uploader_id': '8KVIDEO',
494 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
495 'description': '',
496 'uploader': '8KVIDEO',
497 'license': 'Standard YouTube License',
498 'title': 'UHDTV TEST 8K VIDEO.mp4'
499 },
500 'params': {
501 'youtube_include_dash_manifest': True,
502 'format': '141',
503 },
504 'skip': 'format 141 not served anymore',
505 },
506 # DASH manifest with encrypted signature
507 {
508 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
509 'info_dict': {
510 'id': 'IB3lcPjvWLA',
511 'ext': 'm4a',
512 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
513 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
514 'uploader': 'AfrojackVEVO',
515 'uploader_id': 'AfrojackVEVO',
516 'upload_date': '20131011',
517 'license': 'Standard YouTube License',
518 },
519 'params': {
520 'youtube_include_dash_manifest': True,
521 'format': '141/bestaudio[ext=m4a]',
522 },
523 },
524 # JS player signature function name containing $
525 {
526 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
527 'info_dict': {
528 'id': 'nfWlot6h_JM',
529 'ext': 'm4a',
530 'title': 'Taylor Swift - Shake It Off',
531 'alt_title': 'Shake It Off',
532 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
533 'uploader': 'TaylorSwiftVEVO',
534 'uploader_id': 'TaylorSwiftVEVO',
535 'upload_date': '20140818',
536 'license': 'Standard YouTube License',
537 'creator': 'Taylor Swift',
538 },
539 'params': {
540 'youtube_include_dash_manifest': True,
541 'format': '141/bestaudio[ext=m4a]',
542 },
543 },
544 # Controversy video
545 {
546 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
547 'info_dict': {
548 'id': 'T4XJQO3qol8',
549 'ext': 'mp4',
550 'upload_date': '20100909',
551 'uploader': 'The Amazing Atheist',
552 'uploader_id': 'TheAmazingAtheist',
553 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
554 'license': 'Standard YouTube License',
555 'title': 'Burning Everyone\'s Koran',
556 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
557 }
558 },
559 # Normal age-gate video (No vevo, embed allowed)
560 {
561 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
562 'info_dict': {
563 'id': 'HtVdAasjOgU',
564 'ext': 'mp4',
565 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
566 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
567 'uploader': 'The Witcher',
568 'uploader_id': 'WitcherGame',
569 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
570 'upload_date': '20140605',
571 'license': 'Standard YouTube License',
572 'age_limit': 18,
573 },
574 },
575 # Age-gate video with encrypted signature
576 {
577 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
578 'info_dict': {
579 'id': '6kLq3WMV1nU',
580 'ext': 'mp4',
581 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
582 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
583 'uploader': 'LloydVEVO',
584 'uploader_id': 'LloydVEVO',
585 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
586 'upload_date': '20110629',
587 'license': 'Standard YouTube License',
588 'age_limit': 18,
589 },
590 },
591 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
592 {
593 'url': '__2ABJjxzNo',
594 'info_dict': {
595 'id': '__2ABJjxzNo',
596 'ext': 'mp4',
597 'upload_date': '20100430',
598 'uploader_id': 'deadmau5',
599 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
600 'creator': 'deadmau5',
601 'description': 'md5:12c56784b8032162bb936a5f76d55360',
602 'uploader': 'deadmau5',
603 'license': 'Standard YouTube License',
604 'title': 'Deadmau5 - Some Chords (HD)',
605 'alt_title': 'Some Chords',
606 },
607 'expected_warnings': [
608 'DASH manifest missing',
609 ]
610 },
611 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
612 {
613 'url': 'lqQg6PlCWgI',
614 'info_dict': {
615 'id': 'lqQg6PlCWgI',
616 'ext': 'mp4',
617 'upload_date': '20150827',
618 'uploader_id': 'olympic',
619 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
620 'license': 'Standard YouTube License',
621 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
622 'uploader': 'Olympic',
623 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
624 },
625 'params': {
626 'skip_download': 'requires avconv',
627 }
628 },
629 # Non-square pixels
630 {
631 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
632 'info_dict': {
633 'id': '_b-2C3KPAM0',
634 'ext': 'mp4',
635 'stretched_ratio': 16 / 9.,
636 'upload_date': '20110310',
637 'uploader_id': 'AllenMeow',
638 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
639 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
640 'uploader': '孫艾倫',
641 'license': 'Standard YouTube License',
642 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
643 },
644 },
645 # url_encoded_fmt_stream_map is empty string
646 {
647 'url': 'qEJwOuvDf7I',
648 'info_dict': {
649 'id': 'qEJwOuvDf7I',
650 'ext': 'webm',
651 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
652 'description': '',
653 'upload_date': '20150404',
654 'uploader_id': 'spbelect',
655 'uploader': 'Наблюдатели Петербурга',
656 },
657 'params': {
658 'skip_download': 'requires avconv',
659 },
660 'skip': 'This live event has ended.',
661 },
662 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
663 {
664 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
665 'info_dict': {
666 'id': 'FIl7x6_3R5Y',
667 'ext': 'mp4',
668 'title': 'md5:7b81415841e02ecd4313668cde88737a',
669 'description': 'md5:116377fd2963b81ec4ce64b542173306',
670 'upload_date': '20150625',
671 'uploader_id': 'dorappi2000',
672 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
673 'uploader': 'dorappi2000',
674 'license': 'Standard YouTube License',
675 'formats': 'mincount:32',
676 },
677 },
678 # DASH manifest with segment_list
679 {
680 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
681 'md5': '8ce563a1d667b599d21064e982ab9e31',
682 'info_dict': {
683 'id': 'CsmdDsKjzN8',
684 'ext': 'mp4',
685 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
686 'uploader': 'Airtek',
687 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
688 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
689 'license': 'Standard YouTube License',
690 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
691 },
692 'params': {
693 'youtube_include_dash_manifest': True,
694 'format': '135', # bestvideo
695 },
696 'skip': 'This live event has ended.',
697 },
698 {
699 # Multifeed videos (multiple cameras), URL is for Main Camera
700 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
701 'info_dict': {
702 'id': 'jqWvoWXjCVs',
703 'title': 'teamPGP: Rocket League Noob Stream',
704 'description': 'md5:dc7872fb300e143831327f1bae3af010',
705 },
706 'playlist': [{
707 'info_dict': {
708 'id': 'jqWvoWXjCVs',
709 'ext': 'mp4',
710 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
711 'description': 'md5:dc7872fb300e143831327f1bae3af010',
712 'upload_date': '20150721',
713 'uploader': 'Beer Games Beer',
714 'uploader_id': 'beergamesbeer',
715 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
716 'license': 'Standard YouTube License',
717 },
718 }, {
719 'info_dict': {
720 'id': '6h8e8xoXJzg',
721 'ext': 'mp4',
722 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
723 'description': 'md5:dc7872fb300e143831327f1bae3af010',
724 'upload_date': '20150721',
725 'uploader': 'Beer Games Beer',
726 'uploader_id': 'beergamesbeer',
727 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
728 'license': 'Standard YouTube License',
729 },
730 }, {
731 'info_dict': {
732 'id': 'PUOgX5z9xZw',
733 'ext': 'mp4',
734 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
735 'description': 'md5:dc7872fb300e143831327f1bae3af010',
736 'upload_date': '20150721',
737 'uploader': 'Beer Games Beer',
738 'uploader_id': 'beergamesbeer',
739 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
740 'license': 'Standard YouTube License',
741 },
742 }, {
743 'info_dict': {
744 'id': 'teuwxikvS5k',
745 'ext': 'mp4',
746 'title': 'teamPGP: Rocket League Noob Stream (zim)',
747 'description': 'md5:dc7872fb300e143831327f1bae3af010',
748 'upload_date': '20150721',
749 'uploader': 'Beer Games Beer',
750 'uploader_id': 'beergamesbeer',
751 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
752 'license': 'Standard YouTube License',
753 },
754 }],
755 'params': {
756 'skip_download': True,
757 },
758 },
759 {
760 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
761 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
762 'info_dict': {
763 'id': 'gVfLd0zydlo',
764 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
765 },
766 'playlist_count': 2,
767 'skip': 'Not multifeed anymore',
768 },
769 {
770 'url': 'http://vid.plus/FlRa-iH7PGw',
771 'only_matching': True,
772 },
773 {
774 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
775 'only_matching': True,
776 },
777 {
778 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
779 # Also tests cut-off URL expansion in video description (see
780 # https://github.com/rg3/youtube-dl/issues/1892,
781 # https://github.com/rg3/youtube-dl/issues/8164)
782 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
783 'info_dict': {
784 'id': 'lsguqyKfVQg',
785 'ext': 'mp4',
786 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
787 'alt_title': 'Dark Walk',
788 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
789 'upload_date': '20151119',
790 'uploader_id': 'IronSoulElf',
791 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
792 'uploader': 'IronSoulElf',
793 'license': 'Standard YouTube License',
794 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
795 },
796 'params': {
797 'skip_download': True,
798 },
799 },
800 {
801 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
802 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
803 'only_matching': True,
804 },
805 {
806 # Video with yt:stretch=17:0
807 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
808 'info_dict': {
809 'id': 'Q39EVAstoRM',
810 'ext': 'mp4',
811 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
812 'description': 'md5:ee18a25c350637c8faff806845bddee9',
813 'upload_date': '20151107',
814 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
815 'uploader': 'CH GAMER DROID',
816 },
817 'params': {
818 'skip_download': True,
819 },
820 'skip': 'This video does not exist.',
821 },
822 {
823 # Video licensed under Creative Commons
824 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
825 'info_dict': {
826 'id': 'M4gD1WSo5mA',
827 'ext': 'mp4',
828 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
829 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
830 'upload_date': '20150127',
831 'uploader_id': 'BerkmanCenter',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
833 'uploader': 'BerkmanCenter',
834 'license': 'Creative Commons Attribution license (reuse allowed)',
835 },
836 'params': {
837 'skip_download': True,
838 },
839 },
840 {
841 # Channel-like uploader_url
842 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
843 'info_dict': {
844 'id': 'eQcmzGIKrzg',
845 'ext': 'mp4',
846 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
847 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
848 'upload_date': '20151119',
849 'uploader': 'Bernie 2016',
850 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
851 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
852 'license': 'Creative Commons Attribution license (reuse allowed)',
853 },
854 'params': {
855 'skip_download': True,
856 },
857 },
858 {
859 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
860 'only_matching': True,
861 }
862 ]
863
864 def __init__(self, *args, **kwargs):
865 super(YoutubeIE, self).__init__(*args, **kwargs)
866 self._player_cache = {}
867
868 def report_video_info_webpage_download(self, video_id):
869 """Report attempt to download video info webpage."""
870 self.to_screen('%s: Downloading video info webpage' % video_id)
871
872 def report_information_extraction(self, video_id):
873 """Report attempt to extract video information."""
874 self.to_screen('%s: Extracting video information' % video_id)
875
876 def report_unavailable_format(self, video_id, format):
877 """Report extracted video URL."""
878 self.to_screen('%s: Format %s not available' % (video_id, format))
879
880 def report_rtmp_download(self):
881 """Indicate the download will use the RTMP protocol."""
882 self.to_screen('RTMP download detected')
883
884 def _signature_cache_id(self, example_sig):
885 """ Return a string representation of a signature """
886 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
887
888 def _extract_signature_function(self, video_id, player_url, example_sig):
889 id_m = re.match(
890 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
891 player_url)
892 if not id_m:
893 raise ExtractorError('Cannot identify player %r' % player_url)
894 player_type = id_m.group('ext')
895 player_id = id_m.group('id')
896
897 # Read from filesystem cache
898 func_id = '%s_%s_%s' % (
899 player_type, player_id, self._signature_cache_id(example_sig))
900 assert os.path.basename(func_id) == func_id
901
902 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
903 if cache_spec is not None:
904 return lambda s: ''.join(s[i] for i in cache_spec)
905
906 download_note = (
907 'Downloading player %s' % player_url
908 if self._downloader.params.get('verbose') else
909 'Downloading %s player %s' % (player_type, player_id)
910 )
911 if player_type == 'js':
912 code = self._download_webpage(
913 player_url, video_id,
914 note=download_note,
915 errnote='Download of %s failed' % player_url)
916 res = self._parse_sig_js(code)
917 elif player_type == 'swf':
918 urlh = self._request_webpage(
919 player_url, video_id,
920 note=download_note,
921 errnote='Download of %s failed' % player_url)
922 code = urlh.read()
923 res = self._parse_sig_swf(code)
924 else:
925 assert False, 'Invalid player type %r' % player_type
926
927 test_string = ''.join(map(compat_chr, range(len(example_sig))))
928 cache_res = res(test_string)
929 cache_spec = [ord(c) for c in cache_res]
930
931 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
932 return res
933
934 def _print_sig_code(self, func, example_sig):
935 def gen_sig_code(idxs):
936 def _genslice(start, end, step):
937 starts = '' if start == 0 else str(start)
938 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
939 steps = '' if step == 1 else (':%d' % step)
940 return 's[%s%s%s]' % (starts, ends, steps)
941
942 step = None
943 # Quelch pyflakes warnings - start will be set when step is set
944 start = '(Never used)'
945 for i, prev in zip(idxs[1:], idxs[:-1]):
946 if step is not None:
947 if i - prev == step:
948 continue
949 yield _genslice(start, prev, step)
950 step = None
951 continue
952 if i - prev in [-1, 1]:
953 step = i - prev
954 start = prev
955 continue
956 else:
957 yield 's[%d]' % prev
958 if step is None:
959 yield 's[%d]' % i
960 else:
961 yield _genslice(start, i, step)
962
963 test_string = ''.join(map(compat_chr, range(len(example_sig))))
964 cache_res = func(test_string)
965 cache_spec = [ord(c) for c in cache_res]
966 expr_code = ' + '.join(gen_sig_code(cache_spec))
967 signature_id_tuple = '(%s)' % (
968 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
969 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
970 ' return %s\n') % (signature_id_tuple, expr_code)
971 self.to_screen('Extracted signature function:\n' + code)
972
973 def _parse_sig_js(self, jscode):
974 funcname = self._search_regex(
975 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
976 'Initial JS player signature function name')
977
978 jsi = JSInterpreter(jscode)
979 initial_function = jsi.extract_function(funcname)
980 return lambda s: initial_function([s])
981
982 def _parse_sig_swf(self, file_contents):
983 swfi = SWFInterpreter(file_contents)
984 TARGET_CLASSNAME = 'SignatureDecipher'
985 searched_class = swfi.extract_class(TARGET_CLASSNAME)
986 initial_function = swfi.extract_function(searched_class, 'decipher')
987 return lambda s: initial_function([s])
988
989 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
990 """Turn the encrypted s field into a working signature"""
991
992 if player_url is None:
993 raise ExtractorError('Cannot decrypt signature without player_url')
994
995 if player_url.startswith('//'):
996 player_url = 'https:' + player_url
997 try:
998 player_id = (player_url, self._signature_cache_id(s))
999 if player_id not in self._player_cache:
1000 func = self._extract_signature_function(
1001 video_id, player_url, s
1002 )
1003 self._player_cache[player_id] = func
1004 func = self._player_cache[player_id]
1005 if self._downloader.params.get('youtube_print_sig_code'):
1006 self._print_sig_code(func, s)
1007 return func(s)
1008 except Exception as e:
1009 tb = traceback.format_exc()
1010 raise ExtractorError(
1011 'Signature extraction failed: ' + tb, cause=e)
1012
1013 def _get_subtitles(self, video_id, webpage):
1014 try:
1015 subs_doc = self._download_xml(
1016 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1017 video_id, note=False)
1018 except ExtractorError as err:
1019 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1020 return {}
1021
1022 sub_lang_list = {}
1023 for track in subs_doc.findall('track'):
1024 lang = track.attrib['lang_code']
1025 if lang in sub_lang_list:
1026 continue
1027 sub_formats = []
1028 for ext in self._SUBTITLE_FORMATS:
1029 params = compat_urllib_parse_urlencode({
1030 'lang': lang,
1031 'v': video_id,
1032 'fmt': ext,
1033 'name': track.attrib['name'].encode('utf-8'),
1034 })
1035 sub_formats.append({
1036 'url': 'https://www.youtube.com/api/timedtext?' + params,
1037 'ext': ext,
1038 })
1039 sub_lang_list[lang] = sub_formats
1040 if not sub_lang_list:
1041 self._downloader.report_warning('video doesn\'t have subtitles')
1042 return {}
1043 return sub_lang_list
1044
1045 def _get_ytplayer_config(self, video_id, webpage):
1046 patterns = (
1047 # User data may contain arbitrary character sequences that may affect
1048 # JSON extraction with regex, e.g. when '};' is contained the second
1049 # regex won't capture the whole JSON. Yet working around by trying more
1050 # concrete regex first keeping in mind proper quoted string handling
1051 # to be implemented in future that will replace this workaround (see
1052 # https://github.com/rg3/youtube-dl/issues/7468,
1053 # https://github.com/rg3/youtube-dl/pull/7599)
1054 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1055 r';ytplayer\.config\s*=\s*({.+?});',
1056 )
1057 config = self._search_regex(
1058 patterns, webpage, 'ytplayer.config', default=None)
1059 if config:
1060 return self._parse_json(
1061 uppercase_escape(config), video_id, fatal=False)
1062
1063 def _get_automatic_captions(self, video_id, webpage):
1064 """We need the webpage for getting the captions url, pass it as an
1065 argument to speed up the process."""
1066 self.to_screen('%s: Looking for automatic captions' % video_id)
1067 player_config = self._get_ytplayer_config(video_id, webpage)
1068 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1069 if not player_config:
1070 self._downloader.report_warning(err_msg)
1071 return {}
1072 try:
1073 args = player_config['args']
1074 caption_url = args.get('ttsurl')
1075 if caption_url:
1076 timestamp = args['timestamp']
1077 # We get the available subtitles
1078 list_params = compat_urllib_parse_urlencode({
1079 'type': 'list',
1080 'tlangs': 1,
1081 'asrs': 1,
1082 })
1083 list_url = caption_url + '&' + list_params
1084 caption_list = self._download_xml(list_url, video_id)
1085 original_lang_node = caption_list.find('track')
1086 if original_lang_node is None:
1087 self._downloader.report_warning('Video doesn\'t have automatic captions')
1088 return {}
1089 original_lang = original_lang_node.attrib['lang_code']
1090 caption_kind = original_lang_node.attrib.get('kind', '')
1091
1092 sub_lang_list = {}
1093 for lang_node in caption_list.findall('target'):
1094 sub_lang = lang_node.attrib['lang_code']
1095 sub_formats = []
1096 for ext in self._SUBTITLE_FORMATS:
1097 params = compat_urllib_parse_urlencode({
1098 'lang': original_lang,
1099 'tlang': sub_lang,
1100 'fmt': ext,
1101 'ts': timestamp,
1102 'kind': caption_kind,
1103 })
1104 sub_formats.append({
1105 'url': caption_url + '&' + params,
1106 'ext': ext,
1107 })
1108 sub_lang_list[sub_lang] = sub_formats
1109 return sub_lang_list
1110
1111 # Some videos don't provide ttsurl but rather caption_tracks and
1112 # caption_translation_languages (e.g. 20LmZk1hakA)
1113 caption_tracks = args['caption_tracks']
1114 caption_translation_languages = args['caption_translation_languages']
1115 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1116 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1117 caption_qs = compat_parse_qs(parsed_caption_url.query)
1118
1119 sub_lang_list = {}
1120 for lang in caption_translation_languages.split(','):
1121 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1122 sub_lang = lang_qs.get('lc', [None])[0]
1123 if not sub_lang:
1124 continue
1125 sub_formats = []
1126 for ext in self._SUBTITLE_FORMATS:
1127 caption_qs.update({
1128 'tlang': [sub_lang],
1129 'fmt': [ext],
1130 })
1131 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1132 query=compat_urllib_parse_urlencode(caption_qs, True)))
1133 sub_formats.append({
1134 'url': sub_url,
1135 'ext': ext,
1136 })
1137 sub_lang_list[sub_lang] = sub_formats
1138 return sub_lang_list
1139 # An extractor error can be raise by the download process if there are
1140 # no automatic captions but there are subtitles
1141 except (KeyError, ExtractorError):
1142 self._downloader.report_warning(err_msg)
1143 return {}
1144
1145 def _mark_watched(self, video_id, video_info):
1146 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1147 if not playback_url:
1148 return
1149 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1150 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1151
1152 # cpn generation algorithm is reverse engineered from base.js.
1153 # In fact it works even with dummy cpn.
1154 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1155 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1156
1157 qs.update({
1158 'ver': ['2'],
1159 'cpn': [cpn],
1160 })
1161 playback_url = compat_urlparse.urlunparse(
1162 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1163
1164 self._download_webpage(
1165 playback_url, video_id, 'Marking watched',
1166 'Unable to mark watched', fatal=False)
1167
1168 @classmethod
1169 def extract_id(cls, url):
1170 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1171 if mobj is None:
1172 raise ExtractorError('Invalid URL: %s' % url)
1173 video_id = mobj.group(2)
1174 return video_id
1175
1176 def _extract_from_m3u8(self, manifest_url, video_id):
1177 url_map = {}
1178
1179 def _get_urls(_manifest):
1180 lines = _manifest.split('\n')
1181 urls = filter(lambda l: l and not l.startswith('#'),
1182 lines)
1183 return urls
1184 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1185 formats_urls = _get_urls(manifest)
1186 for format_url in formats_urls:
1187 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1188 url_map[itag] = format_url
1189 return url_map
1190
1191 def _extract_annotations(self, video_id):
1192 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1193 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1194
1195 def _real_extract(self, url):
1196 url, smuggled_data = unsmuggle_url(url, {})
1197
1198 proto = (
1199 'http' if self._downloader.params.get('prefer_insecure', False)
1200 else 'https')
1201
1202 start_time = None
1203 end_time = None
1204 parsed_url = compat_urllib_parse_urlparse(url)
1205 for component in [parsed_url.fragment, parsed_url.query]:
1206 query = compat_parse_qs(component)
1207 if start_time is None and 't' in query:
1208 start_time = parse_duration(query['t'][0])
1209 if start_time is None and 'start' in query:
1210 start_time = parse_duration(query['start'][0])
1211 if end_time is None and 'end' in query:
1212 end_time = parse_duration(query['end'][0])
1213
1214 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1215 mobj = re.search(self._NEXT_URL_RE, url)
1216 if mobj:
1217 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1218 video_id = self.extract_id(url)
1219
1220 # Get video webpage
1221 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1222 video_webpage = self._download_webpage(url, video_id)
1223
1224 # Attempt to extract SWF player URL
1225 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1226 if mobj is not None:
1227 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1228 else:
1229 player_url = None
1230
1231 dash_mpds = []
1232
1233 def add_dash_mpd(video_info):
1234 dash_mpd = video_info.get('dashmpd')
1235 if dash_mpd and dash_mpd[0] not in dash_mpds:
1236 dash_mpds.append(dash_mpd[0])
1237
1238 # Get video info
1239 embed_webpage = None
1240 is_live = None
1241 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1242 age_gate = True
1243 # We simulate the access to the video from www.youtube.com/v/{video_id}
1244 # this can be viewed without login into Youtube
1245 url = proto + '://www.youtube.com/embed/%s' % video_id
1246 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1247 data = compat_urllib_parse_urlencode({
1248 'video_id': video_id,
1249 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1250 'sts': self._search_regex(
1251 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1252 })
1253 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1254 video_info_webpage = self._download_webpage(
1255 video_info_url, video_id,
1256 note='Refetching age-gated info webpage',
1257 errnote='unable to download video info webpage')
1258 video_info = compat_parse_qs(video_info_webpage)
1259 add_dash_mpd(video_info)
1260 else:
1261 age_gate = False
1262 video_info = None
1263 # Try looking directly into the video webpage
1264 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1265 if ytplayer_config:
1266 args = ytplayer_config['args']
1267 if args.get('url_encoded_fmt_stream_map'):
1268 # Convert to the same format returned by compat_parse_qs
1269 video_info = dict((k, [v]) for k, v in args.items())
1270 add_dash_mpd(video_info)
1271 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1272 is_live = True
1273 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1274 # We also try looking in get_video_info since it may contain different dashmpd
1275 # URL that points to a DASH manifest with possibly different itag set (some itags
1276 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1277 # manifest pointed by get_video_info's dashmpd).
1278 # The general idea is to take a union of itags of both DASH manifests (for example
1279 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1280 self.report_video_info_webpage_download(video_id)
1281 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1282 video_info_url = (
1283 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1284 % (proto, video_id, el_type))
1285 video_info_webpage = self._download_webpage(
1286 video_info_url,
1287 video_id, note=False,
1288 errnote='unable to download video info webpage')
1289 get_video_info = compat_parse_qs(video_info_webpage)
1290 if get_video_info.get('use_cipher_signature') != ['True']:
1291 add_dash_mpd(get_video_info)
1292 if not video_info:
1293 video_info = get_video_info
1294 if 'token' in get_video_info:
1295 # Different get_video_info requests may report different results, e.g.
1296 # some may report video unavailability, but some may serve it without
1297 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1298 # the original webpage as well as el=info and el=embedded get_video_info
1299 # requests report video unavailability due to geo restriction while
1300 # el=detailpage succeeds and returns valid data). This is probably
1301 # due to YouTube measures against IP ranges of hosting providers.
1302 # Working around by preferring the first succeeded video_info containing
1303 # the token if no such video_info yet was found.
1304 if 'token' not in video_info:
1305 video_info = get_video_info
1306 break
1307 if 'token' not in video_info:
1308 if 'reason' in video_info:
1309 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1310 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1311 if regions_allowed:
1312 raise ExtractorError('YouTube said: This video is available in %s only' % (
1313 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1314 expected=True)
1315 raise ExtractorError(
1316 'YouTube said: %s' % video_info['reason'][0],
1317 expected=True, video_id=video_id)
1318 else:
1319 raise ExtractorError(
1320 '"token" parameter not in video info for unknown reason',
1321 video_id=video_id)
1322
1323 # title
1324 if 'title' in video_info:
1325 video_title = video_info['title'][0]
1326 else:
1327 self._downloader.report_warning('Unable to extract video title')
1328 video_title = '_'
1329
1330 # description
1331 video_description = get_element_by_id("eow-description", video_webpage)
1332 if video_description:
1333 video_description = re.sub(r'''(?x)
1334 <a\s+
1335 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1336 (?:title|href)="([^"]+)"\s+
1337 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1338 class="[^"]*"[^>]*>
1339 [^<]+\.{3}\s*
1340 </a>
1341 ''', r'\1', video_description)
1342 video_description = clean_html(video_description)
1343 else:
1344 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1345 if fd_mobj:
1346 video_description = unescapeHTML(fd_mobj.group(1))
1347 else:
1348 video_description = ''
1349
1350 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1351 if not self._downloader.params.get('noplaylist'):
1352 entries = []
1353 feed_ids = []
1354 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1355 for feed in multifeed_metadata_list.split(','):
1356 # Unquote should take place before split on comma (,) since textual
1357 # fields may contain comma as well (see
1358 # https://github.com/rg3/youtube-dl/issues/8536)
1359 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1360 entries.append({
1361 '_type': 'url_transparent',
1362 'ie_key': 'Youtube',
1363 'url': smuggle_url(
1364 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1365 {'force_singlefeed': True}),
1366 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1367 })
1368 feed_ids.append(feed_data['id'][0])
1369 self.to_screen(
1370 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1371 % (', '.join(feed_ids), video_id))
1372 return self.playlist_result(entries, video_id, video_title, video_description)
1373 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1374
1375 if 'view_count' in video_info:
1376 view_count = int(video_info['view_count'][0])
1377 else:
1378 view_count = None
1379
1380 # Check for "rental" videos
1381 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1382 raise ExtractorError('"rental" videos not supported')
1383
1384 # Start extracting information
1385 self.report_information_extraction(video_id)
1386
1387 # uploader
1388 if 'author' not in video_info:
1389 raise ExtractorError('Unable to extract uploader name')
1390 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1391
1392 # uploader_id
1393 video_uploader_id = None
1394 video_uploader_url = None
1395 mobj = re.search(
1396 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1397 video_webpage)
1398 if mobj is not None:
1399 video_uploader_id = mobj.group('uploader_id')
1400 video_uploader_url = mobj.group('uploader_url')
1401 else:
1402 self._downloader.report_warning('unable to extract uploader nickname')
1403
1404 # thumbnail image
1405 # We try first to get a high quality image:
1406 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1407 video_webpage, re.DOTALL)
1408 if m_thumb is not None:
1409 video_thumbnail = m_thumb.group(1)
1410 elif 'thumbnail_url' not in video_info:
1411 self._downloader.report_warning('unable to extract video thumbnail')
1412 video_thumbnail = None
1413 else: # don't panic if we can't find it
1414 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1415
1416 # upload date
1417 upload_date = self._html_search_meta(
1418 'datePublished', video_webpage, 'upload date', default=None)
1419 if not upload_date:
1420 upload_date = self._search_regex(
1421 [r'(?s)id="eow-date.*?>(.*?)</span>',
1422 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1423 video_webpage, 'upload date', default=None)
1424 if upload_date:
1425 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1426 upload_date = unified_strdate(upload_date)
1427
1428 video_license = self._html_search_regex(
1429 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1430 video_webpage, 'license', default=None)
1431
1432 m_music = re.search(
1433 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1434 video_webpage)
1435 if m_music:
1436 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1437 video_creator = clean_html(m_music.group('creator'))
1438 else:
1439 video_alt_title = video_creator = None
1440
1441 m_cat_container = self._search_regex(
1442 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1443 video_webpage, 'categories', default=None)
1444 if m_cat_container:
1445 category = self._html_search_regex(
1446 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1447 default=None)
1448 video_categories = None if category is None else [category]
1449 else:
1450 video_categories = None
1451
1452 video_tags = [
1453 unescapeHTML(m.group('content'))
1454 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1455
1456 def _extract_count(count_name):
1457 return str_to_int(self._search_regex(
1458 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1459 % re.escape(count_name),
1460 video_webpage, count_name, default=None))
1461
1462 like_count = _extract_count('like')
1463 dislike_count = _extract_count('dislike')
1464
1465 # subtitles
1466 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1467 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1468
1469 if 'length_seconds' not in video_info:
1470 self._downloader.report_warning('unable to extract video duration')
1471 video_duration = None
1472 else:
1473 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1474
1475 # annotations
1476 video_annotations = None
1477 if self._downloader.params.get('writeannotations', False):
1478 video_annotations = self._extract_annotations(video_id)
1479
1480 def _map_to_format_list(urlmap):
1481 formats = []
1482 for itag, video_real_url in urlmap.items():
1483 dct = {
1484 'format_id': itag,
1485 'url': video_real_url,
1486 'player_url': player_url,
1487 }
1488 if itag in self._formats:
1489 dct.update(self._formats[itag])
1490 formats.append(dct)
1491 return formats
1492
1493 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1494 self.report_rtmp_download()
1495 formats = [{
1496 'format_id': '_rtmp',
1497 'protocol': 'rtmp',
1498 'url': video_info['conn'][0],
1499 'player_url': player_url,
1500 }]
1501 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1502 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1503 if 'rtmpe%3Dyes' in encoded_url_map:
1504 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1505 formats_spec = {}
1506 fmt_list = video_info.get('fmt_list', [''])[0]
1507 if fmt_list:
1508 for fmt in fmt_list.split(','):
1509 spec = fmt.split('/')
1510 if len(spec) > 1:
1511 width_height = spec[1].split('x')
1512 if len(width_height) == 2:
1513 formats_spec[spec[0]] = {
1514 'resolution': spec[1],
1515 'width': int_or_none(width_height[0]),
1516 'height': int_or_none(width_height[1]),
1517 }
1518 formats = []
1519 for url_data_str in encoded_url_map.split(','):
1520 url_data = compat_parse_qs(url_data_str)
1521 if 'itag' not in url_data or 'url' not in url_data:
1522 continue
1523 format_id = url_data['itag'][0]
1524 url = url_data['url'][0]
1525
1526 if 'sig' in url_data:
1527 url += '&signature=' + url_data['sig'][0]
1528 elif 's' in url_data:
1529 encrypted_sig = url_data['s'][0]
1530 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1531
1532 jsplayer_url_json = self._search_regex(
1533 ASSETS_RE,
1534 embed_webpage if age_gate else video_webpage,
1535 'JS player URL (1)', default=None)
1536 if not jsplayer_url_json and not age_gate:
1537 # We need the embed website after all
1538 if embed_webpage is None:
1539 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1540 embed_webpage = self._download_webpage(
1541 embed_url, video_id, 'Downloading embed webpage')
1542 jsplayer_url_json = self._search_regex(
1543 ASSETS_RE, embed_webpage, 'JS player URL')
1544
1545 player_url = json.loads(jsplayer_url_json)
1546 if player_url is None:
1547 player_url_json = self._search_regex(
1548 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1549 video_webpage, 'age gate player URL')
1550 player_url = json.loads(player_url_json)
1551
1552 if self._downloader.params.get('verbose'):
1553 if player_url is None:
1554 player_version = 'unknown'
1555 player_desc = 'unknown'
1556 else:
1557 if player_url.endswith('swf'):
1558 player_version = self._search_regex(
1559 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1560 'flash player', fatal=False)
1561 player_desc = 'flash player %s' % player_version
1562 else:
1563 player_version = self._search_regex(
1564 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1565 player_url,
1566 'html5 player', fatal=False)
1567 player_desc = 'html5 player %s' % player_version
1568
1569 parts_sizes = self._signature_cache_id(encrypted_sig)
1570 self.to_screen('{%s} signature length %s, %s' %
1571 (format_id, parts_sizes, player_desc))
1572
1573 signature = self._decrypt_signature(
1574 encrypted_sig, video_id, player_url, age_gate)
1575 url += '&signature=' + signature
1576 if 'ratebypass' not in url:
1577 url += '&ratebypass=yes'
1578
1579 dct = {
1580 'format_id': format_id,
1581 'url': url,
1582 'player_url': player_url,
1583 }
1584 if format_id in self._formats:
1585 dct.update(self._formats[format_id])
1586 if format_id in formats_spec:
1587 dct.update(formats_spec[format_id])
1588
1589 # Some itags are not included in DASH manifest thus corresponding formats will
1590 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1591 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1592 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1593 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1594
1595 more_fields = {
1596 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1597 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1598 'width': width,
1599 'height': height,
1600 'fps': int_or_none(url_data.get('fps', [None])[0]),
1601 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1602 }
1603 for key, value in more_fields.items():
1604 if value:
1605 dct[key] = value
1606 type_ = url_data.get('type', [None])[0]
1607 if type_:
1608 type_split = type_.split(';')
1609 kind_ext = type_split[0].split('/')
1610 if len(kind_ext) == 2:
1611 kind, _ = kind_ext
1612 dct['ext'] = mimetype2ext(type_split[0])
1613 if kind in ('audio', 'video'):
1614 codecs = None
1615 for mobj in re.finditer(
1616 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1617 if mobj.group('key') == 'codecs':
1618 codecs = mobj.group('val')
1619 break
1620 if codecs:
1621 codecs = codecs.split(',')
1622 if len(codecs) == 2:
1623 acodec, vcodec = codecs[1], codecs[0]
1624 else:
1625 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1626 dct.update({
1627 'acodec': acodec,
1628 'vcodec': vcodec,
1629 })
1630 formats.append(dct)
1631 elif video_info.get('hlsvp'):
1632 manifest_url = video_info['hlsvp'][0]
1633 url_map = self._extract_from_m3u8(manifest_url, video_id)
1634 formats = _map_to_format_list(url_map)
1635 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1636 for a_format in formats:
1637 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1638 else:
1639 unavailable_message = self._html_search_regex(
1640 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1641 video_webpage, 'unavailable message', default=None)
1642 if unavailable_message:
1643 raise ExtractorError(unavailable_message, expected=True)
1644 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1645
1646 # Look for the DASH manifest
1647 if self._downloader.params.get('youtube_include_dash_manifest', True):
1648 dash_mpd_fatal = True
1649 for mpd_url in dash_mpds:
1650 dash_formats = {}
1651 try:
1652 def decrypt_sig(mobj):
1653 s = mobj.group(1)
1654 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1655 return '/signature/%s' % dec_s
1656
1657 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1658
1659 for df in self._extract_mpd_formats(
1660 mpd_url, video_id, fatal=dash_mpd_fatal,
1661 formats_dict=self._formats):
1662 # Do not overwrite DASH format found in some previous DASH manifest
1663 if df['format_id'] not in dash_formats:
1664 dash_formats[df['format_id']] = df
1665 # Additional DASH manifests may end up in HTTP Error 403 therefore
1666 # allow them to fail without bug report message if we already have
1667 # some DASH manifest succeeded. This is temporary workaround to reduce
1668 # burst of bug reports until we figure out the reason and whether it
1669 # can be fixed at all.
1670 dash_mpd_fatal = False
1671 except (ExtractorError, KeyError) as e:
1672 self.report_warning(
1673 'Skipping DASH manifest: %r' % e, video_id)
1674 if dash_formats:
1675 # Remove the formats we found through non-DASH, they
1676 # contain less info and it can be wrong, because we use
1677 # fixed values (for example the resolution). See
1678 # https://github.com/rg3/youtube-dl/issues/5774 for an
1679 # example.
1680 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1681 formats.extend(dash_formats.values())
1682
1683 # Check for malformed aspect ratio
1684 stretched_m = re.search(
1685 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1686 video_webpage)
1687 if stretched_m:
1688 w = float(stretched_m.group('w'))
1689 h = float(stretched_m.group('h'))
1690 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1691 # We will only process correct ratios.
1692 if w > 0 and h > 0:
1693 ratio = w / h
1694 for f in formats:
1695 if f.get('vcodec') != 'none':
1696 f['stretched_ratio'] = ratio
1697
1698 self._sort_formats(formats)
1699
1700 self.mark_watched(video_id, video_info)
1701
1702 return {
1703 'id': video_id,
1704 'uploader': video_uploader,
1705 'uploader_id': video_uploader_id,
1706 'uploader_url': video_uploader_url,
1707 'upload_date': upload_date,
1708 'license': video_license,
1709 'creator': video_creator,
1710 'title': video_title,
1711 'alt_title': video_alt_title,
1712 'thumbnail': video_thumbnail,
1713 'description': video_description,
1714 'categories': video_categories,
1715 'tags': video_tags,
1716 'subtitles': video_subtitles,
1717 'automatic_captions': automatic_captions,
1718 'duration': video_duration,
1719 'age_limit': 18 if age_gate else 0,
1720 'annotations': video_annotations,
1721 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1722 'view_count': view_count,
1723 'like_count': like_count,
1724 'dislike_count': dislike_count,
1725 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1726 'formats': formats,
1727 'is_live': is_live,
1728 'start_time': start_time,
1729 'end_time': end_time,
1730 }
1731
1732
1733 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1734 IE_DESC = 'YouTube.com playlists'
1735 _VALID_URL = r"""(?x)(?:
1736 (?:https?://)?
1737 (?:\w+\.)?
1738 youtube\.com/
1739 (?:
1740 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1741 \? (?:.*?[&;])*? (?:p|a|list)=
1742 | p/
1743 )
1744 (
1745 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1746 # Top tracks, they can also include dots
1747 |(?:MC)[\w\.]*
1748 )
1749 .*
1750 |
1751 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1752 )"""
1753 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1754 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1755 IE_NAME = 'youtube:playlist'
1756 _TESTS = [{
1757 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1758 'info_dict': {
1759 'title': 'ytdl test PL',
1760 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1761 },
1762 'playlist_count': 3,
1763 }, {
1764 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1765 'info_dict': {
1766 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1767 'title': 'YDL_Empty_List',
1768 },
1769 'playlist_count': 0,
1770 }, {
1771 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1772 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1773 'info_dict': {
1774 'title': '29C3: Not my department',
1775 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1776 },
1777 'playlist_count': 95,
1778 }, {
1779 'note': 'issue #673',
1780 'url': 'PLBB231211A4F62143',
1781 'info_dict': {
1782 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1783 'id': 'PLBB231211A4F62143',
1784 },
1785 'playlist_mincount': 26,
1786 }, {
1787 'note': 'Large playlist',
1788 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1789 'info_dict': {
1790 'title': 'Uploads from Cauchemar',
1791 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1792 },
1793 'playlist_mincount': 799,
1794 }, {
1795 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1796 'info_dict': {
1797 'title': 'YDL_safe_search',
1798 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1799 },
1800 'playlist_count': 2,
1801 }, {
1802 'note': 'embedded',
1803 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1804 'playlist_count': 4,
1805 'info_dict': {
1806 'title': 'JODA15',
1807 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1808 }
1809 }, {
1810 'note': 'Embedded SWF player',
1811 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1812 'playlist_count': 4,
1813 'info_dict': {
1814 'title': 'JODA7',
1815 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1816 }
1817 }, {
1818 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1819 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1820 'info_dict': {
1821 'title': 'Uploads from Interstellar Movie',
1822 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1823 },
1824 'playlist_mincout': 21,
1825 }]
1826
1827 def _real_initialize(self):
1828 self._login()
1829
1830 def _extract_mix(self, playlist_id):
1831 # The mixes are generated from a single video
1832 # the id of the playlist is just 'RD' + video_id
1833 ids = []
1834 last_id = playlist_id[-11:]
1835 for n in itertools.count(1):
1836 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1837 webpage = self._download_webpage(
1838 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1839 new_ids = orderedSet(re.findall(
1840 r'''(?xs)data-video-username=".*?".*?
1841 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
1842 webpage))
1843 # Fetch new pages until all the videos are repeated, it seems that
1844 # there are always 51 unique videos.
1845 new_ids = [_id for _id in new_ids if _id not in ids]
1846 if not new_ids:
1847 break
1848 ids.extend(new_ids)
1849 last_id = ids[-1]
1850
1851 url_results = self._ids_to_results(ids)
1852
1853 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1854 title_span = (
1855 search_title('playlist-title') or
1856 search_title('title long-title') or
1857 search_title('title'))
1858 title = clean_html(title_span)
1859
1860 return self.playlist_result(url_results, playlist_id, title)
1861
1862 def _extract_playlist(self, playlist_id):
1863 url = self._TEMPLATE_URL % playlist_id
1864 page = self._download_webpage(url, playlist_id)
1865
1866 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1867 match = match.strip()
1868 # Check if the playlist exists or is private
1869 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1870 raise ExtractorError(
1871 'The playlist doesn\'t exist or is private, use --username or '
1872 '--netrc to access it.',
1873 expected=True)
1874 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1875 raise ExtractorError(
1876 'Invalid parameters. Maybe URL is incorrect.',
1877 expected=True)
1878 elif re.match(r'[^<]*Choose your language[^<]*', match):
1879 continue
1880 else:
1881 self.report_warning('Youtube gives an alert message: ' + match)
1882
1883 playlist_title = self._html_search_regex(
1884 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1885 page, 'title')
1886
1887 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1888
1889 def _check_download_just_video(self, url, playlist_id):
1890 # Check if it's a video-specific URL
1891 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1892 if 'v' in query_dict:
1893 video_id = query_dict['v'][0]
1894 if self._downloader.params.get('noplaylist'):
1895 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1896 return self.url_result(video_id, 'Youtube', video_id=video_id)
1897 else:
1898 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1899
1900 def _real_extract(self, url):
1901 # Extract playlist id
1902 mobj = re.match(self._VALID_URL, url)
1903 if mobj is None:
1904 raise ExtractorError('Invalid URL: %s' % url)
1905 playlist_id = mobj.group(1) or mobj.group(2)
1906
1907 video = self._check_download_just_video(url, playlist_id)
1908 if video:
1909 return video
1910
1911 if playlist_id.startswith(('RD', 'UL', 'PU')):
1912 # Mixes require a custom extraction process
1913 return self._extract_mix(playlist_id)
1914
1915 return self._extract_playlist(playlist_id)
1916
1917
1918 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1919 IE_DESC = 'YouTube.com channels'
1920 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1921 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1922 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1923 IE_NAME = 'youtube:channel'
1924 _TESTS = [{
1925 'note': 'paginated channel',
1926 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1927 'playlist_mincount': 91,
1928 'info_dict': {
1929 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1930 'title': 'Uploads from lex will',
1931 }
1932 }, {
1933 'note': 'Age restricted channel',
1934 # from https://www.youtube.com/user/DeusExOfficial
1935 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1936 'playlist_mincount': 64,
1937 'info_dict': {
1938 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1939 'title': 'Uploads from Deus Ex',
1940 },
1941 }]
1942
1943 @classmethod
1944 def suitable(cls, url):
1945 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1946 else super(YoutubeChannelIE, cls).suitable(url))
1947
1948 def _real_extract(self, url):
1949 channel_id = self._match_id(url)
1950
1951 url = self._TEMPLATE_URL % channel_id
1952
1953 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1954 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1955 # otherwise fallback on channel by page extraction
1956 channel_page = self._download_webpage(
1957 url + '?view=57', channel_id,
1958 'Downloading channel page', fatal=False)
1959 if channel_page is False:
1960 channel_playlist_id = False
1961 else:
1962 channel_playlist_id = self._html_search_meta(
1963 'channelId', channel_page, 'channel id', default=None)
1964 if not channel_playlist_id:
1965 channel_playlist_id = self._search_regex(
1966 r'data-(?:channel-external-|yt)id="([^"]+)"',
1967 channel_page, 'channel id', default=None)
1968 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1969 playlist_id = 'UU' + channel_playlist_id[2:]
1970 return self.url_result(
1971 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1972
1973 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1974 autogenerated = re.search(r'''(?x)
1975 class="[^"]*?(?:
1976 channel-header-autogenerated-label|
1977 yt-channel-title-autogenerated
1978 )[^"]*"''', channel_page) is not None
1979
1980 if autogenerated:
1981 # The videos are contained in a single page
1982 # the ajax pages can't be used, they are empty
1983 entries = [
1984 self.url_result(
1985 video_id, 'Youtube', video_id=video_id,
1986 video_title=video_title)
1987 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1988 return self.playlist_result(entries, channel_id)
1989
1990 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1991
1992
1993 class YoutubeUserIE(YoutubeChannelIE):
1994 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1995 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/|c/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1996 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1997 IE_NAME = 'youtube:user'
1998
1999 _TESTS = [{
2000 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2001 'playlist_mincount': 320,
2002 'info_dict': {
2003 'title': 'TheLinuxFoundation',
2004 }
2005 }, {
2006 'url': 'ytuser:phihag',
2007 'only_matching': True,
2008 }, {
2009 'url': 'https://www.youtube.com/c/gametrailers',
2010 'only_matching': True,
2011 }]
2012
2013 @classmethod
2014 def suitable(cls, url):
2015 # Don't return True if the url can be extracted with other youtube
2016 # extractor, the regex would is too permissive and it would match.
2017 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2018 if any(ie.suitable(url) for ie in other_yt_ies):
2019 return False
2020 else:
2021 return super(YoutubeUserIE, cls).suitable(url)
2022
2023
2024 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2025 IE_DESC = 'YouTube.com live streams'
2026 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2027 IE_NAME = 'youtube:live'
2028
2029 _TESTS = [{
2030 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2031 'info_dict': {
2032 'id': 'a48o2S1cPoo',
2033 'ext': 'mp4',
2034 'title': 'The Young Turks - Live Main Show',
2035 'uploader': 'The Young Turks',
2036 'uploader_id': 'TheYoungTurks',
2037 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2038 'upload_date': '20150715',
2039 'license': 'Standard YouTube License',
2040 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2041 'categories': ['News & Politics'],
2042 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2043 'like_count': int,
2044 'dislike_count': int,
2045 },
2046 'params': {
2047 'skip_download': True,
2048 },
2049 }, {
2050 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2051 'only_matching': True,
2052 }]
2053
2054 def _real_extract(self, url):
2055 mobj = re.match(self._VALID_URL, url)
2056 channel_id = mobj.group('id')
2057 base_url = mobj.group('base_url')
2058 webpage = self._download_webpage(url, channel_id, fatal=False)
2059 if webpage:
2060 page_type = self._og_search_property(
2061 'type', webpage, 'page type', default=None)
2062 video_id = self._html_search_meta(
2063 'videoId', webpage, 'video id', default=None)
2064 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2065 return self.url_result(video_id, YoutubeIE.ie_key())
2066 return self.url_result(base_url)
2067
2068
2069 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2070 IE_DESC = 'YouTube.com user/channel playlists'
2071 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2072 IE_NAME = 'youtube:playlists'
2073
2074 _TESTS = [{
2075 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2076 'playlist_mincount': 4,
2077 'info_dict': {
2078 'id': 'ThirstForScience',
2079 'title': 'Thirst for Science',
2080 },
2081 }, {
2082 # with "Load more" button
2083 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2084 'playlist_mincount': 70,
2085 'info_dict': {
2086 'id': 'igorkle1',
2087 'title': 'Игорь Клейнер',
2088 },
2089 }, {
2090 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2091 'playlist_mincount': 17,
2092 'info_dict': {
2093 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2094 'title': 'Chem Player',
2095 },
2096 }]
2097
2098
2099 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2100 IE_DESC = 'YouTube.com searches'
2101 # there doesn't appear to be a real limit, for example if you search for
2102 # 'python' you get more than 8.000.000 results
2103 _MAX_RESULTS = float('inf')
2104 IE_NAME = 'youtube:search'
2105 _SEARCH_KEY = 'ytsearch'
2106 _EXTRA_QUERY_ARGS = {}
2107 _TESTS = []
2108
2109 def _get_n_results(self, query, n):
2110 """Get a specified number of results for a query"""
2111
2112 videos = []
2113 limit = n
2114
2115 for pagenum in itertools.count(1):
2116 url_query = {
2117 'search_query': query.encode('utf-8'),
2118 'page': pagenum,
2119 'spf': 'navigate',
2120 }
2121 url_query.update(self._EXTRA_QUERY_ARGS)
2122 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2123 data = self._download_json(
2124 result_url, video_id='query "%s"' % query,
2125 note='Downloading page %s' % pagenum,
2126 errnote='Unable to download API page')
2127 html_content = data[1]['body']['content']
2128
2129 if 'class="search-message' in html_content:
2130 raise ExtractorError(
2131 '[youtube] No video results', expected=True)
2132
2133 new_videos = self._ids_to_results(orderedSet(re.findall(
2134 r'href="/watch\?v=(.{11})', html_content)))
2135 videos += new_videos
2136 if not new_videos or len(videos) > limit:
2137 break
2138
2139 if len(videos) > n:
2140 videos = videos[:n]
2141 return self.playlist_result(videos, query)
2142
2143
2144 class YoutubeSearchDateIE(YoutubeSearchIE):
2145 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2146 _SEARCH_KEY = 'ytsearchdate'
2147 IE_DESC = 'YouTube.com searches, newest videos first'
2148 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2149
2150
2151 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2152 IE_DESC = 'YouTube.com search URLs'
2153 IE_NAME = 'youtube:search_url'
2154 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2155 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2156 _TESTS = [{
2157 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2158 'playlist_mincount': 5,
2159 'info_dict': {
2160 'title': 'youtube-dl test video',
2161 }
2162 }, {
2163 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2164 'only_matching': True,
2165 }]
2166
2167 def _real_extract(self, url):
2168 mobj = re.match(self._VALID_URL, url)
2169 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2170 webpage = self._download_webpage(url, query)
2171 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2172
2173
2174 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2175 IE_DESC = 'YouTube.com (multi-season) shows'
2176 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2177 IE_NAME = 'youtube:show'
2178 _TESTS = [{
2179 'url': 'https://www.youtube.com/show/airdisasters',
2180 'playlist_mincount': 5,
2181 'info_dict': {
2182 'id': 'airdisasters',
2183 'title': 'Air Disasters',
2184 }
2185 }]
2186
2187 def _real_extract(self, url):
2188 playlist_id = self._match_id(url)
2189 return super(YoutubeShowIE, self)._real_extract(
2190 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2191
2192
2193 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2194 """
2195 Base class for feed extractors
2196 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2197 """
2198 _LOGIN_REQUIRED = True
2199
2200 @property
2201 def IE_NAME(self):
2202 return 'youtube:%s' % self._FEED_NAME
2203
2204 def _real_initialize(self):
2205 self._login()
2206
2207 def _real_extract(self, url):
2208 page = self._download_webpage(
2209 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2210
2211 # The extraction process is the same as for playlists, but the regex
2212 # for the video ids doesn't contain an index
2213 ids = []
2214 more_widget_html = content_html = page
2215 for page_num in itertools.count(1):
2216 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2217
2218 # 'recommended' feed has infinite 'load more' and each new portion spins
2219 # the same videos in (sometimes) slightly different order, so we'll check
2220 # for unicity and break when portion has no new videos
2221 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2222 if not new_ids:
2223 break
2224
2225 ids.extend(new_ids)
2226
2227 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2228 if not mobj:
2229 break
2230
2231 more = self._download_json(
2232 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2233 'Downloading page #%s' % page_num,
2234 transform_source=uppercase_escape)
2235 content_html = more['content_html']
2236 more_widget_html = more['load_more_widget_html']
2237
2238 return self.playlist_result(
2239 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2240
2241
2242 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2243 IE_NAME = 'youtube:watchlater'
2244 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2245 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2246
2247 _TESTS = [{
2248 'url': 'https://www.youtube.com/playlist?list=WL',
2249 'only_matching': True,
2250 }, {
2251 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2252 'only_matching': True,
2253 }]
2254
2255 def _real_extract(self, url):
2256 video = self._check_download_just_video(url, 'WL')
2257 if video:
2258 return video
2259 return self._extract_playlist('WL')
2260
2261
2262 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2263 IE_NAME = 'youtube:favorites'
2264 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2265 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2266 _LOGIN_REQUIRED = True
2267
2268 def _real_extract(self, url):
2269 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2270 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2271 return self.url_result(playlist_id, 'YoutubePlaylist')
2272
2273
2274 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2275 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2276 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2277 _FEED_NAME = 'recommended'
2278 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2279
2280
2281 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2282 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2283 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2284 _FEED_NAME = 'subscriptions'
2285 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2286
2287
2288 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2289 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2290 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2291 _FEED_NAME = 'history'
2292 _PLAYLIST_TITLE = 'Youtube History'
2293
2294
2295 class YoutubeTruncatedURLIE(InfoExtractor):
2296 IE_NAME = 'youtube:truncated_url'
2297 IE_DESC = False # Do not list
2298 _VALID_URL = r'''(?x)
2299 (?:https?://)?
2300 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2301 (?:watch\?(?:
2302 feature=[a-z_]+|
2303 annotation_id=annotation_[^&]+|
2304 x-yt-cl=[0-9]+|
2305 hl=[^&]*|
2306 t=[0-9]+
2307 )?
2308 |
2309 attribution_link\?a=[^&]+
2310 )
2311 $
2312 '''
2313
2314 _TESTS = [{
2315 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2316 'only_matching': True,
2317 }, {
2318 'url': 'http://www.youtube.com/watch?',
2319 'only_matching': True,
2320 }, {
2321 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2322 'only_matching': True,
2323 }, {
2324 'url': 'https://www.youtube.com/watch?feature=foo',
2325 'only_matching': True,
2326 }, {
2327 'url': 'https://www.youtube.com/watch?hl=en-GB',
2328 'only_matching': True,
2329 }, {
2330 'url': 'https://www.youtube.com/watch?t=2372',
2331 'only_matching': True,
2332 }]
2333
2334 def _real_extract(self, url):
2335 raise ExtractorError(
2336 'Did you forget to quote the URL? Remember that & is a meta '
2337 'character in most shells, so you want to put the URL in quotes, '
2338 'like youtube-dl '
2339 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2340 ' or simply youtube-dl BaW_jenozKc .',
2341 expected=True)
2342
2343
2344 class YoutubeTruncatedIDIE(InfoExtractor):
2345 IE_NAME = 'youtube:truncated_id'
2346 IE_DESC = False # Do not list
2347 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2348
2349 _TESTS = [{
2350 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2351 'only_matching': True,
2352 }]
2353
2354 def _real_extract(self, url):
2355 video_id = self._match_id(url)
2356 raise ExtractorError(
2357 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2358 expected=True)