]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
debian/copyright: Fix lintian's dep5-copyright-license-name-not-unique.
[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 re
10 import time
11 import traceback
12
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
17 compat_chr,
18 compat_parse_qs,
19 compat_urllib_parse,
20 compat_urllib_request,
21 compat_urlparse,
22 compat_str,
23 )
24 from ..utils import (
25 clean_html,
26 ExtractorError,
27 float_or_none,
28 get_element_by_attribute,
29 get_element_by_id,
30 int_or_none,
31 orderedSet,
32 unescapeHTML,
33 unified_strdate,
34 uppercase_escape,
35 )
36
37
38 class YoutubeBaseInfoExtractor(InfoExtractor):
39 """Provide base functions for Youtube extractors"""
40 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
41 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
42 _NETRC_MACHINE = 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED = False
45
46 def _set_language(self):
47 self._set_cookie(
48 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
49 # YouTube sets the expire time to about two months
50 expire_time=time.time() + 2 * 30 * 24 * 3600)
51
52 def _login(self):
53 """
54 Attempt to log in to YouTube.
55 True is returned if successful or skipped.
56 False is returned if login failed.
57
58 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
59 """
60 (username, password) = self._get_login_info()
61 # No authentication to be performed
62 if username is None:
63 if self._LOGIN_REQUIRED:
64 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
65 return True
66
67 login_page = self._download_webpage(
68 self._LOGIN_URL, None,
69 note='Downloading login page',
70 errnote='unable to fetch login page', fatal=False)
71 if login_page is False:
72 return
73
74 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
75 login_page, 'Login GALX parameter')
76
77 # Log in
78 login_form_strs = {
79 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
80 'Email': username,
81 'GALX': galx,
82 'Passwd': password,
83
84 'PersistentCookie': 'yes',
85 '_utf8': '霱',
86 'bgresponse': 'js_disabled',
87 'checkConnection': '',
88 'checkedDomains': 'youtube',
89 'dnConn': '',
90 'pstMsg': '0',
91 'rmShown': '1',
92 'secTok': '',
93 'signIn': 'Sign in',
94 'timeStmp': '',
95 'service': 'youtube',
96 'uilel': '3',
97 'hl': 'en_US',
98 }
99
100 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
101 # chokes on unicode
102 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
103 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
104
105 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
106 login_results = self._download_webpage(
107 req, None,
108 note='Logging in', errnote='unable to log in', fatal=False)
109 if login_results is False:
110 return False
111
112 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
113 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
114
115 # Two-Factor
116 # TODO add SMS and phone call support - these require making a request and then prompting the user
117
118 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
119 tfa_code = self._get_tfa_info()
120
121 if tfa_code is None:
122 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
123 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
124 return False
125
126 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
127
128 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
129 if match is None:
130 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
131 secTok = match.group(1)
132 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
133 if match is None:
134 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
135 timeStmp = match.group(1)
136
137 tfa_form_strs = {
138 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
139 'smsToken': '',
140 'smsUserPin': tfa_code,
141 'smsVerifyPin': 'Verify',
142
143 'PersistentCookie': 'yes',
144 'checkConnection': '',
145 'checkedDomains': 'youtube',
146 'pstMsg': '1',
147 'secTok': secTok,
148 'timeStmp': timeStmp,
149 'service': 'youtube',
150 'hl': 'en_US',
151 }
152 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
153 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
154
155 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
156 tfa_results = self._download_webpage(
157 tfa_req, None,
158 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
159
160 if tfa_results is False:
161 return False
162
163 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
164 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
165 return False
166 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
167 self._downloader.report_warning('unable to log in - did the page structure change?')
168 return False
169 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
170 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
171 return False
172
173 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
174 self._downloader.report_warning('unable to log in: bad username or password')
175 return False
176 return True
177
178 def _real_initialize(self):
179 if self._downloader is None:
180 return
181 self._set_language()
182 if not self._login():
183 return
184
185
186 class YoutubeIE(YoutubeBaseInfoExtractor):
187 IE_DESC = 'YouTube.com'
188 _VALID_URL = r"""(?x)^
189 (
190 (?:https?://|//) # http(s):// or protocol-independent URL
191 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
192 (?:www\.)?deturl\.com/www\.youtube\.com/|
193 (?:www\.)?pwnyoutube\.com/|
194 (?:www\.)?yourepeat\.com/|
195 tube\.majestyc\.net/|
196 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
197 (?:.*?\#/)? # handle anchor (#/) redirect urls
198 (?: # the various things that can precede the ID:
199 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
200 |(?: # or the v= param in all its forms
201 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
202 (?:\?|\#!?) # the params delimiter ? or # or #!
203 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
204 v=
205 )
206 ))
207 |youtu\.be/ # just youtu.be/xxxx
208 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
209 )
210 )? # all until now is optional -> you can pass the naked ID
211 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
212 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
213 (?(1).+)? # if we found the ID, everything can follow
214 $"""
215 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
216 _formats = {
217 '5': {'ext': 'flv', 'width': 400, 'height': 240},
218 '6': {'ext': 'flv', 'width': 450, 'height': 270},
219 '13': {'ext': '3gp'},
220 '17': {'ext': '3gp', 'width': 176, 'height': 144},
221 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
222 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
223 '34': {'ext': 'flv', 'width': 640, 'height': 360},
224 '35': {'ext': 'flv', 'width': 854, 'height': 480},
225 '36': {'ext': '3gp', 'width': 320, 'height': 240},
226 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
227 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
228 '43': {'ext': 'webm', 'width': 640, 'height': 360},
229 '44': {'ext': 'webm', 'width': 854, 'height': 480},
230 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
231 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
232
233
234 # 3d videos
235 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
236 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
237 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
238 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
239 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
240 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
241 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
242
243 # Apple HTTP Live Streaming
244 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
245 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
246 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
247 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
248 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
249 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
250 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
251
252 # DASH mp4 video
253 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
254 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
259 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
262 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
264
265 # Dash mp4 audio
266 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
267 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
268 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
269
270 # Dash webm
271 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
272 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
278 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
279 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
288 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
290 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
291 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
292
293 # Dash webm audio
294 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
295 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
296
297 # Dash webm audio with opus inside
298 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
299 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
300 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
301
302 # RTMP (unnamed)
303 '_rtmp': {'protocol': 'rtmp'},
304 }
305
306 IE_NAME = 'youtube'
307 _TESTS = [
308 {
309 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
310 'info_dict': {
311 'id': 'BaW_jenozKc',
312 'ext': 'mp4',
313 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
314 'uploader': 'Philipp Hagemeister',
315 'uploader_id': 'phihag',
316 'upload_date': '20121002',
317 '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 .',
318 'categories': ['Science & Technology'],
319 'like_count': int,
320 'dislike_count': int,
321 }
322 },
323 {
324 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
325 'note': 'Test generic use_cipher_signature video (#897)',
326 'info_dict': {
327 'id': 'UxxajLWwzqY',
328 'ext': 'mp4',
329 'upload_date': '20120506',
330 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
331 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
332 'uploader': 'Icona Pop',
333 'uploader_id': 'IconaPop',
334 }
335 },
336 {
337 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
338 'note': 'Test VEVO video with age protection (#956)',
339 'info_dict': {
340 'id': '07FYdnEawAQ',
341 'ext': 'mp4',
342 'upload_date': '20130703',
343 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
344 'description': 'md5:64249768eec3bc4276236606ea996373',
345 'uploader': 'justintimberlakeVEVO',
346 'uploader_id': 'justintimberlakeVEVO',
347 }
348 },
349 {
350 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
351 'note': 'Embed-only video (#1746)',
352 'info_dict': {
353 'id': 'yZIXLfi8CZQ',
354 'ext': 'mp4',
355 'upload_date': '20120608',
356 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
357 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
358 'uploader': 'SET India',
359 'uploader_id': 'setindia'
360 }
361 },
362 {
363 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
364 'note': '256k DASH audio (format 141) via DASH manifest',
365 'info_dict': {
366 'id': 'a9LDPn-MO4I',
367 'ext': 'm4a',
368 'upload_date': '20121002',
369 'uploader_id': '8KVIDEO',
370 'description': '',
371 'uploader': '8KVIDEO',
372 'title': 'UHDTV TEST 8K VIDEO.mp4'
373 },
374 'params': {
375 'youtube_include_dash_manifest': True,
376 'format': '141',
377 },
378 },
379 # DASH manifest with encrypted signature
380 {
381 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
382 'info_dict': {
383 'id': 'IB3lcPjvWLA',
384 'ext': 'm4a',
385 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
386 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
387 'uploader': 'AfrojackVEVO',
388 'uploader_id': 'AfrojackVEVO',
389 'upload_date': '20131011',
390 },
391 'params': {
392 'youtube_include_dash_manifest': True,
393 'format': '141',
394 },
395 },
396 # JS player signature function name containing $
397 {
398 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
399 'info_dict': {
400 'id': 'nfWlot6h_JM',
401 'ext': 'm4a',
402 'title': 'Taylor Swift - Shake It Off',
403 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
404 'uploader': 'TaylorSwiftVEVO',
405 'uploader_id': 'TaylorSwiftVEVO',
406 'upload_date': '20140818',
407 },
408 'params': {
409 'youtube_include_dash_manifest': True,
410 'format': '141',
411 },
412 },
413 # Controversy video
414 {
415 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
416 'info_dict': {
417 'id': 'T4XJQO3qol8',
418 'ext': 'mp4',
419 'upload_date': '20100909',
420 'uploader': 'The Amazing Atheist',
421 'uploader_id': 'TheAmazingAtheist',
422 'title': 'Burning Everyone\'s Koran',
423 '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',
424 }
425 },
426 # Normal age-gate video (No vevo, embed allowed)
427 {
428 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
429 'info_dict': {
430 'id': 'HtVdAasjOgU',
431 'ext': 'mp4',
432 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
433 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
434 'uploader': 'The Witcher',
435 'uploader_id': 'WitcherGame',
436 'upload_date': '20140605',
437 },
438 },
439 # Age-gate video with encrypted signature
440 {
441 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
442 'info_dict': {
443 'id': '6kLq3WMV1nU',
444 'ext': 'mp4',
445 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
446 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
447 'uploader': 'LloydVEVO',
448 'uploader_id': 'LloydVEVO',
449 'upload_date': '20110629',
450 },
451 },
452 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
453 {
454 'url': '__2ABJjxzNo',
455 'info_dict': {
456 'id': '__2ABJjxzNo',
457 'ext': 'mp4',
458 'upload_date': '20100430',
459 'uploader_id': 'deadmau5',
460 'description': 'md5:12c56784b8032162bb936a5f76d55360',
461 'uploader': 'deadmau5',
462 'title': 'Deadmau5 - Some Chords (HD)',
463 },
464 'expected_warnings': [
465 'DASH manifest missing',
466 ]
467 },
468 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
469 {
470 'url': 'lqQg6PlCWgI',
471 'info_dict': {
472 'id': 'lqQg6PlCWgI',
473 'ext': 'mp4',
474 'upload_date': '20120731',
475 'uploader_id': 'olympic',
476 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
477 'uploader': 'Olympics',
478 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
479 },
480 'params': {
481 'skip_download': 'requires avconv',
482 }
483 },
484 # Non-square pixels
485 {
486 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
487 'info_dict': {
488 'id': '_b-2C3KPAM0',
489 'ext': 'mp4',
490 'stretched_ratio': 16 / 9.,
491 'upload_date': '20110310',
492 'uploader_id': 'AllenMeow',
493 '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',
494 'uploader': '孫艾倫',
495 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
496 },
497 },
498 # url_encoded_fmt_stream_map is empty string
499 {
500 'url': 'qEJwOuvDf7I',
501 'info_dict': {
502 'id': 'qEJwOuvDf7I',
503 'ext': 'mp4',
504 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
505 'description': '',
506 'upload_date': '20150404',
507 'uploader_id': 'spbelect',
508 'uploader': 'Наблюдатели Петербурга',
509 },
510 'params': {
511 'skip_download': 'requires avconv',
512 }
513 },
514 ]
515
516 def __init__(self, *args, **kwargs):
517 super(YoutubeIE, self).__init__(*args, **kwargs)
518 self._player_cache = {}
519
520 def report_video_info_webpage_download(self, video_id):
521 """Report attempt to download video info webpage."""
522 self.to_screen('%s: Downloading video info webpage' % video_id)
523
524 def report_information_extraction(self, video_id):
525 """Report attempt to extract video information."""
526 self.to_screen('%s: Extracting video information' % video_id)
527
528 def report_unavailable_format(self, video_id, format):
529 """Report extracted video URL."""
530 self.to_screen('%s: Format %s not available' % (video_id, format))
531
532 def report_rtmp_download(self):
533 """Indicate the download will use the RTMP protocol."""
534 self.to_screen('RTMP download detected')
535
536 def _signature_cache_id(self, example_sig):
537 """ Return a string representation of a signature """
538 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
539
540 def _extract_signature_function(self, video_id, player_url, example_sig):
541 id_m = re.match(
542 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
543 player_url)
544 if not id_m:
545 raise ExtractorError('Cannot identify player %r' % player_url)
546 player_type = id_m.group('ext')
547 player_id = id_m.group('id')
548
549 # Read from filesystem cache
550 func_id = '%s_%s_%s' % (
551 player_type, player_id, self._signature_cache_id(example_sig))
552 assert os.path.basename(func_id) == func_id
553
554 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
555 if cache_spec is not None:
556 return lambda s: ''.join(s[i] for i in cache_spec)
557
558 download_note = (
559 'Downloading player %s' % player_url
560 if self._downloader.params.get('verbose') else
561 'Downloading %s player %s' % (player_type, player_id)
562 )
563 if player_type == 'js':
564 code = self._download_webpage(
565 player_url, video_id,
566 note=download_note,
567 errnote='Download of %s failed' % player_url)
568 res = self._parse_sig_js(code)
569 elif player_type == 'swf':
570 urlh = self._request_webpage(
571 player_url, video_id,
572 note=download_note,
573 errnote='Download of %s failed' % player_url)
574 code = urlh.read()
575 res = self._parse_sig_swf(code)
576 else:
577 assert False, 'Invalid player type %r' % player_type
578
579 test_string = ''.join(map(compat_chr, range(len(example_sig))))
580 cache_res = res(test_string)
581 cache_spec = [ord(c) for c in cache_res]
582
583 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
584 return res
585
586 def _print_sig_code(self, func, example_sig):
587 def gen_sig_code(idxs):
588 def _genslice(start, end, step):
589 starts = '' if start == 0 else str(start)
590 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
591 steps = '' if step == 1 else (':%d' % step)
592 return 's[%s%s%s]' % (starts, ends, steps)
593
594 step = None
595 # Quelch pyflakes warnings - start will be set when step is set
596 start = '(Never used)'
597 for i, prev in zip(idxs[1:], idxs[:-1]):
598 if step is not None:
599 if i - prev == step:
600 continue
601 yield _genslice(start, prev, step)
602 step = None
603 continue
604 if i - prev in [-1, 1]:
605 step = i - prev
606 start = prev
607 continue
608 else:
609 yield 's[%d]' % prev
610 if step is None:
611 yield 's[%d]' % i
612 else:
613 yield _genslice(start, i, step)
614
615 test_string = ''.join(map(compat_chr, range(len(example_sig))))
616 cache_res = func(test_string)
617 cache_spec = [ord(c) for c in cache_res]
618 expr_code = ' + '.join(gen_sig_code(cache_spec))
619 signature_id_tuple = '(%s)' % (
620 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
621 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
622 ' return %s\n') % (signature_id_tuple, expr_code)
623 self.to_screen('Extracted signature function:\n' + code)
624
625 def _parse_sig_js(self, jscode):
626 funcname = self._search_regex(
627 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
628 'Initial JS player signature function name')
629
630 jsi = JSInterpreter(jscode)
631 initial_function = jsi.extract_function(funcname)
632 return lambda s: initial_function([s])
633
634 def _parse_sig_swf(self, file_contents):
635 swfi = SWFInterpreter(file_contents)
636 TARGET_CLASSNAME = 'SignatureDecipher'
637 searched_class = swfi.extract_class(TARGET_CLASSNAME)
638 initial_function = swfi.extract_function(searched_class, 'decipher')
639 return lambda s: initial_function([s])
640
641 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
642 """Turn the encrypted s field into a working signature"""
643
644 if player_url is None:
645 raise ExtractorError('Cannot decrypt signature without player_url')
646
647 if player_url.startswith('//'):
648 player_url = 'https:' + player_url
649 try:
650 player_id = (player_url, self._signature_cache_id(s))
651 if player_id not in self._player_cache:
652 func = self._extract_signature_function(
653 video_id, player_url, s
654 )
655 self._player_cache[player_id] = func
656 func = self._player_cache[player_id]
657 if self._downloader.params.get('youtube_print_sig_code'):
658 self._print_sig_code(func, s)
659 return func(s)
660 except Exception as e:
661 tb = traceback.format_exc()
662 raise ExtractorError(
663 'Signature extraction failed: ' + tb, cause=e)
664
665 def _get_subtitles(self, video_id, webpage):
666 try:
667 subs_doc = self._download_xml(
668 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
669 video_id, note=False)
670 except ExtractorError as err:
671 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
672 return {}
673
674 sub_lang_list = {}
675 for track in subs_doc.findall('track'):
676 lang = track.attrib['lang_code']
677 if lang in sub_lang_list:
678 continue
679 sub_formats = []
680 for ext in ['sbv', 'vtt', 'srt']:
681 params = compat_urllib_parse.urlencode({
682 'lang': lang,
683 'v': video_id,
684 'fmt': ext,
685 'name': track.attrib['name'].encode('utf-8'),
686 })
687 sub_formats.append({
688 'url': 'https://www.youtube.com/api/timedtext?' + params,
689 'ext': ext,
690 })
691 sub_lang_list[lang] = sub_formats
692 if not sub_lang_list:
693 self._downloader.report_warning('video doesn\'t have subtitles')
694 return {}
695 return sub_lang_list
696
697 def _get_automatic_captions(self, video_id, webpage):
698 """We need the webpage for getting the captions url, pass it as an
699 argument to speed up the process."""
700 self.to_screen('%s: Looking for automatic captions' % video_id)
701 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
702 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
703 if mobj is None:
704 self._downloader.report_warning(err_msg)
705 return {}
706 player_config = json.loads(mobj.group(1))
707 try:
708 args = player_config['args']
709 caption_url = args['ttsurl']
710 timestamp = args['timestamp']
711 # We get the available subtitles
712 list_params = compat_urllib_parse.urlencode({
713 'type': 'list',
714 'tlangs': 1,
715 'asrs': 1,
716 })
717 list_url = caption_url + '&' + list_params
718 caption_list = self._download_xml(list_url, video_id)
719 original_lang_node = caption_list.find('track')
720 if original_lang_node is None:
721 self._downloader.report_warning('Video doesn\'t have automatic captions')
722 return {}
723 original_lang = original_lang_node.attrib['lang_code']
724 caption_kind = original_lang_node.attrib.get('kind', '')
725
726 sub_lang_list = {}
727 for lang_node in caption_list.findall('target'):
728 sub_lang = lang_node.attrib['lang_code']
729 sub_formats = []
730 for ext in ['sbv', 'vtt', 'srt']:
731 params = compat_urllib_parse.urlencode({
732 'lang': original_lang,
733 'tlang': sub_lang,
734 'fmt': ext,
735 'ts': timestamp,
736 'kind': caption_kind,
737 })
738 sub_formats.append({
739 'url': caption_url + '&' + params,
740 'ext': ext,
741 })
742 sub_lang_list[sub_lang] = sub_formats
743 return sub_lang_list
744 # An extractor error can be raise by the download process if there are
745 # no automatic captions but there are subtitles
746 except (KeyError, ExtractorError):
747 self._downloader.report_warning(err_msg)
748 return {}
749
750 @classmethod
751 def extract_id(cls, url):
752 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
753 if mobj is None:
754 raise ExtractorError('Invalid URL: %s' % url)
755 video_id = mobj.group(2)
756 return video_id
757
758 def _extract_from_m3u8(self, manifest_url, video_id):
759 url_map = {}
760
761 def _get_urls(_manifest):
762 lines = _manifest.split('\n')
763 urls = filter(lambda l: l and not l.startswith('#'),
764 lines)
765 return urls
766 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
767 formats_urls = _get_urls(manifest)
768 for format_url in formats_urls:
769 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
770 url_map[itag] = format_url
771 return url_map
772
773 def _extract_annotations(self, video_id):
774 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
775 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
776
777 def _parse_dash_manifest(
778 self, video_id, dash_manifest_url, player_url, age_gate):
779 def decrypt_sig(mobj):
780 s = mobj.group(1)
781 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
782 return '/signature/%s' % dec_s
783 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
784 dash_doc = self._download_xml(
785 dash_manifest_url, video_id,
786 note='Downloading DASH manifest',
787 errnote='Could not download DASH manifest')
788
789 formats = []
790 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
791 mime_type = a.attrib.get('mimeType')
792 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
793 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
794 if url_el is None:
795 continue
796 if mime_type == 'text/vtt':
797 # TODO implement WebVTT downloading
798 pass
799 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
800 format_id = r.attrib['id']
801 video_url = url_el.text
802 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
803 f = {
804 'format_id': format_id,
805 'url': video_url,
806 'width': int_or_none(r.attrib.get('width')),
807 'height': int_or_none(r.attrib.get('height')),
808 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
809 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
810 'filesize': filesize,
811 'fps': int_or_none(r.attrib.get('frameRate')),
812 }
813 try:
814 existing_format = next(
815 fo for fo in formats
816 if fo['format_id'] == format_id)
817 except StopIteration:
818 full_info = self._formats.get(format_id, {}).copy()
819 full_info.update(f)
820 formats.append(full_info)
821 else:
822 existing_format.update(f)
823 else:
824 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
825 return formats
826
827 def _real_extract(self, url):
828 proto = (
829 'http' if self._downloader.params.get('prefer_insecure', False)
830 else 'https')
831
832 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
833 mobj = re.search(self._NEXT_URL_RE, url)
834 if mobj:
835 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
836 video_id = self.extract_id(url)
837
838 # Get video webpage
839 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
840 video_webpage = self._download_webpage(url, video_id)
841
842 # Attempt to extract SWF player URL
843 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
844 if mobj is not None:
845 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
846 else:
847 player_url = None
848
849 # Get video info
850 embed_webpage = None
851 if re.search(r'player-age-gate-content">', video_webpage) is not None:
852 age_gate = True
853 # We simulate the access to the video from www.youtube.com/v/{video_id}
854 # this can be viewed without login into Youtube
855 url = proto + '://www.youtube.com/embed/%s' % video_id
856 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
857 data = compat_urllib_parse.urlencode({
858 'video_id': video_id,
859 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
860 'sts': self._search_regex(
861 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
862 })
863 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
864 video_info_webpage = self._download_webpage(
865 video_info_url, video_id,
866 note='Refetching age-gated info webpage',
867 errnote='unable to download video info webpage')
868 video_info = compat_parse_qs(video_info_webpage)
869 else:
870 age_gate = False
871 try:
872 # Try looking directly into the video webpage
873 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
874 if not mobj:
875 raise ValueError('Could not find ytplayer.config') # caught below
876 json_code = uppercase_escape(mobj.group(1))
877 ytplayer_config = json.loads(json_code)
878 args = ytplayer_config['args']
879 # Convert to the same format returned by compat_parse_qs
880 video_info = dict((k, [v]) for k, v in args.items())
881 if not args.get('url_encoded_fmt_stream_map'):
882 raise ValueError('No stream_map present') # caught below
883 except ValueError:
884 # We fallback to the get_video_info pages (used by the embed page)
885 self.report_video_info_webpage_download(video_id)
886 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
887 video_info_url = (
888 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
889 % (proto, video_id, el_type))
890 video_info_webpage = self._download_webpage(
891 video_info_url,
892 video_id, note=False,
893 errnote='unable to download video info webpage')
894 video_info = compat_parse_qs(video_info_webpage)
895 if 'token' in video_info:
896 break
897 if 'token' not in video_info:
898 if 'reason' in video_info:
899 raise ExtractorError(
900 'YouTube said: %s' % video_info['reason'][0],
901 expected=True, video_id=video_id)
902 else:
903 raise ExtractorError(
904 '"token" parameter not in video info for unknown reason',
905 video_id=video_id)
906
907 if 'view_count' in video_info:
908 view_count = int(video_info['view_count'][0])
909 else:
910 view_count = None
911
912 # Check for "rental" videos
913 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
914 raise ExtractorError('"rental" videos not supported')
915
916 # Start extracting information
917 self.report_information_extraction(video_id)
918
919 # uploader
920 if 'author' not in video_info:
921 raise ExtractorError('Unable to extract uploader name')
922 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
923
924 # uploader_id
925 video_uploader_id = None
926 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
927 if mobj is not None:
928 video_uploader_id = mobj.group(1)
929 else:
930 self._downloader.report_warning('unable to extract uploader nickname')
931
932 # title
933 if 'title' in video_info:
934 video_title = video_info['title'][0]
935 else:
936 self._downloader.report_warning('Unable to extract video title')
937 video_title = '_'
938
939 # thumbnail image
940 # We try first to get a high quality image:
941 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
942 video_webpage, re.DOTALL)
943 if m_thumb is not None:
944 video_thumbnail = m_thumb.group(1)
945 elif 'thumbnail_url' not in video_info:
946 self._downloader.report_warning('unable to extract video thumbnail')
947 video_thumbnail = None
948 else: # don't panic if we can't find it
949 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
950
951 # upload date
952 upload_date = None
953 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
954 if mobj is None:
955 mobj = re.search(
956 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
957 video_webpage)
958 if mobj is not None:
959 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
960 upload_date = unified_strdate(upload_date)
961
962 m_cat_container = self._search_regex(
963 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
964 video_webpage, 'categories', default=None)
965 if m_cat_container:
966 category = self._html_search_regex(
967 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
968 default=None)
969 video_categories = None if category is None else [category]
970 else:
971 video_categories = None
972
973 # description
974 video_description = get_element_by_id("eow-description", video_webpage)
975 if video_description:
976 video_description = re.sub(r'''(?x)
977 <a\s+
978 (?:[a-zA-Z-]+="[^"]+"\s+)*?
979 title="([^"]+)"\s+
980 (?:[a-zA-Z-]+="[^"]+"\s+)*?
981 class="yt-uix-redirect-link"\s*>
982 [^<]+
983 </a>
984 ''', r'\1', video_description)
985 video_description = clean_html(video_description)
986 else:
987 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
988 if fd_mobj:
989 video_description = unescapeHTML(fd_mobj.group(1))
990 else:
991 video_description = ''
992
993 def _extract_count(count_name):
994 count = self._search_regex(
995 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
996 video_webpage, count_name, default=None)
997 if count is not None:
998 return int(count.replace(',', ''))
999 return None
1000 like_count = _extract_count('like')
1001 dislike_count = _extract_count('dislike')
1002
1003 # subtitles
1004 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1005 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1006
1007 if 'length_seconds' not in video_info:
1008 self._downloader.report_warning('unable to extract video duration')
1009 video_duration = None
1010 else:
1011 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
1012
1013 # annotations
1014 video_annotations = None
1015 if self._downloader.params.get('writeannotations', False):
1016 video_annotations = self._extract_annotations(video_id)
1017
1018 def _map_to_format_list(urlmap):
1019 formats = []
1020 for itag, video_real_url in urlmap.items():
1021 dct = {
1022 'format_id': itag,
1023 'url': video_real_url,
1024 'player_url': player_url,
1025 }
1026 if itag in self._formats:
1027 dct.update(self._formats[itag])
1028 formats.append(dct)
1029 return formats
1030
1031 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1032 self.report_rtmp_download()
1033 formats = [{
1034 'format_id': '_rtmp',
1035 'protocol': 'rtmp',
1036 'url': video_info['conn'][0],
1037 'player_url': player_url,
1038 }]
1039 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1040 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1041 if 'rtmpe%3Dyes' in encoded_url_map:
1042 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1043 url_map = {}
1044 for url_data_str in encoded_url_map.split(','):
1045 url_data = compat_parse_qs(url_data_str)
1046 if 'itag' not in url_data or 'url' not in url_data:
1047 continue
1048 format_id = url_data['itag'][0]
1049 url = url_data['url'][0]
1050
1051 if 'sig' in url_data:
1052 url += '&signature=' + url_data['sig'][0]
1053 elif 's' in url_data:
1054 encrypted_sig = url_data['s'][0]
1055 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1056
1057 jsplayer_url_json = self._search_regex(
1058 ASSETS_RE,
1059 embed_webpage if age_gate else video_webpage,
1060 'JS player URL (1)', default=None)
1061 if not jsplayer_url_json and not age_gate:
1062 # We need the embed website after all
1063 if embed_webpage is None:
1064 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1065 embed_webpage = self._download_webpage(
1066 embed_url, video_id, 'Downloading embed webpage')
1067 jsplayer_url_json = self._search_regex(
1068 ASSETS_RE, embed_webpage, 'JS player URL')
1069
1070 player_url = json.loads(jsplayer_url_json)
1071 if player_url is None:
1072 player_url_json = self._search_regex(
1073 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1074 video_webpage, 'age gate player URL')
1075 player_url = json.loads(player_url_json)
1076
1077 if self._downloader.params.get('verbose'):
1078 if player_url is None:
1079 player_version = 'unknown'
1080 player_desc = 'unknown'
1081 else:
1082 if player_url.endswith('swf'):
1083 player_version = self._search_regex(
1084 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1085 'flash player', fatal=False)
1086 player_desc = 'flash player %s' % player_version
1087 else:
1088 player_version = self._search_regex(
1089 r'html5player-([^/]+?)(?:/html5player)?\.js',
1090 player_url,
1091 'html5 player', fatal=False)
1092 player_desc = 'html5 player %s' % player_version
1093
1094 parts_sizes = self._signature_cache_id(encrypted_sig)
1095 self.to_screen('{%s} signature length %s, %s' %
1096 (format_id, parts_sizes, player_desc))
1097
1098 signature = self._decrypt_signature(
1099 encrypted_sig, video_id, player_url, age_gate)
1100 url += '&signature=' + signature
1101 if 'ratebypass' not in url:
1102 url += '&ratebypass=yes'
1103 url_map[format_id] = url
1104 formats = _map_to_format_list(url_map)
1105 elif video_info.get('hlsvp'):
1106 manifest_url = video_info['hlsvp'][0]
1107 url_map = self._extract_from_m3u8(manifest_url, video_id)
1108 formats = _map_to_format_list(url_map)
1109 else:
1110 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1111
1112 # Look for the DASH manifest
1113 if self._downloader.params.get('youtube_include_dash_manifest', True):
1114 dash_mpd = video_info.get('dashmpd')
1115 if dash_mpd:
1116 dash_manifest_url = dash_mpd[0]
1117 try:
1118 dash_formats = self._parse_dash_manifest(
1119 video_id, dash_manifest_url, player_url, age_gate)
1120 except (ExtractorError, KeyError) as e:
1121 self.report_warning(
1122 'Skipping DASH manifest: %r' % e, video_id)
1123 else:
1124 # Hide the formats we found through non-DASH
1125 dash_keys = set(df['format_id'] for df in dash_formats)
1126 for f in formats:
1127 if f['format_id'] in dash_keys:
1128 f['format_id'] = 'nondash-%s' % f['format_id']
1129 f['preference'] = f.get('preference', 0) - 10000
1130 formats.extend(dash_formats)
1131
1132 # Check for malformed aspect ratio
1133 stretched_m = re.search(
1134 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1135 video_webpage)
1136 if stretched_m:
1137 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1138 for f in formats:
1139 if f.get('vcodec') != 'none':
1140 f['stretched_ratio'] = ratio
1141
1142 self._sort_formats(formats)
1143
1144 return {
1145 'id': video_id,
1146 'uploader': video_uploader,
1147 'uploader_id': video_uploader_id,
1148 'upload_date': upload_date,
1149 'title': video_title,
1150 'thumbnail': video_thumbnail,
1151 'description': video_description,
1152 'categories': video_categories,
1153 'subtitles': video_subtitles,
1154 'automatic_captions': automatic_captions,
1155 'duration': video_duration,
1156 'age_limit': 18 if age_gate else 0,
1157 'annotations': video_annotations,
1158 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1159 'view_count': view_count,
1160 'like_count': like_count,
1161 'dislike_count': dislike_count,
1162 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1163 'formats': formats,
1164 }
1165
1166
1167 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1168 IE_DESC = 'YouTube.com playlists'
1169 _VALID_URL = r"""(?x)(?:
1170 (?:https?://)?
1171 (?:\w+\.)?
1172 youtube\.com/
1173 (?:
1174 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1175 \? (?:.*?&)*? (?:p|a|list)=
1176 | p/
1177 )
1178 (
1179 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1180 # Top tracks, they can also include dots
1181 |(?:MC)[\w\.]*
1182 )
1183 .*
1184 |
1185 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1186 )"""
1187 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1188 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
1189 IE_NAME = 'youtube:playlist'
1190 _TESTS = [{
1191 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1192 'info_dict': {
1193 'title': 'ytdl test PL',
1194 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1195 },
1196 'playlist_count': 3,
1197 }, {
1198 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1199 'info_dict': {
1200 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1201 'title': 'YDL_Empty_List',
1202 },
1203 'playlist_count': 0,
1204 }, {
1205 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1206 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1207 'info_dict': {
1208 'title': '29C3: Not my department',
1209 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1210 },
1211 'playlist_count': 95,
1212 }, {
1213 'note': 'issue #673',
1214 'url': 'PLBB231211A4F62143',
1215 'info_dict': {
1216 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1217 'id': 'PLBB231211A4F62143',
1218 },
1219 'playlist_mincount': 26,
1220 }, {
1221 'note': 'Large playlist',
1222 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1223 'info_dict': {
1224 'title': 'Uploads from Cauchemar',
1225 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1226 },
1227 'playlist_mincount': 799,
1228 }, {
1229 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1230 'info_dict': {
1231 'title': 'YDL_safe_search',
1232 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1233 },
1234 'playlist_count': 2,
1235 }, {
1236 'note': 'embedded',
1237 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1238 'playlist_count': 4,
1239 'info_dict': {
1240 'title': 'JODA15',
1241 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1242 }
1243 }, {
1244 'note': 'Embedded SWF player',
1245 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1246 'playlist_count': 4,
1247 'info_dict': {
1248 'title': 'JODA7',
1249 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1250 }
1251 }, {
1252 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1253 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1254 'info_dict': {
1255 'title': 'Uploads from Interstellar Movie',
1256 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1257 },
1258 'playlist_mincout': 21,
1259 }]
1260
1261 def _real_initialize(self):
1262 self._login()
1263
1264 def _ids_to_results(self, ids):
1265 return [
1266 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1267 for vid_id in ids]
1268
1269 def _extract_mix(self, playlist_id):
1270 # The mixes are generated from a single video
1271 # the id of the playlist is just 'RD' + video_id
1272 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1273 webpage = self._download_webpage(
1274 url, playlist_id, 'Downloading Youtube mix')
1275 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1276 title_span = (
1277 search_title('playlist-title') or
1278 search_title('title long-title') or
1279 search_title('title'))
1280 title = clean_html(title_span)
1281 ids = orderedSet(re.findall(
1282 r'''(?xs)data-video-username=".*?".*?
1283 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
1284 webpage))
1285 url_results = self._ids_to_results(ids)
1286
1287 return self.playlist_result(url_results, playlist_id, title)
1288
1289 def _extract_playlist(self, playlist_id):
1290 url = self._TEMPLATE_URL % playlist_id
1291 page = self._download_webpage(url, playlist_id)
1292 more_widget_html = content_html = page
1293
1294 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1295 match = match.strip()
1296 # Check if the playlist exists or is private
1297 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1298 raise ExtractorError(
1299 'The playlist doesn\'t exist or is private, use --username or '
1300 '--netrc to access it.',
1301 expected=True)
1302 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1303 raise ExtractorError(
1304 'Invalid parameters. Maybe URL is incorrect.',
1305 expected=True)
1306 elif re.match(r'[^<]*Choose your language[^<]*', match):
1307 continue
1308 else:
1309 self.report_warning('Youtube gives an alert message: ' + match)
1310
1311 # Extract the video ids from the playlist pages
1312 ids = []
1313
1314 for page_num in itertools.count(1):
1315 matches = re.finditer(self._VIDEO_RE, content_html)
1316 # We remove the duplicates and the link with index 0
1317 # (it's not the first video of the playlist)
1318 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1319 ids.extend(new_ids)
1320
1321 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1322 if not mobj:
1323 break
1324
1325 more = self._download_json(
1326 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1327 'Downloading page #%s' % page_num,
1328 transform_source=uppercase_escape)
1329 content_html = more['content_html']
1330 if not content_html.strip():
1331 # Some webpages show a "Load more" button but they don't
1332 # have more videos
1333 break
1334 more_widget_html = more['load_more_widget_html']
1335
1336 playlist_title = self._html_search_regex(
1337 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1338 page, 'title')
1339
1340 url_results = self._ids_to_results(ids)
1341 return self.playlist_result(url_results, playlist_id, playlist_title)
1342
1343 def _real_extract(self, url):
1344 # Extract playlist id
1345 mobj = re.match(self._VALID_URL, url)
1346 if mobj is None:
1347 raise ExtractorError('Invalid URL: %s' % url)
1348 playlist_id = mobj.group(1) or mobj.group(2)
1349
1350 # Check if it's a video-specific URL
1351 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1352 if 'v' in query_dict:
1353 video_id = query_dict['v'][0]
1354 if self._downloader.params.get('noplaylist'):
1355 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1356 return self.url_result(video_id, 'Youtube', video_id=video_id)
1357 else:
1358 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1359
1360 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1361 # Mixes require a custom extraction process
1362 return self._extract_mix(playlist_id)
1363
1364 return self._extract_playlist(playlist_id)
1365
1366
1367 class YoutubeChannelIE(InfoExtractor):
1368 IE_DESC = 'YouTube.com channels'
1369 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1370 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1371 IE_NAME = 'youtube:channel'
1372 _TESTS = [{
1373 'note': 'paginated channel',
1374 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1375 'playlist_mincount': 91,
1376 'info_dict': {
1377 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1378 }
1379 }]
1380
1381 @staticmethod
1382 def extract_videos_from_page(page):
1383 ids_in_page = []
1384 titles_in_page = []
1385 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1386 video_id = mobj.group('id')
1387 video_title = unescapeHTML(mobj.group('title'))
1388 try:
1389 idx = ids_in_page.index(video_id)
1390 if video_title and not titles_in_page[idx]:
1391 titles_in_page[idx] = video_title
1392 except ValueError:
1393 ids_in_page.append(video_id)
1394 titles_in_page.append(video_title)
1395 return zip(ids_in_page, titles_in_page)
1396
1397 def _real_extract(self, url):
1398 channel_id = self._match_id(url)
1399
1400 url = self._TEMPLATE_URL % channel_id
1401 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1402 autogenerated = re.search(r'''(?x)
1403 class="[^"]*?(?:
1404 channel-header-autogenerated-label|
1405 yt-channel-title-autogenerated
1406 )[^"]*"''', channel_page) is not None
1407
1408 if autogenerated:
1409 # The videos are contained in a single page
1410 # the ajax pages can't be used, they are empty
1411 entries = [
1412 self.url_result(
1413 video_id, 'Youtube', video_id=video_id,
1414 video_title=video_title)
1415 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1416 return self.playlist_result(entries, channel_id)
1417
1418 def _entries():
1419 more_widget_html = content_html = channel_page
1420 for pagenum in itertools.count(1):
1421
1422 for video_id, video_title in self.extract_videos_from_page(content_html):
1423 yield self.url_result(
1424 video_id, 'Youtube', video_id=video_id,
1425 video_title=video_title)
1426
1427 mobj = re.search(
1428 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1429 more_widget_html)
1430 if not mobj:
1431 break
1432
1433 more = self._download_json(
1434 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1435 'Downloading page #%s' % (pagenum + 1),
1436 transform_source=uppercase_escape)
1437 content_html = more['content_html']
1438 more_widget_html = more['load_more_widget_html']
1439
1440 return self.playlist_result(_entries(), channel_id)
1441
1442
1443 class YoutubeUserIE(YoutubeChannelIE):
1444 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1445 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1446 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1447 IE_NAME = 'youtube:user'
1448
1449 _TESTS = [{
1450 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1451 'playlist_mincount': 320,
1452 'info_dict': {
1453 'title': 'TheLinuxFoundation',
1454 }
1455 }, {
1456 'url': 'ytuser:phihag',
1457 'only_matching': True,
1458 }]
1459
1460 @classmethod
1461 def suitable(cls, url):
1462 # Don't return True if the url can be extracted with other youtube
1463 # extractor, the regex would is too permissive and it would match.
1464 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1465 if any(ie.suitable(url) for ie in other_ies):
1466 return False
1467 else:
1468 return super(YoutubeUserIE, cls).suitable(url)
1469
1470
1471 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1472 IE_DESC = 'YouTube.com searches'
1473 # there doesn't appear to be a real limit, for example if you search for
1474 # 'python' you get more than 8.000.000 results
1475 _MAX_RESULTS = float('inf')
1476 IE_NAME = 'youtube:search'
1477 _SEARCH_KEY = 'ytsearch'
1478 _EXTRA_QUERY_ARGS = {}
1479 _TESTS = []
1480
1481 def _get_n_results(self, query, n):
1482 """Get a specified number of results for a query"""
1483
1484 videos = []
1485 limit = n
1486
1487 for pagenum in itertools.count(1):
1488 url_query = {
1489 'search_query': query,
1490 'page': pagenum,
1491 'spf': 'navigate',
1492 }
1493 url_query.update(self._EXTRA_QUERY_ARGS)
1494 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1495 data = self._download_json(
1496 result_url, video_id='query "%s"' % query,
1497 note='Downloading page %s' % pagenum,
1498 errnote='Unable to download API page')
1499 html_content = data[1]['body']['content']
1500
1501 if 'class="search-message' in html_content:
1502 raise ExtractorError(
1503 '[youtube] No video results', expected=True)
1504
1505 new_videos = self._ids_to_results(orderedSet(re.findall(
1506 r'href="/watch\?v=(.{11})', html_content)))
1507 videos += new_videos
1508 if not new_videos or len(videos) > limit:
1509 break
1510
1511 if len(videos) > n:
1512 videos = videos[:n]
1513 return self.playlist_result(videos, query)
1514
1515
1516 class YoutubeSearchDateIE(YoutubeSearchIE):
1517 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1518 _SEARCH_KEY = 'ytsearchdate'
1519 IE_DESC = 'YouTube.com searches, newest videos first'
1520 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1521
1522
1523 class YoutubeSearchURLIE(InfoExtractor):
1524 IE_DESC = 'YouTube.com search URLs'
1525 IE_NAME = 'youtube:search_url'
1526 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1527 _TESTS = [{
1528 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1529 'playlist_mincount': 5,
1530 'info_dict': {
1531 'title': 'youtube-dl test video',
1532 }
1533 }]
1534
1535 def _real_extract(self, url):
1536 mobj = re.match(self._VALID_URL, url)
1537 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1538
1539 webpage = self._download_webpage(url, query)
1540 result_code = self._search_regex(
1541 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1542
1543 part_codes = re.findall(
1544 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1545 entries = []
1546 for part_code in part_codes:
1547 part_title = self._html_search_regex(
1548 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1549 part_url_snippet = self._html_search_regex(
1550 r'(?s)href="([^"]+)"', part_code, 'item URL')
1551 part_url = compat_urlparse.urljoin(
1552 'https://www.youtube.com/', part_url_snippet)
1553 entries.append({
1554 '_type': 'url',
1555 'url': part_url,
1556 'title': part_title,
1557 })
1558
1559 return {
1560 '_type': 'playlist',
1561 'entries': entries,
1562 'title': query,
1563 }
1564
1565
1566 class YoutubeShowIE(InfoExtractor):
1567 IE_DESC = 'YouTube.com (multi-season) shows'
1568 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1569 IE_NAME = 'youtube:show'
1570 _TESTS = [{
1571 'url': 'http://www.youtube.com/show/airdisasters',
1572 'playlist_mincount': 3,
1573 'info_dict': {
1574 'id': 'airdisasters',
1575 'title': 'Air Disasters',
1576 }
1577 }]
1578
1579 def _real_extract(self, url):
1580 mobj = re.match(self._VALID_URL, url)
1581 playlist_id = mobj.group('id')
1582 webpage = self._download_webpage(
1583 url, playlist_id, 'Downloading show webpage')
1584 # There's one playlist for each season of the show
1585 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1586 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1587 entries = [
1588 self.url_result(
1589 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1590 for season in m_seasons
1591 ]
1592 title = self._og_search_title(webpage, fatal=False)
1593
1594 return {
1595 '_type': 'playlist',
1596 'id': playlist_id,
1597 'title': title,
1598 'entries': entries,
1599 }
1600
1601
1602 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1603 """
1604 Base class for extractors that fetch info from
1605 http://www.youtube.com/feed_ajax
1606 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1607 """
1608 _LOGIN_REQUIRED = True
1609 # use action_load_personal_feed instead of action_load_system_feed
1610 _PERSONAL_FEED = False
1611
1612 @property
1613 def _FEED_TEMPLATE(self):
1614 action = 'action_load_system_feed'
1615 if self._PERSONAL_FEED:
1616 action = 'action_load_personal_feed'
1617 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1618
1619 @property
1620 def IE_NAME(self):
1621 return 'youtube:%s' % self._FEED_NAME
1622
1623 def _real_initialize(self):
1624 self._login()
1625
1626 def _real_extract(self, url):
1627 feed_entries = []
1628 paging = 0
1629 for i in itertools.count(1):
1630 info = self._download_json(
1631 self._FEED_TEMPLATE % paging,
1632 '%s feed' % self._FEED_NAME,
1633 'Downloading page %s' % i,
1634 transform_source=uppercase_escape)
1635 feed_html = info.get('feed_html') or info.get('content_html')
1636 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1637 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1638 ids = orderedSet(m.group(1) for m in m_ids)
1639 feed_entries.extend(
1640 self.url_result(video_id, 'Youtube', video_id=video_id)
1641 for video_id in ids)
1642 mobj = re.search(
1643 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1644 load_more_widget_html)
1645 if mobj is None:
1646 break
1647 paging = mobj.group('paging')
1648 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1649
1650
1651 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1652 IE_NAME = 'youtube:recommended'
1653 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1654 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1655 _FEED_NAME = 'recommended'
1656 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1657
1658
1659 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1660 IE_NAME = 'youtube:watchlater'
1661 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1662 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1663
1664 _TESTS = [] # override PlaylistIE tests
1665
1666 def _real_extract(self, url):
1667 return self._extract_playlist('WL')
1668
1669
1670 class YoutubeHistoryIE(YoutubePlaylistIE):
1671 IE_NAME = 'youtube:history'
1672 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1673 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1674 _TESTS = []
1675
1676 def _real_extract(self, url):
1677 title = 'Youtube History'
1678 page = self._download_webpage('https://www.youtube.com/feed/history', title)
1679
1680 # The extraction process is the same as for playlists, but the regex
1681 # for the video ids doesn't contain an index
1682 ids = []
1683 more_widget_html = content_html = page
1684
1685 for page_num in itertools.count(1):
1686 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1687 new_ids = orderedSet(matches)
1688 ids.extend(new_ids)
1689
1690 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1691 if not mobj:
1692 break
1693
1694 more = self._download_json(
1695 'https://youtube.com/%s' % mobj.group('more'), title,
1696 'Downloading page #%s' % page_num,
1697 transform_source=uppercase_escape)
1698 content_html = more['content_html']
1699 more_widget_html = more['load_more_widget_html']
1700
1701 return {
1702 '_type': 'playlist',
1703 'title': title,
1704 'entries': self._ids_to_results(ids),
1705 }
1706
1707
1708 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1709 IE_NAME = 'youtube:favorites'
1710 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1711 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1712 _LOGIN_REQUIRED = True
1713
1714 def _real_extract(self, url):
1715 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1716 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1717 return self.url_result(playlist_id, 'YoutubePlaylist')
1718
1719
1720 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1721 IE_NAME = 'youtube:subscriptions'
1722 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1723 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1724 _TESTS = []
1725
1726 def _real_extract(self, url):
1727 title = 'Youtube Subscriptions'
1728 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1729
1730 # The extraction process is the same as for playlists, but the regex
1731 # for the video ids doesn't contain an index
1732 ids = []
1733 more_widget_html = content_html = page
1734
1735 for page_num in itertools.count(1):
1736 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1737 new_ids = orderedSet(matches)
1738 ids.extend(new_ids)
1739
1740 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1741 if not mobj:
1742 break
1743
1744 more = self._download_json(
1745 'https://youtube.com/%s' % mobj.group('more'), title,
1746 'Downloading page #%s' % page_num,
1747 transform_source=uppercase_escape)
1748 content_html = more['content_html']
1749 more_widget_html = more['load_more_widget_html']
1750
1751 return {
1752 '_type': 'playlist',
1753 'title': title,
1754 'entries': self._ids_to_results(ids),
1755 }
1756
1757
1758 class YoutubeTruncatedURLIE(InfoExtractor):
1759 IE_NAME = 'youtube:truncated_url'
1760 IE_DESC = False # Do not list
1761 _VALID_URL = r'''(?x)
1762 (?:https?://)?
1763 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1764 (?:watch\?(?:
1765 feature=[a-z_]+|
1766 annotation_id=annotation_[^&]+|
1767 x-yt-cl=[0-9]+|
1768 hl=[^&]*|
1769 )?
1770 |
1771 attribution_link\?a=[^&]+
1772 )
1773 $
1774 '''
1775
1776 _TESTS = [{
1777 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1778 'only_matching': True,
1779 }, {
1780 'url': 'http://www.youtube.com/watch?',
1781 'only_matching': True,
1782 }, {
1783 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1784 'only_matching': True,
1785 }, {
1786 'url': 'https://www.youtube.com/watch?feature=foo',
1787 'only_matching': True,
1788 }, {
1789 'url': 'https://www.youtube.com/watch?hl=en-GB',
1790 'only_matching': True,
1791 }]
1792
1793 def _real_extract(self, url):
1794 raise ExtractorError(
1795 'Did you forget to quote the URL? Remember that & is a meta '
1796 'character in most shells, so you want to put the URL in quotes, '
1797 'like youtube-dl '
1798 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1799 ' or simply youtube-dl BaW_jenozKc .',
1800 expected=True)
1801
1802
1803 class YoutubeTruncatedIDIE(InfoExtractor):
1804 IE_NAME = 'youtube:truncated_id'
1805 IE_DESC = False # Do not list
1806 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1807
1808 _TESTS = [{
1809 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1810 'only_matching': True,
1811 }]
1812
1813 def _real_extract(self, url):
1814 video_id = self._match_id(url)
1815 raise ExtractorError(
1816 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
1817 expected=True)