]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
3690f8021267b30171be4f2e7a019133aaeaaca9
[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 OnDemandPagedList,
32 orderedSet,
33 unescapeHTML,
34 unified_strdate,
35 uppercase_escape,
36 )
37
38
39 class YoutubeBaseInfoExtractor(InfoExtractor):
40 """Provide base functions for Youtube extractors"""
41 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
42 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
43 _NETRC_MACHINE = 'youtube'
44 # If True it will raise an error if no login info is provided
45 _LOGIN_REQUIRED = False
46
47 def _set_language(self):
48 self._set_cookie(
49 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
50 # YouTube sets the expire time to about two months
51 expire_time=time.time() + 2 * 30 * 24 * 3600)
52
53 def _login(self):
54 """
55 Attempt to log in to YouTube.
56 True is returned if successful or skipped.
57 False is returned if login failed.
58
59 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
60 """
61 (username, password) = self._get_login_info()
62 # No authentication to be performed
63 if username is None:
64 if self._LOGIN_REQUIRED:
65 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
66 return True
67
68 login_page = self._download_webpage(
69 self._LOGIN_URL, None,
70 note='Downloading login page',
71 errnote='unable to fetch login page', fatal=False)
72 if login_page is False:
73 return
74
75 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
76 login_page, 'Login GALX parameter')
77
78 # Log in
79 login_form_strs = {
80 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
81 'Email': username,
82 'GALX': galx,
83 'Passwd': password,
84
85 'PersistentCookie': 'yes',
86 '_utf8': '霱',
87 'bgresponse': 'js_disabled',
88 'checkConnection': '',
89 'checkedDomains': 'youtube',
90 'dnConn': '',
91 'pstMsg': '0',
92 'rmShown': '1',
93 'secTok': '',
94 'signIn': 'Sign in',
95 'timeStmp': '',
96 'service': 'youtube',
97 'uilel': '3',
98 'hl': 'en_US',
99 }
100
101 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
102 # chokes on unicode
103 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
104 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
105
106 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
107 login_results = self._download_webpage(
108 req, None,
109 note='Logging in', errnote='unable to log in', fatal=False)
110 if login_results is False:
111 return False
112
113 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
114 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
115
116 # Two-Factor
117 # TODO add SMS and phone call support - these require making a request and then prompting the user
118
119 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
120 tfa_code = self._get_tfa_info()
121
122 if tfa_code is None:
123 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
124 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
125 return False
126
127 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
128
129 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
130 if match is None:
131 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
132 secTok = match.group(1)
133 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
134 if match is None:
135 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
136 timeStmp = match.group(1)
137
138 tfa_form_strs = {
139 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
140 'smsToken': '',
141 'smsUserPin': tfa_code,
142 'smsVerifyPin': 'Verify',
143
144 'PersistentCookie': 'yes',
145 'checkConnection': '',
146 'checkedDomains': 'youtube',
147 'pstMsg': '1',
148 'secTok': secTok,
149 'timeStmp': timeStmp,
150 'service': 'youtube',
151 'hl': 'en_US',
152 }
153 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
154 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
155
156 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
157 tfa_results = self._download_webpage(
158 tfa_req, None,
159 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
160
161 if tfa_results is False:
162 return False
163
164 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
165 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
166 return False
167 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
168 self._downloader.report_warning('unable to log in - did the page structure change?')
169 return False
170 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
171 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
172 return False
173
174 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
175 self._downloader.report_warning('unable to log in: bad username or password')
176 return False
177 return True
178
179 def _real_initialize(self):
180 if self._downloader is None:
181 return
182 self._set_language()
183 if not self._login():
184 return
185
186
187 class YoutubeIE(YoutubeBaseInfoExtractor):
188 IE_DESC = 'YouTube.com'
189 _VALID_URL = r"""(?x)^
190 (
191 (?:https?://|//) # http(s):// or protocol-independent URL
192 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
193 (?:www\.)?deturl\.com/www\.youtube\.com/|
194 (?:www\.)?pwnyoutube\.com/|
195 (?:www\.)?yourepeat\.com/|
196 tube\.majestyc\.net/|
197 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
198 (?:.*?\#/)? # handle anchor (#/) redirect urls
199 (?: # the various things that can precede the ID:
200 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
201 |(?: # or the v= param in all its forms
202 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
203 (?:\?|\#!?) # the params delimiter ? or # or #!
204 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
205 v=
206 )
207 ))
208 |youtu\.be/ # just youtu.be/xxxx
209 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
210 )
211 )? # all until now is optional -> you can pass the naked ID
212 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
213 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
214 (?(1).+)? # if we found the ID, everything can follow
215 $"""
216 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
217 _formats = {
218 '5': {'ext': 'flv', 'width': 400, 'height': 240},
219 '6': {'ext': 'flv', 'width': 450, 'height': 270},
220 '13': {'ext': '3gp'},
221 '17': {'ext': '3gp', 'width': 176, 'height': 144},
222 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
223 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
224 '34': {'ext': 'flv', 'width': 640, 'height': 360},
225 '35': {'ext': 'flv', 'width': 854, 'height': 480},
226 '36': {'ext': '3gp', 'width': 320, 'height': 240},
227 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
228 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
229 '43': {'ext': 'webm', 'width': 640, 'height': 360},
230 '44': {'ext': 'webm', 'width': 854, 'height': 480},
231 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
232 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
233
234
235 # 3d videos
236 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
237 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
238 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
239 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
240 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
241 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
242 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
243
244 # Apple HTTP Live Streaming
245 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
246 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
247 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
248 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
249 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
250 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
251 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
252
253 # DASH mp4 video
254 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
260 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
264 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
265
266 # Dash mp4 audio
267 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
268 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
269 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
270
271 # Dash webm
272 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
278 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
279 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
290 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
291 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
292 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
293
294 # Dash webm audio
295 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
296 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
297
298 # Dash webm audio with opus inside
299 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
300 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
301 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
302
303 # RTMP (unnamed)
304 '_rtmp': {'protocol': 'rtmp'},
305 }
306
307 IE_NAME = 'youtube'
308 _TESTS = [
309 {
310 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
311 'info_dict': {
312 'id': 'BaW_jenozKc',
313 'ext': 'mp4',
314 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
315 'uploader': 'Philipp Hagemeister',
316 'uploader_id': 'phihag',
317 'upload_date': '20121002',
318 '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 .',
319 'categories': ['Science & Technology'],
320 'like_count': int,
321 'dislike_count': int,
322 }
323 },
324 {
325 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
326 'note': 'Test generic use_cipher_signature video (#897)',
327 'info_dict': {
328 'id': 'UxxajLWwzqY',
329 'ext': 'mp4',
330 'upload_date': '20120506',
331 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
332 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
333 'uploader': 'Icona Pop',
334 'uploader_id': 'IconaPop',
335 }
336 },
337 {
338 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
339 'note': 'Test VEVO video with age protection (#956)',
340 'info_dict': {
341 'id': '07FYdnEawAQ',
342 'ext': 'mp4',
343 'upload_date': '20130703',
344 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
345 'description': 'md5:64249768eec3bc4276236606ea996373',
346 'uploader': 'justintimberlakeVEVO',
347 'uploader_id': 'justintimberlakeVEVO',
348 }
349 },
350 {
351 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
352 'note': 'Embed-only video (#1746)',
353 'info_dict': {
354 'id': 'yZIXLfi8CZQ',
355 'ext': 'mp4',
356 'upload_date': '20120608',
357 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
358 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
359 'uploader': 'SET India',
360 'uploader_id': 'setindia'
361 }
362 },
363 {
364 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
365 'note': '256k DASH audio (format 141) via DASH manifest',
366 'info_dict': {
367 'id': 'a9LDPn-MO4I',
368 'ext': 'm4a',
369 'upload_date': '20121002',
370 'uploader_id': '8KVIDEO',
371 'description': '',
372 'uploader': '8KVIDEO',
373 'title': 'UHDTV TEST 8K VIDEO.mp4'
374 },
375 'params': {
376 'youtube_include_dash_manifest': True,
377 'format': '141',
378 },
379 },
380 # DASH manifest with encrypted signature
381 {
382 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
383 'info_dict': {
384 'id': 'IB3lcPjvWLA',
385 'ext': 'm4a',
386 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
387 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
388 'uploader': 'AfrojackVEVO',
389 'uploader_id': 'AfrojackVEVO',
390 'upload_date': '20131011',
391 },
392 'params': {
393 'youtube_include_dash_manifest': True,
394 'format': '141',
395 },
396 },
397 # JS player signature function name containing $
398 {
399 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
400 'info_dict': {
401 'id': 'nfWlot6h_JM',
402 'ext': 'm4a',
403 'title': 'Taylor Swift - Shake It Off',
404 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
405 'uploader': 'TaylorSwiftVEVO',
406 'uploader_id': 'TaylorSwiftVEVO',
407 'upload_date': '20140818',
408 },
409 'params': {
410 'youtube_include_dash_manifest': True,
411 'format': '141',
412 },
413 },
414 # Controversy video
415 {
416 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
417 'info_dict': {
418 'id': 'T4XJQO3qol8',
419 'ext': 'mp4',
420 'upload_date': '20100909',
421 'uploader': 'The Amazing Atheist',
422 'uploader_id': 'TheAmazingAtheist',
423 'title': 'Burning Everyone\'s Koran',
424 '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',
425 }
426 },
427 # Normal age-gate video (No vevo, embed allowed)
428 {
429 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
430 'info_dict': {
431 'id': 'HtVdAasjOgU',
432 'ext': 'mp4',
433 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
434 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
435 'uploader': 'The Witcher',
436 'uploader_id': 'WitcherGame',
437 'upload_date': '20140605',
438 },
439 },
440 # Age-gate video with encrypted signature
441 {
442 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
443 'info_dict': {
444 'id': '6kLq3WMV1nU',
445 'ext': 'mp4',
446 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
447 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
448 'uploader': 'LloydVEVO',
449 'uploader_id': 'LloydVEVO',
450 'upload_date': '20110629',
451 },
452 },
453 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
454 {
455 'url': '__2ABJjxzNo',
456 'info_dict': {
457 'id': '__2ABJjxzNo',
458 'ext': 'mp4',
459 'upload_date': '20100430',
460 'uploader_id': 'deadmau5',
461 'description': 'md5:12c56784b8032162bb936a5f76d55360',
462 'uploader': 'deadmau5',
463 'title': 'Deadmau5 - Some Chords (HD)',
464 },
465 'expected_warnings': [
466 'DASH manifest missing',
467 ]
468 },
469 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
470 {
471 'url': 'lqQg6PlCWgI',
472 'info_dict': {
473 'id': 'lqQg6PlCWgI',
474 'ext': 'mp4',
475 'upload_date': '20120731',
476 'uploader_id': 'olympic',
477 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
478 'uploader': 'Olympics',
479 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
480 },
481 'params': {
482 'skip_download': 'requires avconv',
483 }
484 },
485 # Non-square pixels
486 {
487 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
488 'info_dict': {
489 'id': '_b-2C3KPAM0',
490 'ext': 'mp4',
491 'stretched_ratio': 16 / 9.,
492 'upload_date': '20110310',
493 'uploader_id': 'AllenMeow',
494 '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',
495 'uploader': '孫艾倫',
496 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
497 },
498 }
499 ]
500
501 def __init__(self, *args, **kwargs):
502 super(YoutubeIE, self).__init__(*args, **kwargs)
503 self._player_cache = {}
504
505 def report_video_info_webpage_download(self, video_id):
506 """Report attempt to download video info webpage."""
507 self.to_screen('%s: Downloading video info webpage' % video_id)
508
509 def report_information_extraction(self, video_id):
510 """Report attempt to extract video information."""
511 self.to_screen('%s: Extracting video information' % video_id)
512
513 def report_unavailable_format(self, video_id, format):
514 """Report extracted video URL."""
515 self.to_screen('%s: Format %s not available' % (video_id, format))
516
517 def report_rtmp_download(self):
518 """Indicate the download will use the RTMP protocol."""
519 self.to_screen('RTMP download detected')
520
521 def _signature_cache_id(self, example_sig):
522 """ Return a string representation of a signature """
523 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
524
525 def _extract_signature_function(self, video_id, player_url, example_sig):
526 id_m = re.match(
527 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
528 player_url)
529 if not id_m:
530 raise ExtractorError('Cannot identify player %r' % player_url)
531 player_type = id_m.group('ext')
532 player_id = id_m.group('id')
533
534 # Read from filesystem cache
535 func_id = '%s_%s_%s' % (
536 player_type, player_id, self._signature_cache_id(example_sig))
537 assert os.path.basename(func_id) == func_id
538
539 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
540 if cache_spec is not None:
541 return lambda s: ''.join(s[i] for i in cache_spec)
542
543 download_note = (
544 'Downloading player %s' % player_url
545 if self._downloader.params.get('verbose') else
546 'Downloading %s player %s' % (player_type, player_id)
547 )
548 if player_type == 'js':
549 code = self._download_webpage(
550 player_url, video_id,
551 note=download_note,
552 errnote='Download of %s failed' % player_url)
553 res = self._parse_sig_js(code)
554 elif player_type == 'swf':
555 urlh = self._request_webpage(
556 player_url, video_id,
557 note=download_note,
558 errnote='Download of %s failed' % player_url)
559 code = urlh.read()
560 res = self._parse_sig_swf(code)
561 else:
562 assert False, 'Invalid player type %r' % player_type
563
564 test_string = ''.join(map(compat_chr, range(len(example_sig))))
565 cache_res = res(test_string)
566 cache_spec = [ord(c) for c in cache_res]
567
568 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
569 return res
570
571 def _print_sig_code(self, func, example_sig):
572 def gen_sig_code(idxs):
573 def _genslice(start, end, step):
574 starts = '' if start == 0 else str(start)
575 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
576 steps = '' if step == 1 else (':%d' % step)
577 return 's[%s%s%s]' % (starts, ends, steps)
578
579 step = None
580 # Quelch pyflakes warnings - start will be set when step is set
581 start = '(Never used)'
582 for i, prev in zip(idxs[1:], idxs[:-1]):
583 if step is not None:
584 if i - prev == step:
585 continue
586 yield _genslice(start, prev, step)
587 step = None
588 continue
589 if i - prev in [-1, 1]:
590 step = i - prev
591 start = prev
592 continue
593 else:
594 yield 's[%d]' % prev
595 if step is None:
596 yield 's[%d]' % i
597 else:
598 yield _genslice(start, i, step)
599
600 test_string = ''.join(map(compat_chr, range(len(example_sig))))
601 cache_res = func(test_string)
602 cache_spec = [ord(c) for c in cache_res]
603 expr_code = ' + '.join(gen_sig_code(cache_spec))
604 signature_id_tuple = '(%s)' % (
605 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
606 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
607 ' return %s\n') % (signature_id_tuple, expr_code)
608 self.to_screen('Extracted signature function:\n' + code)
609
610 def _parse_sig_js(self, jscode):
611 funcname = self._search_regex(
612 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
613 'Initial JS player signature function name')
614
615 jsi = JSInterpreter(jscode)
616 initial_function = jsi.extract_function(funcname)
617 return lambda s: initial_function([s])
618
619 def _parse_sig_swf(self, file_contents):
620 swfi = SWFInterpreter(file_contents)
621 TARGET_CLASSNAME = 'SignatureDecipher'
622 searched_class = swfi.extract_class(TARGET_CLASSNAME)
623 initial_function = swfi.extract_function(searched_class, 'decipher')
624 return lambda s: initial_function([s])
625
626 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
627 """Turn the encrypted s field into a working signature"""
628
629 if player_url is None:
630 raise ExtractorError('Cannot decrypt signature without player_url')
631
632 if player_url.startswith('//'):
633 player_url = 'https:' + player_url
634 try:
635 player_id = (player_url, self._signature_cache_id(s))
636 if player_id not in self._player_cache:
637 func = self._extract_signature_function(
638 video_id, player_url, s
639 )
640 self._player_cache[player_id] = func
641 func = self._player_cache[player_id]
642 if self._downloader.params.get('youtube_print_sig_code'):
643 self._print_sig_code(func, s)
644 return func(s)
645 except Exception as e:
646 tb = traceback.format_exc()
647 raise ExtractorError(
648 'Signature extraction failed: ' + tb, cause=e)
649
650 def _get_subtitles(self, video_id, webpage):
651 try:
652 subs_doc = self._download_xml(
653 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
654 video_id, note=False)
655 except ExtractorError as err:
656 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
657 return {}
658
659 sub_lang_list = {}
660 for track in subs_doc.findall('track'):
661 lang = track.attrib['lang_code']
662 if lang in sub_lang_list:
663 continue
664 sub_formats = []
665 for ext in ['sbv', 'vtt', 'srt']:
666 params = compat_urllib_parse.urlencode({
667 'lang': lang,
668 'v': video_id,
669 'fmt': ext,
670 'name': track.attrib['name'].encode('utf-8'),
671 })
672 sub_formats.append({
673 'url': 'https://www.youtube.com/api/timedtext?' + params,
674 'ext': ext,
675 })
676 sub_lang_list[lang] = sub_formats
677 if not sub_lang_list:
678 self._downloader.report_warning('video doesn\'t have subtitles')
679 return {}
680 return sub_lang_list
681
682 def _get_automatic_captions(self, video_id, webpage):
683 """We need the webpage for getting the captions url, pass it as an
684 argument to speed up the process."""
685 self.to_screen('%s: Looking for automatic captions' % video_id)
686 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
687 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
688 if mobj is None:
689 self._downloader.report_warning(err_msg)
690 return {}
691 player_config = json.loads(mobj.group(1))
692 try:
693 args = player_config['args']
694 caption_url = args['ttsurl']
695 timestamp = args['timestamp']
696 # We get the available subtitles
697 list_params = compat_urllib_parse.urlencode({
698 'type': 'list',
699 'tlangs': 1,
700 'asrs': 1,
701 })
702 list_url = caption_url + '&' + list_params
703 caption_list = self._download_xml(list_url, video_id)
704 original_lang_node = caption_list.find('track')
705 if original_lang_node is None:
706 self._downloader.report_warning('Video doesn\'t have automatic captions')
707 return {}
708 original_lang = original_lang_node.attrib['lang_code']
709 caption_kind = original_lang_node.attrib.get('kind', '')
710
711 sub_lang_list = {}
712 for lang_node in caption_list.findall('target'):
713 sub_lang = lang_node.attrib['lang_code']
714 sub_formats = []
715 for ext in ['sbv', 'vtt', 'srt']:
716 params = compat_urllib_parse.urlencode({
717 'lang': original_lang,
718 'tlang': sub_lang,
719 'fmt': ext,
720 'ts': timestamp,
721 'kind': caption_kind,
722 })
723 sub_formats.append({
724 'url': caption_url + '&' + params,
725 'ext': ext,
726 })
727 sub_lang_list[sub_lang] = sub_formats
728 return sub_lang_list
729 # An extractor error can be raise by the download process if there are
730 # no automatic captions but there are subtitles
731 except (KeyError, ExtractorError):
732 self._downloader.report_warning(err_msg)
733 return {}
734
735 @classmethod
736 def extract_id(cls, url):
737 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
738 if mobj is None:
739 raise ExtractorError('Invalid URL: %s' % url)
740 video_id = mobj.group(2)
741 return video_id
742
743 def _extract_from_m3u8(self, manifest_url, video_id):
744 url_map = {}
745
746 def _get_urls(_manifest):
747 lines = _manifest.split('\n')
748 urls = filter(lambda l: l and not l.startswith('#'),
749 lines)
750 return urls
751 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
752 formats_urls = _get_urls(manifest)
753 for format_url in formats_urls:
754 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
755 url_map[itag] = format_url
756 return url_map
757
758 def _extract_annotations(self, video_id):
759 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
760 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
761
762 def _parse_dash_manifest(
763 self, video_id, dash_manifest_url, player_url, age_gate):
764 def decrypt_sig(mobj):
765 s = mobj.group(1)
766 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
767 return '/signature/%s' % dec_s
768 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
769 dash_doc = self._download_xml(
770 dash_manifest_url, video_id,
771 note='Downloading DASH manifest',
772 errnote='Could not download DASH manifest')
773
774 formats = []
775 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
776 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
777 if url_el is None:
778 continue
779 format_id = r.attrib['id']
780 video_url = url_el.text
781 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
782 f = {
783 'format_id': format_id,
784 'url': video_url,
785 'width': int_or_none(r.attrib.get('width')),
786 'height': int_or_none(r.attrib.get('height')),
787 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
788 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
789 'filesize': filesize,
790 'fps': int_or_none(r.attrib.get('frameRate')),
791 }
792 try:
793 existing_format = next(
794 fo for fo in formats
795 if fo['format_id'] == format_id)
796 except StopIteration:
797 full_info = self._formats.get(format_id, {}).copy()
798 full_info.update(f)
799 formats.append(full_info)
800 else:
801 existing_format.update(f)
802 return formats
803
804 def _real_extract(self, url):
805 proto = (
806 'http' if self._downloader.params.get('prefer_insecure', False)
807 else 'https')
808
809 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
810 mobj = re.search(self._NEXT_URL_RE, url)
811 if mobj:
812 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
813 video_id = self.extract_id(url)
814
815 # Get video webpage
816 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
817 video_webpage = self._download_webpage(url, video_id)
818
819 # Attempt to extract SWF player URL
820 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
821 if mobj is not None:
822 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
823 else:
824 player_url = None
825
826 # Get video info
827 embed_webpage = None
828 if re.search(r'player-age-gate-content">', video_webpage) is not None:
829 age_gate = True
830 # We simulate the access to the video from www.youtube.com/v/{video_id}
831 # this can be viewed without login into Youtube
832 url = proto + '://www.youtube.com/embed/%s' % video_id
833 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
834 data = compat_urllib_parse.urlencode({
835 'video_id': video_id,
836 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
837 'sts': self._search_regex(
838 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
839 })
840 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
841 video_info_webpage = self._download_webpage(
842 video_info_url, video_id,
843 note='Refetching age-gated info webpage',
844 errnote='unable to download video info webpage')
845 video_info = compat_parse_qs(video_info_webpage)
846 else:
847 age_gate = False
848 try:
849 # Try looking directly into the video webpage
850 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
851 if not mobj:
852 raise ValueError('Could not find ytplayer.config') # caught below
853 json_code = uppercase_escape(mobj.group(1))
854 ytplayer_config = json.loads(json_code)
855 args = ytplayer_config['args']
856 # Convert to the same format returned by compat_parse_qs
857 video_info = dict((k, [v]) for k, v in args.items())
858 if 'url_encoded_fmt_stream_map' not in args:
859 raise ValueError('No stream_map present') # caught below
860 except ValueError:
861 # We fallback to the get_video_info pages (used by the embed page)
862 self.report_video_info_webpage_download(video_id)
863 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
864 video_info_url = (
865 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
866 % (proto, video_id, el_type))
867 video_info_webpage = self._download_webpage(
868 video_info_url,
869 video_id, note=False,
870 errnote='unable to download video info webpage')
871 video_info = compat_parse_qs(video_info_webpage)
872 if 'token' in video_info:
873 break
874 if 'token' not in video_info:
875 if 'reason' in video_info:
876 raise ExtractorError(
877 'YouTube said: %s' % video_info['reason'][0],
878 expected=True, video_id=video_id)
879 else:
880 raise ExtractorError(
881 '"token" parameter not in video info for unknown reason',
882 video_id=video_id)
883
884 if 'view_count' in video_info:
885 view_count = int(video_info['view_count'][0])
886 else:
887 view_count = None
888
889 # Check for "rental" videos
890 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
891 raise ExtractorError('"rental" videos not supported')
892
893 # Start extracting information
894 self.report_information_extraction(video_id)
895
896 # uploader
897 if 'author' not in video_info:
898 raise ExtractorError('Unable to extract uploader name')
899 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
900
901 # uploader_id
902 video_uploader_id = None
903 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
904 if mobj is not None:
905 video_uploader_id = mobj.group(1)
906 else:
907 self._downloader.report_warning('unable to extract uploader nickname')
908
909 # title
910 if 'title' in video_info:
911 video_title = video_info['title'][0]
912 else:
913 self._downloader.report_warning('Unable to extract video title')
914 video_title = '_'
915
916 # thumbnail image
917 # We try first to get a high quality image:
918 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
919 video_webpage, re.DOTALL)
920 if m_thumb is not None:
921 video_thumbnail = m_thumb.group(1)
922 elif 'thumbnail_url' not in video_info:
923 self._downloader.report_warning('unable to extract video thumbnail')
924 video_thumbnail = None
925 else: # don't panic if we can't find it
926 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
927
928 # upload date
929 upload_date = None
930 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
931 if mobj is None:
932 mobj = re.search(
933 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
934 video_webpage)
935 if mobj is not None:
936 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
937 upload_date = unified_strdate(upload_date)
938
939 m_cat_container = self._search_regex(
940 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
941 video_webpage, 'categories', default=None)
942 if m_cat_container:
943 category = self._html_search_regex(
944 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
945 default=None)
946 video_categories = None if category is None else [category]
947 else:
948 video_categories = None
949
950 # description
951 video_description = get_element_by_id("eow-description", video_webpage)
952 if video_description:
953 video_description = re.sub(r'''(?x)
954 <a\s+
955 (?:[a-zA-Z-]+="[^"]+"\s+)*?
956 title="([^"]+)"\s+
957 (?:[a-zA-Z-]+="[^"]+"\s+)*?
958 class="yt-uix-redirect-link"\s*>
959 [^<]+
960 </a>
961 ''', r'\1', video_description)
962 video_description = clean_html(video_description)
963 else:
964 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
965 if fd_mobj:
966 video_description = unescapeHTML(fd_mobj.group(1))
967 else:
968 video_description = ''
969
970 def _extract_count(count_name):
971 count = self._search_regex(
972 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
973 video_webpage, count_name, default=None)
974 if count is not None:
975 return int(count.replace(',', ''))
976 return None
977 like_count = _extract_count('like')
978 dislike_count = _extract_count('dislike')
979
980 # subtitles
981 video_subtitles = self.extract_subtitles(video_id, video_webpage)
982 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
983
984 if 'length_seconds' not in video_info:
985 self._downloader.report_warning('unable to extract video duration')
986 video_duration = None
987 else:
988 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
989
990 # annotations
991 video_annotations = None
992 if self._downloader.params.get('writeannotations', False):
993 video_annotations = self._extract_annotations(video_id)
994
995 def _map_to_format_list(urlmap):
996 formats = []
997 for itag, video_real_url in urlmap.items():
998 dct = {
999 'format_id': itag,
1000 'url': video_real_url,
1001 'player_url': player_url,
1002 }
1003 if itag in self._formats:
1004 dct.update(self._formats[itag])
1005 formats.append(dct)
1006 return formats
1007
1008 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1009 self.report_rtmp_download()
1010 formats = [{
1011 'format_id': '_rtmp',
1012 'protocol': 'rtmp',
1013 'url': video_info['conn'][0],
1014 'player_url': player_url,
1015 }]
1016 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1017 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1018 if 'rtmpe%3Dyes' in encoded_url_map:
1019 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1020 url_map = {}
1021 for url_data_str in encoded_url_map.split(','):
1022 url_data = compat_parse_qs(url_data_str)
1023 if 'itag' not in url_data or 'url' not in url_data:
1024 continue
1025 format_id = url_data['itag'][0]
1026 url = url_data['url'][0]
1027
1028 if 'sig' in url_data:
1029 url += '&signature=' + url_data['sig'][0]
1030 elif 's' in url_data:
1031 encrypted_sig = url_data['s'][0]
1032 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1033
1034 jsplayer_url_json = self._search_regex(
1035 ASSETS_RE,
1036 embed_webpage if age_gate else video_webpage,
1037 'JS player URL (1)', default=None)
1038 if not jsplayer_url_json and not age_gate:
1039 # We need the embed website after all
1040 if embed_webpage is None:
1041 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1042 embed_webpage = self._download_webpage(
1043 embed_url, video_id, 'Downloading embed webpage')
1044 jsplayer_url_json = self._search_regex(
1045 ASSETS_RE, embed_webpage, 'JS player URL')
1046
1047 player_url = json.loads(jsplayer_url_json)
1048 if player_url is None:
1049 player_url_json = self._search_regex(
1050 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1051 video_webpage, 'age gate player URL')
1052 player_url = json.loads(player_url_json)
1053
1054 if self._downloader.params.get('verbose'):
1055 if player_url is None:
1056 player_version = 'unknown'
1057 player_desc = 'unknown'
1058 else:
1059 if player_url.endswith('swf'):
1060 player_version = self._search_regex(
1061 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1062 'flash player', fatal=False)
1063 player_desc = 'flash player %s' % player_version
1064 else:
1065 player_version = self._search_regex(
1066 r'html5player-([^/]+?)(?:/html5player)?\.js',
1067 player_url,
1068 'html5 player', fatal=False)
1069 player_desc = 'html5 player %s' % player_version
1070
1071 parts_sizes = self._signature_cache_id(encrypted_sig)
1072 self.to_screen('{%s} signature length %s, %s' %
1073 (format_id, parts_sizes, player_desc))
1074
1075 signature = self._decrypt_signature(
1076 encrypted_sig, video_id, player_url, age_gate)
1077 url += '&signature=' + signature
1078 if 'ratebypass' not in url:
1079 url += '&ratebypass=yes'
1080 url_map[format_id] = url
1081 formats = _map_to_format_list(url_map)
1082 elif video_info.get('hlsvp'):
1083 manifest_url = video_info['hlsvp'][0]
1084 url_map = self._extract_from_m3u8(manifest_url, video_id)
1085 formats = _map_to_format_list(url_map)
1086 else:
1087 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1088
1089 # Look for the DASH manifest
1090 if self._downloader.params.get('youtube_include_dash_manifest', True):
1091 dash_mpd = video_info.get('dashmpd')
1092 if dash_mpd:
1093 dash_manifest_url = dash_mpd[0]
1094 try:
1095 dash_formats = self._parse_dash_manifest(
1096 video_id, dash_manifest_url, player_url, age_gate)
1097 except (ExtractorError, KeyError) as e:
1098 self.report_warning(
1099 'Skipping DASH manifest: %r' % e, video_id)
1100 else:
1101 # Hide the formats we found through non-DASH
1102 dash_keys = set(df['format_id'] for df in dash_formats)
1103 for f in formats:
1104 if f['format_id'] in dash_keys:
1105 f['format_id'] = 'nondash-%s' % f['format_id']
1106 f['preference'] = f.get('preference', 0) - 10000
1107 formats.extend(dash_formats)
1108
1109 # Check for malformed aspect ratio
1110 stretched_m = re.search(
1111 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1112 video_webpage)
1113 if stretched_m:
1114 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1115 for f in formats:
1116 if f.get('vcodec') != 'none':
1117 f['stretched_ratio'] = ratio
1118
1119 self._sort_formats(formats)
1120
1121 return {
1122 'id': video_id,
1123 'uploader': video_uploader,
1124 'uploader_id': video_uploader_id,
1125 'upload_date': upload_date,
1126 'title': video_title,
1127 'thumbnail': video_thumbnail,
1128 'description': video_description,
1129 'categories': video_categories,
1130 'subtitles': video_subtitles,
1131 'automatic_captions': automatic_captions,
1132 'duration': video_duration,
1133 'age_limit': 18 if age_gate else 0,
1134 'annotations': video_annotations,
1135 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1136 'view_count': view_count,
1137 'like_count': like_count,
1138 'dislike_count': dislike_count,
1139 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1140 'formats': formats,
1141 }
1142
1143
1144 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1145 IE_DESC = 'YouTube.com playlists'
1146 _VALID_URL = r"""(?x)(?:
1147 (?:https?://)?
1148 (?:\w+\.)?
1149 youtube\.com/
1150 (?:
1151 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1152 \? (?:.*?&)*? (?:p|a|list)=
1153 | p/
1154 )
1155 (
1156 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1157 # Top tracks, they can also include dots
1158 |(?:MC)[\w\.]*
1159 )
1160 .*
1161 |
1162 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1163 )"""
1164 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1165 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
1166 IE_NAME = 'youtube:playlist'
1167 _TESTS = [{
1168 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1169 'info_dict': {
1170 'title': 'ytdl test PL',
1171 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1172 },
1173 'playlist_count': 3,
1174 }, {
1175 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1176 'info_dict': {
1177 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1178 'title': 'YDL_Empty_List',
1179 },
1180 'playlist_count': 0,
1181 }, {
1182 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1183 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1184 'info_dict': {
1185 'title': '29C3: Not my department',
1186 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1187 },
1188 'playlist_count': 95,
1189 }, {
1190 'note': 'issue #673',
1191 'url': 'PLBB231211A4F62143',
1192 'info_dict': {
1193 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1194 'id': 'PLBB231211A4F62143',
1195 },
1196 'playlist_mincount': 26,
1197 }, {
1198 'note': 'Large playlist',
1199 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1200 'info_dict': {
1201 'title': 'Uploads from Cauchemar',
1202 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1203 },
1204 'playlist_mincount': 799,
1205 }, {
1206 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1207 'info_dict': {
1208 'title': 'YDL_safe_search',
1209 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1210 },
1211 'playlist_count': 2,
1212 }, {
1213 'note': 'embedded',
1214 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1215 'playlist_count': 4,
1216 'info_dict': {
1217 'title': 'JODA15',
1218 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1219 }
1220 }, {
1221 'note': 'Embedded SWF player',
1222 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1223 'playlist_count': 4,
1224 'info_dict': {
1225 'title': 'JODA7',
1226 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1227 }
1228 }, {
1229 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1230 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1231 'info_dict': {
1232 'title': 'Uploads from Interstellar Movie',
1233 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1234 },
1235 'playlist_mincout': 21,
1236 }]
1237
1238 def _real_initialize(self):
1239 self._login()
1240
1241 def _ids_to_results(self, ids):
1242 return [
1243 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1244 for vid_id in ids]
1245
1246 def _extract_mix(self, playlist_id):
1247 # The mixes are generated from a single video
1248 # the id of the playlist is just 'RD' + video_id
1249 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1250 webpage = self._download_webpage(
1251 url, playlist_id, 'Downloading Youtube mix')
1252 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1253 title_span = (
1254 search_title('playlist-title') or
1255 search_title('title long-title') or
1256 search_title('title'))
1257 title = clean_html(title_span)
1258 ids = orderedSet(re.findall(
1259 r'''(?xs)data-video-username=".*?".*?
1260 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
1261 webpage))
1262 url_results = self._ids_to_results(ids)
1263
1264 return self.playlist_result(url_results, playlist_id, title)
1265
1266 def _real_extract(self, url):
1267 # Extract playlist id
1268 mobj = re.match(self._VALID_URL, url)
1269 if mobj is None:
1270 raise ExtractorError('Invalid URL: %s' % url)
1271 playlist_id = mobj.group(1) or mobj.group(2)
1272
1273 # Check if it's a video-specific URL
1274 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1275 if 'v' in query_dict:
1276 video_id = query_dict['v'][0]
1277 if self._downloader.params.get('noplaylist'):
1278 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1279 return self.url_result(video_id, 'Youtube', video_id=video_id)
1280 else:
1281 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1282
1283 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1284 # Mixes require a custom extraction process
1285 return self._extract_mix(playlist_id)
1286
1287 url = self._TEMPLATE_URL % playlist_id
1288 page = self._download_webpage(url, playlist_id)
1289 more_widget_html = content_html = page
1290
1291 # Check if the playlist exists or is private
1292 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1293 raise ExtractorError(
1294 'The playlist doesn\'t exist or is private, use --username or '
1295 '--netrc to access it.',
1296 expected=True)
1297
1298 # Extract the video ids from the playlist pages
1299 ids = []
1300
1301 for page_num in itertools.count(1):
1302 matches = re.finditer(self._VIDEO_RE, content_html)
1303 # We remove the duplicates and the link with index 0
1304 # (it's not the first video of the playlist)
1305 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1306 ids.extend(new_ids)
1307
1308 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1309 if not mobj:
1310 break
1311
1312 more = self._download_json(
1313 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1314 'Downloading page #%s' % page_num,
1315 transform_source=uppercase_escape)
1316 content_html = more['content_html']
1317 if not content_html.strip():
1318 # Some webpages show a "Load more" button but they don't
1319 # have more videos
1320 break
1321 more_widget_html = more['load_more_widget_html']
1322
1323 playlist_title = self._html_search_regex(
1324 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1325 page, 'title')
1326
1327 url_results = self._ids_to_results(ids)
1328 return self.playlist_result(url_results, playlist_id, playlist_title)
1329
1330
1331 class YoutubeChannelIE(InfoExtractor):
1332 IE_DESC = 'YouTube.com channels'
1333 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1334 IE_NAME = 'youtube:channel'
1335 _TESTS = [{
1336 'note': 'paginated channel',
1337 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1338 'playlist_mincount': 91,
1339 'info_dict': {
1340 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1341 }
1342 }]
1343
1344 def extract_videos_from_page(self, page):
1345 ids_in_page = []
1346 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1347 if mobj.group(1) not in ids_in_page:
1348 ids_in_page.append(mobj.group(1))
1349 return ids_in_page
1350
1351 def _real_extract(self, url):
1352 channel_id = self._match_id(url)
1353
1354 video_ids = []
1355 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1356 channel_page = self._download_webpage(url, channel_id)
1357 autogenerated = re.search(r'''(?x)
1358 class="[^"]*?(?:
1359 channel-header-autogenerated-label|
1360 yt-channel-title-autogenerated
1361 )[^"]*"''', channel_page) is not None
1362
1363 if autogenerated:
1364 # The videos are contained in a single page
1365 # the ajax pages can't be used, they are empty
1366 video_ids = self.extract_videos_from_page(channel_page)
1367 entries = [
1368 self.url_result(video_id, 'Youtube', video_id=video_id)
1369 for video_id in video_ids]
1370 return self.playlist_result(entries, channel_id)
1371
1372 def _entries():
1373 more_widget_html = content_html = channel_page
1374 for pagenum in itertools.count(1):
1375
1376 ids_in_page = self.extract_videos_from_page(content_html)
1377 for video_id in ids_in_page:
1378 yield self.url_result(
1379 video_id, 'Youtube', video_id=video_id)
1380
1381 mobj = re.search(
1382 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1383 more_widget_html)
1384 if not mobj:
1385 break
1386
1387 more = self._download_json(
1388 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1389 'Downloading page #%s' % (pagenum + 1),
1390 transform_source=uppercase_escape)
1391 content_html = more['content_html']
1392 more_widget_html = more['load_more_widget_html']
1393
1394 return self.playlist_result(_entries(), channel_id)
1395
1396
1397 class YoutubeUserIE(InfoExtractor):
1398 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1399 _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_-]+)'
1400 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1401 _GDATA_PAGE_SIZE = 50
1402 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1403 IE_NAME = 'youtube:user'
1404
1405 _TESTS = [{
1406 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1407 'playlist_mincount': 320,
1408 'info_dict': {
1409 'title': 'TheLinuxFoundation',
1410 }
1411 }, {
1412 'url': 'ytuser:phihag',
1413 'only_matching': True,
1414 }]
1415
1416 @classmethod
1417 def suitable(cls, url):
1418 # Don't return True if the url can be extracted with other youtube
1419 # extractor, the regex would is too permissive and it would match.
1420 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1421 if any(ie.suitable(url) for ie in other_ies):
1422 return False
1423 else:
1424 return super(YoutubeUserIE, cls).suitable(url)
1425
1426 def _real_extract(self, url):
1427 username = self._match_id(url)
1428
1429 # Download video ids using YouTube Data API. Result size per
1430 # query is limited (currently to 50 videos) so we need to query
1431 # page by page until there are no video ids - it means we got
1432 # all of them.
1433
1434 def download_page(pagenum):
1435 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1436
1437 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1438 page = self._download_webpage(
1439 gdata_url, username,
1440 'Downloading video ids from %d to %d' % (
1441 start_index, start_index + self._GDATA_PAGE_SIZE))
1442
1443 try:
1444 response = json.loads(page)
1445 except ValueError as err:
1446 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1447 if 'entry' not in response['feed']:
1448 return
1449
1450 # Extract video identifiers
1451 entries = response['feed']['entry']
1452 for entry in entries:
1453 title = entry['title']['$t']
1454 video_id = entry['id']['$t'].split('/')[-1]
1455 yield {
1456 '_type': 'url',
1457 'url': video_id,
1458 'ie_key': 'Youtube',
1459 'id': video_id,
1460 'title': title,
1461 }
1462 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1463
1464 return self.playlist_result(url_results, playlist_title=username)
1465
1466
1467 class YoutubeSearchIE(SearchInfoExtractor):
1468 IE_DESC = 'YouTube.com searches'
1469 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1470 _MAX_RESULTS = 1000
1471 IE_NAME = 'youtube:search'
1472 _SEARCH_KEY = 'ytsearch'
1473
1474 def _get_n_results(self, query, n):
1475 """Get a specified number of results for a query"""
1476
1477 video_ids = []
1478 pagenum = 0
1479 limit = n
1480 PAGE_SIZE = 50
1481
1482 while (PAGE_SIZE * pagenum) < limit:
1483 result_url = self._API_URL % (
1484 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1485 (PAGE_SIZE * pagenum) + 1)
1486 data_json = self._download_webpage(
1487 result_url, video_id='query "%s"' % query,
1488 note='Downloading page %s' % (pagenum + 1),
1489 errnote='Unable to download API page')
1490 data = json.loads(data_json)
1491 api_response = data['data']
1492
1493 if 'items' not in api_response:
1494 raise ExtractorError(
1495 '[youtube] No video results', expected=True)
1496
1497 new_ids = list(video['id'] for video in api_response['items'])
1498 video_ids += new_ids
1499
1500 limit = min(n, api_response['totalItems'])
1501 pagenum += 1
1502
1503 if len(video_ids) > n:
1504 video_ids = video_ids[:n]
1505 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1506 for video_id in video_ids]
1507 return self.playlist_result(videos, query)
1508
1509
1510 class YoutubeSearchDateIE(YoutubeSearchIE):
1511 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1512 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1513 _SEARCH_KEY = 'ytsearchdate'
1514 IE_DESC = 'YouTube.com searches, newest videos first'
1515
1516
1517 class YoutubeSearchURLIE(InfoExtractor):
1518 IE_DESC = 'YouTube.com search URLs'
1519 IE_NAME = 'youtube:search_url'
1520 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1521 _TESTS = [{
1522 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1523 'playlist_mincount': 5,
1524 'info_dict': {
1525 'title': 'youtube-dl test video',
1526 }
1527 }]
1528
1529 def _real_extract(self, url):
1530 mobj = re.match(self._VALID_URL, url)
1531 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1532
1533 webpage = self._download_webpage(url, query)
1534 result_code = self._search_regex(
1535 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1536
1537 part_codes = re.findall(
1538 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1539 entries = []
1540 for part_code in part_codes:
1541 part_title = self._html_search_regex(
1542 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1543 part_url_snippet = self._html_search_regex(
1544 r'(?s)href="([^"]+)"', part_code, 'item URL')
1545 part_url = compat_urlparse.urljoin(
1546 'https://www.youtube.com/', part_url_snippet)
1547 entries.append({
1548 '_type': 'url',
1549 'url': part_url,
1550 'title': part_title,
1551 })
1552
1553 return {
1554 '_type': 'playlist',
1555 'entries': entries,
1556 'title': query,
1557 }
1558
1559
1560 class YoutubeShowIE(InfoExtractor):
1561 IE_DESC = 'YouTube.com (multi-season) shows'
1562 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1563 IE_NAME = 'youtube:show'
1564 _TESTS = [{
1565 'url': 'http://www.youtube.com/show/airdisasters',
1566 'playlist_mincount': 3,
1567 'info_dict': {
1568 'id': 'airdisasters',
1569 'title': 'Air Disasters',
1570 }
1571 }]
1572
1573 def _real_extract(self, url):
1574 mobj = re.match(self._VALID_URL, url)
1575 playlist_id = mobj.group('id')
1576 webpage = self._download_webpage(
1577 url, playlist_id, 'Downloading show webpage')
1578 # There's one playlist for each season of the show
1579 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1580 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1581 entries = [
1582 self.url_result(
1583 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1584 for season in m_seasons
1585 ]
1586 title = self._og_search_title(webpage, fatal=False)
1587
1588 return {
1589 '_type': 'playlist',
1590 'id': playlist_id,
1591 'title': title,
1592 'entries': entries,
1593 }
1594
1595
1596 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1597 """
1598 Base class for extractors that fetch info from
1599 http://www.youtube.com/feed_ajax
1600 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1601 """
1602 _LOGIN_REQUIRED = True
1603 # use action_load_personal_feed instead of action_load_system_feed
1604 _PERSONAL_FEED = False
1605
1606 @property
1607 def _FEED_TEMPLATE(self):
1608 action = 'action_load_system_feed'
1609 if self._PERSONAL_FEED:
1610 action = 'action_load_personal_feed'
1611 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1612
1613 @property
1614 def IE_NAME(self):
1615 return 'youtube:%s' % self._FEED_NAME
1616
1617 def _real_initialize(self):
1618 self._login()
1619
1620 def _real_extract(self, url):
1621 feed_entries = []
1622 paging = 0
1623 for i in itertools.count(1):
1624 info = self._download_json(
1625 self._FEED_TEMPLATE % paging,
1626 '%s feed' % self._FEED_NAME,
1627 'Downloading page %s' % i,
1628 transform_source=uppercase_escape)
1629 feed_html = info.get('feed_html') or info.get('content_html')
1630 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1631 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1632 ids = orderedSet(m.group(1) for m in m_ids)
1633 feed_entries.extend(
1634 self.url_result(video_id, 'Youtube', video_id=video_id)
1635 for video_id in ids)
1636 mobj = re.search(
1637 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1638 load_more_widget_html)
1639 if mobj is None:
1640 break
1641 paging = mobj.group('paging')
1642 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1643
1644
1645 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1646 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1647 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1648 _FEED_NAME = 'recommended'
1649 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1650
1651
1652 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1653 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1654 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1655 _FEED_NAME = 'watch_later'
1656 _PLAYLIST_TITLE = 'Youtube Watch Later'
1657 _PERSONAL_FEED = True
1658
1659
1660 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1661 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1662 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1663 _FEED_NAME = 'history'
1664 _PERSONAL_FEED = True
1665 _PLAYLIST_TITLE = 'Youtube Watch History'
1666
1667
1668 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1669 IE_NAME = 'youtube:favorites'
1670 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1671 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1672 _LOGIN_REQUIRED = True
1673
1674 def _real_extract(self, url):
1675 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1676 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1677 return self.url_result(playlist_id, 'YoutubePlaylist')
1678
1679
1680 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1681 IE_NAME = 'youtube:subscriptions'
1682 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1683 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1684 _TESTS = []
1685
1686 def _real_extract(self, url):
1687 title = 'Youtube Subscriptions'
1688 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1689
1690 # The extraction process is the same as for playlists, but the regex
1691 # for the video ids doesn't contain an index
1692 ids = []
1693 more_widget_html = content_html = page
1694
1695 for page_num in itertools.count(1):
1696 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1697 new_ids = orderedSet(matches)
1698 ids.extend(new_ids)
1699
1700 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1701 if not mobj:
1702 break
1703
1704 more = self._download_json(
1705 'https://youtube.com/%s' % mobj.group('more'), title,
1706 'Downloading page #%s' % page_num,
1707 transform_source=uppercase_escape)
1708 content_html = more['content_html']
1709 more_widget_html = more['load_more_widget_html']
1710
1711 return {
1712 '_type': 'playlist',
1713 'title': title,
1714 'entries': self._ids_to_results(ids),
1715 }
1716
1717
1718 class YoutubeTruncatedURLIE(InfoExtractor):
1719 IE_NAME = 'youtube:truncated_url'
1720 IE_DESC = False # Do not list
1721 _VALID_URL = r'''(?x)
1722 (?:https?://)?
1723 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1724 (?:watch\?(?:
1725 feature=[a-z_]+|
1726 annotation_id=annotation_[^&]+|
1727 x-yt-cl=[0-9]+|
1728 hl=[^&]*|
1729 )?
1730 |
1731 attribution_link\?a=[^&]+
1732 )
1733 $
1734 '''
1735
1736 _TESTS = [{
1737 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1738 'only_matching': True,
1739 }, {
1740 'url': 'http://www.youtube.com/watch?',
1741 'only_matching': True,
1742 }, {
1743 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1744 'only_matching': True,
1745 }, {
1746 'url': 'https://www.youtube.com/watch?feature=foo',
1747 'only_matching': True,
1748 }, {
1749 'url': 'https://www.youtube.com/watch?hl=en-GB',
1750 'only_matching': True,
1751 }]
1752
1753 def _real_extract(self, url):
1754 raise ExtractorError(
1755 'Did you forget to quote the URL? Remember that & is a meta '
1756 'character in most shells, so you want to put the URL in quotes, '
1757 'like youtube-dl '
1758 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1759 ' or simply youtube-dl BaW_jenozKc .',
1760 expected=True)
1761
1762
1763 class YoutubeTruncatedIDIE(InfoExtractor):
1764 IE_NAME = 'youtube:truncated_id'
1765 IE_DESC = False # Do not list
1766 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1767
1768 _TESTS = [{
1769 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1770 'only_matching': True,
1771 }]
1772
1773 def _real_extract(self, url):
1774 video_id = self._match_id(url)
1775 raise ExtractorError(
1776 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
1777 expected=True)