]> jfr.im git - yt-dlp.git/blame - youtube_dl/extractor/youtube.py
[youtube:channel] Specify first page download message
[yt-dlp.git] / youtube_dl / extractor / youtube.py
CommitLineData
c5e8d7af 1# coding: utf-8
c5e8d7af 2
78caa52a
PH
3from __future__ import unicode_literals
4
5
0ca96d48 6import itertools
c5e8d7af 7import json
c4417ddb 8import os.path
c5e8d7af 9import re
42939b61 10import time
e0df6211 11import traceback
c5e8d7af 12
b05654f0 13from .common import InfoExtractor, SearchInfoExtractor
2b25cb5d 14from ..jsinterp import JSInterpreter
54256267 15from ..swfinterp import SWFInterpreter
4bb4a188 16from ..compat import (
edf3e38e 17 compat_chr,
c5e8d7af 18 compat_parse_qs,
c5e8d7af
PH
19 compat_urllib_parse,
20 compat_urllib_request,
7c61bd36 21 compat_urlparse,
c5e8d7af 22 compat_str,
4bb4a188
PH
23)
24from ..utils import (
c5e8d7af 25 clean_html,
c5e8d7af 26 ExtractorError,
2d30521a 27 float_or_none,
4bb4a188
PH
28 get_element_by_attribute,
29 get_element_by_id,
dd27fd17 30 int_or_none,
9c44d242 31 OnDemandPagedList,
4bb4a188 32 orderedSet,
c5e8d7af
PH
33 unescapeHTML,
34 unified_strdate,
81c2f20b 35 uppercase_escape,
c5e8d7af
PH
36)
37
5f6a1245 38
de7f3446 39class YoutubeBaseInfoExtractor(InfoExtractor):
b2e8bc1b
JMF
40 """Provide base functions for Youtube extractors"""
41 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
83317f69 42 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
b2e8bc1b
JMF
43 _NETRC_MACHINE = 'youtube'
44 # If True it will raise an error if no login info is provided
45 _LOGIN_REQUIRED = False
46
b2e8bc1b 47 def _set_language(self):
810fb84d
PH
48 self._set_cookie(
49 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
42939b61 50 # YouTube sets the expire time to about two months
810fb84d 51 expire_time=time.time() + 2 * 30 * 24 * 3600)
b2e8bc1b
JMF
52
53 def _login(self):
83317f69 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 """
b2e8bc1b
JMF
61 (username, password) = self._get_login_info()
62 # No authentication to be performed
63 if username is None:
64 if self._LOGIN_REQUIRED:
69ea8ca4 65 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
83317f69 66 return True
b2e8bc1b 67
7cc3570e
PH
68 login_page = self._download_webpage(
69 self._LOGIN_URL, None,
69ea8ca4
PH
70 note='Downloading login page',
71 errnote='unable to fetch login page', fatal=False)
7cc3570e
PH
72 if login_page is False:
73 return
b2e8bc1b 74
795f28f8 75 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
78caa52a 76 login_page, 'Login GALX parameter')
c5e8d7af 77
b2e8bc1b
JMF
78 # Log in
79 login_form_strs = {
8bcc8756
JW
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',
b2e8bc1b 99 }
83317f69 100
b2e8bc1b
JMF
101 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
102 # chokes on unicode
5f6a1245 103 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
b2e8bc1b 104 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
7cc3570e
PH
105
106 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
107 login_results = self._download_webpage(
108 req, None,
69ea8ca4 109 note='Logging in', errnote='unable to log in', fatal=False)
7cc3570e
PH
110 if login_results is False:
111 return False
83317f69 112
113 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
69ea8ca4 114 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
83317f69 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:
69ea8ca4
PH
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.)')
83317f69 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:
69ea8ca4 131 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
83317f69 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:
69ea8ca4 135 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
83317f69 136 timeStmp = match.group(1)
137
138 tfa_form_strs = {
78caa52a
PH
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',
83317f69 152 }
5f6a1245 153 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
83317f69 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,
69ea8ca4 159 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
83317f69 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:
69ea8ca4 165 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
83317f69 166 return False
167 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
69ea8ca4 168 self._downloader.report_warning('unable to log in - did the page structure change?')
83317f69 169 return False
170 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
69ea8ca4 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.')
83317f69 172 return False
173
7cc3570e 174 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
69ea8ca4 175 self._downloader.report_warning('unable to log in: bad username or password')
b2e8bc1b
JMF
176 return False
177 return True
178
b2e8bc1b
JMF
179 def _real_initialize(self):
180 if self._downloader is None:
181 return
42939b61 182 self._set_language()
b2e8bc1b
JMF
183 if not self._login():
184 return
c5e8d7af 185
8377574c 186
360e1ca5 187class YoutubeIE(YoutubeBaseInfoExtractor):
78caa52a 188 IE_DESC = 'YouTube.com'
cb7dfeea 189 _VALID_URL = r"""(?x)^
c5e8d7af 190 (
edb53e2d 191 (?:https?://|//) # http(s):// or protocol-independent URL
cb7dfeea 192 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
484aaeb2 193 (?:www\.)?deturl\.com/www\.youtube\.com/|
e70dc1d1 194 (?:www\.)?pwnyoutube\.com/|
f7000f3a 195 (?:www\.)?yourepeat\.com/|
e69ae5b9
JMF
196 tube\.majestyc\.net/|
197 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
c5e8d7af
PH
198 (?:.*?\#/)? # handle anchor (#/) redirect urls
199 (?: # the various things that can precede the ID:
ac7553d0 200 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
c5e8d7af 201 |(?: # or the v= param in all its forms
f7000f3a 202 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
c5e8d7af
PH
203 (?:\?|\#!?) # the params delimiter ? or # or #!
204 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
205 v=
206 )
f4b05232
JMF
207 ))
208 |youtu\.be/ # just youtu.be/xxxx
edb53e2d 209 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
f4b05232 210 )
c5e8d7af 211 )? # all until now is optional -> you can pass the naked ID
8963d9c2 212 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
9291475f 213 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
c5e8d7af
PH
214 (?(1).+)? # if we found the ID, everything can follow
215 $"""
c5e8d7af 216 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
2c62dc26
PH
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
1d043b93 234
86fe61c8 235 # 3d videos
43b81eb9
PH
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},
836a086c 243
96fb5605 244 # Apple HTTP Live Streaming
43b81eb9
PH
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},
2c62dc26
PH
252
253 # DASH mp4 video
43b81eb9
PH
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},
e65566a9 259 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
43b81eb9
PH
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},
0d2c1418
PH
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'},
836a086c 265
f6f1fc92 266 # Dash mp4 audio
62cd676c
PH
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'},
836a086c
AZ
270
271 # Dash webm
e75cafe9
A
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},
1cc887cb 278 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
e75cafe9
A
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},
3c80377b 286 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
18061bba 287 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
a51d3aa0
PH
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'},
76b3c610 290 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
10a404c3 291 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
76b3c610 292 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
2c62dc26
PH
293
294 # Dash webm audio
55db73ef 295 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
e75cafe9 296 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
ce6b9a2d 297
0857baad
PH
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
ce6b9a2d
PH
303 # RTMP (unnamed)
304 '_rtmp': {'protocol': 'rtmp'},
c5e8d7af 305 }
836a086c 306
78caa52a 307 IE_NAME = 'youtube'
2eb88d95
PH
308 _TESTS = [
309 {
4bc3a23e
PH
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'],
3e7c1224
PH
320 'like_count': int,
321 'dislike_count': int,
2eb88d95 322 }
0e853ca4 323 },
0e853ca4 324 {
4bc3a23e
PH
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',
2eb88d95 335 }
c108eb73
JMF
336 },
337 {
4bc3a23e
PH
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',
c108eb73
JMF
348 }
349 },
fccd3771 350 {
4bc3a23e
PH
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'
fccd3771
PH
361 }
362 },
dd27fd17 363 {
4bc3a23e
PH
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'
4919603f 374 },
4bc3a23e
PH
375 'params': {
376 'youtube_include_dash_manifest': True,
377 'format': '141',
4919603f 378 },
dd27fd17 379 },
3489b7d2
JMF
380 # DASH manifest with encrypted signature
381 {
78caa52a
PH
382 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
383 'info_dict': {
384 'id': 'IB3lcPjvWLA',
385 'ext': 'm4a',
b766eb27
JMF
386 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
387 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
78caa52a
PH
388 'uploader': 'AfrojackVEVO',
389 'uploader_id': 'AfrojackVEVO',
390 'upload_date': '20131011',
3489b7d2 391 },
4bc3a23e 392 'params': {
78caa52a
PH
393 'youtube_include_dash_manifest': True,
394 'format': '141',
3489b7d2
JMF
395 },
396 },
aaeb86f6
S
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 },
aa79ac0c
PH
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 }
c522adb1
JMF
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',
9ed99402 434 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
c522adb1
JMF
435 'uploader': 'The Witcher',
436 'uploader_id': 'WitcherGame',
437 'upload_date': '20140605',
438 },
439 },
fccae2b9
S
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 },
774e208f
PH
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 ]
e52a40ab
PH
468 },
469 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
470 {
471 'url': 'lqQg6PlCWgI',
472 'info_dict': {
473 'id': 'lqQg6PlCWgI',
474 'ext': 'mp4',
cbe2bd91
PH
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',
e52a40ab 483 }
cbe2bd91 484 },
6271f1ca
PH
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 },
06b491eb
S
498 },
499 # url_encoded_fmt_stream_map is empty string
500 {
501 'url': 'qEJwOuvDf7I',
502 'info_dict': {
503 'id': 'qEJwOuvDf7I',
504 'ext': 'mp4',
505 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
506 'description': '',
507 'upload_date': '20150404',
508 'uploader_id': 'spbelect',
509 'uploader': 'Наблюдатели Петербурга',
510 },
511 'params': {
512 'skip_download': 'requires avconv',
513 }
514 },
2eb88d95
PH
515 ]
516
e0df6211
PH
517 def __init__(self, *args, **kwargs):
518 super(YoutubeIE, self).__init__(*args, **kwargs)
83799698 519 self._player_cache = {}
e0df6211 520
c5e8d7af
PH
521 def report_video_info_webpage_download(self, video_id):
522 """Report attempt to download video info webpage."""
69ea8ca4 523 self.to_screen('%s: Downloading video info webpage' % video_id)
c5e8d7af 524
c5e8d7af
PH
525 def report_information_extraction(self, video_id):
526 """Report attempt to extract video information."""
69ea8ca4 527 self.to_screen('%s: Extracting video information' % video_id)
c5e8d7af
PH
528
529 def report_unavailable_format(self, video_id, format):
530 """Report extracted video URL."""
69ea8ca4 531 self.to_screen('%s: Format %s not available' % (video_id, format))
c5e8d7af
PH
532
533 def report_rtmp_download(self):
534 """Indicate the download will use the RTMP protocol."""
69ea8ca4 535 self.to_screen('RTMP download detected')
c5e8d7af 536
60064c53
PH
537 def _signature_cache_id(self, example_sig):
538 """ Return a string representation of a signature """
78caa52a 539 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
60064c53
PH
540
541 def _extract_signature_function(self, video_id, player_url, example_sig):
cf010131 542 id_m = re.match(
60620368 543 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
cf010131 544 player_url)
c081b35c
PH
545 if not id_m:
546 raise ExtractorError('Cannot identify player %r' % player_url)
e0df6211
PH
547 player_type = id_m.group('ext')
548 player_id = id_m.group('id')
549
c4417ddb 550 # Read from filesystem cache
60064c53
PH
551 func_id = '%s_%s_%s' % (
552 player_type, player_id, self._signature_cache_id(example_sig))
c4417ddb 553 assert os.path.basename(func_id) == func_id
a0e07d31 554
69ea8ca4 555 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
a0e07d31 556 if cache_spec is not None:
78caa52a 557 return lambda s: ''.join(s[i] for i in cache_spec)
83799698 558
6d1a55a5
PH
559 download_note = (
560 'Downloading player %s' % player_url
561 if self._downloader.params.get('verbose') else
562 'Downloading %s player %s' % (player_type, player_id)
563 )
e0df6211
PH
564 if player_type == 'js':
565 code = self._download_webpage(
566 player_url, video_id,
6d1a55a5 567 note=download_note,
69ea8ca4 568 errnote='Download of %s failed' % player_url)
83799698 569 res = self._parse_sig_js(code)
c4417ddb 570 elif player_type == 'swf':
e0df6211
PH
571 urlh = self._request_webpage(
572 player_url, video_id,
6d1a55a5 573 note=download_note,
69ea8ca4 574 errnote='Download of %s failed' % player_url)
e0df6211 575 code = urlh.read()
83799698 576 res = self._parse_sig_swf(code)
e0df6211
PH
577 else:
578 assert False, 'Invalid player type %r' % player_type
579
785521bf
PH
580 test_string = ''.join(map(compat_chr, range(len(example_sig))))
581 cache_res = res(test_string)
582 cache_spec = [ord(c) for c in cache_res]
83799698 583
69ea8ca4 584 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
83799698
PH
585 return res
586
60064c53 587 def _print_sig_code(self, func, example_sig):
edf3e38e
PH
588 def gen_sig_code(idxs):
589 def _genslice(start, end, step):
78caa52a 590 starts = '' if start == 0 else str(start)
8bcc8756 591 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
69ea8ca4 592 steps = '' if step == 1 else (':%d' % step)
78caa52a 593 return 's[%s%s%s]' % (starts, ends, steps)
edf3e38e
PH
594
595 step = None
7af808a5
PH
596 # Quelch pyflakes warnings - start will be set when step is set
597 start = '(Never used)'
edf3e38e
PH
598 for i, prev in zip(idxs[1:], idxs[:-1]):
599 if step is not None:
600 if i - prev == step:
601 continue
602 yield _genslice(start, prev, step)
603 step = None
604 continue
605 if i - prev in [-1, 1]:
606 step = i - prev
607 start = prev
608 continue
609 else:
78caa52a 610 yield 's[%d]' % prev
edf3e38e 611 if step is None:
78caa52a 612 yield 's[%d]' % i
edf3e38e
PH
613 else:
614 yield _genslice(start, i, step)
615
78caa52a 616 test_string = ''.join(map(compat_chr, range(len(example_sig))))
c705320f 617 cache_res = func(test_string)
edf3e38e 618 cache_spec = [ord(c) for c in cache_res]
78caa52a 619 expr_code = ' + '.join(gen_sig_code(cache_spec))
60064c53
PH
620 signature_id_tuple = '(%s)' % (
621 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
69ea8ca4 622 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
78caa52a 623 ' return %s\n') % (signature_id_tuple, expr_code)
69ea8ca4 624 self.to_screen('Extracted signature function:\n' + code)
edf3e38e 625
e0df6211
PH
626 def _parse_sig_js(self, jscode):
627 funcname = self._search_regex(
aaeb86f6 628 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
8bcc8756 629 'Initial JS player signature function name')
2b25cb5d
PH
630
631 jsi = JSInterpreter(jscode)
632 initial_function = jsi.extract_function(funcname)
e0df6211
PH
633 return lambda s: initial_function([s])
634
635 def _parse_sig_swf(self, file_contents):
54256267 636 swfi = SWFInterpreter(file_contents)
78caa52a 637 TARGET_CLASSNAME = 'SignatureDecipher'
54256267 638 searched_class = swfi.extract_class(TARGET_CLASSNAME)
78caa52a 639 initial_function = swfi.extract_function(searched_class, 'decipher')
e0df6211
PH
640 return lambda s: initial_function([s])
641
83799698 642 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
257a2501 643 """Turn the encrypted s field into a working signature"""
6b37f0be 644
c8bf86d5 645 if player_url is None:
69ea8ca4 646 raise ExtractorError('Cannot decrypt signature without player_url')
920de7a2 647
69ea8ca4 648 if player_url.startswith('//'):
78caa52a 649 player_url = 'https:' + player_url
c8bf86d5 650 try:
62af3a0e 651 player_id = (player_url, self._signature_cache_id(s))
c8bf86d5
PH
652 if player_id not in self._player_cache:
653 func = self._extract_signature_function(
60064c53 654 video_id, player_url, s
c8bf86d5
PH
655 )
656 self._player_cache[player_id] = func
657 func = self._player_cache[player_id]
658 if self._downloader.params.get('youtube_print_sig_code'):
60064c53 659 self._print_sig_code(func, s)
c8bf86d5
PH
660 return func(s)
661 except Exception as e:
662 tb = traceback.format_exc()
663 raise ExtractorError(
78caa52a 664 'Signature extraction failed: ' + tb, cause=e)
e0df6211 665
360e1ca5 666 def _get_subtitles(self, video_id, webpage):
de7f3446 667 try:
60e47a26 668 subs_doc = self._download_xml(
38c2e5b8 669 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
7fad1c63
JMF
670 video_id, note=False)
671 except ExtractorError as err:
69ea8ca4 672 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
de7f3446 673 return {}
de7f3446
JMF
674
675 sub_lang_list = {}
60e47a26
JMF
676 for track in subs_doc.findall('track'):
677 lang = track.attrib['lang_code']
7e660ac1
LD
678 if lang in sub_lang_list:
679 continue
360e1ca5
JMF
680 sub_formats = []
681 for ext in ['sbv', 'vtt', 'srt']:
682 params = compat_urllib_parse.urlencode({
683 'lang': lang,
684 'v': video_id,
685 'fmt': ext,
686 'name': track.attrib['name'].encode('utf-8'),
687 })
688 sub_formats.append({
689 'url': 'https://www.youtube.com/api/timedtext?' + params,
690 'ext': ext,
691 })
692 sub_lang_list[lang] = sub_formats
de7f3446 693 if not sub_lang_list:
69ea8ca4 694 self._downloader.report_warning('video doesn\'t have subtitles')
de7f3446
JMF
695 return {}
696 return sub_lang_list
697
360e1ca5 698 def _get_automatic_captions(self, video_id, webpage):
de7f3446
JMF
699 """We need the webpage for getting the captions url, pass it as an
700 argument to speed up the process."""
69ea8ca4 701 self.to_screen('%s: Looking for automatic captions' % video_id)
de7f3446 702 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
78caa52a 703 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
de7f3446
JMF
704 if mobj is None:
705 self._downloader.report_warning(err_msg)
706 return {}
707 player_config = json.loads(mobj.group(1))
708 try:
0792d563
PH
709 args = player_config['args']
710 caption_url = args['ttsurl']
711 timestamp = args['timestamp']
055e6f36
JMF
712 # We get the available subtitles
713 list_params = compat_urllib_parse.urlencode({
714 'type': 'list',
715 'tlangs': 1,
716 'asrs': 1,
de7f3446 717 })
055e6f36 718 list_url = caption_url + '&' + list_params
e26f8712 719 caption_list = self._download_xml(list_url, video_id)
e3dc22ca 720 original_lang_node = caption_list.find('track')
7d900ef1 721 if original_lang_node is None:
69ea8ca4 722 self._downloader.report_warning('Video doesn\'t have automatic captions')
e3dc22ca
JMF
723 return {}
724 original_lang = original_lang_node.attrib['lang_code']
7d900ef1 725 caption_kind = original_lang_node.attrib.get('kind', '')
055e6f36
JMF
726
727 sub_lang_list = {}
728 for lang_node in caption_list.findall('target'):
729 sub_lang = lang_node.attrib['lang_code']
360e1ca5
JMF
730 sub_formats = []
731 for ext in ['sbv', 'vtt', 'srt']:
732 params = compat_urllib_parse.urlencode({
733 'lang': original_lang,
734 'tlang': sub_lang,
735 'fmt': ext,
736 'ts': timestamp,
737 'kind': caption_kind,
738 })
739 sub_formats.append({
740 'url': caption_url + '&' + params,
741 'ext': ext,
742 })
743 sub_lang_list[sub_lang] = sub_formats
055e6f36 744 return sub_lang_list
de7f3446
JMF
745 # An extractor error can be raise by the download process if there are
746 # no automatic captions but there are subtitles
747 except (KeyError, ExtractorError):
748 self._downloader.report_warning(err_msg)
749 return {}
750
97665381
PH
751 @classmethod
752 def extract_id(cls, url):
753 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
c5e8d7af 754 if mobj is None:
69ea8ca4 755 raise ExtractorError('Invalid URL: %s' % url)
c5e8d7af
PH
756 video_id = mobj.group(2)
757 return video_id
758
1d043b93
JMF
759 def _extract_from_m3u8(self, manifest_url, video_id):
760 url_map = {}
5f6a1245 761
1d043b93
JMF
762 def _get_urls(_manifest):
763 lines = _manifest.split('\n')
764 urls = filter(lambda l: l and not l.startswith('#'),
8bcc8756 765 lines)
1d043b93 766 return urls
78caa52a 767 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1d043b93
JMF
768 formats_urls = _get_urls(manifest)
769 for format_url in formats_urls:
890f62e8 770 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1d043b93
JMF
771 url_map[itag] = format_url
772 return url_map
773
1fb07d10
JG
774 def _extract_annotations(self, video_id):
775 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
69ea8ca4 776 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1fb07d10 777
da276600
PH
778 def _parse_dash_manifest(
779 self, video_id, dash_manifest_url, player_url, age_gate):
774e208f
PH
780 def decrypt_sig(mobj):
781 s = mobj.group(1)
782 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
783 return '/signature/%s' % dec_s
784 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
785 dash_doc = self._download_xml(
786 dash_manifest_url, video_id,
787 note='Downloading DASH manifest',
788 errnote='Could not download DASH manifest')
789
790 formats = []
de5c5456
YCH
791 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
792 mime_type = a.attrib.get('mimeType')
793 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
794 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
795 if url_el is None:
796 continue
797 if mime_type == 'text/vtt':
798 # TODO implement WebVTT downloading
799 pass
800 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
801 format_id = r.attrib['id']
802 video_url = url_el.text
803 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
804 f = {
805 'format_id': format_id,
806 'url': video_url,
807 'width': int_or_none(r.attrib.get('width')),
808 'height': int_or_none(r.attrib.get('height')),
809 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
810 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
811 'filesize': filesize,
812 'fps': int_or_none(r.attrib.get('frameRate')),
813 }
814 try:
815 existing_format = next(
816 fo for fo in formats
817 if fo['format_id'] == format_id)
818 except StopIteration:
819 full_info = self._formats.get(format_id, {}).copy()
820 full_info.update(f)
821 formats.append(full_info)
822 else:
823 existing_format.update(f)
824 else:
825 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
774e208f
PH
826 return formats
827
c5e8d7af 828 def _real_extract(self, url):
7e8c0af0 829 proto = (
78caa52a
PH
830 'http' if self._downloader.params.get('prefer_insecure', False)
831 else 'https')
7e8c0af0 832
c5e8d7af
PH
833 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
834 mobj = re.search(self._NEXT_URL_RE, url)
835 if mobj:
7e8c0af0 836 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
97665381 837 video_id = self.extract_id(url)
c5e8d7af
PH
838
839 # Get video webpage
aa79ac0c 840 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
a1f934b1 841 video_webpage = self._download_webpage(url, video_id)
c5e8d7af
PH
842
843 # Attempt to extract SWF player URL
e0df6211 844 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
c5e8d7af
PH
845 if mobj is not None:
846 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
847 else:
848 player_url = None
849
850 # Get video info
6449cd80 851 embed_webpage = None
c108eb73 852 if re.search(r'player-age-gate-content">', video_webpage) is not None:
c108eb73
JMF
853 age_gate = True
854 # We simulate the access to the video from www.youtube.com/v/{video_id}
855 # this can be viewed without login into Youtube
beb95e77
CL
856 url = proto + '://www.youtube.com/embed/%s' % video_id
857 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
2c57c7fa
JMF
858 data = compat_urllib_parse.urlencode({
859 'video_id': video_id,
860 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
c084c934 861 'sts': self._search_regex(
beb95e77 862 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
2c57c7fa 863 })
7e8c0af0 864 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
94bd3613
PH
865 video_info_webpage = self._download_webpage(
866 video_info_url, video_id,
20436c30 867 note='Refetching age-gated info webpage',
94bd3613 868 errnote='unable to download video info webpage')
c5e8d7af 869 video_info = compat_parse_qs(video_info_webpage)
c108eb73
JMF
870 else:
871 age_gate = False
4e62ebe2
JMF
872 try:
873 # Try looking directly into the video webpage
874 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
875 if not mobj:
876 raise ValueError('Could not find ytplayer.config') # caught below
877 json_code = uppercase_escape(mobj.group(1))
878 ytplayer_config = json.loads(json_code)
879 args = ytplayer_config['args']
880 # Convert to the same format returned by compat_parse_qs
881 video_info = dict((k, [v]) for k, v in args.items())
e40bd5f0 882 if not args.get('url_encoded_fmt_stream_map'):
4e62ebe2
JMF
883 raise ValueError('No stream_map present') # caught below
884 except ValueError:
885 # We fallback to the get_video_info pages (used by the embed page)
886 self.report_video_info_webpage_download(video_id)
887 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
810fb84d
PH
888 video_info_url = (
889 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
890 % (proto, video_id, el_type))
891 video_info_webpage = self._download_webpage(
892 video_info_url,
4e62ebe2
JMF
893 video_id, note=False,
894 errnote='unable to download video info webpage')
895 video_info = compat_parse_qs(video_info_webpage)
896 if 'token' in video_info:
897 break
c5e8d7af
PH
898 if 'token' not in video_info:
899 if 'reason' in video_info:
d11271dd 900 raise ExtractorError(
78caa52a 901 'YouTube said: %s' % video_info['reason'][0],
d11271dd 902 expected=True, video_id=video_id)
c5e8d7af 903 else:
d11271dd 904 raise ExtractorError(
78caa52a 905 '"token" parameter not in video info for unknown reason',
d11271dd 906 video_id=video_id)
c5e8d7af 907
1d699755
PH
908 if 'view_count' in video_info:
909 view_count = int(video_info['view_count'][0])
910 else:
911 view_count = None
912
c5e8d7af
PH
913 # Check for "rental" videos
914 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
69ea8ca4 915 raise ExtractorError('"rental" videos not supported')
c5e8d7af
PH
916
917 # Start extracting information
918 self.report_information_extraction(video_id)
919
920 # uploader
921 if 'author' not in video_info:
69ea8ca4 922 raise ExtractorError('Unable to extract uploader name')
c5e8d7af
PH
923 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
924
925 # uploader_id
926 video_uploader_id = None
927 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
928 if mobj is not None:
929 video_uploader_id = mobj.group(1)
930 else:
69ea8ca4 931 self._downloader.report_warning('unable to extract uploader nickname')
c5e8d7af
PH
932
933 # title
a8c6b241 934 if 'title' in video_info:
aa92f063 935 video_title = video_info['title'][0]
a8c6b241 936 else:
69ea8ca4 937 self._downloader.report_warning('Unable to extract video title')
78caa52a 938 video_title = '_'
c5e8d7af
PH
939
940 # thumbnail image
7763b04e
JMF
941 # We try first to get a high quality image:
942 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
943 video_webpage, re.DOTALL)
944 if m_thumb is not None:
945 video_thumbnail = m_thumb.group(1)
946 elif 'thumbnail_url' not in video_info:
69ea8ca4 947 self._downloader.report_warning('unable to extract video thumbnail')
f490e77e 948 video_thumbnail = None
c5e8d7af
PH
949 else: # don't panic if we can't find it
950 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
951
952 # upload date
953 upload_date = None
ad3bc6ac 954 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
beee53de
PH
955 if mobj is None:
956 mobj = re.search(
263bd4ec 957 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
beee53de 958 video_webpage)
c5e8d7af
PH
959 if mobj is not None:
960 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
961 upload_date = unified_strdate(upload_date)
962
55f7bd2d
PH
963 m_cat_container = self._search_regex(
964 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
624dcebf 965 video_webpage, 'categories', default=None)
ec8deefc 966 if m_cat_container:
ad3bc6ac 967 category = self._html_search_regex(
01ed5c9b 968 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
ad3bc6ac
PH
969 default=None)
970 video_categories = None if category is None else [category]
971 else:
972 video_categories = None
ec8deefc 973
c5e8d7af
PH
974 # description
975 video_description = get_element_by_id("eow-description", video_webpage)
976 if video_description:
27dcce19
PH
977 video_description = re.sub(r'''(?x)
978 <a\s+
979 (?:[a-zA-Z-]+="[^"]+"\s+)*?
980 title="([^"]+)"\s+
981 (?:[a-zA-Z-]+="[^"]+"\s+)*?
982 class="yt-uix-redirect-link"\s*>
983 [^<]+
984 </a>
985 ''', r'\1', video_description)
c5e8d7af
PH
986 video_description = clean_html(video_description)
987 else:
988 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
989 if fd_mobj:
990 video_description = unescapeHTML(fd_mobj.group(1))
991 else:
78caa52a 992 video_description = ''
c5e8d7af 993
f30a38be 994 def _extract_count(count_name):
46374a56 995 count = self._search_regex(
f30a38be
JMF
996 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
997 video_webpage, count_name, default=None)
336c3a69
JMF
998 if count is not None:
999 return int(count.replace(',', ''))
1000 return None
69ea8ca4
PH
1001 like_count = _extract_count('like')
1002 dislike_count = _extract_count('dislike')
336c3a69 1003
c5e8d7af 1004 # subtitles
d82134c3 1005 video_subtitles = self.extract_subtitles(video_id, video_webpage)
360e1ca5 1006 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
c5e8d7af
PH
1007
1008 if 'length_seconds' not in video_info:
69ea8ca4 1009 self._downloader.report_warning('unable to extract video duration')
b466b702 1010 video_duration = None
c5e8d7af 1011 else:
b466b702 1012 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
c5e8d7af 1013
1fb07d10
JG
1014 # annotations
1015 video_annotations = None
1016 if self._downloader.params.get('writeannotations', False):
5f6a1245 1017 video_annotations = self._extract_annotations(video_id)
1fb07d10 1018
dd27fd17
PH
1019 def _map_to_format_list(urlmap):
1020 formats = []
1021 for itag, video_real_url in urlmap.items():
1022 dct = {
1023 'format_id': itag,
1024 'url': video_real_url,
1025 'player_url': player_url,
1026 }
0b65e5d4
PH
1027 if itag in self._formats:
1028 dct.update(self._formats[itag])
dd27fd17
PH
1029 formats.append(dct)
1030 return formats
1031
c5e8d7af
PH
1032 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1033 self.report_rtmp_download()
dd27fd17
PH
1034 formats = [{
1035 'format_id': '_rtmp',
1036 'protocol': 'rtmp',
1037 'url': video_info['conn'][0],
1038 'player_url': player_url,
1039 }]
24270b03 1040 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
5f6a1245 1041 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
00fe14fc 1042 if 'rtmpe%3Dyes' in encoded_url_map:
a7055eb9 1043 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
c5e8d7af 1044 url_map = {}
00fe14fc 1045 for url_data_str in encoded_url_map.split(','):
c5e8d7af 1046 url_data = compat_parse_qs(url_data_str)
201e9eaa
PH
1047 if 'itag' not in url_data or 'url' not in url_data:
1048 continue
1049 format_id = url_data['itag'][0]
1050 url = url_data['url'][0]
1051
1052 if 'sig' in url_data:
1053 url += '&signature=' + url_data['sig'][0]
1054 elif 's' in url_data:
1055 encrypted_sig = url_data['s'][0]
6449cd80 1056 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
201e9eaa 1057
beb95e77 1058 jsplayer_url_json = self._search_regex(
6449cd80
PH
1059 ASSETS_RE,
1060 embed_webpage if age_gate else video_webpage,
1061 'JS player URL (1)', default=None)
1062 if not jsplayer_url_json and not age_gate:
1063 # We need the embed website after all
1064 if embed_webpage is None:
1065 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1066 embed_webpage = self._download_webpage(
1067 embed_url, video_id, 'Downloading embed webpage')
1068 jsplayer_url_json = self._search_regex(
1069 ASSETS_RE, embed_webpage, 'JS player URL')
1070
beb95e77 1071 player_url = json.loads(jsplayer_url_json)
201e9eaa
PH
1072 if player_url is None:
1073 player_url_json = self._search_regex(
1074 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
78caa52a 1075 video_webpage, 'age gate player URL')
201e9eaa
PH
1076 player_url = json.loads(player_url_json)
1077
1078 if self._downloader.params.get('verbose'):
cf010131 1079 if player_url is None:
201e9eaa
PH
1080 player_version = 'unknown'
1081 player_desc = 'unknown'
1082 else:
1083 if player_url.endswith('swf'):
1084 player_version = self._search_regex(
1085 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
78caa52a 1086 'flash player', fatal=False)
201e9eaa 1087 player_desc = 'flash player %s' % player_version
cf010131 1088 else:
201e9eaa
PH
1089 player_version = self._search_regex(
1090 r'html5player-([^/]+?)(?:/html5player)?\.js',
1091 player_url,
1092 'html5 player', fatal=False)
78caa52a 1093 player_desc = 'html5 player %s' % player_version
201e9eaa 1094
60064c53 1095 parts_sizes = self._signature_cache_id(encrypted_sig)
69ea8ca4 1096 self.to_screen('{%s} signature length %s, %s' %
9e1a5b84 1097 (format_id, parts_sizes, player_desc))
201e9eaa
PH
1098
1099 signature = self._decrypt_signature(
1100 encrypted_sig, video_id, player_url, age_gate)
1101 url += '&signature=' + signature
1102 if 'ratebypass' not in url:
1103 url += '&ratebypass=yes'
1104 url_map[format_id] = url
dd27fd17 1105 formats = _map_to_format_list(url_map)
1d043b93
JMF
1106 elif video_info.get('hlsvp'):
1107 manifest_url = video_info['hlsvp'][0]
1108 url_map = self._extract_from_m3u8(manifest_url, video_id)
dd27fd17 1109 formats = _map_to_format_list(url_map)
c5e8d7af 1110 else:
69ea8ca4 1111 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
c5e8d7af 1112
dd27fd17 1113 # Look for the DASH manifest
203fb43f 1114 if self._downloader.params.get('youtube_include_dash_manifest', True):
774e208f 1115 dash_mpd = video_info.get('dashmpd')
75111274 1116 if dash_mpd:
774e208f
PH
1117 dash_manifest_url = dash_mpd[0]
1118 try:
1119 dash_formats = self._parse_dash_manifest(
da276600 1120 video_id, dash_manifest_url, player_url, age_gate)
774e208f
PH
1121 except (ExtractorError, KeyError) as e:
1122 self.report_warning(
1123 'Skipping DASH manifest: %r' % e, video_id)
1124 else:
e65566a9
PH
1125 # Hide the formats we found through non-DASH
1126 dash_keys = set(df['format_id'] for df in dash_formats)
1127 for f in formats:
1128 if f['format_id'] in dash_keys:
1129 f['format_id'] = 'nondash-%s' % f['format_id']
ee61f6f3 1130 f['preference'] = f.get('preference', 0) - 10000
774e208f 1131 formats.extend(dash_formats)
d80044c2 1132
6271f1ca
PH
1133 # Check for malformed aspect ratio
1134 stretched_m = re.search(
1135 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1136 video_webpage)
1137 if stretched_m:
1138 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1139 for f in formats:
1140 if f.get('vcodec') != 'none':
1141 f['stretched_ratio'] = ratio
1142
4bcc7bd1 1143 self._sort_formats(formats)
4ea3be0a 1144
1145 return {
8bcc8756
JW
1146 'id': video_id,
1147 'uploader': video_uploader,
1148 'uploader_id': video_uploader_id,
1149 'upload_date': upload_date,
1150 'title': video_title,
1151 'thumbnail': video_thumbnail,
1152 'description': video_description,
1153 'categories': video_categories,
1154 'subtitles': video_subtitles,
360e1ca5 1155 'automatic_captions': automatic_captions,
8bcc8756
JW
1156 'duration': video_duration,
1157 'age_limit': 18 if age_gate else 0,
1158 'annotations': video_annotations,
7e8c0af0 1159 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
8bcc8756 1160 'view_count': view_count,
4ea3be0a 1161 'like_count': like_count,
1162 'dislike_count': dislike_count,
2d30521a 1163 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
8bcc8756 1164 'formats': formats,
4ea3be0a 1165 }
c5e8d7af 1166
5f6a1245 1167
880e1c52 1168class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
78caa52a 1169 IE_DESC = 'YouTube.com playlists'
d67cc9fa 1170 _VALID_URL = r"""(?x)(?:
c5e8d7af
PH
1171 (?:https?://)?
1172 (?:\w+\.)?
1173 youtube\.com/
1174 (?:
ac7553d0 1175 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
c5e8d7af
PH
1176 \? (?:.*?&)*? (?:p|a|list)=
1177 | p/
1178 )
d67cc9fa 1179 (
99209c29 1180 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
5f6a1245 1181 # Top tracks, they can also include dots
d67cc9fa
JMF
1182 |(?:MC)[\w\.]*
1183 )
c5e8d7af
PH
1184 .*
1185 |
99209c29 1186 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
c5e8d7af 1187 )"""
dbb94fb0 1188 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
dbb94fb0 1189 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)'
78caa52a 1190 IE_NAME = 'youtube:playlist'
81127aa5
PH
1191 _TESTS = [{
1192 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1193 'info_dict': {
1194 'title': 'ytdl test PL',
a1cf99d0 1195 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
81127aa5
PH
1196 },
1197 'playlist_count': 3,
9291475f
PH
1198 }, {
1199 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1200 'info_dict': {
acf757f4 1201 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
9291475f
PH
1202 'title': 'YDL_Empty_List',
1203 },
1204 'playlist_count': 0,
1205 }, {
1206 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1207 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1208 'info_dict': {
1209 'title': '29C3: Not my department',
acf757f4 1210 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
9291475f
PH
1211 },
1212 'playlist_count': 95,
1213 }, {
1214 'note': 'issue #673',
1215 'url': 'PLBB231211A4F62143',
1216 'info_dict': {
f46a8702 1217 'title': '[OLD]Team Fortress 2 (Class-based LP)',
acf757f4 1218 'id': 'PLBB231211A4F62143',
9291475f
PH
1219 },
1220 'playlist_mincount': 26,
1221 }, {
1222 'note': 'Large playlist',
1223 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1224 'info_dict': {
1225 'title': 'Uploads from Cauchemar',
acf757f4 1226 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
9291475f
PH
1227 },
1228 'playlist_mincount': 799,
1229 }, {
1230 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1231 'info_dict': {
1232 'title': 'YDL_safe_search',
acf757f4 1233 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
9291475f
PH
1234 },
1235 'playlist_count': 2,
ac7553d0
PH
1236 }, {
1237 'note': 'embedded',
1238 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1239 'playlist_count': 4,
1240 'info_dict': {
1241 'title': 'JODA15',
acf757f4 1242 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
ac7553d0 1243 }
6b08cdf6
PH
1244 }, {
1245 'note': 'Embedded SWF player',
1246 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1247 'playlist_count': 4,
1248 'info_dict': {
1249 'title': 'JODA7',
acf757f4 1250 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
6b08cdf6 1251 }
4b7df0d3
JMF
1252 }, {
1253 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1254 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1255 'info_dict': {
acf757f4
PH
1256 'title': 'Uploads from Interstellar Movie',
1257 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
4b7df0d3
JMF
1258 },
1259 'playlist_mincout': 21,
81127aa5 1260 }]
c5e8d7af 1261
880e1c52
JMF
1262 def _real_initialize(self):
1263 self._login()
1264
652cdaa2 1265 def _ids_to_results(self, ids):
c9cc0bf5
PH
1266 return [
1267 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1268 for vid_id in ids]
652cdaa2
JMF
1269
1270 def _extract_mix(self, playlist_id):
99209c29 1271 # The mixes are generated from a single video
652cdaa2 1272 # the id of the playlist is just 'RD' + video_id
7d4afc55 1273 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
c9cc0bf5 1274 webpage = self._download_webpage(
78caa52a 1275 url, playlist_id, 'Downloading Youtube mix')
bc2f773b 1276 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
c9cc0bf5
PH
1277 title_span = (
1278 search_title('playlist-title') or
1279 search_title('title long-title') or
1280 search_title('title'))
76d1700b 1281 title = clean_html(title_span)
c9cc0bf5
PH
1282 ids = orderedSet(re.findall(
1283 r'''(?xs)data-video-username=".*?".*?
1284 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
1285 webpage))
652cdaa2
JMF
1286 url_results = self._ids_to_results(ids)
1287
1288 return self.playlist_result(url_results, playlist_id, title)
1289
448830ce 1290 def _extract_playlist(self, playlist_id):
dbb94fb0
S
1291 url = self._TEMPLATE_URL % playlist_id
1292 page = self._download_webpage(url, playlist_id)
1293 more_widget_html = content_html = page
1294
10c0e2d8 1295 # Check if the playlist exists or is private
e399853d 1296 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
10c0e2d8 1297 raise ExtractorError(
78caa52a 1298 'The playlist doesn\'t exist or is private, use --username or '
10c0e2d8
JMF
1299 '--netrc to access it.',
1300 expected=True)
1301
dcbb4580
JMF
1302 # Extract the video ids from the playlist pages
1303 ids = []
c5e8d7af 1304
755eb032 1305 for page_num in itertools.count(1):
dbb94fb0 1306 matches = re.finditer(self._VIDEO_RE, content_html)
6e47b51e
JMF
1307 # We remove the duplicates and the link with index 0
1308 # (it's not the first video of the playlist)
1309 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
dcbb4580 1310 ids.extend(new_ids)
c5e8d7af 1311
dbb94fb0
S
1312 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1313 if not mobj:
c5e8d7af
PH
1314 break
1315
dbb94fb0 1316 more = self._download_json(
5912c639
PH
1317 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1318 'Downloading page #%s' % page_num,
1319 transform_source=uppercase_escape)
dbb94fb0 1320 content_html = more['content_html']
4b7df0d3
JMF
1321 if not content_html.strip():
1322 # Some webpages show a "Load more" button but they don't
1323 # have more videos
1324 break
dbb94fb0
S
1325 more_widget_html = more['load_more_widget_html']
1326
1327 playlist_title = self._html_search_regex(
68eb8e90 1328 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
78caa52a 1329 page, 'title')
c5e8d7af 1330
652cdaa2 1331 url_results = self._ids_to_results(ids)
dcbb4580 1332 return self.playlist_result(url_results, playlist_id, playlist_title)
c5e8d7af 1333
448830ce
S
1334 def _real_extract(self, url):
1335 # Extract playlist id
1336 mobj = re.match(self._VALID_URL, url)
1337 if mobj is None:
1338 raise ExtractorError('Invalid URL: %s' % url)
1339 playlist_id = mobj.group(1) or mobj.group(2)
1340
1341 # Check if it's a video-specific URL
1342 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1343 if 'v' in query_dict:
1344 video_id = query_dict['v'][0]
1345 if self._downloader.params.get('noplaylist'):
1346 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1347 return self.url_result(video_id, 'Youtube', video_id=video_id)
1348 else:
1349 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1350
1351 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1352 # Mixes require a custom extraction process
1353 return self._extract_mix(playlist_id)
1354
1355 return self._extract_playlist(playlist_id)
1356
c5e8d7af
PH
1357
1358class YoutubeChannelIE(InfoExtractor):
78caa52a 1359 IE_DESC = 'YouTube.com channels'
9ff67727 1360 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
eb0f3e7e 1361 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
78caa52a 1362 IE_NAME = 'youtube:channel'
cdc628a4
PH
1363 _TESTS = [{
1364 'note': 'paginated channel',
1365 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1366 'playlist_mincount': 91,
acf757f4
PH
1367 'info_dict': {
1368 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1369 }
cdc628a4 1370 }]
c5e8d7af
PH
1371
1372 def extract_videos_from_page(self, page):
1373 ids_in_page = []
fb69240c
S
1374 titles_in_page = []
1375 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1376 video_id = mobj.group('id')
1377 video_title = unescapeHTML(mobj.group('title'))
1378 try:
1379 idx = ids_in_page.index(video_id)
1380 if video_title and not titles_in_page[idx]:
1381 titles_in_page[idx] = video_title
1382 except ValueError:
1383 ids_in_page.append(video_id)
1384 titles_in_page.append(video_title)
1385 return zip(ids_in_page, titles_in_page)
c5e8d7af
PH
1386
1387 def _real_extract(self, url):
9ff67727 1388 channel_id = self._match_id(url)
c5e8d7af 1389
eb0f3e7e 1390 url = self._TEMPLATE_URL % channel_id
60bf45c8 1391 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
31812a9e
PH
1392 autogenerated = re.search(r'''(?x)
1393 class="[^"]*?(?:
1394 channel-header-autogenerated-label|
1395 yt-channel-title-autogenerated
1396 )[^"]*"''', channel_page) is not None
c5e8d7af 1397
b9643eed
JMF
1398 if autogenerated:
1399 # The videos are contained in a single page
1400 # the ajax pages can't be used, they are empty
b82f815f 1401 entries = [
fb69240c
S
1402 self.url_result(
1403 video_id, 'Youtube', video_id=video_id,
1404 video_title=video_title)
8f02ad4f 1405 for video_id, video_title in self.extract_videos_from_page(channel_page)]
b82f815f
PH
1406 return self.playlist_result(entries, channel_id)
1407
1408 def _entries():
23d3608c 1409 more_widget_html = content_html = channel_page
b9643eed 1410 for pagenum in itertools.count(1):
81c2f20b 1411
8f02ad4f 1412 for video_id, video_title in self.extract_videos_from_page(content_html):
b82f815f 1413 yield self.url_result(
fb69240c
S
1414 video_id, 'Youtube', video_id=video_id,
1415 video_title=video_title)
5f6a1245 1416
23d3608c
JMF
1417 mobj = re.search(
1418 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1419 more_widget_html)
1420 if not mobj:
b9643eed 1421 break
c5e8d7af 1422
23d3608c
JMF
1423 more = self._download_json(
1424 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1425 'Downloading page #%s' % (pagenum + 1),
1426 transform_source=uppercase_escape)
1427 content_html = more['content_html']
1428 more_widget_html = more['load_more_widget_html']
1429
b82f815f 1430 return self.playlist_result(_entries(), channel_id)
c5e8d7af
PH
1431
1432
eb0f3e7e 1433class YoutubeUserIE(YoutubeChannelIE):
78caa52a 1434 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
9ff67727 1435 _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_-]+)'
eb0f3e7e 1436 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
78caa52a 1437 IE_NAME = 'youtube:user'
c5e8d7af 1438
cdc628a4
PH
1439 _TESTS = [{
1440 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1441 'playlist_mincount': 320,
1442 'info_dict': {
1443 'title': 'TheLinuxFoundation',
1444 }
1445 }, {
1446 'url': 'ytuser:phihag',
1447 'only_matching': True,
1448 }]
1449
e3ea4790 1450 @classmethod
f4b05232 1451 def suitable(cls, url):
e3ea4790
JMF
1452 # Don't return True if the url can be extracted with other youtube
1453 # extractor, the regex would is too permissive and it would match.
1454 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
5f6a1245
JW
1455 if any(ie.suitable(url) for ie in other_ies):
1456 return False
1457 else:
1458 return super(YoutubeUserIE, cls).suitable(url)
f4b05232 1459
b05654f0
PH
1460
1461class YoutubeSearchIE(SearchInfoExtractor):
78caa52a
PH
1462 IE_DESC = 'YouTube.com searches'
1463 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
b05654f0 1464 _MAX_RESULTS = 1000
78caa52a 1465 IE_NAME = 'youtube:search'
b05654f0
PH
1466 _SEARCH_KEY = 'ytsearch'
1467
b05654f0
PH
1468 def _get_n_results(self, query, n):
1469 """Get a specified number of results for a query"""
1470
1471 video_ids = []
1472 pagenum = 0
1473 limit = n
83d548ef 1474 PAGE_SIZE = 50
b05654f0 1475
83d548ef
PH
1476 while (PAGE_SIZE * pagenum) < limit:
1477 result_url = self._API_URL % (
1478 compat_urllib_parse.quote_plus(query.encode('utf-8')),
ed553379 1479 max((PAGE_SIZE * pagenum) + 1), 2)
7cc3570e 1480 data_json = self._download_webpage(
69ea8ca4
PH
1481 result_url, video_id='query "%s"' % query,
1482 note='Downloading page %s' % (pagenum + 1),
1483 errnote='Unable to download API page')
7cc3570e
PH
1484 data = json.loads(data_json)
1485 api_response = data['data']
1486
1487 if 'items' not in api_response:
07ad22b8 1488 raise ExtractorError(
78caa52a 1489 '[youtube] No video results', expected=True)
b05654f0
PH
1490
1491 new_ids = list(video['id'] for video in api_response['items'])
1492 video_ids += new_ids
1493
1494 limit = min(n, api_response['totalItems'])
1495 pagenum += 1
1496
1497 if len(video_ids) > n:
1498 video_ids = video_ids[:n]
7012b23c
PH
1499 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1500 for video_id in video_ids]
b05654f0 1501 return self.playlist_result(videos, query)
75dff0ee 1502
c9ae7b95 1503
a3dd9248 1504class YoutubeSearchDateIE(YoutubeSearchIE):
cb7fb546 1505 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
a3dd9248
CM
1506 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1507 _SEARCH_KEY = 'ytsearchdate'
78caa52a 1508 IE_DESC = 'YouTube.com searches, newest videos first'
75dff0ee 1509
c9ae7b95
PH
1510
1511class YoutubeSearchURLIE(InfoExtractor):
78caa52a
PH
1512 IE_DESC = 'YouTube.com search URLs'
1513 IE_NAME = 'youtube:search_url'
c9ae7b95 1514 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
cdc628a4
PH
1515 _TESTS = [{
1516 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1517 'playlist_mincount': 5,
1518 'info_dict': {
1519 'title': 'youtube-dl test video',
1520 }
1521 }]
c9ae7b95
PH
1522
1523 def _real_extract(self, url):
1524 mobj = re.match(self._VALID_URL, url)
1525 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1526
1527 webpage = self._download_webpage(url, query)
1528 result_code = self._search_regex(
98998cde 1529 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
c9ae7b95
PH
1530
1531 part_codes = re.findall(
1532 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1533 entries = []
1534 for part_code in part_codes:
1535 part_title = self._html_search_regex(
6feb2d5e 1536 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
c9ae7b95
PH
1537 part_url_snippet = self._html_search_regex(
1538 r'(?s)href="([^"]+)"', part_code, 'item URL')
1539 part_url = compat_urlparse.urljoin(
1540 'https://www.youtube.com/', part_url_snippet)
1541 entries.append({
1542 '_type': 'url',
1543 'url': part_url,
1544 'title': part_title,
1545 })
1546
1547 return {
1548 '_type': 'playlist',
1549 'entries': entries,
1550 'title': query,
1551 }
1552
1553
75dff0ee 1554class YoutubeShowIE(InfoExtractor):
78caa52a 1555 IE_DESC = 'YouTube.com (multi-season) shows'
cdc628a4 1556 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
78caa52a 1557 IE_NAME = 'youtube:show'
cdc628a4
PH
1558 _TESTS = [{
1559 'url': 'http://www.youtube.com/show/airdisasters',
1560 'playlist_mincount': 3,
1561 'info_dict': {
1562 'id': 'airdisasters',
1563 'title': 'Air Disasters',
1564 }
1565 }]
75dff0ee
JMF
1566
1567 def _real_extract(self, url):
1568 mobj = re.match(self._VALID_URL, url)
cdc628a4
PH
1569 playlist_id = mobj.group('id')
1570 webpage = self._download_webpage(
1571 url, playlist_id, 'Downloading show webpage')
75dff0ee
JMF
1572 # There's one playlist for each season of the show
1573 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
cdc628a4
PH
1574 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1575 entries = [
1576 self.url_result(
1577 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1578 for season in m_seasons
1579 ]
1580 title = self._og_search_title(webpage, fatal=False)
1581
1582 return {
1583 '_type': 'playlist',
1584 'id': playlist_id,
1585 'title': title,
1586 'entries': entries,
1587 }
04cc9617
JMF
1588
1589
b2e8bc1b 1590class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
d7ae0639
JMF
1591 """
1592 Base class for extractors that fetch info from
1593 http://www.youtube.com/feed_ajax
1594 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1595 """
b2e8bc1b 1596 _LOGIN_REQUIRED = True
43ba5456
JMF
1597 # use action_load_personal_feed instead of action_load_system_feed
1598 _PERSONAL_FEED = False
04cc9617 1599
d7ae0639
JMF
1600 @property
1601 def _FEED_TEMPLATE(self):
43ba5456
JMF
1602 action = 'action_load_system_feed'
1603 if self._PERSONAL_FEED:
1604 action = 'action_load_personal_feed'
38c2e5b8 1605 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
d7ae0639
JMF
1606
1607 @property
1608 def IE_NAME(self):
78caa52a 1609 return 'youtube:%s' % self._FEED_NAME
04cc9617 1610
81f0259b 1611 def _real_initialize(self):
b2e8bc1b 1612 self._login()
81f0259b 1613
04cc9617
JMF
1614 def _real_extract(self, url):
1615 feed_entries = []
0e44d838
JMF
1616 paging = 0
1617 for i in itertools.count(1):
84d84211
PH
1618 info = self._download_json(
1619 self._FEED_TEMPLATE % paging,
1620 '%s feed' % self._FEED_NAME,
1621 'Downloading page %s' % i,
1622 transform_source=uppercase_escape)
f6177462 1623 feed_html = info.get('feed_html') or info.get('content_html')
1a9b9649 1624 load_more_widget_html = info.get('load_more_widget_html') or feed_html
43ba5456 1625 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
04cc9617 1626 ids = orderedSet(m.group(1) for m in m_ids)
7012b23c
PH
1627 feed_entries.extend(
1628 self.url_result(video_id, 'Youtube', video_id=video_id)
1629 for video_id in ids)
05ee2b6d
JMF
1630 mobj = re.search(
1631 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1a9b9649 1632 load_more_widget_html)
05ee2b6d 1633 if mobj is None:
04cc9617 1634 break
05ee2b6d 1635 paging = mobj.group('paging')
d7ae0639
JMF
1636 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1637
5f6a1245 1638
d7ae0639 1639class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
7e17ec8c 1640 IE_NAME = 'youtube:recommended'
f3a34072 1641 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
d7ae0639
JMF
1642 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1643 _FEED_NAME = 'recommended'
78caa52a 1644 _PLAYLIST_TITLE = 'Youtube Recommended videos'
c626a3d9 1645
5f6a1245 1646
448830ce 1647class YoutubeWatchLaterIE(YoutubePlaylistIE):
7e17ec8c 1648 IE_NAME = 'youtube:watchlater'
f3a34072 1649 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
448830ce 1650 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
157e9e5a
JMF
1651
1652 _TESTS = [] # override PlaylistIE tests
c626a3d9 1653
448830ce
S
1654 def _real_extract(self, url):
1655 return self._extract_playlist('WL')
1656
5f6a1245 1657
f459d170 1658class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
7e17ec8c 1659 IE_NAME = 'youtube:history'
f3a34072 1660 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
78caa52a 1661 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
f459d170
JMF
1662 _FEED_NAME = 'history'
1663 _PERSONAL_FEED = True
78caa52a 1664 _PLAYLIST_TITLE = 'Youtube Watch History'
f459d170 1665
5f6a1245 1666
c626a3d9 1667class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
78caa52a 1668 IE_NAME = 'youtube:favorites'
f3a34072 1669 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
c7a7750d 1670 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
c626a3d9
JMF
1671 _LOGIN_REQUIRED = True
1672
1673 def _real_extract(self, url):
1674 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
78caa52a 1675 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
c626a3d9 1676 return self.url_result(playlist_id, 'YoutubePlaylist')
15870e90
PH
1677
1678
1ed5b5c9 1679class YoutubeSubscriptionsIE(YoutubePlaylistIE):
78caa52a
PH
1680 IE_NAME = 'youtube:subscriptions'
1681 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1ed5b5c9 1682 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
81127aa5 1683 _TESTS = []
1ed5b5c9
JMF
1684
1685 def _real_extract(self, url):
78caa52a 1686 title = 'Youtube Subscriptions'
1ed5b5c9
JMF
1687 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1688
1689 # The extraction process is the same as for playlists, but the regex
1690 # for the video ids doesn't contain an index
1691 ids = []
1692 more_widget_html = content_html = page
1693
1694 for page_num in itertools.count(1):
1695 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1696 new_ids = orderedSet(matches)
1697 ids.extend(new_ids)
1698
1699 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1700 if not mobj:
1701 break
1702
1703 more = self._download_json(
1704 'https://youtube.com/%s' % mobj.group('more'), title,
1705 'Downloading page #%s' % page_num,
1706 transform_source=uppercase_escape)
1707 content_html = more['content_html']
1708 more_widget_html = more['load_more_widget_html']
1709
1710 return {
1711 '_type': 'playlist',
1712 'title': title,
1713 'entries': self._ids_to_results(ids),
1714 }
1715
1716
15870e90
PH
1717class YoutubeTruncatedURLIE(InfoExtractor):
1718 IE_NAME = 'youtube:truncated_url'
1719 IE_DESC = False # Do not list
975d35db 1720 _VALID_URL = r'''(?x)
b95aab84
PH
1721 (?:https?://)?
1722 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1723 (?:watch\?(?:
c4808c60 1724 feature=[a-z_]+|
b95aab84
PH
1725 annotation_id=annotation_[^&]+|
1726 x-yt-cl=[0-9]+|
c1708b89 1727 hl=[^&]*|
b95aab84
PH
1728 )?
1729 |
1730 attribution_link\?a=[^&]+
1731 )
1732 $
975d35db 1733 '''
15870e90 1734
c4808c60
PH
1735 _TESTS = [{
1736 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1737 'only_matching': True,
dc2fc736
PH
1738 }, {
1739 'url': 'http://www.youtube.com/watch?',
1740 'only_matching': True,
b95aab84
PH
1741 }, {
1742 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1743 'only_matching': True,
1744 }, {
1745 'url': 'https://www.youtube.com/watch?feature=foo',
1746 'only_matching': True,
c1708b89
PH
1747 }, {
1748 'url': 'https://www.youtube.com/watch?hl=en-GB',
1749 'only_matching': True,
c4808c60
PH
1750 }]
1751
15870e90
PH
1752 def _real_extract(self, url):
1753 raise ExtractorError(
78caa52a
PH
1754 'Did you forget to quote the URL? Remember that & is a meta '
1755 'character in most shells, so you want to put the URL in quotes, '
1756 'like youtube-dl '
1757 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1758 ' or simply youtube-dl BaW_jenozKc .',
15870e90 1759 expected=True)
772fd5cc
PH
1760
1761
1762class YoutubeTruncatedIDIE(InfoExtractor):
1763 IE_NAME = 'youtube:truncated_id'
1764 IE_DESC = False # Do not list
b95aab84 1765 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
772fd5cc
PH
1766
1767 _TESTS = [{
1768 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1769 'only_matching': True,
1770 }]
1771
1772 def _real_extract(self, url):
1773 video_id = self._match_id(url)
1774 raise ExtractorError(
1775 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
1776 expected=True)