]>
Commit | Line | Data |
---|---|---|
c5e8d7af | 1 | # coding: utf-8 |
c5e8d7af | 2 | |
78caa52a PH |
3 | from __future__ import unicode_literals |
4 | ||
5 | ||
0ca96d48 | 6 | import itertools |
c5e8d7af | 7 | import json |
c4417ddb | 8 | import os.path |
c5e8d7af | 9 | import re |
e0df6211 | 10 | import traceback |
c5e8d7af | 11 | |
b05654f0 | 12 | from .common import InfoExtractor, SearchInfoExtractor |
54d39d8b | 13 | from .subtitles import SubtitlesInfoExtractor |
2b25cb5d | 14 | from ..jsinterp import JSInterpreter |
54256267 | 15 | from ..swfinterp import SWFInterpreter |
c5e8d7af | 16 | from ..utils 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 PH |
22 | compat_str, |
23 | ||
24 | clean_html, | |
25 | get_element_by_id, | |
652cdaa2 | 26 | get_element_by_attribute, |
c5e8d7af | 27 | ExtractorError, |
dd27fd17 | 28 | int_or_none, |
b7ab0590 | 29 | PagedList, |
c5e8d7af PH |
30 | unescapeHTML, |
31 | unified_strdate, | |
04cc9617 | 32 | orderedSet, |
81c2f20b | 33 | uppercase_escape, |
c5e8d7af PH |
34 | ) |
35 | ||
de7f3446 | 36 | class YoutubeBaseInfoExtractor(InfoExtractor): |
b2e8bc1b JMF |
37 | """Provide base functions for Youtube extractors""" |
38 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
83317f69 | 39 | _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' |
b2e8bc1b | 40 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' |
38c2e5b8 | 41 | _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' |
b2e8bc1b JMF |
42 | _NETRC_MACHINE = 'youtube' |
43 | # If True it will raise an error if no login info is provided | |
44 | _LOGIN_REQUIRED = False | |
45 | ||
b2e8bc1b | 46 | def _set_language(self): |
7cc3570e PH |
47 | return bool(self._download_webpage( |
48 | self._LANG_URL, None, | |
69ea8ca4 | 49 | note='Setting language', errnote='unable to set language', |
7cc3570e | 50 | fatal=False)) |
b2e8bc1b JMF |
51 | |
52 | def _login(self): | |
83317f69 | 53 | """ |
54 | Attempt to log in to YouTube. | |
55 | True is returned if successful or skipped. | |
56 | False is returned if login failed. | |
57 | ||
58 | If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. | |
59 | """ | |
b2e8bc1b JMF |
60 | (username, password) = self._get_login_info() |
61 | # No authentication to be performed | |
62 | if username is None: | |
63 | if self._LOGIN_REQUIRED: | |
69ea8ca4 | 64 | raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) |
83317f69 | 65 | return True |
b2e8bc1b | 66 | |
7cc3570e PH |
67 | login_page = self._download_webpage( |
68 | self._LOGIN_URL, None, | |
69ea8ca4 PH |
69 | note='Downloading login page', |
70 | errnote='unable to fetch login page', fatal=False) | |
7cc3570e PH |
71 | if login_page is False: |
72 | return | |
b2e8bc1b | 73 | |
795f28f8 | 74 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', |
78caa52a | 75 | login_page, 'Login GALX parameter') |
c5e8d7af | 76 | |
b2e8bc1b JMF |
77 | # Log in |
78 | login_form_strs = { | |
78caa52a PH |
79 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', |
80 | 'Email': username, | |
81 | 'GALX': galx, | |
82 | 'Passwd': password, | |
83 | ||
84 | 'PersistentCookie': 'yes', | |
85 | '_utf8': '霱', | |
86 | 'bgresponse': 'js_disabled', | |
87 | 'checkConnection': '', | |
88 | 'checkedDomains': 'youtube', | |
89 | 'dnConn': '', | |
90 | 'pstMsg': '0', | |
91 | 'rmShown': '1', | |
92 | 'secTok': '', | |
93 | 'signIn': 'Sign in', | |
94 | 'timeStmp': '', | |
95 | 'service': 'youtube', | |
96 | 'uilel': '3', | |
97 | 'hl': 'en_US', | |
b2e8bc1b | 98 | } |
83317f69 | 99 | |
b2e8bc1b JMF |
100 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode |
101 | # chokes on unicode | |
102 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
103 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
7cc3570e PH |
104 | |
105 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
106 | login_results = self._download_webpage( | |
107 | req, None, | |
69ea8ca4 | 108 | note='Logging in', errnote='unable to log in', fatal=False) |
7cc3570e PH |
109 | if login_results is False: |
110 | return False | |
83317f69 | 111 | |
112 | if re.search(r'id="errormsg_0_Passwd"', login_results) is not None: | |
69ea8ca4 | 113 | raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True) |
83317f69 | 114 | |
115 | # Two-Factor | |
116 | # TODO add SMS and phone call support - these require making a request and then prompting the user | |
117 | ||
118 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None: | |
119 | tfa_code = self._get_tfa_info() | |
120 | ||
121 | if tfa_code is None: | |
69ea8ca4 PH |
122 | self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>') |
123 | self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)') | |
83317f69 | 124 | return False |
125 | ||
126 | # Unlike the first login form, secTok and timeStmp are both required for the TFA form | |
127 | ||
128 | match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
129 | if match is None: | |
69ea8ca4 | 130 | self._downloader.report_warning('Failed to get secTok - did the page structure change?') |
83317f69 | 131 | secTok = match.group(1) |
132 | match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
133 | if match is None: | |
69ea8ca4 | 134 | self._downloader.report_warning('Failed to get timeStmp - did the page structure change?') |
83317f69 | 135 | timeStmp = match.group(1) |
136 | ||
137 | tfa_form_strs = { | |
78caa52a PH |
138 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', |
139 | 'smsToken': '', | |
140 | 'smsUserPin': tfa_code, | |
141 | 'smsVerifyPin': 'Verify', | |
142 | ||
143 | 'PersistentCookie': 'yes', | |
144 | 'checkConnection': '', | |
145 | 'checkedDomains': 'youtube', | |
146 | 'pstMsg': '1', | |
147 | 'secTok': secTok, | |
148 | 'timeStmp': timeStmp, | |
149 | 'service': 'youtube', | |
150 | 'hl': 'en_US', | |
83317f69 | 151 | } |
152 | tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in tfa_form_strs.items()) | |
153 | tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii') | |
154 | ||
155 | tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) | |
156 | tfa_results = self._download_webpage( | |
157 | tfa_req, None, | |
69ea8ca4 | 158 | note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) |
83317f69 | 159 | |
160 | if tfa_results is False: | |
161 | return False | |
162 | ||
163 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None: | |
69ea8ca4 | 164 | self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.') |
83317f69 | 165 | return False |
166 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None: | |
69ea8ca4 | 167 | self._downloader.report_warning('unable to log in - did the page structure change?') |
83317f69 | 168 | return False |
169 | if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None: | |
69ea8ca4 | 170 | self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.') |
83317f69 | 171 | return False |
172 | ||
7cc3570e | 173 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: |
69ea8ca4 | 174 | self._downloader.report_warning('unable to log in: bad username or password') |
b2e8bc1b JMF |
175 | return False |
176 | return True | |
177 | ||
178 | def _confirm_age(self): | |
179 | age_form = { | |
7cc3570e PH |
180 | 'next_url': '/', |
181 | 'action_confirm': 'Confirm', | |
182 | } | |
5700e779 JMF |
183 | req = compat_urllib_request.Request(self._AGE_URL, |
184 | compat_urllib_parse.urlencode(age_form).encode('ascii')) | |
7cc3570e PH |
185 | |
186 | self._download_webpage( | |
187 | req, None, | |
69ea8ca4 | 188 | note='Confirming age', errnote='Unable to confirm age') |
b2e8bc1b JMF |
189 | return True |
190 | ||
191 | def _real_initialize(self): | |
192 | if self._downloader is None: | |
193 | return | |
194 | if not self._set_language(): | |
195 | return | |
196 | if not self._login(): | |
197 | return | |
198 | self._confirm_age() | |
c5e8d7af | 199 | |
8377574c | 200 | |
de7f3446 | 201 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
78caa52a | 202 | IE_DESC = 'YouTube.com' |
cb7dfeea | 203 | _VALID_URL = r"""(?x)^ |
c5e8d7af | 204 | ( |
edb53e2d | 205 | (?:https?://|//) # http(s):// or protocol-independent URL |
cb7dfeea | 206 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| |
484aaeb2 | 207 | (?:www\.)?deturl\.com/www\.youtube\.com/| |
e70dc1d1 | 208 | (?:www\.)?pwnyoutube\.com/| |
f7000f3a | 209 | (?:www\.)?yourepeat\.com/| |
e69ae5b9 JMF |
210 | tube\.majestyc\.net/| |
211 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
c5e8d7af PH |
212 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
213 | (?: # the various things that can precede the ID: | |
214 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
215 | |(?: # or the v= param in all its forms | |
f7000f3a | 216 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
217 | (?:\?|\#!?) # the params delimiter ? or # or #! |
218 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
219 | v= | |
220 | ) | |
f4b05232 JMF |
221 | )) |
222 | |youtu\.be/ # just youtu.be/xxxx | |
edb53e2d | 223 | |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= |
f4b05232 | 224 | ) |
c5e8d7af | 225 | )? # all until now is optional -> you can pass the naked ID |
8963d9c2 | 226 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
9291475f | 227 | (?!.*?&list=) # combined list/video URLs are handled by the playlist IE |
c5e8d7af PH |
228 | (?(1).+)? # if we found the ID, everything can follow |
229 | $""" | |
c5e8d7af | 230 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
2c62dc26 PH |
231 | _formats = { |
232 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
233 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
234 | '13': {'ext': '3gp'}, | |
235 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
236 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
237 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
238 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
239 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
240 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
241 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
242 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
243 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
244 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
245 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
246 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
247 | ||
1d043b93 | 248 | |
86fe61c8 | 249 | # 3d videos |
43b81eb9 PH |
250 | '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}, |
251 | '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
252 | '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
253 | '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}, | |
254 | '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
255 | '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
256 | '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
836a086c | 257 | |
96fb5605 | 258 | # Apple HTTP Live Streaming |
43b81eb9 PH |
259 | '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, |
260 | '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}, | |
261 | '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}, | |
262 | '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}, | |
263 | '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}, | |
264 | '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
265 | '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}, | |
2c62dc26 PH |
266 | |
267 | # DASH mp4 video | |
43b81eb9 PH |
268 | '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
269 | '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
270 | '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
271 | '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
272 | '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
273 | '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
274 | '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
275 | '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
836a086c | 276 | |
f6f1fc92 | 277 | # Dash mp4 audio |
2c62dc26 PH |
278 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, |
279 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, | |
280 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, | |
836a086c AZ |
281 | |
282 | # Dash webm | |
e75cafe9 A |
283 | '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, |
284 | '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
285 | '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
286 | '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
287 | '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
288 | '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
289 | '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
290 | '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
291 | '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
292 | '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
293 | '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
294 | '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
295 | '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
3c80377b | 296 | '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
18061bba | 297 | '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, |
2c62dc26 PH |
298 | |
299 | # Dash webm audio | |
55db73ef | 300 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, |
e75cafe9 | 301 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, |
ce6b9a2d PH |
302 | |
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', | |
386 | 'title': 'Afrojack - The Spark ft. Spree Wilson', | |
387 | 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8', | |
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 | }, | |
2eb88d95 PH |
397 | ] |
398 | ||
e0df6211 PH |
399 | def __init__(self, *args, **kwargs): |
400 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
83799698 | 401 | self._player_cache = {} |
e0df6211 | 402 | |
c5e8d7af PH |
403 | def report_video_info_webpage_download(self, video_id): |
404 | """Report attempt to download video info webpage.""" | |
69ea8ca4 | 405 | self.to_screen('%s: Downloading video info webpage' % video_id) |
c5e8d7af | 406 | |
c5e8d7af PH |
407 | def report_information_extraction(self, video_id): |
408 | """Report attempt to extract video information.""" | |
69ea8ca4 | 409 | self.to_screen('%s: Extracting video information' % video_id) |
c5e8d7af PH |
410 | |
411 | def report_unavailable_format(self, video_id, format): | |
412 | """Report extracted video URL.""" | |
69ea8ca4 | 413 | self.to_screen('%s: Format %s not available' % (video_id, format)) |
c5e8d7af PH |
414 | |
415 | def report_rtmp_download(self): | |
416 | """Indicate the download will use the RTMP protocol.""" | |
69ea8ca4 | 417 | self.to_screen('RTMP download detected') |
c5e8d7af | 418 | |
60064c53 PH |
419 | def _signature_cache_id(self, example_sig): |
420 | """ Return a string representation of a signature """ | |
78caa52a | 421 | return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) |
60064c53 PH |
422 | |
423 | def _extract_signature_function(self, video_id, player_url, example_sig): | |
cf010131 | 424 | id_m = re.match( |
c081b35c | 425 | r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$', |
cf010131 | 426 | player_url) |
c081b35c PH |
427 | if not id_m: |
428 | raise ExtractorError('Cannot identify player %r' % player_url) | |
e0df6211 PH |
429 | player_type = id_m.group('ext') |
430 | player_id = id_m.group('id') | |
431 | ||
c4417ddb | 432 | # Read from filesystem cache |
60064c53 PH |
433 | func_id = '%s_%s_%s' % ( |
434 | player_type, player_id, self._signature_cache_id(example_sig)) | |
c4417ddb | 435 | assert os.path.basename(func_id) == func_id |
a0e07d31 | 436 | |
69ea8ca4 | 437 | cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id) |
a0e07d31 | 438 | if cache_spec is not None: |
78caa52a | 439 | return lambda s: ''.join(s[i] for i in cache_spec) |
83799698 | 440 | |
e0df6211 PH |
441 | if player_type == 'js': |
442 | code = self._download_webpage( | |
443 | player_url, video_id, | |
69ea8ca4 PH |
444 | note='Downloading %s player %s' % (player_type, player_id), |
445 | errnote='Download of %s failed' % player_url) | |
83799698 | 446 | res = self._parse_sig_js(code) |
c4417ddb | 447 | elif player_type == 'swf': |
e0df6211 PH |
448 | urlh = self._request_webpage( |
449 | player_url, video_id, | |
69ea8ca4 PH |
450 | note='Downloading %s player %s' % (player_type, player_id), |
451 | errnote='Download of %s failed' % player_url) | |
e0df6211 | 452 | code = urlh.read() |
83799698 | 453 | res = self._parse_sig_swf(code) |
e0df6211 PH |
454 | else: |
455 | assert False, 'Invalid player type %r' % player_type | |
456 | ||
a0e07d31 | 457 | if cache_spec is None: |
78caa52a | 458 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) |
a0e07d31 PH |
459 | cache_res = res(test_string) |
460 | cache_spec = [ord(c) for c in cache_res] | |
83799698 | 461 | |
69ea8ca4 | 462 | self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) |
83799698 PH |
463 | return res |
464 | ||
60064c53 | 465 | def _print_sig_code(self, func, example_sig): |
edf3e38e PH |
466 | def gen_sig_code(idxs): |
467 | def _genslice(start, end, step): | |
78caa52a | 468 | starts = '' if start == 0 else str(start) |
69ea8ca4 PH |
469 | ends = (':%d' % (end+step)) if end + step >= 0 else ':' |
470 | steps = '' if step == 1 else (':%d' % step) | |
78caa52a | 471 | return 's[%s%s%s]' % (starts, ends, steps) |
edf3e38e PH |
472 | |
473 | step = None | |
0ca96d48 PH |
474 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
475 | # set as soon as step is set | |
edf3e38e PH |
476 | for i, prev in zip(idxs[1:], idxs[:-1]): |
477 | if step is not None: | |
478 | if i - prev == step: | |
479 | continue | |
480 | yield _genslice(start, prev, step) | |
481 | step = None | |
482 | continue | |
483 | if i - prev in [-1, 1]: | |
484 | step = i - prev | |
485 | start = prev | |
486 | continue | |
487 | else: | |
78caa52a | 488 | yield 's[%d]' % prev |
edf3e38e | 489 | if step is None: |
78caa52a | 490 | yield 's[%d]' % i |
edf3e38e PH |
491 | else: |
492 | yield _genslice(start, i, step) | |
493 | ||
78caa52a | 494 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) |
c705320f | 495 | cache_res = func(test_string) |
edf3e38e | 496 | cache_spec = [ord(c) for c in cache_res] |
78caa52a | 497 | expr_code = ' + '.join(gen_sig_code(cache_spec)) |
60064c53 PH |
498 | signature_id_tuple = '(%s)' % ( |
499 | ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) | |
69ea8ca4 | 500 | code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n' |
78caa52a | 501 | ' return %s\n') % (signature_id_tuple, expr_code) |
69ea8ca4 | 502 | self.to_screen('Extracted signature function:\n' + code) |
edf3e38e | 503 | |
e0df6211 PH |
504 | def _parse_sig_js(self, jscode): |
505 | funcname = self._search_regex( | |
c26e9ac4 | 506 | r'signature=([$a-zA-Z]+)', jscode, |
78caa52a | 507 | 'Initial JS player signature function name') |
2b25cb5d PH |
508 | |
509 | jsi = JSInterpreter(jscode) | |
510 | initial_function = jsi.extract_function(funcname) | |
e0df6211 PH |
511 | return lambda s: initial_function([s]) |
512 | ||
513 | def _parse_sig_swf(self, file_contents): | |
54256267 | 514 | swfi = SWFInterpreter(file_contents) |
78caa52a | 515 | TARGET_CLASSNAME = 'SignatureDecipher' |
54256267 | 516 | searched_class = swfi.extract_class(TARGET_CLASSNAME) |
78caa52a | 517 | initial_function = swfi.extract_function(searched_class, 'decipher') |
e0df6211 PH |
518 | return lambda s: initial_function([s]) |
519 | ||
83799698 | 520 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
257a2501 | 521 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 522 | |
c8bf86d5 | 523 | if player_url is None: |
69ea8ca4 | 524 | raise ExtractorError('Cannot decrypt signature without player_url') |
920de7a2 | 525 | |
69ea8ca4 | 526 | if player_url.startswith('//'): |
78caa52a | 527 | player_url = 'https:' + player_url |
c8bf86d5 | 528 | try: |
62af3a0e | 529 | player_id = (player_url, self._signature_cache_id(s)) |
c8bf86d5 PH |
530 | if player_id not in self._player_cache: |
531 | func = self._extract_signature_function( | |
60064c53 | 532 | video_id, player_url, s |
c8bf86d5 PH |
533 | ) |
534 | self._player_cache[player_id] = func | |
535 | func = self._player_cache[player_id] | |
536 | if self._downloader.params.get('youtube_print_sig_code'): | |
60064c53 | 537 | self._print_sig_code(func, s) |
c8bf86d5 PH |
538 | return func(s) |
539 | except Exception as e: | |
540 | tb = traceback.format_exc() | |
541 | raise ExtractorError( | |
78caa52a | 542 | 'Signature extraction failed: ' + tb, cause=e) |
e0df6211 | 543 | |
1f343eaa | 544 | def _get_available_subtitles(self, video_id, webpage): |
de7f3446 | 545 | try: |
7fad1c63 | 546 | sub_list = self._download_webpage( |
38c2e5b8 | 547 | 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, |
7fad1c63 JMF |
548 | video_id, note=False) |
549 | except ExtractorError as err: | |
69ea8ca4 | 550 | self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err)) |
de7f3446 JMF |
551 | return {} |
552 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
553 | ||
554 | sub_lang_list = {} | |
555 | for l in lang_list: | |
556 | lang = l[1] | |
7e660ac1 LD |
557 | if lang in sub_lang_list: |
558 | continue | |
de7f3446 JMF |
559 | params = compat_urllib_parse.urlencode({ |
560 | 'lang': lang, | |
561 | 'v': video_id, | |
ca715127 | 562 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
c3197e3e | 563 | 'name': unescapeHTML(l[0]).encode('utf-8'), |
de7f3446 | 564 | }) |
78caa52a | 565 | url = 'https://www.youtube.com/api/timedtext?' + params |
de7f3446 JMF |
566 | sub_lang_list[lang] = url |
567 | if not sub_lang_list: | |
69ea8ca4 | 568 | self._downloader.report_warning('video doesn\'t have subtitles') |
de7f3446 JMF |
569 | return {} |
570 | return sub_lang_list | |
571 | ||
055e6f36 | 572 | def _get_available_automatic_caption(self, video_id, webpage): |
de7f3446 JMF |
573 | """We need the webpage for getting the captions url, pass it as an |
574 | argument to speed up the process.""" | |
ca715127 | 575 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
69ea8ca4 | 576 | self.to_screen('%s: Looking for automatic captions' % video_id) |
de7f3446 | 577 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) |
78caa52a | 578 | err_msg = 'Couldn\'t find automatic captions for %s' % video_id |
de7f3446 JMF |
579 | if mobj is None: |
580 | self._downloader.report_warning(err_msg) | |
581 | return {} | |
582 | player_config = json.loads(mobj.group(1)) | |
583 | try: | |
584 | args = player_config[u'args'] | |
585 | caption_url = args[u'ttsurl'] | |
586 | timestamp = args[u'timestamp'] | |
055e6f36 JMF |
587 | # We get the available subtitles |
588 | list_params = compat_urllib_parse.urlencode({ | |
589 | 'type': 'list', | |
590 | 'tlangs': 1, | |
591 | 'asrs': 1, | |
de7f3446 | 592 | }) |
055e6f36 | 593 | list_url = caption_url + '&' + list_params |
e26f8712 | 594 | caption_list = self._download_xml(list_url, video_id) |
e3dc22ca | 595 | original_lang_node = caption_list.find('track') |
f6a54188 | 596 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
69ea8ca4 | 597 | self._downloader.report_warning('Video doesn\'t have automatic captions') |
e3dc22ca JMF |
598 | return {} |
599 | original_lang = original_lang_node.attrib['lang_code'] | |
055e6f36 JMF |
600 | |
601 | sub_lang_list = {} | |
602 | for lang_node in caption_list.findall('target'): | |
603 | sub_lang = lang_node.attrib['lang_code'] | |
604 | params = compat_urllib_parse.urlencode({ | |
605 | 'lang': original_lang, | |
606 | 'tlang': sub_lang, | |
607 | 'fmt': sub_format, | |
608 | 'ts': timestamp, | |
609 | 'kind': 'asr', | |
610 | }) | |
611 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
612 | return sub_lang_list | |
de7f3446 JMF |
613 | # An extractor error can be raise by the download process if there are |
614 | # no automatic captions but there are subtitles | |
615 | except (KeyError, ExtractorError): | |
616 | self._downloader.report_warning(err_msg) | |
617 | return {} | |
618 | ||
97665381 PH |
619 | @classmethod |
620 | def extract_id(cls, url): | |
621 | mobj = re.match(cls._VALID_URL, url, re.VERBOSE) | |
c5e8d7af | 622 | if mobj is None: |
69ea8ca4 | 623 | raise ExtractorError('Invalid URL: %s' % url) |
c5e8d7af PH |
624 | video_id = mobj.group(2) |
625 | return video_id | |
626 | ||
1d043b93 JMF |
627 | def _extract_from_m3u8(self, manifest_url, video_id): |
628 | url_map = {} | |
629 | def _get_urls(_manifest): | |
630 | lines = _manifest.split('\n') | |
631 | urls = filter(lambda l: l and not l.startswith('#'), | |
632 | lines) | |
633 | return urls | |
78caa52a | 634 | manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest') |
1d043b93 JMF |
635 | formats_urls = _get_urls(manifest) |
636 | for format_url in formats_urls: | |
890f62e8 | 637 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
1d043b93 JMF |
638 | url_map[itag] = format_url |
639 | return url_map | |
640 | ||
1fb07d10 JG |
641 | def _extract_annotations(self, video_id): |
642 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
69ea8ca4 | 643 | return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') |
1fb07d10 | 644 | |
c5e8d7af | 645 | def _real_extract(self, url): |
7e8c0af0 | 646 | proto = ( |
78caa52a PH |
647 | 'http' if self._downloader.params.get('prefer_insecure', False) |
648 | else 'https') | |
7e8c0af0 | 649 | |
c5e8d7af PH |
650 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter |
651 | mobj = re.search(self._NEXT_URL_RE, url) | |
652 | if mobj: | |
7e8c0af0 | 653 | url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') |
97665381 | 654 | video_id = self.extract_id(url) |
c5e8d7af PH |
655 | |
656 | # Get video webpage | |
7e8c0af0 | 657 | url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
336c3a69 | 658 | video_webpage = self._download_webpage(url, video_id) |
c5e8d7af PH |
659 | |
660 | # Attempt to extract SWF player URL | |
e0df6211 | 661 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
c5e8d7af PH |
662 | if mobj is not None: |
663 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
664 | else: | |
665 | player_url = None | |
666 | ||
667 | # Get video info | |
668 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
669 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
670 | self.report_age_confirmation() | |
671 | age_gate = True | |
672 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
673 | # this can be viewed without login into Youtube | |
2c57c7fa JMF |
674 | data = compat_urllib_parse.urlencode({ |
675 | 'video_id': video_id, | |
676 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
c084c934 JMF |
677 | 'sts': self._search_regex( |
678 | r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'), | |
2c57c7fa | 679 | }) |
7e8c0af0 | 680 | video_info_url = proto + '://www.youtube.com/get_video_info?' + data |
c5e8d7af PH |
681 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
682 | note=False, | |
683 | errnote='unable to download video info webpage') | |
684 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
685 | else: |
686 | age_gate = False | |
687 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
7e8c0af0 | 688 | video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' |
c108eb73 JMF |
689 | % (video_id, el_type)) |
690 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
691 | note=False, | |
692 | errnote='unable to download video info webpage') | |
693 | video_info = compat_parse_qs(video_info_webpage) | |
694 | if 'token' in video_info: | |
695 | break | |
c5e8d7af PH |
696 | if 'token' not in video_info: |
697 | if 'reason' in video_info: | |
d11271dd | 698 | raise ExtractorError( |
78caa52a | 699 | 'YouTube said: %s' % video_info['reason'][0], |
d11271dd | 700 | expected=True, video_id=video_id) |
c5e8d7af | 701 | else: |
d11271dd | 702 | raise ExtractorError( |
78caa52a | 703 | '"token" parameter not in video info for unknown reason', |
d11271dd | 704 | video_id=video_id) |
c5e8d7af | 705 | |
1d699755 PH |
706 | if 'view_count' in video_info: |
707 | view_count = int(video_info['view_count'][0]) | |
708 | else: | |
709 | view_count = None | |
710 | ||
c5e8d7af PH |
711 | # Check for "rental" videos |
712 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
69ea8ca4 | 713 | raise ExtractorError('"rental" videos not supported') |
c5e8d7af PH |
714 | |
715 | # Start extracting information | |
716 | self.report_information_extraction(video_id) | |
717 | ||
718 | # uploader | |
719 | if 'author' not in video_info: | |
69ea8ca4 | 720 | raise ExtractorError('Unable to extract uploader name') |
c5e8d7af PH |
721 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) |
722 | ||
723 | # uploader_id | |
724 | video_uploader_id = None | |
725 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
726 | if mobj is not None: | |
727 | video_uploader_id = mobj.group(1) | |
728 | else: | |
69ea8ca4 | 729 | self._downloader.report_warning('unable to extract uploader nickname') |
c5e8d7af PH |
730 | |
731 | # title | |
a8c6b241 | 732 | if 'title' in video_info: |
aa92f063 | 733 | video_title = video_info['title'][0] |
a8c6b241 | 734 | else: |
69ea8ca4 | 735 | self._downloader.report_warning('Unable to extract video title') |
78caa52a | 736 | video_title = '_' |
c5e8d7af PH |
737 | |
738 | # thumbnail image | |
7763b04e JMF |
739 | # We try first to get a high quality image: |
740 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
741 | video_webpage, re.DOTALL) | |
742 | if m_thumb is not None: | |
743 | video_thumbnail = m_thumb.group(1) | |
744 | elif 'thumbnail_url' not in video_info: | |
69ea8ca4 | 745 | self._downloader.report_warning('unable to extract video thumbnail') |
f490e77e | 746 | video_thumbnail = None |
c5e8d7af PH |
747 | else: # don't panic if we can't find it |
748 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
749 | ||
750 | # upload date | |
751 | upload_date = None | |
ad3bc6ac | 752 | mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage) |
beee53de PH |
753 | if mobj is None: |
754 | mobj = re.search( | |
263bd4ec | 755 | r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>', |
beee53de | 756 | video_webpage) |
c5e8d7af PH |
757 | if mobj is not None: |
758 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
759 | upload_date = unified_strdate(upload_date) | |
760 | ||
55f7bd2d PH |
761 | m_cat_container = self._search_regex( |
762 | r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>', | |
763 | video_webpage, 'categories', fatal=False) | |
ec8deefc | 764 | if m_cat_container: |
ad3bc6ac | 765 | category = self._html_search_regex( |
01ed5c9b | 766 | r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', |
ad3bc6ac PH |
767 | default=None) |
768 | video_categories = None if category is None else [category] | |
769 | else: | |
770 | video_categories = None | |
ec8deefc | 771 | |
c5e8d7af PH |
772 | # description |
773 | video_description = get_element_by_id("eow-description", video_webpage) | |
774 | if video_description: | |
27dcce19 PH |
775 | video_description = re.sub(r'''(?x) |
776 | <a\s+ | |
777 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
778 | title="([^"]+)"\s+ | |
779 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
780 | class="yt-uix-redirect-link"\s*> | |
781 | [^<]+ | |
782 | </a> | |
783 | ''', r'\1', video_description) | |
c5e8d7af PH |
784 | video_description = clean_html(video_description) |
785 | else: | |
786 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
787 | if fd_mobj: | |
788 | video_description = unescapeHTML(fd_mobj.group(1)) | |
789 | else: | |
78caa52a | 790 | video_description = '' |
c5e8d7af | 791 | |
f30a38be | 792 | def _extract_count(count_name): |
46374a56 | 793 | count = self._search_regex( |
f30a38be JMF |
794 | r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name), |
795 | video_webpage, count_name, default=None) | |
336c3a69 JMF |
796 | if count is not None: |
797 | return int(count.replace(',', '')) | |
798 | return None | |
69ea8ca4 PH |
799 | like_count = _extract_count('like') |
800 | dislike_count = _extract_count('dislike') | |
336c3a69 | 801 | |
c5e8d7af | 802 | # subtitles |
d82134c3 | 803 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
c5e8d7af | 804 | |
c5e8d7af | 805 | if self._downloader.params.get('listsubtitles', False): |
d665f8d3 | 806 | self._list_available_subtitles(video_id, video_webpage) |
c5e8d7af PH |
807 | return |
808 | ||
809 | if 'length_seconds' not in video_info: | |
69ea8ca4 | 810 | self._downloader.report_warning('unable to extract video duration') |
b466b702 | 811 | video_duration = None |
c5e8d7af | 812 | else: |
b466b702 | 813 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
c5e8d7af | 814 | |
1fb07d10 JG |
815 | # annotations |
816 | video_annotations = None | |
817 | if self._downloader.params.get('writeannotations', False): | |
818 | video_annotations = self._extract_annotations(video_id) | |
819 | ||
c5e8d7af | 820 | # Decide which formats to download |
c5e8d7af | 821 | try: |
ae7ed920 | 822 | mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) |
50be92c1 PH |
823 | if not mobj: |
824 | raise ValueError('Could not find vevo ID') | |
ae7ed920 PH |
825 | json_code = uppercase_escape(mobj.group(1)) |
826 | ytplayer_config = json.loads(json_code) | |
3489b7d2 | 827 | args = ytplayer_config['args'] |
7ce7e394 JMF |
828 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
829 | # this signatures are encrypted | |
44d46655 | 830 | if 'url_encoded_fmt_stream_map' not in args: |
69ea8ca4 | 831 | raise ValueError('No stream_map present') # caught below |
00fe14fc JMF |
832 | re_signature = re.compile(r'[&,]s=') |
833 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
7ce7e394 | 834 | if m_s is not None: |
69ea8ca4 | 835 | self.to_screen('%s: Encrypted signatures detected.' % video_id) |
c5e8d7af | 836 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
78caa52a | 837 | m_s = re_signature.search(args.get('adaptive_fmts', '')) |
b7a68384 | 838 | if m_s is not None: |
00fe14fc JMF |
839 | if 'adaptive_fmts' in video_info: |
840 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
37b6d5f6 | 841 | else: |
00fe14fc | 842 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
c5e8d7af PH |
843 | except ValueError: |
844 | pass | |
845 | ||
dd27fd17 PH |
846 | def _map_to_format_list(urlmap): |
847 | formats = [] | |
848 | for itag, video_real_url in urlmap.items(): | |
849 | dct = { | |
850 | 'format_id': itag, | |
851 | 'url': video_real_url, | |
852 | 'player_url': player_url, | |
853 | } | |
0b65e5d4 PH |
854 | if itag in self._formats: |
855 | dct.update(self._formats[itag]) | |
dd27fd17 PH |
856 | formats.append(dct) |
857 | return formats | |
858 | ||
c5e8d7af PH |
859 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): |
860 | self.report_rtmp_download() | |
dd27fd17 PH |
861 | formats = [{ |
862 | 'format_id': '_rtmp', | |
863 | 'protocol': 'rtmp', | |
864 | 'url': video_info['conn'][0], | |
865 | 'player_url': player_url, | |
866 | }] | |
00fe14fc JMF |
867 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
868 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
869 | if 'rtmpe%3Dyes' in encoded_url_map: | |
a7055eb9 | 870 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
c5e8d7af | 871 | url_map = {} |
00fe14fc | 872 | for url_data_str in encoded_url_map.split(','): |
c5e8d7af | 873 | url_data = compat_parse_qs(url_data_str) |
201e9eaa PH |
874 | if 'itag' not in url_data or 'url' not in url_data: |
875 | continue | |
876 | format_id = url_data['itag'][0] | |
877 | url = url_data['url'][0] | |
878 | ||
879 | if 'sig' in url_data: | |
880 | url += '&signature=' + url_data['sig'][0] | |
881 | elif 's' in url_data: | |
882 | encrypted_sig = url_data['s'][0] | |
883 | ||
884 | if not age_gate: | |
885 | jsplayer_url_json = self._search_regex( | |
886 | r'"assets":.+?"js":\s*("[^"]+")', | |
78caa52a | 887 | video_webpage, 'JS player URL') |
201e9eaa PH |
888 | player_url = json.loads(jsplayer_url_json) |
889 | if player_url is None: | |
890 | player_url_json = self._search_regex( | |
891 | r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', | |
78caa52a | 892 | video_webpage, 'age gate player URL') |
201e9eaa PH |
893 | player_url = json.loads(player_url_json) |
894 | ||
895 | if self._downloader.params.get('verbose'): | |
cf010131 | 896 | if player_url is None: |
201e9eaa PH |
897 | player_version = 'unknown' |
898 | player_desc = 'unknown' | |
899 | else: | |
900 | if player_url.endswith('swf'): | |
901 | player_version = self._search_regex( | |
902 | r'-(.+?)(?:/watch_as3)?\.swf$', player_url, | |
78caa52a | 903 | 'flash player', fatal=False) |
201e9eaa | 904 | player_desc = 'flash player %s' % player_version |
cf010131 | 905 | else: |
201e9eaa PH |
906 | player_version = self._search_regex( |
907 | r'html5player-([^/]+?)(?:/html5player)?\.js', | |
908 | player_url, | |
909 | 'html5 player', fatal=False) | |
78caa52a | 910 | player_desc = 'html5 player %s' % player_version |
201e9eaa | 911 | |
60064c53 | 912 | parts_sizes = self._signature_cache_id(encrypted_sig) |
69ea8ca4 | 913 | self.to_screen('{%s} signature length %s, %s' % |
98eb1c3f | 914 | (format_id, parts_sizes, player_desc)) |
201e9eaa PH |
915 | |
916 | signature = self._decrypt_signature( | |
917 | encrypted_sig, video_id, player_url, age_gate) | |
918 | url += '&signature=' + signature | |
919 | if 'ratebypass' not in url: | |
920 | url += '&ratebypass=yes' | |
921 | url_map[format_id] = url | |
dd27fd17 | 922 | formats = _map_to_format_list(url_map) |
1d043b93 JMF |
923 | elif video_info.get('hlsvp'): |
924 | manifest_url = video_info['hlsvp'][0] | |
925 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
dd27fd17 | 926 | formats = _map_to_format_list(url_map) |
c5e8d7af | 927 | else: |
69ea8ca4 | 928 | raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
c5e8d7af | 929 | |
dd27fd17 | 930 | # Look for the DASH manifest |
d68f0cdb | 931 | if (self._downloader.params.get('youtube_include_dash_manifest', False)): |
dd27fd17 | 932 | try: |
d68f0cdb | 933 | # The DASH manifest used needs to be the one from the original video_webpage. |
934 | # The one found in get_video_info seems to be using different signatures. | |
935 | # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage. | |
936 | # Luckily, it seems, this case uses some kind of default signature (len == 86), so the | |
937 | # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here. | |
938 | if age_gate: | |
3489b7d2 | 939 | dash_manifest_url = video_info.get('dashmpd')[0] |
d68f0cdb | 940 | else: |
3489b7d2 | 941 | dash_manifest_url = ytplayer_config['args']['dashmpd'] |
d68f0cdb | 942 | def decrypt_sig(mobj): |
943 | s = mobj.group(1) | |
944 | dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) | |
945 | return '/signature/%s' % dec_s | |
946 | dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) | |
dd27fd17 | 947 | dash_doc = self._download_xml( |
d68f0cdb | 948 | dash_manifest_url, video_id, |
69ea8ca4 PH |
949 | note='Downloading DASH manifest', |
950 | errnote='Could not download DASH manifest') | |
951 | for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): | |
dd27fd17 PH |
952 | url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') |
953 | if url_el is None: | |
954 | continue | |
955 | format_id = r.attrib['id'] | |
956 | video_url = url_el.text | |
957 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) | |
958 | f = { | |
959 | 'format_id': format_id, | |
960 | 'url': video_url, | |
961 | 'width': int_or_none(r.attrib.get('width')), | |
962 | 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), | |
963 | 'asr': int_or_none(r.attrib.get('audioSamplingRate')), | |
964 | 'filesize': filesize, | |
965 | } | |
966 | try: | |
967 | existing_format = next( | |
968 | fo for fo in formats | |
969 | if fo['format_id'] == format_id) | |
970 | except StopIteration: | |
971 | f.update(self._formats.get(format_id, {})) | |
972 | formats.append(f) | |
973 | else: | |
974 | existing_format.update(f) | |
975 | ||
976 | except (ExtractorError, KeyError) as e: | |
69ea8ca4 | 977 | self.report_warning('Skipping DASH manifest: %s' % e, video_id) |
d80044c2 | 978 | |
4bcc7bd1 | 979 | self._sort_formats(formats) |
4ea3be0a | 980 | |
981 | return { | |
982 | 'id': video_id, | |
983 | 'uploader': video_uploader, | |
984 | 'uploader_id': video_uploader_id, | |
985 | 'upload_date': upload_date, | |
986 | 'title': video_title, | |
987 | 'thumbnail': video_thumbnail, | |
988 | 'description': video_description, | |
ec8deefc | 989 | 'categories': video_categories, |
4ea3be0a | 990 | 'subtitles': video_subtitles, |
991 | 'duration': video_duration, | |
992 | 'age_limit': 18 if age_gate else 0, | |
993 | 'annotations': video_annotations, | |
7e8c0af0 | 994 | 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, |
4ea3be0a | 995 | 'view_count': view_count, |
996 | 'like_count': like_count, | |
997 | 'dislike_count': dislike_count, | |
998 | 'formats': formats, | |
999 | } | |
c5e8d7af | 1000 | |
880e1c52 | 1001 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
78caa52a | 1002 | IE_DESC = 'YouTube.com playlists' |
d67cc9fa | 1003 | _VALID_URL = r"""(?x)(?: |
c5e8d7af PH |
1004 | (?:https?://)? |
1005 | (?:\w+\.)? | |
1006 | youtube\.com/ | |
1007 | (?: | |
1008 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
1009 | \? (?:.*?&)*? (?:p|a|list)= | |
1010 | | p/ | |
1011 | ) | |
d67cc9fa | 1012 | ( |
7d568f5a | 1013 | (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,} |
d67cc9fa JMF |
1014 | # Top tracks, they can also include dots |
1015 | |(?:MC)[\w\.]* | |
1016 | ) | |
c5e8d7af PH |
1017 | .* |
1018 | | | |
7d568f5a | 1019 | ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
c5e8d7af | 1020 | )""" |
dbb94fb0 | 1021 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' |
dcbb4580 | 1022 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' |
dbb94fb0 | 1023 | _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
78caa52a | 1024 | IE_NAME = 'youtube:playlist' |
81127aa5 PH |
1025 | _TESTS = [{ |
1026 | 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', | |
1027 | 'info_dict': { | |
1028 | 'title': 'ytdl test PL', | |
1029 | }, | |
1030 | 'playlist_count': 3, | |
9291475f PH |
1031 | }, { |
1032 | 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', | |
1033 | 'info_dict': { | |
1034 | 'title': 'YDL_Empty_List', | |
1035 | }, | |
1036 | 'playlist_count': 0, | |
1037 | }, { | |
1038 | 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', | |
1039 | 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', | |
1040 | 'info_dict': { | |
1041 | 'title': '29C3: Not my department', | |
1042 | }, | |
1043 | 'playlist_count': 95, | |
1044 | }, { | |
1045 | 'note': 'issue #673', | |
1046 | 'url': 'PLBB231211A4F62143', | |
1047 | 'info_dict': { | |
1048 | 'title': 'Team Fortress 2 (Class-based LP)', | |
1049 | }, | |
1050 | 'playlist_mincount': 26, | |
1051 | }, { | |
1052 | 'note': 'Large playlist', | |
1053 | 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', | |
1054 | 'info_dict': { | |
1055 | 'title': 'Uploads from Cauchemar', | |
1056 | }, | |
1057 | 'playlist_mincount': 799, | |
1058 | }, { | |
1059 | 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', | |
1060 | 'info_dict': { | |
1061 | 'title': 'YDL_safe_search', | |
1062 | }, | |
1063 | 'playlist_count': 2, | |
81127aa5 | 1064 | }] |
c5e8d7af | 1065 | |
880e1c52 JMF |
1066 | def _real_initialize(self): |
1067 | self._login() | |
1068 | ||
652cdaa2 | 1069 | def _ids_to_results(self, ids): |
c9cc0bf5 PH |
1070 | return [ |
1071 | self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
1072 | for vid_id in ids] | |
652cdaa2 JMF |
1073 | |
1074 | def _extract_mix(self, playlist_id): | |
1075 | # The mixes are generated from a a single video | |
1076 | # the id of the playlist is just 'RD' + video_id | |
7d4afc55 | 1077 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
c9cc0bf5 | 1078 | webpage = self._download_webpage( |
78caa52a | 1079 | url, playlist_id, 'Downloading Youtube mix') |
bc2f773b | 1080 | search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) |
c9cc0bf5 PH |
1081 | title_span = ( |
1082 | search_title('playlist-title') or | |
1083 | search_title('title long-title') or | |
1084 | search_title('title')) | |
76d1700b | 1085 | title = clean_html(title_span) |
c9cc0bf5 PH |
1086 | ids = orderedSet(re.findall( |
1087 | r'''(?xs)data-video-username=".*?".*? | |
1088 | href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id), | |
1089 | webpage)) | |
652cdaa2 JMF |
1090 | url_results = self._ids_to_results(ids) |
1091 | ||
1092 | return self.playlist_result(url_results, playlist_id, title) | |
1093 | ||
c5e8d7af PH |
1094 | def _real_extract(self, url): |
1095 | # Extract playlist id | |
d67cc9fa | 1096 | mobj = re.match(self._VALID_URL, url) |
c5e8d7af | 1097 | if mobj is None: |
69ea8ca4 | 1098 | raise ExtractorError('Invalid URL: %s' % url) |
47192f92 FV |
1099 | playlist_id = mobj.group(1) or mobj.group(2) |
1100 | ||
1101 | # Check if it's a video-specific URL | |
7c61bd36 | 1102 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
47192f92 FV |
1103 | if 'v' in query_dict: |
1104 | video_id = query_dict['v'][0] | |
1105 | if self._downloader.params.get('noplaylist'): | |
69ea8ca4 | 1106 | self.to_screen('Downloading just video %s because of --no-playlist' % video_id) |
7012b23c | 1107 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
47192f92 | 1108 | else: |
69ea8ca4 | 1109 | self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) |
c5e8d7af | 1110 | |
7d4afc55 | 1111 | if playlist_id.startswith('RD'): |
652cdaa2 JMF |
1112 | # Mixes require a custom extraction process |
1113 | return self._extract_mix(playlist_id) | |
0a688bc0 | 1114 | if playlist_id.startswith('TL'): |
69ea8ca4 | 1115 | raise ExtractorError('For downloading YouTube.com top lists, use ' |
78caa52a | 1116 | 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) |
652cdaa2 | 1117 | |
dbb94fb0 S |
1118 | url = self._TEMPLATE_URL % playlist_id |
1119 | page = self._download_webpage(url, playlist_id) | |
1120 | more_widget_html = content_html = page | |
1121 | ||
10c0e2d8 | 1122 | # Check if the playlist exists or is private |
e399853d | 1123 | if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None: |
10c0e2d8 | 1124 | raise ExtractorError( |
78caa52a | 1125 | 'The playlist doesn\'t exist or is private, use --username or ' |
10c0e2d8 JMF |
1126 | '--netrc to access it.', |
1127 | expected=True) | |
1128 | ||
dcbb4580 JMF |
1129 | # Extract the video ids from the playlist pages |
1130 | ids = [] | |
c5e8d7af | 1131 | |
755eb032 | 1132 | for page_num in itertools.count(1): |
dbb94fb0 | 1133 | matches = re.finditer(self._VIDEO_RE, content_html) |
6e47b51e JMF |
1134 | # We remove the duplicates and the link with index 0 |
1135 | # (it's not the first video of the playlist) | |
1136 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
dcbb4580 | 1137 | ids.extend(new_ids) |
c5e8d7af | 1138 | |
dbb94fb0 S |
1139 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) |
1140 | if not mobj: | |
c5e8d7af PH |
1141 | break |
1142 | ||
dbb94fb0 | 1143 | more = self._download_json( |
5912c639 PH |
1144 | 'https://youtube.com/%s' % mobj.group('more'), playlist_id, |
1145 | 'Downloading page #%s' % page_num, | |
1146 | transform_source=uppercase_escape) | |
dbb94fb0 S |
1147 | content_html = more['content_html'] |
1148 | more_widget_html = more['load_more_widget_html'] | |
1149 | ||
1150 | playlist_title = self._html_search_regex( | |
68eb8e90 | 1151 | r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', |
78caa52a | 1152 | page, 'title') |
c5e8d7af | 1153 | |
652cdaa2 | 1154 | url_results = self._ids_to_results(ids) |
dcbb4580 | 1155 | return self.playlist_result(url_results, playlist_id, playlist_title) |
c5e8d7af PH |
1156 | |
1157 | ||
0a688bc0 | 1158 | class YoutubeTopListIE(YoutubePlaylistIE): |
78caa52a | 1159 | IE_NAME = 'youtube:toplist' |
69ea8ca4 | 1160 | IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"' |
78caa52a | 1161 | ' (Example: "yttoplist:music:Top Tracks")') |
0a688bc0 | 1162 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' |
81127aa5 | 1163 | _TESTS = [] |
0a688bc0 JMF |
1164 | |
1165 | def _real_extract(self, url): | |
1166 | mobj = re.match(self._VALID_URL, url) | |
1167 | channel = mobj.group('chann') | |
1168 | title = mobj.group('title') | |
1169 | query = compat_urllib_parse.urlencode({'title': title}) | |
beddbc2a | 1170 | playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query) |
0a688bc0 | 1171 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) |
78caa52a | 1172 | link = self._html_search_regex(playlist_re, channel_page, 'list') |
0a688bc0 JMF |
1173 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) |
1174 | ||
1175 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
1176 | ids = [] | |
1177 | # sometimes the webpage doesn't contain the videos | |
1178 | # retry until we get them | |
1179 | for i in itertools.count(0): | |
78caa52a | 1180 | msg = 'Downloading Youtube mix' |
0a688bc0 JMF |
1181 | if i > 0: |
1182 | msg += ', retry #%d' % i | |
c9cc0bf5 | 1183 | |
0a688bc0 JMF |
1184 | webpage = self._download_webpage(url, title, msg) |
1185 | ids = orderedSet(re.findall(video_re, webpage)) | |
1186 | if ids: | |
1187 | break | |
1188 | url_results = self._ids_to_results(ids) | |
1189 | return self.playlist_result(url_results, playlist_title=title) | |
1190 | ||
1191 | ||
c5e8d7af | 1192 | class YoutubeChannelIE(InfoExtractor): |
78caa52a | 1193 | IE_DESC = 'YouTube.com channels' |
c5e8d7af | 1194 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
c5e8d7af | 1195 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
38c2e5b8 | 1196 | _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' |
78caa52a | 1197 | IE_NAME = 'youtube:channel' |
c5e8d7af PH |
1198 | |
1199 | def extract_videos_from_page(self, page): | |
1200 | ids_in_page = [] | |
1201 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
1202 | if mobj.group(1) not in ids_in_page: | |
1203 | ids_in_page.append(mobj.group(1)) | |
1204 | return ids_in_page | |
1205 | ||
1206 | def _real_extract(self, url): | |
1207 | # Extract channel id | |
1208 | mobj = re.match(self._VALID_URL, url) | |
1209 | if mobj is None: | |
69ea8ca4 | 1210 | raise ExtractorError('Invalid URL: %s' % url) |
c5e8d7af PH |
1211 | |
1212 | # Download channel page | |
1213 | channel_id = mobj.group(1) | |
1214 | video_ids = [] | |
b9643eed JMF |
1215 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
1216 | channel_page = self._download_webpage(url, channel_id) | |
31812a9e PH |
1217 | autogenerated = re.search(r'''(?x) |
1218 | class="[^"]*?(?: | |
1219 | channel-header-autogenerated-label| | |
1220 | yt-channel-title-autogenerated | |
1221 | )[^"]*"''', channel_page) is not None | |
c5e8d7af | 1222 | |
b9643eed JMF |
1223 | if autogenerated: |
1224 | # The videos are contained in a single page | |
1225 | # the ajax pages can't be used, they are empty | |
1226 | video_ids = self.extract_videos_from_page(channel_page) | |
1227 | else: | |
1228 | # Download all channel pages using the json-based channel_ajax query | |
1229 | for pagenum in itertools.count(1): | |
1230 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
81c2f20b | 1231 | page = self._download_json( |
69ea8ca4 | 1232 | url, channel_id, note='Downloading page #%s' % pagenum, |
81c2f20b PH |
1233 | transform_source=uppercase_escape) |
1234 | ||
b9643eed JMF |
1235 | ids_in_page = self.extract_videos_from_page(page['content_html']) |
1236 | video_ids.extend(ids_in_page) | |
1237 | ||
1238 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
1239 | break | |
c5e8d7af | 1240 | |
69ea8ca4 | 1241 | self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) |
c5e8d7af | 1242 | |
7012b23c PH |
1243 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1244 | for video_id in video_ids] | |
1245 | return self.playlist_result(url_entries, channel_id) | |
c5e8d7af PH |
1246 | |
1247 | ||
1248 | class YoutubeUserIE(InfoExtractor): | |
78caa52a | 1249 | IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' |
c9ae7b95 | 1250 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
38c2e5b8 | 1251 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' |
c5e8d7af | 1252 | _GDATA_PAGE_SIZE = 50 |
38c2e5b8 | 1253 | _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
78caa52a | 1254 | IE_NAME = 'youtube:user' |
c5e8d7af | 1255 | |
e3ea4790 | 1256 | @classmethod |
f4b05232 | 1257 | def suitable(cls, url): |
e3ea4790 JMF |
1258 | # Don't return True if the url can be extracted with other youtube |
1259 | # extractor, the regex would is too permissive and it would match. | |
1260 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1261 | if any(ie.suitable(url) for ie in other_ies): return False | |
f4b05232 JMF |
1262 | else: return super(YoutubeUserIE, cls).suitable(url) |
1263 | ||
c5e8d7af PH |
1264 | def _real_extract(self, url): |
1265 | # Extract username | |
1266 | mobj = re.match(self._VALID_URL, url) | |
1267 | if mobj is None: | |
69ea8ca4 | 1268 | raise ExtractorError('Invalid URL: %s' % url) |
c5e8d7af PH |
1269 | |
1270 | username = mobj.group(1) | |
1271 | ||
1272 | # Download video ids using YouTube Data API. Result size per | |
1273 | # query is limited (currently to 50 videos) so we need to query | |
1274 | # page by page until there are no video ids - it means we got | |
1275 | # all of them. | |
1276 | ||
b7ab0590 | 1277 | def download_page(pagenum): |
c5e8d7af PH |
1278 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
1279 | ||
1280 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
b7ab0590 PH |
1281 | page = self._download_webpage( |
1282 | gdata_url, username, | |
78caa52a | 1283 | 'Downloading video ids from %d to %d' % ( |
b7ab0590 | 1284 | start_index, start_index + self._GDATA_PAGE_SIZE)) |
c5e8d7af | 1285 | |
fd9cf738 JMF |
1286 | try: |
1287 | response = json.loads(page) | |
1288 | except ValueError as err: | |
69ea8ca4 | 1289 | raise ExtractorError('Invalid JSON in API response: ' + compat_str(err)) |
71c82637 | 1290 | if 'entry' not in response['feed']: |
b7ab0590 | 1291 | return |
fd9cf738 | 1292 | |
c5e8d7af | 1293 | # Extract video identifiers |
e302f9ce PH |
1294 | entries = response['feed']['entry'] |
1295 | for entry in entries: | |
1296 | title = entry['title']['$t'] | |
1297 | video_id = entry['id']['$t'].split('/')[-1] | |
b7ab0590 | 1298 | yield { |
e302f9ce PH |
1299 | '_type': 'url', |
1300 | 'url': video_id, | |
1301 | 'ie_key': 'Youtube', | |
b11cec41 | 1302 | 'id': video_id, |
e302f9ce | 1303 | 'title': title, |
b7ab0590 PH |
1304 | } |
1305 | url_results = PagedList(download_page, self._GDATA_PAGE_SIZE) | |
c5e8d7af | 1306 | |
7012b23c PH |
1307 | return self.playlist_result(url_results, playlist_title=username) |
1308 | ||
b05654f0 PH |
1309 | |
1310 | class YoutubeSearchIE(SearchInfoExtractor): | |
78caa52a PH |
1311 | IE_DESC = 'YouTube.com searches' |
1312 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' | |
b05654f0 | 1313 | _MAX_RESULTS = 1000 |
78caa52a | 1314 | IE_NAME = 'youtube:search' |
b05654f0 PH |
1315 | _SEARCH_KEY = 'ytsearch' |
1316 | ||
b05654f0 PH |
1317 | def _get_n_results(self, query, n): |
1318 | """Get a specified number of results for a query""" | |
1319 | ||
1320 | video_ids = [] | |
1321 | pagenum = 0 | |
1322 | limit = n | |
83d548ef | 1323 | PAGE_SIZE = 50 |
b05654f0 | 1324 | |
83d548ef PH |
1325 | while (PAGE_SIZE * pagenum) < limit: |
1326 | result_url = self._API_URL % ( | |
1327 | compat_urllib_parse.quote_plus(query.encode('utf-8')), | |
1328 | (PAGE_SIZE * pagenum) + 1) | |
7cc3570e | 1329 | data_json = self._download_webpage( |
69ea8ca4 PH |
1330 | result_url, video_id='query "%s"' % query, |
1331 | note='Downloading page %s' % (pagenum + 1), | |
1332 | errnote='Unable to download API page') | |
7cc3570e PH |
1333 | data = json.loads(data_json) |
1334 | api_response = data['data'] | |
1335 | ||
1336 | if 'items' not in api_response: | |
07ad22b8 | 1337 | raise ExtractorError( |
78caa52a | 1338 | '[youtube] No video results', expected=True) |
b05654f0 PH |
1339 | |
1340 | new_ids = list(video['id'] for video in api_response['items']) | |
1341 | video_ids += new_ids | |
1342 | ||
1343 | limit = min(n, api_response['totalItems']) | |
1344 | pagenum += 1 | |
1345 | ||
1346 | if len(video_ids) > n: | |
1347 | video_ids = video_ids[:n] | |
7012b23c PH |
1348 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1349 | for video_id in video_ids] | |
b05654f0 | 1350 | return self.playlist_result(videos, query) |
75dff0ee | 1351 | |
c9ae7b95 | 1352 | |
a3dd9248 | 1353 | class YoutubeSearchDateIE(YoutubeSearchIE): |
cb7fb546 | 1354 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
a3dd9248 CM |
1355 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
1356 | _SEARCH_KEY = 'ytsearchdate' | |
78caa52a | 1357 | IE_DESC = 'YouTube.com searches, newest videos first' |
75dff0ee | 1358 | |
c9ae7b95 PH |
1359 | |
1360 | class YoutubeSearchURLIE(InfoExtractor): | |
78caa52a PH |
1361 | IE_DESC = 'YouTube.com search URLs' |
1362 | IE_NAME = 'youtube:search_url' | |
c9ae7b95 PH |
1363 | _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)' |
1364 | ||
1365 | def _real_extract(self, url): | |
1366 | mobj = re.match(self._VALID_URL, url) | |
1367 | query = compat_urllib_parse.unquote_plus(mobj.group('query')) | |
1368 | ||
1369 | webpage = self._download_webpage(url, query) | |
1370 | result_code = self._search_regex( | |
78caa52a | 1371 | r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML') |
c9ae7b95 PH |
1372 | |
1373 | part_codes = re.findall( | |
1374 | r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code) | |
1375 | entries = [] | |
1376 | for part_code in part_codes: | |
1377 | part_title = self._html_search_regex( | |
6feb2d5e | 1378 | [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) |
c9ae7b95 PH |
1379 | part_url_snippet = self._html_search_regex( |
1380 | r'(?s)href="([^"]+)"', part_code, 'item URL') | |
1381 | part_url = compat_urlparse.urljoin( | |
1382 | 'https://www.youtube.com/', part_url_snippet) | |
1383 | entries.append({ | |
1384 | '_type': 'url', | |
1385 | 'url': part_url, | |
1386 | 'title': part_title, | |
1387 | }) | |
1388 | ||
1389 | return { | |
1390 | '_type': 'playlist', | |
1391 | 'entries': entries, | |
1392 | 'title': query, | |
1393 | } | |
1394 | ||
1395 | ||
75dff0ee | 1396 | class YoutubeShowIE(InfoExtractor): |
78caa52a | 1397 | IE_DESC = 'YouTube.com (multi-season) shows' |
75dff0ee | 1398 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
78caa52a | 1399 | IE_NAME = 'youtube:show' |
75dff0ee JMF |
1400 | |
1401 | def _real_extract(self, url): | |
1402 | mobj = re.match(self._VALID_URL, url) | |
1403 | show_name = mobj.group(1) | |
78caa52a | 1404 | webpage = self._download_webpage(url, show_name, 'Downloading show webpage') |
75dff0ee JMF |
1405 | # There's one playlist for each season of the show |
1406 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
69ea8ca4 | 1407 | self.to_screen('%s: Found %s seasons' % (show_name, len(m_seasons))) |
75dff0ee | 1408 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] |
04cc9617 JMF |
1409 | |
1410 | ||
b2e8bc1b | 1411 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1412 | """ |
1413 | Base class for extractors that fetch info from | |
1414 | http://www.youtube.com/feed_ajax | |
1415 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1416 | """ | |
b2e8bc1b | 1417 | _LOGIN_REQUIRED = True |
43ba5456 JMF |
1418 | # use action_load_personal_feed instead of action_load_system_feed |
1419 | _PERSONAL_FEED = False | |
04cc9617 | 1420 | |
d7ae0639 JMF |
1421 | @property |
1422 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1423 | action = 'action_load_system_feed' |
1424 | if self._PERSONAL_FEED: | |
1425 | action = 'action_load_personal_feed' | |
38c2e5b8 | 1426 | return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) |
d7ae0639 JMF |
1427 | |
1428 | @property | |
1429 | def IE_NAME(self): | |
78caa52a | 1430 | return 'youtube:%s' % self._FEED_NAME |
04cc9617 | 1431 | |
81f0259b | 1432 | def _real_initialize(self): |
b2e8bc1b | 1433 | self._login() |
81f0259b | 1434 | |
04cc9617 JMF |
1435 | def _real_extract(self, url): |
1436 | feed_entries = [] | |
0e44d838 JMF |
1437 | paging = 0 |
1438 | for i in itertools.count(1): | |
f6177462 | 1439 | info = self._download_json(self._FEED_TEMPLATE % paging, |
78caa52a PH |
1440 | '%s feed' % self._FEED_NAME, |
1441 | 'Downloading page %s' % i) | |
f6177462 | 1442 | feed_html = info.get('feed_html') or info.get('content_html') |
1a9b9649 | 1443 | load_more_widget_html = info.get('load_more_widget_html') or feed_html |
43ba5456 | 1444 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 | 1445 | ids = orderedSet(m.group(1) for m in m_ids) |
7012b23c PH |
1446 | feed_entries.extend( |
1447 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1448 | for video_id in ids) | |
05ee2b6d JMF |
1449 | mobj = re.search( |
1450 | r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)', | |
1a9b9649 | 1451 | load_more_widget_html) |
05ee2b6d | 1452 | if mobj is None: |
04cc9617 | 1453 | break |
05ee2b6d | 1454 | paging = mobj.group('paging') |
d7ae0639 JMF |
1455 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1456 | ||
d7ae0639 | 1457 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): |
78caa52a | 1458 | IE_DESC = 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' |
d7ae0639 JMF |
1459 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' |
1460 | _FEED_NAME = 'recommended' | |
78caa52a | 1461 | _PLAYLIST_TITLE = 'Youtube Recommended videos' |
c626a3d9 | 1462 | |
43ba5456 | 1463 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
78caa52a | 1464 | IE_DESC = 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' |
43ba5456 JMF |
1465 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' |
1466 | _FEED_NAME = 'watch_later' | |
78caa52a | 1467 | _PLAYLIST_TITLE = 'Youtube Watch Later' |
43ba5456 | 1468 | _PERSONAL_FEED = True |
c626a3d9 | 1469 | |
f459d170 | 1470 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
78caa52a PH |
1471 | IE_DESC = 'Youtube watch history, "ythistory" keyword (requires authentication)' |
1472 | _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory' | |
f459d170 JMF |
1473 | _FEED_NAME = 'history' |
1474 | _PERSONAL_FEED = True | |
78caa52a | 1475 | _PLAYLIST_TITLE = 'Youtube Watch History' |
f459d170 | 1476 | |
c626a3d9 | 1477 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
78caa52a PH |
1478 | IE_NAME = 'youtube:favorites' |
1479 | IE_DESC = 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
c7a7750d | 1480 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
c626a3d9 JMF |
1481 | _LOGIN_REQUIRED = True |
1482 | ||
1483 | def _real_extract(self, url): | |
1484 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
78caa52a | 1485 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id') |
c626a3d9 | 1486 | return self.url_result(playlist_id, 'YoutubePlaylist') |
15870e90 PH |
1487 | |
1488 | ||
1ed5b5c9 | 1489 | class YoutubeSubscriptionsIE(YoutubePlaylistIE): |
78caa52a PH |
1490 | IE_NAME = 'youtube:subscriptions' |
1491 | IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' | |
1ed5b5c9 | 1492 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' |
81127aa5 | 1493 | _TESTS = [] |
1ed5b5c9 JMF |
1494 | |
1495 | def _real_extract(self, url): | |
78caa52a | 1496 | title = 'Youtube Subscriptions' |
1ed5b5c9 JMF |
1497 | page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title) |
1498 | ||
1499 | # The extraction process is the same as for playlists, but the regex | |
1500 | # for the video ids doesn't contain an index | |
1501 | ids = [] | |
1502 | more_widget_html = content_html = page | |
1503 | ||
1504 | for page_num in itertools.count(1): | |
1505 | matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html) | |
1506 | new_ids = orderedSet(matches) | |
1507 | ids.extend(new_ids) | |
1508 | ||
1509 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) | |
1510 | if not mobj: | |
1511 | break | |
1512 | ||
1513 | more = self._download_json( | |
1514 | 'https://youtube.com/%s' % mobj.group('more'), title, | |
1515 | 'Downloading page #%s' % page_num, | |
1516 | transform_source=uppercase_escape) | |
1517 | content_html = more['content_html'] | |
1518 | more_widget_html = more['load_more_widget_html'] | |
1519 | ||
1520 | return { | |
1521 | '_type': 'playlist', | |
1522 | 'title': title, | |
1523 | 'entries': self._ids_to_results(ids), | |
1524 | } | |
1525 | ||
1526 | ||
15870e90 PH |
1527 | class YoutubeTruncatedURLIE(InfoExtractor): |
1528 | IE_NAME = 'youtube:truncated_url' | |
1529 | IE_DESC = False # Do not list | |
975d35db | 1530 | _VALID_URL = r'''(?x) |
c4808c60 PH |
1531 | (?:https?://)?[^/]+/watch\?(?: |
1532 | feature=[a-z_]+| | |
1533 | annotation_id=annotation_[^&]+ | |
1534 | )?$| | |
975d35db PH |
1535 | (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$ |
1536 | ''' | |
15870e90 | 1537 | |
c4808c60 PH |
1538 | _TESTS = [{ |
1539 | 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041', | |
1540 | 'only_matching': True, | |
dc2fc736 PH |
1541 | }, { |
1542 | 'url': 'http://www.youtube.com/watch?', | |
1543 | 'only_matching': True, | |
c4808c60 PH |
1544 | }] |
1545 | ||
15870e90 PH |
1546 | def _real_extract(self, url): |
1547 | raise ExtractorError( | |
78caa52a PH |
1548 | 'Did you forget to quote the URL? Remember that & is a meta ' |
1549 | 'character in most shells, so you want to put the URL in quotes, ' | |
1550 | 'like youtube-dl ' | |
1551 | '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' | |
1552 | ' or simply youtube-dl BaW_jenozKc .', | |
15870e90 | 1553 | expected=True) |