]>
Commit | Line | Data |
---|---|---|
1 | # coding: utf-8 | |
2 | ||
3 | from __future__ import unicode_literals | |
4 | ||
5 | ||
6 | import itertools | |
7 | import json | |
8 | import os.path | |
9 | import re | |
10 | import time | |
11 | import traceback | |
12 | ||
13 | from .common import InfoExtractor, SearchInfoExtractor | |
14 | from ..jsinterp import JSInterpreter | |
15 | from ..swfinterp import SWFInterpreter | |
16 | from ..compat import ( | |
17 | compat_chr, | |
18 | compat_parse_qs, | |
19 | compat_urllib_parse, | |
20 | compat_urllib_request, | |
21 | compat_urlparse, | |
22 | compat_str, | |
23 | ) | |
24 | from ..utils import ( | |
25 | clean_html, | |
26 | ExtractorError, | |
27 | float_or_none, | |
28 | get_element_by_attribute, | |
29 | get_element_by_id, | |
30 | int_or_none, | |
31 | orderedSet, | |
32 | unescapeHTML, | |
33 | unified_strdate, | |
34 | uppercase_escape, | |
35 | ) | |
36 | ||
37 | ||
38 | class YoutubeBaseInfoExtractor(InfoExtractor): | |
39 | """Provide base functions for Youtube extractors""" | |
40 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
41 | _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' | |
42 | _NETRC_MACHINE = 'youtube' | |
43 | # If True it will raise an error if no login info is provided | |
44 | _LOGIN_REQUIRED = False | |
45 | ||
46 | def _set_language(self): | |
47 | self._set_cookie( | |
48 | '.youtube.com', 'PREF', 'f1=50000000&hl=en', | |
49 | # YouTube sets the expire time to about two months | |
50 | expire_time=time.time() + 2 * 30 * 24 * 3600) | |
51 | ||
52 | def _ids_to_results(self, ids): | |
53 | return [ | |
54 | self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
55 | for vid_id in ids] | |
56 | ||
57 | def _login(self): | |
58 | """ | |
59 | Attempt to log in to YouTube. | |
60 | True is returned if successful or skipped. | |
61 | False is returned if login failed. | |
62 | ||
63 | If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. | |
64 | """ | |
65 | (username, password) = self._get_login_info() | |
66 | # No authentication to be performed | |
67 | if username is None: | |
68 | if self._LOGIN_REQUIRED: | |
69 | raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
70 | return True | |
71 | ||
72 | login_page = self._download_webpage( | |
73 | self._LOGIN_URL, None, | |
74 | note='Downloading login page', | |
75 | errnote='unable to fetch login page', fatal=False) | |
76 | if login_page is False: | |
77 | return | |
78 | ||
79 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', | |
80 | login_page, 'Login GALX parameter') | |
81 | ||
82 | # Log in | |
83 | login_form_strs = { | |
84 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
85 | 'Email': username, | |
86 | 'GALX': galx, | |
87 | 'Passwd': password, | |
88 | ||
89 | 'PersistentCookie': 'yes', | |
90 | '_utf8': '霱', | |
91 | 'bgresponse': 'js_disabled', | |
92 | 'checkConnection': '', | |
93 | 'checkedDomains': 'youtube', | |
94 | 'dnConn': '', | |
95 | 'pstMsg': '0', | |
96 | 'rmShown': '1', | |
97 | 'secTok': '', | |
98 | 'signIn': 'Sign in', | |
99 | 'timeStmp': '', | |
100 | 'service': 'youtube', | |
101 | 'uilel': '3', | |
102 | 'hl': 'en_US', | |
103 | } | |
104 | ||
105 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
106 | # chokes on unicode | |
107 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items()) | |
108 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
109 | ||
110 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
111 | login_results = self._download_webpage( | |
112 | req, None, | |
113 | note='Logging in', errnote='unable to log in', fatal=False) | |
114 | if login_results is False: | |
115 | return False | |
116 | ||
117 | if re.search(r'id="errormsg_0_Passwd"', login_results) is not None: | |
118 | raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True) | |
119 | ||
120 | # Two-Factor | |
121 | # TODO add SMS and phone call support - these require making a request and then prompting the user | |
122 | ||
123 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None: | |
124 | tfa_code = self._get_tfa_info() | |
125 | ||
126 | if tfa_code is None: | |
127 | self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>') | |
128 | self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)') | |
129 | return False | |
130 | ||
131 | # Unlike the first login form, secTok and timeStmp are both required for the TFA form | |
132 | ||
133 | match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
134 | if match is None: | |
135 | self._downloader.report_warning('Failed to get secTok - did the page structure change?') | |
136 | secTok = match.group(1) | |
137 | match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
138 | if match is None: | |
139 | self._downloader.report_warning('Failed to get timeStmp - did the page structure change?') | |
140 | timeStmp = match.group(1) | |
141 | ||
142 | tfa_form_strs = { | |
143 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
144 | 'smsToken': '', | |
145 | 'smsUserPin': tfa_code, | |
146 | 'smsVerifyPin': 'Verify', | |
147 | ||
148 | 'PersistentCookie': 'yes', | |
149 | 'checkConnection': '', | |
150 | 'checkedDomains': 'youtube', | |
151 | 'pstMsg': '1', | |
152 | 'secTok': secTok, | |
153 | 'timeStmp': timeStmp, | |
154 | 'service': 'youtube', | |
155 | 'hl': 'en_US', | |
156 | } | |
157 | tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items()) | |
158 | tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii') | |
159 | ||
160 | tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) | |
161 | tfa_results = self._download_webpage( | |
162 | tfa_req, None, | |
163 | note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) | |
164 | ||
165 | if tfa_results is False: | |
166 | return False | |
167 | ||
168 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None: | |
169 | self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.') | |
170 | return False | |
171 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None: | |
172 | self._downloader.report_warning('unable to log in - did the page structure change?') | |
173 | return False | |
174 | if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None: | |
175 | self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.') | |
176 | return False | |
177 | ||
178 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
179 | self._downloader.report_warning('unable to log in: bad username or password') | |
180 | return False | |
181 | return True | |
182 | ||
183 | def _real_initialize(self): | |
184 | if self._downloader is None: | |
185 | return | |
186 | self._set_language() | |
187 | if not self._login(): | |
188 | return | |
189 | ||
190 | ||
191 | class YoutubeIE(YoutubeBaseInfoExtractor): | |
192 | IE_DESC = 'YouTube.com' | |
193 | _VALID_URL = r"""(?x)^ | |
194 | ( | |
195 | (?:https?://|//) # http(s):// or protocol-independent URL | |
196 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| | |
197 | (?:www\.)?deturl\.com/www\.youtube\.com/| | |
198 | (?:www\.)?pwnyoutube\.com/| | |
199 | (?:www\.)?yourepeat\.com/| | |
200 | tube\.majestyc\.net/| | |
201 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
202 | (?:.*?\#/)? # handle anchor (#/) redirect urls | |
203 | (?: # the various things that can precede the ID: | |
204 | (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/ | |
205 | |(?: # or the v= param in all its forms | |
206 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) | |
207 | (?:\?|\#!?) # the params delimiter ? or # or #! | |
208 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
209 | v= | |
210 | ) | |
211 | )) | |
212 | |youtu\.be/ # just youtu.be/xxxx | |
213 | |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= | |
214 | ) | |
215 | )? # all until now is optional -> you can pass the naked ID | |
216 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID | |
217 | (?!.*?&list=) # combined list/video URLs are handled by the playlist IE | |
218 | (?(1).+)? # if we found the ID, everything can follow | |
219 | $""" | |
220 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' | |
221 | _formats = { | |
222 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
223 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
224 | '13': {'ext': '3gp'}, | |
225 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
226 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
227 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
228 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
229 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
230 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
231 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
232 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
233 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
234 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
235 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
236 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
237 | ||
238 | ||
239 | # 3d videos | |
240 | '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
241 | '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
242 | '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
243 | '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}, | |
244 | '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
245 | '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
246 | '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
247 | ||
248 | # Apple HTTP Live Streaming | |
249 | '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
250 | '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}, | |
251 | '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}, | |
252 | '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}, | |
253 | '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}, | |
254 | '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
255 | '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}, | |
256 | ||
257 | # DASH mp4 video | |
258 | '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
259 | '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
260 | '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
261 | '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
262 | '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
263 | '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559) | |
264 | '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
265 | '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
266 | '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, | |
267 | '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, | |
268 | '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}, | |
269 | ||
270 | # Dash mp4 audio | |
271 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, | |
272 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, | |
273 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, | |
274 | ||
275 | # Dash webm | |
276 | '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
277 | '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
278 | '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
279 | '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
280 | '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
281 | '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
282 | '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'}, | |
283 | '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
284 | '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
285 | '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
286 | '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
287 | '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
288 | '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
289 | '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
290 | '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
291 | '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
292 | '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
293 | '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
294 | '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
295 | '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'}, | |
296 | '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
297 | ||
298 | # Dash webm audio | |
299 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, | |
300 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, | |
301 | ||
302 | # Dash webm audio with opus inside | |
303 | '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, | |
304 | '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, | |
305 | '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, | |
306 | ||
307 | # RTMP (unnamed) | |
308 | '_rtmp': {'protocol': 'rtmp'}, | |
309 | } | |
310 | ||
311 | IE_NAME = 'youtube' | |
312 | _TESTS = [ | |
313 | { | |
314 | 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc', | |
315 | 'info_dict': { | |
316 | 'id': 'BaW_jenozKc', | |
317 | 'ext': 'mp4', | |
318 | 'title': 'youtube-dl test video "\'/\\ä↭𝕐', | |
319 | 'uploader': 'Philipp Hagemeister', | |
320 | 'uploader_id': 'phihag', | |
321 | 'upload_date': '20121002', | |
322 | '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 .', | |
323 | 'categories': ['Science & Technology'], | |
324 | 'like_count': int, | |
325 | 'dislike_count': int, | |
326 | } | |
327 | }, | |
328 | { | |
329 | 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY', | |
330 | 'note': 'Test generic use_cipher_signature video (#897)', | |
331 | 'info_dict': { | |
332 | 'id': 'UxxajLWwzqY', | |
333 | 'ext': 'mp4', | |
334 | 'upload_date': '20120506', | |
335 | 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', | |
336 | 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f', | |
337 | 'uploader': 'Icona Pop', | |
338 | 'uploader_id': 'IconaPop', | |
339 | } | |
340 | }, | |
341 | { | |
342 | 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ', | |
343 | 'note': 'Test VEVO video with age protection (#956)', | |
344 | 'info_dict': { | |
345 | 'id': '07FYdnEawAQ', | |
346 | 'ext': 'mp4', | |
347 | 'upload_date': '20130703', | |
348 | 'title': 'Justin Timberlake - Tunnel Vision (Explicit)', | |
349 | 'description': 'md5:64249768eec3bc4276236606ea996373', | |
350 | 'uploader': 'justintimberlakeVEVO', | |
351 | 'uploader_id': 'justintimberlakeVEVO', | |
352 | } | |
353 | }, | |
354 | { | |
355 | 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ', | |
356 | 'note': 'Embed-only video (#1746)', | |
357 | 'info_dict': { | |
358 | 'id': 'yZIXLfi8CZQ', | |
359 | 'ext': 'mp4', | |
360 | 'upload_date': '20120608', | |
361 | 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012', | |
362 | 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7', | |
363 | 'uploader': 'SET India', | |
364 | 'uploader_id': 'setindia' | |
365 | } | |
366 | }, | |
367 | { | |
368 | 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I', | |
369 | 'note': '256k DASH audio (format 141) via DASH manifest', | |
370 | 'info_dict': { | |
371 | 'id': 'a9LDPn-MO4I', | |
372 | 'ext': 'm4a', | |
373 | 'upload_date': '20121002', | |
374 | 'uploader_id': '8KVIDEO', | |
375 | 'description': '', | |
376 | 'uploader': '8KVIDEO', | |
377 | 'title': 'UHDTV TEST 8K VIDEO.mp4' | |
378 | }, | |
379 | 'params': { | |
380 | 'youtube_include_dash_manifest': True, | |
381 | 'format': '141', | |
382 | }, | |
383 | }, | |
384 | # DASH manifest with encrypted signature | |
385 | { | |
386 | 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA', | |
387 | 'info_dict': { | |
388 | 'id': 'IB3lcPjvWLA', | |
389 | 'ext': 'm4a', | |
390 | 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson', | |
391 | 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d', | |
392 | 'uploader': 'AfrojackVEVO', | |
393 | 'uploader_id': 'AfrojackVEVO', | |
394 | 'upload_date': '20131011', | |
395 | }, | |
396 | 'params': { | |
397 | 'youtube_include_dash_manifest': True, | |
398 | 'format': '141', | |
399 | }, | |
400 | }, | |
401 | # JS player signature function name containing $ | |
402 | { | |
403 | 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM', | |
404 | 'info_dict': { | |
405 | 'id': 'nfWlot6h_JM', | |
406 | 'ext': 'm4a', | |
407 | 'title': 'Taylor Swift - Shake It Off', | |
408 | 'description': 'md5:2acfda1b285bdd478ccec22f9918199d', | |
409 | 'uploader': 'TaylorSwiftVEVO', | |
410 | 'uploader_id': 'TaylorSwiftVEVO', | |
411 | 'upload_date': '20140818', | |
412 | }, | |
413 | 'params': { | |
414 | 'youtube_include_dash_manifest': True, | |
415 | 'format': '141', | |
416 | }, | |
417 | }, | |
418 | # Controversy video | |
419 | { | |
420 | 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8', | |
421 | 'info_dict': { | |
422 | 'id': 'T4XJQO3qol8', | |
423 | 'ext': 'mp4', | |
424 | 'upload_date': '20100909', | |
425 | 'uploader': 'The Amazing Atheist', | |
426 | 'uploader_id': 'TheAmazingAtheist', | |
427 | 'title': 'Burning Everyone\'s Koran', | |
428 | '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', | |
429 | } | |
430 | }, | |
431 | # Normal age-gate video (No vevo, embed allowed) | |
432 | { | |
433 | 'url': 'http://youtube.com/watch?v=HtVdAasjOgU', | |
434 | 'info_dict': { | |
435 | 'id': 'HtVdAasjOgU', | |
436 | 'ext': 'mp4', | |
437 | 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer', | |
438 | 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}', | |
439 | 'uploader': 'The Witcher', | |
440 | 'uploader_id': 'WitcherGame', | |
441 | 'upload_date': '20140605', | |
442 | }, | |
443 | }, | |
444 | # Age-gate video with encrypted signature | |
445 | { | |
446 | 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU', | |
447 | 'info_dict': { | |
448 | 'id': '6kLq3WMV1nU', | |
449 | 'ext': 'mp4', | |
450 | 'title': 'Dedication To My Ex (Miss That) (Lyric Video)', | |
451 | 'description': 'md5:33765bb339e1b47e7e72b5490139bb41', | |
452 | 'uploader': 'LloydVEVO', | |
453 | 'uploader_id': 'LloydVEVO', | |
454 | 'upload_date': '20110629', | |
455 | }, | |
456 | }, | |
457 | # video_info is None (https://github.com/rg3/youtube-dl/issues/4421) | |
458 | { | |
459 | 'url': '__2ABJjxzNo', | |
460 | 'info_dict': { | |
461 | 'id': '__2ABJjxzNo', | |
462 | 'ext': 'mp4', | |
463 | 'upload_date': '20100430', | |
464 | 'uploader_id': 'deadmau5', | |
465 | 'description': 'md5:12c56784b8032162bb936a5f76d55360', | |
466 | 'uploader': 'deadmau5', | |
467 | 'title': 'Deadmau5 - Some Chords (HD)', | |
468 | }, | |
469 | 'expected_warnings': [ | |
470 | 'DASH manifest missing', | |
471 | ] | |
472 | }, | |
473 | # Olympics (https://github.com/rg3/youtube-dl/issues/4431) | |
474 | { | |
475 | 'url': 'lqQg6PlCWgI', | |
476 | 'info_dict': { | |
477 | 'id': 'lqQg6PlCWgI', | |
478 | 'ext': 'mp4', | |
479 | 'upload_date': '20120731', | |
480 | 'uploader_id': 'olympic', | |
481 | 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', | |
482 | 'uploader': 'Olympics', | |
483 | 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', | |
484 | }, | |
485 | 'params': { | |
486 | 'skip_download': 'requires avconv', | |
487 | } | |
488 | }, | |
489 | # Non-square pixels | |
490 | { | |
491 | 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0', | |
492 | 'info_dict': { | |
493 | 'id': '_b-2C3KPAM0', | |
494 | 'ext': 'mp4', | |
495 | 'stretched_ratio': 16 / 9., | |
496 | 'upload_date': '20110310', | |
497 | 'uploader_id': 'AllenMeow', | |
498 | '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', | |
499 | 'uploader': '孫艾倫', | |
500 | 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人', | |
501 | }, | |
502 | }, | |
503 | # url_encoded_fmt_stream_map is empty string | |
504 | { | |
505 | 'url': 'qEJwOuvDf7I', | |
506 | 'info_dict': { | |
507 | 'id': 'qEJwOuvDf7I', | |
508 | 'ext': 'mp4', | |
509 | 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге', | |
510 | 'description': '', | |
511 | 'upload_date': '20150404', | |
512 | 'uploader_id': 'spbelect', | |
513 | 'uploader': 'Наблюдатели Петербурга', | |
514 | }, | |
515 | 'params': { | |
516 | 'skip_download': 'requires avconv', | |
517 | } | |
518 | }, | |
519 | ] | |
520 | ||
521 | def __init__(self, *args, **kwargs): | |
522 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
523 | self._player_cache = {} | |
524 | ||
525 | def report_video_info_webpage_download(self, video_id): | |
526 | """Report attempt to download video info webpage.""" | |
527 | self.to_screen('%s: Downloading video info webpage' % video_id) | |
528 | ||
529 | def report_information_extraction(self, video_id): | |
530 | """Report attempt to extract video information.""" | |
531 | self.to_screen('%s: Extracting video information' % video_id) | |
532 | ||
533 | def report_unavailable_format(self, video_id, format): | |
534 | """Report extracted video URL.""" | |
535 | self.to_screen('%s: Format %s not available' % (video_id, format)) | |
536 | ||
537 | def report_rtmp_download(self): | |
538 | """Indicate the download will use the RTMP protocol.""" | |
539 | self.to_screen('RTMP download detected') | |
540 | ||
541 | def _signature_cache_id(self, example_sig): | |
542 | """ Return a string representation of a signature """ | |
543 | return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) | |
544 | ||
545 | def _extract_signature_function(self, video_id, player_url, example_sig): | |
546 | id_m = re.match( | |
547 | r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$', | |
548 | player_url) | |
549 | if not id_m: | |
550 | raise ExtractorError('Cannot identify player %r' % player_url) | |
551 | player_type = id_m.group('ext') | |
552 | player_id = id_m.group('id') | |
553 | ||
554 | # Read from filesystem cache | |
555 | func_id = '%s_%s_%s' % ( | |
556 | player_type, player_id, self._signature_cache_id(example_sig)) | |
557 | assert os.path.basename(func_id) == func_id | |
558 | ||
559 | cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id) | |
560 | if cache_spec is not None: | |
561 | return lambda s: ''.join(s[i] for i in cache_spec) | |
562 | ||
563 | download_note = ( | |
564 | 'Downloading player %s' % player_url | |
565 | if self._downloader.params.get('verbose') else | |
566 | 'Downloading %s player %s' % (player_type, player_id) | |
567 | ) | |
568 | if player_type == 'js': | |
569 | code = self._download_webpage( | |
570 | player_url, video_id, | |
571 | note=download_note, | |
572 | errnote='Download of %s failed' % player_url) | |
573 | res = self._parse_sig_js(code) | |
574 | elif player_type == 'swf': | |
575 | urlh = self._request_webpage( | |
576 | player_url, video_id, | |
577 | note=download_note, | |
578 | errnote='Download of %s failed' % player_url) | |
579 | code = urlh.read() | |
580 | res = self._parse_sig_swf(code) | |
581 | else: | |
582 | assert False, 'Invalid player type %r' % player_type | |
583 | ||
584 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) | |
585 | cache_res = res(test_string) | |
586 | cache_spec = [ord(c) for c in cache_res] | |
587 | ||
588 | self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) | |
589 | return res | |
590 | ||
591 | def _print_sig_code(self, func, example_sig): | |
592 | def gen_sig_code(idxs): | |
593 | def _genslice(start, end, step): | |
594 | starts = '' if start == 0 else str(start) | |
595 | ends = (':%d' % (end + step)) if end + step >= 0 else ':' | |
596 | steps = '' if step == 1 else (':%d' % step) | |
597 | return 's[%s%s%s]' % (starts, ends, steps) | |
598 | ||
599 | step = None | |
600 | # Quelch pyflakes warnings - start will be set when step is set | |
601 | start = '(Never used)' | |
602 | for i, prev in zip(idxs[1:], idxs[:-1]): | |
603 | if step is not None: | |
604 | if i - prev == step: | |
605 | continue | |
606 | yield _genslice(start, prev, step) | |
607 | step = None | |
608 | continue | |
609 | if i - prev in [-1, 1]: | |
610 | step = i - prev | |
611 | start = prev | |
612 | continue | |
613 | else: | |
614 | yield 's[%d]' % prev | |
615 | if step is None: | |
616 | yield 's[%d]' % i | |
617 | else: | |
618 | yield _genslice(start, i, step) | |
619 | ||
620 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) | |
621 | cache_res = func(test_string) | |
622 | cache_spec = [ord(c) for c in cache_res] | |
623 | expr_code = ' + '.join(gen_sig_code(cache_spec)) | |
624 | signature_id_tuple = '(%s)' % ( | |
625 | ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) | |
626 | code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n' | |
627 | ' return %s\n') % (signature_id_tuple, expr_code) | |
628 | self.to_screen('Extracted signature function:\n' + code) | |
629 | ||
630 | def _parse_sig_js(self, jscode): | |
631 | funcname = self._search_regex( | |
632 | r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode, | |
633 | 'Initial JS player signature function name') | |
634 | ||
635 | jsi = JSInterpreter(jscode) | |
636 | initial_function = jsi.extract_function(funcname) | |
637 | return lambda s: initial_function([s]) | |
638 | ||
639 | def _parse_sig_swf(self, file_contents): | |
640 | swfi = SWFInterpreter(file_contents) | |
641 | TARGET_CLASSNAME = 'SignatureDecipher' | |
642 | searched_class = swfi.extract_class(TARGET_CLASSNAME) | |
643 | initial_function = swfi.extract_function(searched_class, 'decipher') | |
644 | return lambda s: initial_function([s]) | |
645 | ||
646 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): | |
647 | """Turn the encrypted s field into a working signature""" | |
648 | ||
649 | if player_url is None: | |
650 | raise ExtractorError('Cannot decrypt signature without player_url') | |
651 | ||
652 | if player_url.startswith('//'): | |
653 | player_url = 'https:' + player_url | |
654 | try: | |
655 | player_id = (player_url, self._signature_cache_id(s)) | |
656 | if player_id not in self._player_cache: | |
657 | func = self._extract_signature_function( | |
658 | video_id, player_url, s | |
659 | ) | |
660 | self._player_cache[player_id] = func | |
661 | func = self._player_cache[player_id] | |
662 | if self._downloader.params.get('youtube_print_sig_code'): | |
663 | self._print_sig_code(func, s) | |
664 | return func(s) | |
665 | except Exception as e: | |
666 | tb = traceback.format_exc() | |
667 | raise ExtractorError( | |
668 | 'Signature extraction failed: ' + tb, cause=e) | |
669 | ||
670 | def _get_subtitles(self, video_id, webpage): | |
671 | try: | |
672 | subs_doc = self._download_xml( | |
673 | 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
674 | video_id, note=False) | |
675 | except ExtractorError as err: | |
676 | self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err)) | |
677 | return {} | |
678 | ||
679 | sub_lang_list = {} | |
680 | for track in subs_doc.findall('track'): | |
681 | lang = track.attrib['lang_code'] | |
682 | if lang in sub_lang_list: | |
683 | continue | |
684 | sub_formats = [] | |
685 | for ext in ['sbv', 'vtt', 'srt']: | |
686 | params = compat_urllib_parse.urlencode({ | |
687 | 'lang': lang, | |
688 | 'v': video_id, | |
689 | 'fmt': ext, | |
690 | 'name': track.attrib['name'].encode('utf-8'), | |
691 | }) | |
692 | sub_formats.append({ | |
693 | 'url': 'https://www.youtube.com/api/timedtext?' + params, | |
694 | 'ext': ext, | |
695 | }) | |
696 | sub_lang_list[lang] = sub_formats | |
697 | if not sub_lang_list: | |
698 | self._downloader.report_warning('video doesn\'t have subtitles') | |
699 | return {} | |
700 | return sub_lang_list | |
701 | ||
702 | def _get_automatic_captions(self, video_id, webpage): | |
703 | """We need the webpage for getting the captions url, pass it as an | |
704 | argument to speed up the process.""" | |
705 | self.to_screen('%s: Looking for automatic captions' % video_id) | |
706 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
707 | err_msg = 'Couldn\'t find automatic captions for %s' % video_id | |
708 | if mobj is None: | |
709 | self._downloader.report_warning(err_msg) | |
710 | return {} | |
711 | player_config = json.loads(mobj.group(1)) | |
712 | try: | |
713 | args = player_config['args'] | |
714 | caption_url = args['ttsurl'] | |
715 | timestamp = args['timestamp'] | |
716 | # We get the available subtitles | |
717 | list_params = compat_urllib_parse.urlencode({ | |
718 | 'type': 'list', | |
719 | 'tlangs': 1, | |
720 | 'asrs': 1, | |
721 | }) | |
722 | list_url = caption_url + '&' + list_params | |
723 | caption_list = self._download_xml(list_url, video_id) | |
724 | original_lang_node = caption_list.find('track') | |
725 | if original_lang_node is None: | |
726 | self._downloader.report_warning('Video doesn\'t have automatic captions') | |
727 | return {} | |
728 | original_lang = original_lang_node.attrib['lang_code'] | |
729 | caption_kind = original_lang_node.attrib.get('kind', '') | |
730 | ||
731 | sub_lang_list = {} | |
732 | for lang_node in caption_list.findall('target'): | |
733 | sub_lang = lang_node.attrib['lang_code'] | |
734 | sub_formats = [] | |
735 | for ext in ['sbv', 'vtt', 'srt']: | |
736 | params = compat_urllib_parse.urlencode({ | |
737 | 'lang': original_lang, | |
738 | 'tlang': sub_lang, | |
739 | 'fmt': ext, | |
740 | 'ts': timestamp, | |
741 | 'kind': caption_kind, | |
742 | }) | |
743 | sub_formats.append({ | |
744 | 'url': caption_url + '&' + params, | |
745 | 'ext': ext, | |
746 | }) | |
747 | sub_lang_list[sub_lang] = sub_formats | |
748 | return sub_lang_list | |
749 | # An extractor error can be raise by the download process if there are | |
750 | # no automatic captions but there are subtitles | |
751 | except (KeyError, ExtractorError): | |
752 | self._downloader.report_warning(err_msg) | |
753 | return {} | |
754 | ||
755 | @classmethod | |
756 | def extract_id(cls, url): | |
757 | mobj = re.match(cls._VALID_URL, url, re.VERBOSE) | |
758 | if mobj is None: | |
759 | raise ExtractorError('Invalid URL: %s' % url) | |
760 | video_id = mobj.group(2) | |
761 | return video_id | |
762 | ||
763 | def _extract_from_m3u8(self, manifest_url, video_id): | |
764 | url_map = {} | |
765 | ||
766 | def _get_urls(_manifest): | |
767 | lines = _manifest.split('\n') | |
768 | urls = filter(lambda l: l and not l.startswith('#'), | |
769 | lines) | |
770 | return urls | |
771 | manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest') | |
772 | formats_urls = _get_urls(manifest) | |
773 | for format_url in formats_urls: | |
774 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') | |
775 | url_map[itag] = format_url | |
776 | return url_map | |
777 | ||
778 | def _extract_annotations(self, video_id): | |
779 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
780 | return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') | |
781 | ||
782 | def _parse_dash_manifest( | |
783 | self, video_id, dash_manifest_url, player_url, age_gate): | |
784 | def decrypt_sig(mobj): | |
785 | s = mobj.group(1) | |
786 | dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) | |
787 | return '/signature/%s' % dec_s | |
788 | dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) | |
789 | dash_doc = self._download_xml( | |
790 | dash_manifest_url, video_id, | |
791 | note='Downloading DASH manifest', | |
792 | errnote='Could not download DASH manifest') | |
793 | ||
794 | formats = [] | |
795 | for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'): | |
796 | mime_type = a.attrib.get('mimeType') | |
797 | for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'): | |
798 | url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') | |
799 | if url_el is None: | |
800 | continue | |
801 | if mime_type == 'text/vtt': | |
802 | # TODO implement WebVTT downloading | |
803 | pass | |
804 | elif mime_type.startswith('audio/') or mime_type.startswith('video/'): | |
805 | format_id = r.attrib['id'] | |
806 | video_url = url_el.text | |
807 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) | |
808 | f = { | |
809 | 'format_id': format_id, | |
810 | 'url': video_url, | |
811 | 'width': int_or_none(r.attrib.get('width')), | |
812 | 'height': int_or_none(r.attrib.get('height')), | |
813 | 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), | |
814 | 'asr': int_or_none(r.attrib.get('audioSamplingRate')), | |
815 | 'filesize': filesize, | |
816 | 'fps': int_or_none(r.attrib.get('frameRate')), | |
817 | } | |
818 | try: | |
819 | existing_format = next( | |
820 | fo for fo in formats | |
821 | if fo['format_id'] == format_id) | |
822 | except StopIteration: | |
823 | full_info = self._formats.get(format_id, {}).copy() | |
824 | full_info.update(f) | |
825 | formats.append(full_info) | |
826 | else: | |
827 | existing_format.update(f) | |
828 | else: | |
829 | self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) | |
830 | return formats | |
831 | ||
832 | def _real_extract(self, url): | |
833 | proto = ( | |
834 | 'http' if self._downloader.params.get('prefer_insecure', False) | |
835 | else 'https') | |
836 | ||
837 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
838 | mobj = re.search(self._NEXT_URL_RE, url) | |
839 | if mobj: | |
840 | url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
841 | video_id = self.extract_id(url) | |
842 | ||
843 | # Get video webpage | |
844 | url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id | |
845 | video_webpage = self._download_webpage(url, video_id) | |
846 | ||
847 | # Attempt to extract SWF player URL | |
848 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
849 | if mobj is not None: | |
850 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
851 | else: | |
852 | player_url = None | |
853 | ||
854 | # Get video info | |
855 | embed_webpage = None | |
856 | if re.search(r'player-age-gate-content">', video_webpage) is not None: | |
857 | age_gate = True | |
858 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
859 | # this can be viewed without login into Youtube | |
860 | url = proto + '://www.youtube.com/embed/%s' % video_id | |
861 | embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage') | |
862 | data = compat_urllib_parse.urlencode({ | |
863 | 'video_id': video_id, | |
864 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
865 | 'sts': self._search_regex( | |
866 | r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''), | |
867 | }) | |
868 | video_info_url = proto + '://www.youtube.com/get_video_info?' + data | |
869 | video_info_webpage = self._download_webpage( | |
870 | video_info_url, video_id, | |
871 | note='Refetching age-gated info webpage', | |
872 | errnote='unable to download video info webpage') | |
873 | video_info = compat_parse_qs(video_info_webpage) | |
874 | else: | |
875 | age_gate = False | |
876 | try: | |
877 | # Try looking directly into the video webpage | |
878 | mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) | |
879 | if not mobj: | |
880 | raise ValueError('Could not find ytplayer.config') # caught below | |
881 | json_code = uppercase_escape(mobj.group(1)) | |
882 | ytplayer_config = json.loads(json_code) | |
883 | args = ytplayer_config['args'] | |
884 | # Convert to the same format returned by compat_parse_qs | |
885 | video_info = dict((k, [v]) for k, v in args.items()) | |
886 | if not args.get('url_encoded_fmt_stream_map'): | |
887 | raise ValueError('No stream_map present') # caught below | |
888 | except ValueError: | |
889 | # We fallback to the get_video_info pages (used by the embed page) | |
890 | self.report_video_info_webpage_download(video_id) | |
891 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
892 | video_info_url = ( | |
893 | '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
894 | % (proto, video_id, el_type)) | |
895 | video_info_webpage = self._download_webpage( | |
896 | video_info_url, | |
897 | video_id, note=False, | |
898 | errnote='unable to download video info webpage') | |
899 | video_info = compat_parse_qs(video_info_webpage) | |
900 | if 'token' in video_info: | |
901 | break | |
902 | if 'token' not in video_info: | |
903 | if 'reason' in video_info: | |
904 | raise ExtractorError( | |
905 | 'YouTube said: %s' % video_info['reason'][0], | |
906 | expected=True, video_id=video_id) | |
907 | else: | |
908 | raise ExtractorError( | |
909 | '"token" parameter not in video info for unknown reason', | |
910 | video_id=video_id) | |
911 | ||
912 | if 'view_count' in video_info: | |
913 | view_count = int(video_info['view_count'][0]) | |
914 | else: | |
915 | view_count = None | |
916 | ||
917 | # Check for "rental" videos | |
918 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
919 | raise ExtractorError('"rental" videos not supported') | |
920 | ||
921 | # Start extracting information | |
922 | self.report_information_extraction(video_id) | |
923 | ||
924 | # uploader | |
925 | if 'author' not in video_info: | |
926 | raise ExtractorError('Unable to extract uploader name') | |
927 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
928 | ||
929 | # uploader_id | |
930 | video_uploader_id = None | |
931 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
932 | if mobj is not None: | |
933 | video_uploader_id = mobj.group(1) | |
934 | else: | |
935 | self._downloader.report_warning('unable to extract uploader nickname') | |
936 | ||
937 | # title | |
938 | if 'title' in video_info: | |
939 | video_title = video_info['title'][0] | |
940 | else: | |
941 | self._downloader.report_warning('Unable to extract video title') | |
942 | video_title = '_' | |
943 | ||
944 | # thumbnail image | |
945 | # We try first to get a high quality image: | |
946 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
947 | video_webpage, re.DOTALL) | |
948 | if m_thumb is not None: | |
949 | video_thumbnail = m_thumb.group(1) | |
950 | elif 'thumbnail_url' not in video_info: | |
951 | self._downloader.report_warning('unable to extract video thumbnail') | |
952 | video_thumbnail = None | |
953 | else: # don't panic if we can't find it | |
954 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
955 | ||
956 | # upload date | |
957 | upload_date = None | |
958 | mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage) | |
959 | if mobj is None: | |
960 | mobj = re.search( | |
961 | r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>', | |
962 | video_webpage) | |
963 | if mobj is not None: | |
964 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
965 | upload_date = unified_strdate(upload_date) | |
966 | ||
967 | m_cat_container = self._search_regex( | |
968 | r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>', | |
969 | video_webpage, 'categories', default=None) | |
970 | if m_cat_container: | |
971 | category = self._html_search_regex( | |
972 | r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', | |
973 | default=None) | |
974 | video_categories = None if category is None else [category] | |
975 | else: | |
976 | video_categories = None | |
977 | ||
978 | # description | |
979 | video_description = get_element_by_id("eow-description", video_webpage) | |
980 | if video_description: | |
981 | video_description = re.sub(r'''(?x) | |
982 | <a\s+ | |
983 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
984 | title="([^"]+)"\s+ | |
985 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
986 | class="yt-uix-redirect-link"\s*> | |
987 | [^<]+ | |
988 | </a> | |
989 | ''', r'\1', video_description) | |
990 | video_description = clean_html(video_description) | |
991 | else: | |
992 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
993 | if fd_mobj: | |
994 | video_description = unescapeHTML(fd_mobj.group(1)) | |
995 | else: | |
996 | video_description = '' | |
997 | ||
998 | def _extract_count(count_name): | |
999 | count = self._search_regex( | |
1000 | r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name), | |
1001 | video_webpage, count_name, default=None) | |
1002 | if count is not None: | |
1003 | return int(count.replace(',', '')) | |
1004 | return None | |
1005 | like_count = _extract_count('like') | |
1006 | dislike_count = _extract_count('dislike') | |
1007 | ||
1008 | # subtitles | |
1009 | video_subtitles = self.extract_subtitles(video_id, video_webpage) | |
1010 | automatic_captions = self.extract_automatic_captions(video_id, video_webpage) | |
1011 | ||
1012 | if 'length_seconds' not in video_info: | |
1013 | self._downloader.report_warning('unable to extract video duration') | |
1014 | video_duration = None | |
1015 | else: | |
1016 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) | |
1017 | ||
1018 | # annotations | |
1019 | video_annotations = None | |
1020 | if self._downloader.params.get('writeannotations', False): | |
1021 | video_annotations = self._extract_annotations(video_id) | |
1022 | ||
1023 | def _map_to_format_list(urlmap): | |
1024 | formats = [] | |
1025 | for itag, video_real_url in urlmap.items(): | |
1026 | dct = { | |
1027 | 'format_id': itag, | |
1028 | 'url': video_real_url, | |
1029 | 'player_url': player_url, | |
1030 | } | |
1031 | if itag in self._formats: | |
1032 | dct.update(self._formats[itag]) | |
1033 | formats.append(dct) | |
1034 | return formats | |
1035 | ||
1036 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
1037 | self.report_rtmp_download() | |
1038 | formats = [{ | |
1039 | 'format_id': '_rtmp', | |
1040 | 'protocol': 'rtmp', | |
1041 | 'url': video_info['conn'][0], | |
1042 | 'player_url': player_url, | |
1043 | }] | |
1044 | elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1: | |
1045 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0] | |
1046 | if 'rtmpe%3Dyes' in encoded_url_map: | |
1047 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
1048 | url_map = {} | |
1049 | for url_data_str in encoded_url_map.split(','): | |
1050 | url_data = compat_parse_qs(url_data_str) | |
1051 | if 'itag' not in url_data or 'url' not in url_data: | |
1052 | continue | |
1053 | format_id = url_data['itag'][0] | |
1054 | url = url_data['url'][0] | |
1055 | ||
1056 | if 'sig' in url_data: | |
1057 | url += '&signature=' + url_data['sig'][0] | |
1058 | elif 's' in url_data: | |
1059 | encrypted_sig = url_data['s'][0] | |
1060 | ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")' | |
1061 | ||
1062 | jsplayer_url_json = self._search_regex( | |
1063 | ASSETS_RE, | |
1064 | embed_webpage if age_gate else video_webpage, | |
1065 | 'JS player URL (1)', default=None) | |
1066 | if not jsplayer_url_json and not age_gate: | |
1067 | # We need the embed website after all | |
1068 | if embed_webpage is None: | |
1069 | embed_url = proto + '://www.youtube.com/embed/%s' % video_id | |
1070 | embed_webpage = self._download_webpage( | |
1071 | embed_url, video_id, 'Downloading embed webpage') | |
1072 | jsplayer_url_json = self._search_regex( | |
1073 | ASSETS_RE, embed_webpage, 'JS player URL') | |
1074 | ||
1075 | player_url = json.loads(jsplayer_url_json) | |
1076 | if player_url is None: | |
1077 | player_url_json = self._search_regex( | |
1078 | r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', | |
1079 | video_webpage, 'age gate player URL') | |
1080 | player_url = json.loads(player_url_json) | |
1081 | ||
1082 | if self._downloader.params.get('verbose'): | |
1083 | if player_url is None: | |
1084 | player_version = 'unknown' | |
1085 | player_desc = 'unknown' | |
1086 | else: | |
1087 | if player_url.endswith('swf'): | |
1088 | player_version = self._search_regex( | |
1089 | r'-(.+?)(?:/watch_as3)?\.swf$', player_url, | |
1090 | 'flash player', fatal=False) | |
1091 | player_desc = 'flash player %s' % player_version | |
1092 | else: | |
1093 | player_version = self._search_regex( | |
1094 | r'html5player-([^/]+?)(?:/html5player)?\.js', | |
1095 | player_url, | |
1096 | 'html5 player', fatal=False) | |
1097 | player_desc = 'html5 player %s' % player_version | |
1098 | ||
1099 | parts_sizes = self._signature_cache_id(encrypted_sig) | |
1100 | self.to_screen('{%s} signature length %s, %s' % | |
1101 | (format_id, parts_sizes, player_desc)) | |
1102 | ||
1103 | signature = self._decrypt_signature( | |
1104 | encrypted_sig, video_id, player_url, age_gate) | |
1105 | url += '&signature=' + signature | |
1106 | if 'ratebypass' not in url: | |
1107 | url += '&ratebypass=yes' | |
1108 | url_map[format_id] = url | |
1109 | formats = _map_to_format_list(url_map) | |
1110 | elif video_info.get('hlsvp'): | |
1111 | manifest_url = video_info['hlsvp'][0] | |
1112 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
1113 | formats = _map_to_format_list(url_map) | |
1114 | else: | |
1115 | raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') | |
1116 | ||
1117 | # Look for the DASH manifest | |
1118 | if self._downloader.params.get('youtube_include_dash_manifest', True): | |
1119 | dash_mpd = video_info.get('dashmpd') | |
1120 | if dash_mpd: | |
1121 | dash_manifest_url = dash_mpd[0] | |
1122 | try: | |
1123 | dash_formats = self._parse_dash_manifest( | |
1124 | video_id, dash_manifest_url, player_url, age_gate) | |
1125 | except (ExtractorError, KeyError) as e: | |
1126 | self.report_warning( | |
1127 | 'Skipping DASH manifest: %r' % e, video_id) | |
1128 | else: | |
1129 | # Remove the formats we found through non-DASH, they | |
1130 | # contain less info and it can be wrong, because we use | |
1131 | # fixed values (for example the resolution). See | |
1132 | # https://github.com/rg3/youtube-dl/issues/5774 for an | |
1133 | # example. | |
1134 | dash_keys = set(df['format_id'] for df in dash_formats) | |
1135 | formats = [f for f in formats if f['format_id'] not in dash_keys] | |
1136 | formats.extend(dash_formats) | |
1137 | ||
1138 | # Check for malformed aspect ratio | |
1139 | stretched_m = re.search( | |
1140 | r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">', | |
1141 | video_webpage) | |
1142 | if stretched_m: | |
1143 | ratio = float(stretched_m.group('w')) / float(stretched_m.group('h')) | |
1144 | for f in formats: | |
1145 | if f.get('vcodec') != 'none': | |
1146 | f['stretched_ratio'] = ratio | |
1147 | ||
1148 | self._sort_formats(formats) | |
1149 | ||
1150 | return { | |
1151 | 'id': video_id, | |
1152 | 'uploader': video_uploader, | |
1153 | 'uploader_id': video_uploader_id, | |
1154 | 'upload_date': upload_date, | |
1155 | 'title': video_title, | |
1156 | 'thumbnail': video_thumbnail, | |
1157 | 'description': video_description, | |
1158 | 'categories': video_categories, | |
1159 | 'subtitles': video_subtitles, | |
1160 | 'automatic_captions': automatic_captions, | |
1161 | 'duration': video_duration, | |
1162 | 'age_limit': 18 if age_gate else 0, | |
1163 | 'annotations': video_annotations, | |
1164 | 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, | |
1165 | 'view_count': view_count, | |
1166 | 'like_count': like_count, | |
1167 | 'dislike_count': dislike_count, | |
1168 | 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]), | |
1169 | 'formats': formats, | |
1170 | } | |
1171 | ||
1172 | ||
1173 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): | |
1174 | IE_DESC = 'YouTube.com playlists' | |
1175 | _VALID_URL = r"""(?x)(?: | |
1176 | (?:https?://)? | |
1177 | (?:\w+\.)? | |
1178 | youtube\.com/ | |
1179 | (?: | |
1180 | (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) | |
1181 | \? (?:.*?&)*? (?:p|a|list)= | |
1182 | | p/ | |
1183 | ) | |
1184 | ( | |
1185 | (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,} | |
1186 | # Top tracks, they can also include dots | |
1187 | |(?:MC)[\w\.]* | |
1188 | ) | |
1189 | .* | |
1190 | | | |
1191 | ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,}) | |
1192 | )""" | |
1193 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' | |
1194 | _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' | |
1195 | IE_NAME = 'youtube:playlist' | |
1196 | _TESTS = [{ | |
1197 | 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', | |
1198 | 'info_dict': { | |
1199 | 'title': 'ytdl test PL', | |
1200 | 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', | |
1201 | }, | |
1202 | 'playlist_count': 3, | |
1203 | }, { | |
1204 | 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', | |
1205 | 'info_dict': { | |
1206 | 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', | |
1207 | 'title': 'YDL_Empty_List', | |
1208 | }, | |
1209 | 'playlist_count': 0, | |
1210 | }, { | |
1211 | 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', | |
1212 | 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', | |
1213 | 'info_dict': { | |
1214 | 'title': '29C3: Not my department', | |
1215 | 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', | |
1216 | }, | |
1217 | 'playlist_count': 95, | |
1218 | }, { | |
1219 | 'note': 'issue #673', | |
1220 | 'url': 'PLBB231211A4F62143', | |
1221 | 'info_dict': { | |
1222 | 'title': '[OLD]Team Fortress 2 (Class-based LP)', | |
1223 | 'id': 'PLBB231211A4F62143', | |
1224 | }, | |
1225 | 'playlist_mincount': 26, | |
1226 | }, { | |
1227 | 'note': 'Large playlist', | |
1228 | 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', | |
1229 | 'info_dict': { | |
1230 | 'title': 'Uploads from Cauchemar', | |
1231 | 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q', | |
1232 | }, | |
1233 | 'playlist_mincount': 799, | |
1234 | }, { | |
1235 | 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', | |
1236 | 'info_dict': { | |
1237 | 'title': 'YDL_safe_search', | |
1238 | 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', | |
1239 | }, | |
1240 | 'playlist_count': 2, | |
1241 | }, { | |
1242 | 'note': 'embedded', | |
1243 | 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', | |
1244 | 'playlist_count': 4, | |
1245 | 'info_dict': { | |
1246 | 'title': 'JODA15', | |
1247 | 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', | |
1248 | } | |
1249 | }, { | |
1250 | 'note': 'Embedded SWF player', | |
1251 | 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0', | |
1252 | 'playlist_count': 4, | |
1253 | 'info_dict': { | |
1254 | 'title': 'JODA7', | |
1255 | 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ', | |
1256 | } | |
1257 | }, { | |
1258 | 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos', | |
1259 | 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA', | |
1260 | 'info_dict': { | |
1261 | 'title': 'Uploads from Interstellar Movie', | |
1262 | 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', | |
1263 | }, | |
1264 | 'playlist_mincout': 21, | |
1265 | }] | |
1266 | ||
1267 | def _real_initialize(self): | |
1268 | self._login() | |
1269 | ||
1270 | def _extract_mix(self, playlist_id): | |
1271 | # The mixes are generated from a single video | |
1272 | # the id of the playlist is just 'RD' + video_id | |
1273 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) | |
1274 | webpage = self._download_webpage( | |
1275 | url, playlist_id, 'Downloading Youtube mix') | |
1276 | search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) | |
1277 | title_span = ( | |
1278 | search_title('playlist-title') or | |
1279 | search_title('title long-title') or | |
1280 | search_title('title')) | |
1281 | title = clean_html(title_span) | |
1282 | ids = orderedSet(re.findall( | |
1283 | r'''(?xs)data-video-username=".*?".*? | |
1284 | href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id), | |
1285 | webpage)) | |
1286 | url_results = self._ids_to_results(ids) | |
1287 | ||
1288 | return self.playlist_result(url_results, playlist_id, title) | |
1289 | ||
1290 | def _extract_playlist(self, playlist_id): | |
1291 | url = self._TEMPLATE_URL % playlist_id | |
1292 | page = self._download_webpage(url, playlist_id) | |
1293 | more_widget_html = content_html = page | |
1294 | ||
1295 | for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page): | |
1296 | match = match.strip() | |
1297 | # Check if the playlist exists or is private | |
1298 | if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match): | |
1299 | raise ExtractorError( | |
1300 | 'The playlist doesn\'t exist or is private, use --username or ' | |
1301 | '--netrc to access it.', | |
1302 | expected=True) | |
1303 | elif re.match(r'[^<]*Invalid parameters[^<]*', match): | |
1304 | raise ExtractorError( | |
1305 | 'Invalid parameters. Maybe URL is incorrect.', | |
1306 | expected=True) | |
1307 | elif re.match(r'[^<]*Choose your language[^<]*', match): | |
1308 | continue | |
1309 | else: | |
1310 | self.report_warning('Youtube gives an alert message: ' + match) | |
1311 | ||
1312 | # Extract the video ids from the playlist pages | |
1313 | ids = [] | |
1314 | ||
1315 | for page_num in itertools.count(1): | |
1316 | matches = re.finditer(self._VIDEO_RE, content_html) | |
1317 | # We remove the duplicates and the link with index 0 | |
1318 | # (it's not the first video of the playlist) | |
1319 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
1320 | ids.extend(new_ids) | |
1321 | ||
1322 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) | |
1323 | if not mobj: | |
1324 | break | |
1325 | ||
1326 | more = self._download_json( | |
1327 | 'https://youtube.com/%s' % mobj.group('more'), playlist_id, | |
1328 | 'Downloading page #%s' % page_num, | |
1329 | transform_source=uppercase_escape) | |
1330 | content_html = more['content_html'] | |
1331 | if not content_html.strip(): | |
1332 | # Some webpages show a "Load more" button but they don't | |
1333 | # have more videos | |
1334 | break | |
1335 | more_widget_html = more['load_more_widget_html'] | |
1336 | ||
1337 | playlist_title = self._html_search_regex( | |
1338 | r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', | |
1339 | page, 'title') | |
1340 | ||
1341 | url_results = self._ids_to_results(ids) | |
1342 | return self.playlist_result(url_results, playlist_id, playlist_title) | |
1343 | ||
1344 | def _real_extract(self, url): | |
1345 | # Extract playlist id | |
1346 | mobj = re.match(self._VALID_URL, url) | |
1347 | if mobj is None: | |
1348 | raise ExtractorError('Invalid URL: %s' % url) | |
1349 | playlist_id = mobj.group(1) or mobj.group(2) | |
1350 | ||
1351 | # Check if it's a video-specific URL | |
1352 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) | |
1353 | if 'v' in query_dict: | |
1354 | video_id = query_dict['v'][0] | |
1355 | if self._downloader.params.get('noplaylist'): | |
1356 | self.to_screen('Downloading just video %s because of --no-playlist' % video_id) | |
1357 | return self.url_result(video_id, 'Youtube', video_id=video_id) | |
1358 | else: | |
1359 | self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
1360 | ||
1361 | if playlist_id.startswith('RD') or playlist_id.startswith('UL'): | |
1362 | # Mixes require a custom extraction process | |
1363 | return self._extract_mix(playlist_id) | |
1364 | ||
1365 | return self._extract_playlist(playlist_id) | |
1366 | ||
1367 | ||
1368 | class YoutubeChannelIE(InfoExtractor): | |
1369 | IE_DESC = 'YouTube.com channels' | |
1370 | _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' | |
1371 | _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' | |
1372 | IE_NAME = 'youtube:channel' | |
1373 | _TESTS = [{ | |
1374 | 'note': 'paginated channel', | |
1375 | 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', | |
1376 | 'playlist_mincount': 91, | |
1377 | 'info_dict': { | |
1378 | 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', | |
1379 | } | |
1380 | }] | |
1381 | ||
1382 | @staticmethod | |
1383 | def extract_videos_from_page(page): | |
1384 | ids_in_page = [] | |
1385 | titles_in_page = [] | |
1386 | for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page): | |
1387 | video_id = mobj.group('id') | |
1388 | video_title = unescapeHTML(mobj.group('title')) | |
1389 | try: | |
1390 | idx = ids_in_page.index(video_id) | |
1391 | if video_title and not titles_in_page[idx]: | |
1392 | titles_in_page[idx] = video_title | |
1393 | except ValueError: | |
1394 | ids_in_page.append(video_id) | |
1395 | titles_in_page.append(video_title) | |
1396 | return zip(ids_in_page, titles_in_page) | |
1397 | ||
1398 | def _real_extract(self, url): | |
1399 | channel_id = self._match_id(url) | |
1400 | ||
1401 | url = self._TEMPLATE_URL % channel_id | |
1402 | ||
1403 | # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778) | |
1404 | # Workaround by extracting as a playlist if managed to obtain channel playlist URL | |
1405 | # otherwise fallback on channel by page extraction | |
1406 | channel_page = self._download_webpage( | |
1407 | url + '?view=57', channel_id, | |
1408 | 'Downloading channel page', fatal=False) | |
1409 | channel_playlist_id = self._html_search_meta( | |
1410 | 'channelId', channel_page, 'channel id', default=None) | |
1411 | if not channel_playlist_id: | |
1412 | channel_playlist_id = self._search_regex( | |
1413 | r'data-channel-external-id="([^"]+)"', | |
1414 | channel_page, 'channel id', default=None) | |
1415 | if channel_playlist_id and channel_playlist_id.startswith('UC'): | |
1416 | playlist_id = 'UU' + channel_playlist_id[2:] | |
1417 | return self.url_result( | |
1418 | compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist') | |
1419 | ||
1420 | channel_page = self._download_webpage(url, channel_id, 'Downloading page #1') | |
1421 | autogenerated = re.search(r'''(?x) | |
1422 | class="[^"]*?(?: | |
1423 | channel-header-autogenerated-label| | |
1424 | yt-channel-title-autogenerated | |
1425 | )[^"]*"''', channel_page) is not None | |
1426 | ||
1427 | if autogenerated: | |
1428 | # The videos are contained in a single page | |
1429 | # the ajax pages can't be used, they are empty | |
1430 | entries = [ | |
1431 | self.url_result( | |
1432 | video_id, 'Youtube', video_id=video_id, | |
1433 | video_title=video_title) | |
1434 | for video_id, video_title in self.extract_videos_from_page(channel_page)] | |
1435 | return self.playlist_result(entries, channel_id) | |
1436 | ||
1437 | def _entries(): | |
1438 | more_widget_html = content_html = channel_page | |
1439 | for pagenum in itertools.count(1): | |
1440 | ||
1441 | for video_id, video_title in self.extract_videos_from_page(content_html): | |
1442 | yield self.url_result( | |
1443 | video_id, 'Youtube', video_id=video_id, | |
1444 | video_title=video_title) | |
1445 | ||
1446 | mobj = re.search( | |
1447 | r'data-uix-load-more-href="/?(?P<more>[^"]+)"', | |
1448 | more_widget_html) | |
1449 | if not mobj: | |
1450 | break | |
1451 | ||
1452 | more = self._download_json( | |
1453 | 'https://youtube.com/%s' % mobj.group('more'), channel_id, | |
1454 | 'Downloading page #%s' % (pagenum + 1), | |
1455 | transform_source=uppercase_escape) | |
1456 | content_html = more['content_html'] | |
1457 | more_widget_html = more['load_more_widget_html'] | |
1458 | ||
1459 | return self.playlist_result(_entries(), channel_id) | |
1460 | ||
1461 | ||
1462 | class YoutubeUserIE(YoutubeChannelIE): | |
1463 | IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' | |
1464 | _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_-]+)' | |
1465 | _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos' | |
1466 | IE_NAME = 'youtube:user' | |
1467 | ||
1468 | _TESTS = [{ | |
1469 | 'url': 'https://www.youtube.com/user/TheLinuxFoundation', | |
1470 | 'playlist_mincount': 320, | |
1471 | 'info_dict': { | |
1472 | 'title': 'TheLinuxFoundation', | |
1473 | } | |
1474 | }, { | |
1475 | 'url': 'ytuser:phihag', | |
1476 | 'only_matching': True, | |
1477 | }] | |
1478 | ||
1479 | @classmethod | |
1480 | def suitable(cls, url): | |
1481 | # Don't return True if the url can be extracted with other youtube | |
1482 | # extractor, the regex would is too permissive and it would match. | |
1483 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1484 | if any(ie.suitable(url) for ie in other_ies): | |
1485 | return False | |
1486 | else: | |
1487 | return super(YoutubeUserIE, cls).suitable(url) | |
1488 | ||
1489 | ||
1490 | class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): | |
1491 | IE_DESC = 'YouTube.com searches' | |
1492 | # there doesn't appear to be a real limit, for example if you search for | |
1493 | # 'python' you get more than 8.000.000 results | |
1494 | _MAX_RESULTS = float('inf') | |
1495 | IE_NAME = 'youtube:search' | |
1496 | _SEARCH_KEY = 'ytsearch' | |
1497 | _EXTRA_QUERY_ARGS = {} | |
1498 | _TESTS = [] | |
1499 | ||
1500 | def _get_n_results(self, query, n): | |
1501 | """Get a specified number of results for a query""" | |
1502 | ||
1503 | videos = [] | |
1504 | limit = n | |
1505 | ||
1506 | for pagenum in itertools.count(1): | |
1507 | url_query = { | |
1508 | 'search_query': query, | |
1509 | 'page': pagenum, | |
1510 | 'spf': 'navigate', | |
1511 | } | |
1512 | url_query.update(self._EXTRA_QUERY_ARGS) | |
1513 | result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query) | |
1514 | data = self._download_json( | |
1515 | result_url, video_id='query "%s"' % query, | |
1516 | note='Downloading page %s' % pagenum, | |
1517 | errnote='Unable to download API page') | |
1518 | html_content = data[1]['body']['content'] | |
1519 | ||
1520 | if 'class="search-message' in html_content: | |
1521 | raise ExtractorError( | |
1522 | '[youtube] No video results', expected=True) | |
1523 | ||
1524 | new_videos = self._ids_to_results(orderedSet(re.findall( | |
1525 | r'href="/watch\?v=(.{11})', html_content))) | |
1526 | videos += new_videos | |
1527 | if not new_videos or len(videos) > limit: | |
1528 | break | |
1529 | ||
1530 | if len(videos) > n: | |
1531 | videos = videos[:n] | |
1532 | return self.playlist_result(videos, query) | |
1533 | ||
1534 | ||
1535 | class YoutubeSearchDateIE(YoutubeSearchIE): | |
1536 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' | |
1537 | _SEARCH_KEY = 'ytsearchdate' | |
1538 | IE_DESC = 'YouTube.com searches, newest videos first' | |
1539 | _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'} | |
1540 | ||
1541 | ||
1542 | class YoutubeSearchURLIE(InfoExtractor): | |
1543 | IE_DESC = 'YouTube.com search URLs' | |
1544 | IE_NAME = 'youtube:search_url' | |
1545 | _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)' | |
1546 | _TESTS = [{ | |
1547 | 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', | |
1548 | 'playlist_mincount': 5, | |
1549 | 'info_dict': { | |
1550 | 'title': 'youtube-dl test video', | |
1551 | } | |
1552 | }] | |
1553 | ||
1554 | def _real_extract(self, url): | |
1555 | mobj = re.match(self._VALID_URL, url) | |
1556 | query = compat_urllib_parse.unquote_plus(mobj.group('query')) | |
1557 | ||
1558 | webpage = self._download_webpage(url, query) | |
1559 | result_code = self._search_regex( | |
1560 | r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML') | |
1561 | ||
1562 | part_codes = re.findall( | |
1563 | r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code) | |
1564 | entries = [] | |
1565 | for part_code in part_codes: | |
1566 | part_title = self._html_search_regex( | |
1567 | [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) | |
1568 | part_url_snippet = self._html_search_regex( | |
1569 | r'(?s)href="([^"]+)"', part_code, 'item URL') | |
1570 | part_url = compat_urlparse.urljoin( | |
1571 | 'https://www.youtube.com/', part_url_snippet) | |
1572 | entries.append({ | |
1573 | '_type': 'url', | |
1574 | 'url': part_url, | |
1575 | 'title': part_title, | |
1576 | }) | |
1577 | ||
1578 | return { | |
1579 | '_type': 'playlist', | |
1580 | 'entries': entries, | |
1581 | 'title': query, | |
1582 | } | |
1583 | ||
1584 | ||
1585 | class YoutubeShowIE(InfoExtractor): | |
1586 | IE_DESC = 'YouTube.com (multi-season) shows' | |
1587 | _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' | |
1588 | IE_NAME = 'youtube:show' | |
1589 | _TESTS = [{ | |
1590 | 'url': 'http://www.youtube.com/show/airdisasters', | |
1591 | 'playlist_mincount': 3, | |
1592 | 'info_dict': { | |
1593 | 'id': 'airdisasters', | |
1594 | 'title': 'Air Disasters', | |
1595 | } | |
1596 | }] | |
1597 | ||
1598 | def _real_extract(self, url): | |
1599 | mobj = re.match(self._VALID_URL, url) | |
1600 | playlist_id = mobj.group('id') | |
1601 | webpage = self._download_webpage( | |
1602 | url, playlist_id, 'Downloading show webpage') | |
1603 | # There's one playlist for each season of the show | |
1604 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1605 | self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) | |
1606 | entries = [ | |
1607 | self.url_result( | |
1608 | 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') | |
1609 | for season in m_seasons | |
1610 | ] | |
1611 | title = self._og_search_title(webpage, fatal=False) | |
1612 | ||
1613 | return { | |
1614 | '_type': 'playlist', | |
1615 | 'id': playlist_id, | |
1616 | 'title': title, | |
1617 | 'entries': entries, | |
1618 | } | |
1619 | ||
1620 | ||
1621 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): | |
1622 | """ | |
1623 | Base class for feed extractors | |
1624 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1625 | """ | |
1626 | _LOGIN_REQUIRED = True | |
1627 | ||
1628 | @property | |
1629 | def IE_NAME(self): | |
1630 | return 'youtube:%s' % self._FEED_NAME | |
1631 | ||
1632 | def _real_initialize(self): | |
1633 | self._login() | |
1634 | ||
1635 | def _real_extract(self, url): | |
1636 | page = self._download_webpage( | |
1637 | 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE) | |
1638 | ||
1639 | # The extraction process is the same as for playlists, but the regex | |
1640 | # for the video ids doesn't contain an index | |
1641 | ids = [] | |
1642 | more_widget_html = content_html = page | |
1643 | for page_num in itertools.count(1): | |
1644 | matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html) | |
1645 | ||
1646 | # 'recommended' feed has infinite 'load more' and each new portion spins | |
1647 | # the same videos in (sometimes) slightly different order, so we'll check | |
1648 | # for unicity and break when portion has no new videos | |
1649 | new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches)) | |
1650 | if not new_ids: | |
1651 | break | |
1652 | ||
1653 | ids.extend(new_ids) | |
1654 | ||
1655 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) | |
1656 | if not mobj: | |
1657 | break | |
1658 | ||
1659 | more = self._download_json( | |
1660 | 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE, | |
1661 | 'Downloading page #%s' % page_num, | |
1662 | transform_source=uppercase_escape) | |
1663 | content_html = more['content_html'] | |
1664 | more_widget_html = more['load_more_widget_html'] | |
1665 | ||
1666 | return self.playlist_result( | |
1667 | self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE) | |
1668 | ||
1669 | ||
1670 | class YoutubeWatchLaterIE(YoutubePlaylistIE): | |
1671 | IE_NAME = 'youtube:watchlater' | |
1672 | IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)' | |
1673 | _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater' | |
1674 | ||
1675 | _TESTS = [] # override PlaylistIE tests | |
1676 | ||
1677 | def _real_extract(self, url): | |
1678 | return self._extract_playlist('WL') | |
1679 | ||
1680 | ||
1681 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): | |
1682 | IE_NAME = 'youtube:favorites' | |
1683 | IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)' | |
1684 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' | |
1685 | _LOGIN_REQUIRED = True | |
1686 | ||
1687 | def _real_extract(self, url): | |
1688 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1689 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id') | |
1690 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
1691 | ||
1692 | ||
1693 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1694 | IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)' | |
1695 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1696 | _FEED_NAME = 'recommended' | |
1697 | _PLAYLIST_TITLE = 'Youtube Recommended videos' | |
1698 | ||
1699 | ||
1700 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
1701 | IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' | |
1702 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1703 | _FEED_NAME = 'subscriptions' | |
1704 | _PLAYLIST_TITLE = 'Youtube Subscriptions' | |
1705 | ||
1706 | ||
1707 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): | |
1708 | IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)' | |
1709 | _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory' | |
1710 | _FEED_NAME = 'history' | |
1711 | _PLAYLIST_TITLE = 'Youtube History' | |
1712 | ||
1713 | ||
1714 | class YoutubeTruncatedURLIE(InfoExtractor): | |
1715 | IE_NAME = 'youtube:truncated_url' | |
1716 | IE_DESC = False # Do not list | |
1717 | _VALID_URL = r'''(?x) | |
1718 | (?:https?://)? | |
1719 | (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/ | |
1720 | (?:watch\?(?: | |
1721 | feature=[a-z_]+| | |
1722 | annotation_id=annotation_[^&]+| | |
1723 | x-yt-cl=[0-9]+| | |
1724 | hl=[^&]*| | |
1725 | )? | |
1726 | | | |
1727 | attribution_link\?a=[^&]+ | |
1728 | ) | |
1729 | $ | |
1730 | ''' | |
1731 | ||
1732 | _TESTS = [{ | |
1733 | 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041', | |
1734 | 'only_matching': True, | |
1735 | }, { | |
1736 | 'url': 'http://www.youtube.com/watch?', | |
1737 | 'only_matching': True, | |
1738 | }, { | |
1739 | 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534', | |
1740 | 'only_matching': True, | |
1741 | }, { | |
1742 | 'url': 'https://www.youtube.com/watch?feature=foo', | |
1743 | 'only_matching': True, | |
1744 | }, { | |
1745 | 'url': 'https://www.youtube.com/watch?hl=en-GB', | |
1746 | 'only_matching': True, | |
1747 | }] | |
1748 | ||
1749 | def _real_extract(self, url): | |
1750 | raise ExtractorError( | |
1751 | 'Did you forget to quote the URL? Remember that & is a meta ' | |
1752 | 'character in most shells, so you want to put the URL in quotes, ' | |
1753 | 'like youtube-dl ' | |
1754 | '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' | |
1755 | ' or simply youtube-dl BaW_jenozKc .', | |
1756 | expected=True) | |
1757 | ||
1758 | ||
1759 | class YoutubeTruncatedIDIE(InfoExtractor): | |
1760 | IE_NAME = 'youtube:truncated_id' | |
1761 | IE_DESC = False # Do not list | |
1762 | _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$' | |
1763 | ||
1764 | _TESTS = [{ | |
1765 | 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob', | |
1766 | 'only_matching': True, | |
1767 | }] | |
1768 | ||
1769 | def _real_extract(self, url): | |
1770 | video_id = self._match_id(url) | |
1771 | raise ExtractorError( | |
1772 | 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url), | |
1773 | expected=True) |