]>
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 .subtitles import SubtitlesInfoExtractor | |
15 | from ..jsinterp import JSInterpreter | |
16 | from ..swfinterp import SWFInterpreter | |
17 | from ..utils import ( | |
18 | compat_chr, | |
19 | compat_parse_qs, | |
20 | compat_urllib_parse, | |
21 | compat_urllib_request, | |
22 | compat_urlparse, | |
23 | compat_str, | |
24 | ||
25 | clean_html, | |
26 | get_element_by_id, | |
27 | get_element_by_attribute, | |
28 | ExtractorError, | |
29 | int_or_none, | |
30 | OnDemandPagedList, | |
31 | unescapeHTML, | |
32 | unified_strdate, | |
33 | orderedSet, | |
34 | uppercase_escape, | |
35 | ) | |
36 | ||
37 | ||
38 | class YoutubeBaseInfoExtractor(InfoExtractor): | |
39 | """Provide base functions for Youtube extractors""" | |
40 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
41 | _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' | |
42 | _NETRC_MACHINE = 'youtube' | |
43 | # If True it will raise an error if no login info is provided | |
44 | _LOGIN_REQUIRED = False | |
45 | ||
46 | def _set_language(self): | |
47 | self._set_cookie( | |
48 | '.youtube.com', 'PREF', 'f1=50000000&hl=en', | |
49 | # YouTube sets the expire time to about two months | |
50 | expire_time=time.time() + 2 * 30 * 24 * 3600) | |
51 | ||
52 | def _login(self): | |
53 | """ | |
54 | Attempt to log in to YouTube. | |
55 | True is returned if successful or skipped. | |
56 | False is returned if login failed. | |
57 | ||
58 | If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. | |
59 | """ | |
60 | (username, password) = self._get_login_info() | |
61 | # No authentication to be performed | |
62 | if username is None: | |
63 | if self._LOGIN_REQUIRED: | |
64 | raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
65 | return True | |
66 | ||
67 | login_page = self._download_webpage( | |
68 | self._LOGIN_URL, None, | |
69 | note='Downloading login page', | |
70 | errnote='unable to fetch login page', fatal=False) | |
71 | if login_page is False: | |
72 | return | |
73 | ||
74 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', | |
75 | login_page, 'Login GALX parameter') | |
76 | ||
77 | # Log in | |
78 | login_form_strs = { | |
79 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
80 | 'Email': username, | |
81 | 'GALX': galx, | |
82 | 'Passwd': password, | |
83 | ||
84 | 'PersistentCookie': 'yes', | |
85 | '_utf8': '霱', | |
86 | 'bgresponse': 'js_disabled', | |
87 | 'checkConnection': '', | |
88 | 'checkedDomains': 'youtube', | |
89 | 'dnConn': '', | |
90 | 'pstMsg': '0', | |
91 | 'rmShown': '1', | |
92 | 'secTok': '', | |
93 | 'signIn': 'Sign in', | |
94 | 'timeStmp': '', | |
95 | 'service': 'youtube', | |
96 | 'uilel': '3', | |
97 | 'hl': 'en_US', | |
98 | } | |
99 | ||
100 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
101 | # chokes on unicode | |
102 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items()) | |
103 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
104 | ||
105 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
106 | login_results = self._download_webpage( | |
107 | req, None, | |
108 | note='Logging in', errnote='unable to log in', fatal=False) | |
109 | if login_results is False: | |
110 | return False | |
111 | ||
112 | if re.search(r'id="errormsg_0_Passwd"', login_results) is not None: | |
113 | raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True) | |
114 | ||
115 | # Two-Factor | |
116 | # TODO add SMS and phone call support - these require making a request and then prompting the user | |
117 | ||
118 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None: | |
119 | tfa_code = self._get_tfa_info() | |
120 | ||
121 | if tfa_code is None: | |
122 | self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>') | |
123 | self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)') | |
124 | return False | |
125 | ||
126 | # Unlike the first login form, secTok and timeStmp are both required for the TFA form | |
127 | ||
128 | match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
129 | if match is None: | |
130 | self._downloader.report_warning('Failed to get secTok - did the page structure change?') | |
131 | secTok = match.group(1) | |
132 | match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) | |
133 | if match is None: | |
134 | self._downloader.report_warning('Failed to get timeStmp - did the page structure change?') | |
135 | timeStmp = match.group(1) | |
136 | ||
137 | tfa_form_strs = { | |
138 | 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
139 | 'smsToken': '', | |
140 | 'smsUserPin': tfa_code, | |
141 | 'smsVerifyPin': 'Verify', | |
142 | ||
143 | 'PersistentCookie': 'yes', | |
144 | 'checkConnection': '', | |
145 | 'checkedDomains': 'youtube', | |
146 | 'pstMsg': '1', | |
147 | 'secTok': secTok, | |
148 | 'timeStmp': timeStmp, | |
149 | 'service': 'youtube', | |
150 | 'hl': 'en_US', | |
151 | } | |
152 | tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items()) | |
153 | tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii') | |
154 | ||
155 | tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data) | |
156 | tfa_results = self._download_webpage( | |
157 | tfa_req, None, | |
158 | note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) | |
159 | ||
160 | if tfa_results is False: | |
161 | return False | |
162 | ||
163 | if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None: | |
164 | self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.') | |
165 | return False | |
166 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None: | |
167 | self._downloader.report_warning('unable to log in - did the page structure change?') | |
168 | return False | |
169 | if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None: | |
170 | self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.') | |
171 | return False | |
172 | ||
173 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
174 | self._downloader.report_warning('unable to log in: bad username or password') | |
175 | return False | |
176 | return True | |
177 | ||
178 | def _real_initialize(self): | |
179 | if self._downloader is None: | |
180 | return | |
181 | self._set_language() | |
182 | if not self._login(): | |
183 | return | |
184 | ||
185 | ||
186 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): | |
187 | IE_DESC = 'YouTube.com' | |
188 | _VALID_URL = r"""(?x)^ | |
189 | ( | |
190 | (?:https?://|//) # http(s):// or protocol-independent URL | |
191 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| | |
192 | (?:www\.)?deturl\.com/www\.youtube\.com/| | |
193 | (?:www\.)?pwnyoutube\.com/| | |
194 | (?:www\.)?yourepeat\.com/| | |
195 | tube\.majestyc\.net/| | |
196 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
197 | (?:.*?\#/)? # handle anchor (#/) redirect urls | |
198 | (?: # the various things that can precede the ID: | |
199 | (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/ | |
200 | |(?: # or the v= param in all its forms | |
201 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) | |
202 | (?:\?|\#!?) # the params delimiter ? or # or #! | |
203 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
204 | v= | |
205 | ) | |
206 | )) | |
207 | |youtu\.be/ # just youtu.be/xxxx | |
208 | |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= | |
209 | ) | |
210 | )? # all until now is optional -> you can pass the naked ID | |
211 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID | |
212 | (?!.*?&list=) # combined list/video URLs are handled by the playlist IE | |
213 | (?(1).+)? # if we found the ID, everything can follow | |
214 | $""" | |
215 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' | |
216 | _formats = { | |
217 | '5': {'ext': 'flv', 'width': 400, 'height': 240}, | |
218 | '6': {'ext': 'flv', 'width': 450, 'height': 270}, | |
219 | '13': {'ext': '3gp'}, | |
220 | '17': {'ext': '3gp', 'width': 176, 'height': 144}, | |
221 | '18': {'ext': 'mp4', 'width': 640, 'height': 360}, | |
222 | '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, | |
223 | '34': {'ext': 'flv', 'width': 640, 'height': 360}, | |
224 | '35': {'ext': 'flv', 'width': 854, 'height': 480}, | |
225 | '36': {'ext': '3gp', 'width': 320, 'height': 240}, | |
226 | '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, | |
227 | '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, | |
228 | '43': {'ext': 'webm', 'width': 640, 'height': 360}, | |
229 | '44': {'ext': 'webm', 'width': 854, 'height': 480}, | |
230 | '45': {'ext': 'webm', 'width': 1280, 'height': 720}, | |
231 | '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, | |
232 | ||
233 | ||
234 | # 3d videos | |
235 | '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
236 | '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
237 | '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
238 | '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}, | |
239 | '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}, | |
240 | '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}, | |
241 | '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}, | |
242 | ||
243 | # Apple HTTP Live Streaming | |
244 | '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
245 | '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}, | |
246 | '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}, | |
247 | '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}, | |
248 | '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}, | |
249 | '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, | |
250 | '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}, | |
251 | ||
252 | # DASH mp4 video | |
253 | '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
254 | '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
255 | '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
256 | '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
257 | '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
258 | '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
259 | '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
260 | '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
261 | '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, | |
262 | '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, | |
263 | '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}, | |
264 | ||
265 | # Dash mp4 audio | |
266 | '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50}, | |
267 | '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50}, | |
268 | '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50}, | |
269 | ||
270 | # Dash webm | |
271 | '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
272 | '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
273 | '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
274 | '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
275 | '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
276 | '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, | |
277 | '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'}, | |
278 | '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
279 | '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
280 | '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
281 | '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
282 | '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
283 | '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
284 | '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
285 | '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
286 | '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, | |
287 | '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
288 | '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, | |
289 | '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'}, | |
290 | ||
291 | # Dash webm audio | |
292 | '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, | |
293 | '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, | |
294 | ||
295 | # Dash webm audio with opus inside | |
296 | '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, | |
297 | '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, | |
298 | '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, | |
299 | ||
300 | # RTMP (unnamed) | |
301 | '_rtmp': {'protocol': 'rtmp'}, | |
302 | } | |
303 | ||
304 | IE_NAME = 'youtube' | |
305 | _TESTS = [ | |
306 | { | |
307 | 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc', | |
308 | 'info_dict': { | |
309 | 'id': 'BaW_jenozKc', | |
310 | 'ext': 'mp4', | |
311 | 'title': 'youtube-dl test video "\'/\\ä↭𝕐', | |
312 | 'uploader': 'Philipp Hagemeister', | |
313 | 'uploader_id': 'phihag', | |
314 | 'upload_date': '20121002', | |
315 | '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 .', | |
316 | 'categories': ['Science & Technology'], | |
317 | 'like_count': int, | |
318 | 'dislike_count': int, | |
319 | } | |
320 | }, | |
321 | { | |
322 | 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY', | |
323 | 'note': 'Test generic use_cipher_signature video (#897)', | |
324 | 'info_dict': { | |
325 | 'id': 'UxxajLWwzqY', | |
326 | 'ext': 'mp4', | |
327 | 'upload_date': '20120506', | |
328 | 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', | |
329 | 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f', | |
330 | 'uploader': 'Icona Pop', | |
331 | 'uploader_id': 'IconaPop', | |
332 | } | |
333 | }, | |
334 | { | |
335 | 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ', | |
336 | 'note': 'Test VEVO video with age protection (#956)', | |
337 | 'info_dict': { | |
338 | 'id': '07FYdnEawAQ', | |
339 | 'ext': 'mp4', | |
340 | 'upload_date': '20130703', | |
341 | 'title': 'Justin Timberlake - Tunnel Vision (Explicit)', | |
342 | 'description': 'md5:64249768eec3bc4276236606ea996373', | |
343 | 'uploader': 'justintimberlakeVEVO', | |
344 | 'uploader_id': 'justintimberlakeVEVO', | |
345 | } | |
346 | }, | |
347 | { | |
348 | 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ', | |
349 | 'note': 'Embed-only video (#1746)', | |
350 | 'info_dict': { | |
351 | 'id': 'yZIXLfi8CZQ', | |
352 | 'ext': 'mp4', | |
353 | 'upload_date': '20120608', | |
354 | 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012', | |
355 | 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7', | |
356 | 'uploader': 'SET India', | |
357 | 'uploader_id': 'setindia' | |
358 | } | |
359 | }, | |
360 | { | |
361 | 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I', | |
362 | 'note': '256k DASH audio (format 141) via DASH manifest', | |
363 | 'info_dict': { | |
364 | 'id': 'a9LDPn-MO4I', | |
365 | 'ext': 'm4a', | |
366 | 'upload_date': '20121002', | |
367 | 'uploader_id': '8KVIDEO', | |
368 | 'description': '', | |
369 | 'uploader': '8KVIDEO', | |
370 | 'title': 'UHDTV TEST 8K VIDEO.mp4' | |
371 | }, | |
372 | 'params': { | |
373 | 'youtube_include_dash_manifest': True, | |
374 | 'format': '141', | |
375 | }, | |
376 | }, | |
377 | # DASH manifest with encrypted signature | |
378 | { | |
379 | 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA', | |
380 | 'info_dict': { | |
381 | 'id': 'IB3lcPjvWLA', | |
382 | 'ext': 'm4a', | |
383 | 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson', | |
384 | 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d', | |
385 | 'uploader': 'AfrojackVEVO', | |
386 | 'uploader_id': 'AfrojackVEVO', | |
387 | 'upload_date': '20131011', | |
388 | }, | |
389 | 'params': { | |
390 | 'youtube_include_dash_manifest': True, | |
391 | 'format': '141', | |
392 | }, | |
393 | }, | |
394 | # Controversy video | |
395 | { | |
396 | 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8', | |
397 | 'info_dict': { | |
398 | 'id': 'T4XJQO3qol8', | |
399 | 'ext': 'mp4', | |
400 | 'upload_date': '20100909', | |
401 | 'uploader': 'The Amazing Atheist', | |
402 | 'uploader_id': 'TheAmazingAtheist', | |
403 | 'title': 'Burning Everyone\'s Koran', | |
404 | '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', | |
405 | } | |
406 | }, | |
407 | # Normal age-gate video (No vevo, embed allowed) | |
408 | { | |
409 | 'url': 'http://youtube.com/watch?v=HtVdAasjOgU', | |
410 | 'info_dict': { | |
411 | 'id': 'HtVdAasjOgU', | |
412 | 'ext': 'mp4', | |
413 | 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer', | |
414 | 'description': 'md5:eca57043abae25130f58f655ad9a7771', | |
415 | 'uploader': 'The Witcher', | |
416 | 'uploader_id': 'WitcherGame', | |
417 | 'upload_date': '20140605', | |
418 | }, | |
419 | }, | |
420 | # video_info is None (https://github.com/rg3/youtube-dl/issues/4421) | |
421 | { | |
422 | 'url': '__2ABJjxzNo', | |
423 | 'info_dict': { | |
424 | 'id': '__2ABJjxzNo', | |
425 | 'ext': 'mp4', | |
426 | 'upload_date': '20100430', | |
427 | 'uploader_id': 'deadmau5', | |
428 | 'description': 'md5:12c56784b8032162bb936a5f76d55360', | |
429 | 'uploader': 'deadmau5', | |
430 | 'title': 'Deadmau5 - Some Chords (HD)', | |
431 | }, | |
432 | 'expected_warnings': [ | |
433 | 'DASH manifest missing', | |
434 | ] | |
435 | } | |
436 | ] | |
437 | ||
438 | def __init__(self, *args, **kwargs): | |
439 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
440 | self._player_cache = {} | |
441 | ||
442 | def report_video_info_webpage_download(self, video_id): | |
443 | """Report attempt to download video info webpage.""" | |
444 | self.to_screen('%s: Downloading video info webpage' % video_id) | |
445 | ||
446 | def report_information_extraction(self, video_id): | |
447 | """Report attempt to extract video information.""" | |
448 | self.to_screen('%s: Extracting video information' % video_id) | |
449 | ||
450 | def report_unavailable_format(self, video_id, format): | |
451 | """Report extracted video URL.""" | |
452 | self.to_screen('%s: Format %s not available' % (video_id, format)) | |
453 | ||
454 | def report_rtmp_download(self): | |
455 | """Indicate the download will use the RTMP protocol.""" | |
456 | self.to_screen('RTMP download detected') | |
457 | ||
458 | def _signature_cache_id(self, example_sig): | |
459 | """ Return a string representation of a signature """ | |
460 | return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) | |
461 | ||
462 | def _extract_signature_function(self, video_id, player_url, example_sig): | |
463 | id_m = re.match( | |
464 | r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$', | |
465 | player_url) | |
466 | if not id_m: | |
467 | raise ExtractorError('Cannot identify player %r' % player_url) | |
468 | player_type = id_m.group('ext') | |
469 | player_id = id_m.group('id') | |
470 | ||
471 | # Read from filesystem cache | |
472 | func_id = '%s_%s_%s' % ( | |
473 | player_type, player_id, self._signature_cache_id(example_sig)) | |
474 | assert os.path.basename(func_id) == func_id | |
475 | ||
476 | cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id) | |
477 | if cache_spec is not None: | |
478 | return lambda s: ''.join(s[i] for i in cache_spec) | |
479 | ||
480 | if player_type == 'js': | |
481 | code = self._download_webpage( | |
482 | player_url, video_id, | |
483 | note='Downloading %s player %s' % (player_type, player_id), | |
484 | errnote='Download of %s failed' % player_url) | |
485 | res = self._parse_sig_js(code) | |
486 | elif player_type == 'swf': | |
487 | urlh = self._request_webpage( | |
488 | player_url, video_id, | |
489 | note='Downloading %s player %s' % (player_type, player_id), | |
490 | errnote='Download of %s failed' % player_url) | |
491 | code = urlh.read() | |
492 | res = self._parse_sig_swf(code) | |
493 | else: | |
494 | assert False, 'Invalid player type %r' % player_type | |
495 | ||
496 | if cache_spec is None: | |
497 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) | |
498 | cache_res = res(test_string) | |
499 | cache_spec = [ord(c) for c in cache_res] | |
500 | ||
501 | self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) | |
502 | return res | |
503 | ||
504 | def _print_sig_code(self, func, example_sig): | |
505 | def gen_sig_code(idxs): | |
506 | def _genslice(start, end, step): | |
507 | starts = '' if start == 0 else str(start) | |
508 | ends = (':%d' % (end + step)) if end + step >= 0 else ':' | |
509 | steps = '' if step == 1 else (':%d' % step) | |
510 | return 's[%s%s%s]' % (starts, ends, steps) | |
511 | ||
512 | step = None | |
513 | start = '(Never used)' # Quelch pyflakes warnings - start will be | |
514 | # set as soon as step is set | |
515 | for i, prev in zip(idxs[1:], idxs[:-1]): | |
516 | if step is not None: | |
517 | if i - prev == step: | |
518 | continue | |
519 | yield _genslice(start, prev, step) | |
520 | step = None | |
521 | continue | |
522 | if i - prev in [-1, 1]: | |
523 | step = i - prev | |
524 | start = prev | |
525 | continue | |
526 | else: | |
527 | yield 's[%d]' % prev | |
528 | if step is None: | |
529 | yield 's[%d]' % i | |
530 | else: | |
531 | yield _genslice(start, i, step) | |
532 | ||
533 | test_string = ''.join(map(compat_chr, range(len(example_sig)))) | |
534 | cache_res = func(test_string) | |
535 | cache_spec = [ord(c) for c in cache_res] | |
536 | expr_code = ' + '.join(gen_sig_code(cache_spec)) | |
537 | signature_id_tuple = '(%s)' % ( | |
538 | ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) | |
539 | code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n' | |
540 | ' return %s\n') % (signature_id_tuple, expr_code) | |
541 | self.to_screen('Extracted signature function:\n' + code) | |
542 | ||
543 | def _parse_sig_js(self, jscode): | |
544 | funcname = self._search_regex( | |
545 | r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode, | |
546 | 'Initial JS player signature function name') | |
547 | ||
548 | jsi = JSInterpreter(jscode) | |
549 | initial_function = jsi.extract_function(funcname) | |
550 | return lambda s: initial_function([s]) | |
551 | ||
552 | def _parse_sig_swf(self, file_contents): | |
553 | swfi = SWFInterpreter(file_contents) | |
554 | TARGET_CLASSNAME = 'SignatureDecipher' | |
555 | searched_class = swfi.extract_class(TARGET_CLASSNAME) | |
556 | initial_function = swfi.extract_function(searched_class, 'decipher') | |
557 | return lambda s: initial_function([s]) | |
558 | ||
559 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): | |
560 | """Turn the encrypted s field into a working signature""" | |
561 | ||
562 | if player_url is None: | |
563 | raise ExtractorError('Cannot decrypt signature without player_url') | |
564 | ||
565 | if player_url.startswith('//'): | |
566 | player_url = 'https:' + player_url | |
567 | try: | |
568 | player_id = (player_url, self._signature_cache_id(s)) | |
569 | if player_id not in self._player_cache: | |
570 | func = self._extract_signature_function( | |
571 | video_id, player_url, s | |
572 | ) | |
573 | self._player_cache[player_id] = func | |
574 | func = self._player_cache[player_id] | |
575 | if self._downloader.params.get('youtube_print_sig_code'): | |
576 | self._print_sig_code(func, s) | |
577 | return func(s) | |
578 | except Exception as e: | |
579 | tb = traceback.format_exc() | |
580 | raise ExtractorError( | |
581 | 'Signature extraction failed: ' + tb, cause=e) | |
582 | ||
583 | def _get_available_subtitles(self, video_id, webpage): | |
584 | try: | |
585 | sub_list = self._download_webpage( | |
586 | 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
587 | video_id, note=False) | |
588 | except ExtractorError as err: | |
589 | self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err)) | |
590 | return {} | |
591 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
592 | ||
593 | sub_lang_list = {} | |
594 | for l in lang_list: | |
595 | lang = l[1] | |
596 | if lang in sub_lang_list: | |
597 | continue | |
598 | params = compat_urllib_parse.urlencode({ | |
599 | 'lang': lang, | |
600 | 'v': video_id, | |
601 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), | |
602 | 'name': unescapeHTML(l[0]).encode('utf-8'), | |
603 | }) | |
604 | url = 'https://www.youtube.com/api/timedtext?' + params | |
605 | sub_lang_list[lang] = url | |
606 | if not sub_lang_list: | |
607 | self._downloader.report_warning('video doesn\'t have subtitles') | |
608 | return {} | |
609 | return sub_lang_list | |
610 | ||
611 | def _get_available_automatic_caption(self, video_id, webpage): | |
612 | """We need the webpage for getting the captions url, pass it as an | |
613 | argument to speed up the process.""" | |
614 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') | |
615 | self.to_screen('%s: Looking for automatic captions' % video_id) | |
616 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
617 | err_msg = 'Couldn\'t find automatic captions for %s' % video_id | |
618 | if mobj is None: | |
619 | self._downloader.report_warning(err_msg) | |
620 | return {} | |
621 | player_config = json.loads(mobj.group(1)) | |
622 | try: | |
623 | args = player_config['args'] | |
624 | caption_url = args['ttsurl'] | |
625 | timestamp = args['timestamp'] | |
626 | # We get the available subtitles | |
627 | list_params = compat_urllib_parse.urlencode({ | |
628 | 'type': 'list', | |
629 | 'tlangs': 1, | |
630 | 'asrs': 1, | |
631 | }) | |
632 | list_url = caption_url + '&' + list_params | |
633 | caption_list = self._download_xml(list_url, video_id) | |
634 | original_lang_node = caption_list.find('track') | |
635 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr': | |
636 | self._downloader.report_warning('Video doesn\'t have automatic captions') | |
637 | return {} | |
638 | original_lang = original_lang_node.attrib['lang_code'] | |
639 | ||
640 | sub_lang_list = {} | |
641 | for lang_node in caption_list.findall('target'): | |
642 | sub_lang = lang_node.attrib['lang_code'] | |
643 | params = compat_urllib_parse.urlencode({ | |
644 | 'lang': original_lang, | |
645 | 'tlang': sub_lang, | |
646 | 'fmt': sub_format, | |
647 | 'ts': timestamp, | |
648 | 'kind': 'asr', | |
649 | }) | |
650 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
651 | return sub_lang_list | |
652 | # An extractor error can be raise by the download process if there are | |
653 | # no automatic captions but there are subtitles | |
654 | except (KeyError, ExtractorError): | |
655 | self._downloader.report_warning(err_msg) | |
656 | return {} | |
657 | ||
658 | @classmethod | |
659 | def extract_id(cls, url): | |
660 | mobj = re.match(cls._VALID_URL, url, re.VERBOSE) | |
661 | if mobj is None: | |
662 | raise ExtractorError('Invalid URL: %s' % url) | |
663 | video_id = mobj.group(2) | |
664 | return video_id | |
665 | ||
666 | def _extract_from_m3u8(self, manifest_url, video_id): | |
667 | url_map = {} | |
668 | ||
669 | def _get_urls(_manifest): | |
670 | lines = _manifest.split('\n') | |
671 | urls = filter(lambda l: l and not l.startswith('#'), | |
672 | lines) | |
673 | return urls | |
674 | manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest') | |
675 | formats_urls = _get_urls(manifest) | |
676 | for format_url in formats_urls: | |
677 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') | |
678 | url_map[itag] = format_url | |
679 | return url_map | |
680 | ||
681 | def _extract_annotations(self, video_id): | |
682 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
683 | return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') | |
684 | ||
685 | def _parse_dash_manifest( | |
686 | self, video_id, dash_manifest_url, player_url, age_gate): | |
687 | def decrypt_sig(mobj): | |
688 | s = mobj.group(1) | |
689 | dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) | |
690 | return '/signature/%s' % dec_s | |
691 | dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) | |
692 | dash_doc = self._download_xml( | |
693 | dash_manifest_url, video_id, | |
694 | note='Downloading DASH manifest', | |
695 | errnote='Could not download DASH manifest') | |
696 | ||
697 | formats = [] | |
698 | for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): | |
699 | url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') | |
700 | if url_el is None: | |
701 | continue | |
702 | format_id = r.attrib['id'] | |
703 | video_url = url_el.text | |
704 | filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) | |
705 | f = { | |
706 | 'format_id': format_id, | |
707 | 'url': video_url, | |
708 | 'width': int_or_none(r.attrib.get('width')), | |
709 | 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), | |
710 | 'asr': int_or_none(r.attrib.get('audioSamplingRate')), | |
711 | 'filesize': filesize, | |
712 | 'fps': int_or_none(r.attrib.get('frameRate')), | |
713 | } | |
714 | try: | |
715 | existing_format = next( | |
716 | fo for fo in formats | |
717 | if fo['format_id'] == format_id) | |
718 | except StopIteration: | |
719 | f.update(self._formats.get(format_id, {})) | |
720 | formats.append(f) | |
721 | else: | |
722 | existing_format.update(f) | |
723 | return formats | |
724 | ||
725 | def _real_extract(self, url): | |
726 | proto = ( | |
727 | 'http' if self._downloader.params.get('prefer_insecure', False) | |
728 | else 'https') | |
729 | ||
730 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
731 | mobj = re.search(self._NEXT_URL_RE, url) | |
732 | if mobj: | |
733 | url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
734 | video_id = self.extract_id(url) | |
735 | ||
736 | # Get video webpage | |
737 | url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id | |
738 | video_webpage = self._download_webpage(url, video_id) | |
739 | ||
740 | # Attempt to extract SWF player URL | |
741 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) | |
742 | if mobj is not None: | |
743 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
744 | else: | |
745 | player_url = None | |
746 | ||
747 | # Get video info | |
748 | if re.search(r'player-age-gate-content">', video_webpage) is not None: | |
749 | age_gate = True | |
750 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
751 | # this can be viewed without login into Youtube | |
752 | data = compat_urllib_parse.urlencode({ | |
753 | 'video_id': video_id, | |
754 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
755 | 'sts': self._search_regex( | |
756 | r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''), | |
757 | }) | |
758 | video_info_url = proto + '://www.youtube.com/get_video_info?' + data | |
759 | video_info_webpage = self._download_webpage( | |
760 | video_info_url, video_id, | |
761 | note='Refetching age-gated info webpage', | |
762 | errnote='unable to download video info webpage') | |
763 | video_info = compat_parse_qs(video_info_webpage) | |
764 | else: | |
765 | age_gate = False | |
766 | try: | |
767 | # Try looking directly into the video webpage | |
768 | mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) | |
769 | if not mobj: | |
770 | raise ValueError('Could not find ytplayer.config') # caught below | |
771 | json_code = uppercase_escape(mobj.group(1)) | |
772 | ytplayer_config = json.loads(json_code) | |
773 | args = ytplayer_config['args'] | |
774 | # Convert to the same format returned by compat_parse_qs | |
775 | video_info = dict((k, [v]) for k, v in args.items()) | |
776 | if 'url_encoded_fmt_stream_map' not in args: | |
777 | raise ValueError('No stream_map present') # caught below | |
778 | except ValueError: | |
779 | # We fallback to the get_video_info pages (used by the embed page) | |
780 | self.report_video_info_webpage_download(video_id) | |
781 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
782 | video_info_url = ( | |
783 | '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
784 | % (proto, video_id, el_type)) | |
785 | video_info_webpage = self._download_webpage( | |
786 | video_info_url, | |
787 | video_id, note=False, | |
788 | errnote='unable to download video info webpage') | |
789 | video_info = compat_parse_qs(video_info_webpage) | |
790 | if 'token' in video_info: | |
791 | break | |
792 | if 'token' not in video_info: | |
793 | if 'reason' in video_info: | |
794 | raise ExtractorError( | |
795 | 'YouTube said: %s' % video_info['reason'][0], | |
796 | expected=True, video_id=video_id) | |
797 | else: | |
798 | raise ExtractorError( | |
799 | '"token" parameter not in video info for unknown reason', | |
800 | video_id=video_id) | |
801 | ||
802 | if 'view_count' in video_info: | |
803 | view_count = int(video_info['view_count'][0]) | |
804 | else: | |
805 | view_count = None | |
806 | ||
807 | # Check for "rental" videos | |
808 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
809 | raise ExtractorError('"rental" videos not supported') | |
810 | ||
811 | # Start extracting information | |
812 | self.report_information_extraction(video_id) | |
813 | ||
814 | # uploader | |
815 | if 'author' not in video_info: | |
816 | raise ExtractorError('Unable to extract uploader name') | |
817 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
818 | ||
819 | # uploader_id | |
820 | video_uploader_id = None | |
821 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
822 | if mobj is not None: | |
823 | video_uploader_id = mobj.group(1) | |
824 | else: | |
825 | self._downloader.report_warning('unable to extract uploader nickname') | |
826 | ||
827 | # title | |
828 | if 'title' in video_info: | |
829 | video_title = video_info['title'][0] | |
830 | else: | |
831 | self._downloader.report_warning('Unable to extract video title') | |
832 | video_title = '_' | |
833 | ||
834 | # thumbnail image | |
835 | # We try first to get a high quality image: | |
836 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
837 | video_webpage, re.DOTALL) | |
838 | if m_thumb is not None: | |
839 | video_thumbnail = m_thumb.group(1) | |
840 | elif 'thumbnail_url' not in video_info: | |
841 | self._downloader.report_warning('unable to extract video thumbnail') | |
842 | video_thumbnail = None | |
843 | else: # don't panic if we can't find it | |
844 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
845 | ||
846 | # upload date | |
847 | upload_date = None | |
848 | mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage) | |
849 | if mobj is None: | |
850 | mobj = re.search( | |
851 | r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>', | |
852 | video_webpage) | |
853 | if mobj is not None: | |
854 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
855 | upload_date = unified_strdate(upload_date) | |
856 | ||
857 | m_cat_container = self._search_regex( | |
858 | r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>', | |
859 | video_webpage, 'categories', fatal=False) | |
860 | if m_cat_container: | |
861 | category = self._html_search_regex( | |
862 | r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', | |
863 | default=None) | |
864 | video_categories = None if category is None else [category] | |
865 | else: | |
866 | video_categories = None | |
867 | ||
868 | # description | |
869 | video_description = get_element_by_id("eow-description", video_webpage) | |
870 | if video_description: | |
871 | video_description = re.sub(r'''(?x) | |
872 | <a\s+ | |
873 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
874 | title="([^"]+)"\s+ | |
875 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
876 | class="yt-uix-redirect-link"\s*> | |
877 | [^<]+ | |
878 | </a> | |
879 | ''', r'\1', video_description) | |
880 | video_description = clean_html(video_description) | |
881 | else: | |
882 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
883 | if fd_mobj: | |
884 | video_description = unescapeHTML(fd_mobj.group(1)) | |
885 | else: | |
886 | video_description = '' | |
887 | ||
888 | def _extract_count(count_name): | |
889 | count = self._search_regex( | |
890 | r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name), | |
891 | video_webpage, count_name, default=None) | |
892 | if count is not None: | |
893 | return int(count.replace(',', '')) | |
894 | return None | |
895 | like_count = _extract_count('like') | |
896 | dislike_count = _extract_count('dislike') | |
897 | ||
898 | # subtitles | |
899 | video_subtitles = self.extract_subtitles(video_id, video_webpage) | |
900 | ||
901 | if self._downloader.params.get('listsubtitles', False): | |
902 | self._list_available_subtitles(video_id, video_webpage) | |
903 | return | |
904 | ||
905 | if 'length_seconds' not in video_info: | |
906 | self._downloader.report_warning('unable to extract video duration') | |
907 | video_duration = None | |
908 | else: | |
909 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) | |
910 | ||
911 | # annotations | |
912 | video_annotations = None | |
913 | if self._downloader.params.get('writeannotations', False): | |
914 | video_annotations = self._extract_annotations(video_id) | |
915 | ||
916 | def _map_to_format_list(urlmap): | |
917 | formats = [] | |
918 | for itag, video_real_url in urlmap.items(): | |
919 | dct = { | |
920 | 'format_id': itag, | |
921 | 'url': video_real_url, | |
922 | 'player_url': player_url, | |
923 | } | |
924 | if itag in self._formats: | |
925 | dct.update(self._formats[itag]) | |
926 | formats.append(dct) | |
927 | return formats | |
928 | ||
929 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
930 | self.report_rtmp_download() | |
931 | formats = [{ | |
932 | 'format_id': '_rtmp', | |
933 | 'protocol': 'rtmp', | |
934 | 'url': video_info['conn'][0], | |
935 | 'player_url': player_url, | |
936 | }] | |
937 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: | |
938 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0] | |
939 | if 'rtmpe%3Dyes' in encoded_url_map: | |
940 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) | |
941 | url_map = {} | |
942 | for url_data_str in encoded_url_map.split(','): | |
943 | url_data = compat_parse_qs(url_data_str) | |
944 | if 'itag' not in url_data or 'url' not in url_data: | |
945 | continue | |
946 | format_id = url_data['itag'][0] | |
947 | url = url_data['url'][0] | |
948 | ||
949 | if 'sig' in url_data: | |
950 | url += '&signature=' + url_data['sig'][0] | |
951 | elif 's' in url_data: | |
952 | encrypted_sig = url_data['s'][0] | |
953 | ||
954 | if not age_gate: | |
955 | jsplayer_url_json = self._search_regex( | |
956 | r'"assets":.+?"js":\s*("[^"]+")', | |
957 | video_webpage, 'JS player URL') | |
958 | player_url = json.loads(jsplayer_url_json) | |
959 | if player_url is None: | |
960 | player_url_json = self._search_regex( | |
961 | r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', | |
962 | video_webpage, 'age gate player URL') | |
963 | player_url = json.loads(player_url_json) | |
964 | ||
965 | if self._downloader.params.get('verbose'): | |
966 | if player_url is None: | |
967 | player_version = 'unknown' | |
968 | player_desc = 'unknown' | |
969 | else: | |
970 | if player_url.endswith('swf'): | |
971 | player_version = self._search_regex( | |
972 | r'-(.+?)(?:/watch_as3)?\.swf$', player_url, | |
973 | 'flash player', fatal=False) | |
974 | player_desc = 'flash player %s' % player_version | |
975 | else: | |
976 | player_version = self._search_regex( | |
977 | r'html5player-([^/]+?)(?:/html5player)?\.js', | |
978 | player_url, | |
979 | 'html5 player', fatal=False) | |
980 | player_desc = 'html5 player %s' % player_version | |
981 | ||
982 | parts_sizes = self._signature_cache_id(encrypted_sig) | |
983 | self.to_screen('{%s} signature length %s, %s' % | |
984 | (format_id, parts_sizes, player_desc)) | |
985 | ||
986 | signature = self._decrypt_signature( | |
987 | encrypted_sig, video_id, player_url, age_gate) | |
988 | url += '&signature=' + signature | |
989 | if 'ratebypass' not in url: | |
990 | url += '&ratebypass=yes' | |
991 | url_map[format_id] = url | |
992 | formats = _map_to_format_list(url_map) | |
993 | elif video_info.get('hlsvp'): | |
994 | manifest_url = video_info['hlsvp'][0] | |
995 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
996 | formats = _map_to_format_list(url_map) | |
997 | else: | |
998 | raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') | |
999 | ||
1000 | # Look for the DASH manifest | |
1001 | if self._downloader.params.get('youtube_include_dash_manifest', True): | |
1002 | dash_mpd = video_info.get('dashmpd') | |
1003 | if not dash_mpd: | |
1004 | self.report_warning('%s: DASH manifest missing' % video_id) | |
1005 | else: | |
1006 | dash_manifest_url = dash_mpd[0] | |
1007 | try: | |
1008 | dash_formats = self._parse_dash_manifest( | |
1009 | video_id, dash_manifest_url, player_url, age_gate) | |
1010 | except (ExtractorError, KeyError) as e: | |
1011 | self.report_warning( | |
1012 | 'Skipping DASH manifest: %r' % e, video_id) | |
1013 | else: | |
1014 | formats.extend(dash_formats) | |
1015 | ||
1016 | self._sort_formats(formats) | |
1017 | ||
1018 | return { | |
1019 | 'id': video_id, | |
1020 | 'uploader': video_uploader, | |
1021 | 'uploader_id': video_uploader_id, | |
1022 | 'upload_date': upload_date, | |
1023 | 'title': video_title, | |
1024 | 'thumbnail': video_thumbnail, | |
1025 | 'description': video_description, | |
1026 | 'categories': video_categories, | |
1027 | 'subtitles': video_subtitles, | |
1028 | 'duration': video_duration, | |
1029 | 'age_limit': 18 if age_gate else 0, | |
1030 | 'annotations': video_annotations, | |
1031 | 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, | |
1032 | 'view_count': view_count, | |
1033 | 'like_count': like_count, | |
1034 | 'dislike_count': dislike_count, | |
1035 | 'formats': formats, | |
1036 | } | |
1037 | ||
1038 | ||
1039 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): | |
1040 | IE_DESC = 'YouTube.com playlists' | |
1041 | _VALID_URL = r"""(?x)(?: | |
1042 | (?:https?://)? | |
1043 | (?:\w+\.)? | |
1044 | youtube\.com/ | |
1045 | (?: | |
1046 | (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) | |
1047 | \? (?:.*?&)*? (?:p|a|list)= | |
1048 | | p/ | |
1049 | ) | |
1050 | ( | |
1051 | (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,} | |
1052 | # Top tracks, they can also include dots | |
1053 | |(?:MC)[\w\.]* | |
1054 | ) | |
1055 | .* | |
1056 | | | |
1057 | ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) | |
1058 | )""" | |
1059 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' | |
1060 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' | |
1061 | _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' | |
1062 | IE_NAME = 'youtube:playlist' | |
1063 | _TESTS = [{ | |
1064 | 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', | |
1065 | 'info_dict': { | |
1066 | 'title': 'ytdl test PL', | |
1067 | 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', | |
1068 | }, | |
1069 | 'playlist_count': 3, | |
1070 | }, { | |
1071 | 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', | |
1072 | 'info_dict': { | |
1073 | 'title': 'YDL_Empty_List', | |
1074 | }, | |
1075 | 'playlist_count': 0, | |
1076 | }, { | |
1077 | 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', | |
1078 | 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', | |
1079 | 'info_dict': { | |
1080 | 'title': '29C3: Not my department', | |
1081 | }, | |
1082 | 'playlist_count': 95, | |
1083 | }, { | |
1084 | 'note': 'issue #673', | |
1085 | 'url': 'PLBB231211A4F62143', | |
1086 | 'info_dict': { | |
1087 | 'title': '[OLD]Team Fortress 2 (Class-based LP)', | |
1088 | }, | |
1089 | 'playlist_mincount': 26, | |
1090 | }, { | |
1091 | 'note': 'Large playlist', | |
1092 | 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', | |
1093 | 'info_dict': { | |
1094 | 'title': 'Uploads from Cauchemar', | |
1095 | }, | |
1096 | 'playlist_mincount': 799, | |
1097 | }, { | |
1098 | 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', | |
1099 | 'info_dict': { | |
1100 | 'title': 'YDL_safe_search', | |
1101 | }, | |
1102 | 'playlist_count': 2, | |
1103 | }, { | |
1104 | 'note': 'embedded', | |
1105 | 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', | |
1106 | 'playlist_count': 4, | |
1107 | 'info_dict': { | |
1108 | 'title': 'JODA15', | |
1109 | } | |
1110 | }, { | |
1111 | 'note': 'Embedded SWF player', | |
1112 | 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0', | |
1113 | 'playlist_count': 4, | |
1114 | 'info_dict': { | |
1115 | 'title': 'JODA7', | |
1116 | } | |
1117 | }] | |
1118 | ||
1119 | def _real_initialize(self): | |
1120 | self._login() | |
1121 | ||
1122 | def _ids_to_results(self, ids): | |
1123 | return [ | |
1124 | self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
1125 | for vid_id in ids] | |
1126 | ||
1127 | def _extract_mix(self, playlist_id): | |
1128 | # The mixes are generated from a a single video | |
1129 | # the id of the playlist is just 'RD' + video_id | |
1130 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) | |
1131 | webpage = self._download_webpage( | |
1132 | url, playlist_id, 'Downloading Youtube mix') | |
1133 | search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) | |
1134 | title_span = ( | |
1135 | search_title('playlist-title') or | |
1136 | search_title('title long-title') or | |
1137 | search_title('title')) | |
1138 | title = clean_html(title_span) | |
1139 | ids = orderedSet(re.findall( | |
1140 | r'''(?xs)data-video-username=".*?".*? | |
1141 | href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id), | |
1142 | webpage)) | |
1143 | url_results = self._ids_to_results(ids) | |
1144 | ||
1145 | return self.playlist_result(url_results, playlist_id, title) | |
1146 | ||
1147 | def _real_extract(self, url): | |
1148 | # Extract playlist id | |
1149 | mobj = re.match(self._VALID_URL, url) | |
1150 | if mobj is None: | |
1151 | raise ExtractorError('Invalid URL: %s' % url) | |
1152 | playlist_id = mobj.group(1) or mobj.group(2) | |
1153 | ||
1154 | # Check if it's a video-specific URL | |
1155 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) | |
1156 | if 'v' in query_dict: | |
1157 | video_id = query_dict['v'][0] | |
1158 | if self._downloader.params.get('noplaylist'): | |
1159 | self.to_screen('Downloading just video %s because of --no-playlist' % video_id) | |
1160 | return self.url_result(video_id, 'Youtube', video_id=video_id) | |
1161 | else: | |
1162 | self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
1163 | ||
1164 | if playlist_id.startswith('RD'): | |
1165 | # Mixes require a custom extraction process | |
1166 | return self._extract_mix(playlist_id) | |
1167 | if playlist_id.startswith('TL'): | |
1168 | raise ExtractorError('For downloading YouTube.com top lists, use ' | |
1169 | 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
1170 | ||
1171 | url = self._TEMPLATE_URL % playlist_id | |
1172 | page = self._download_webpage(url, playlist_id) | |
1173 | more_widget_html = content_html = page | |
1174 | ||
1175 | # Check if the playlist exists or is private | |
1176 | if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None: | |
1177 | raise ExtractorError( | |
1178 | 'The playlist doesn\'t exist or is private, use --username or ' | |
1179 | '--netrc to access it.', | |
1180 | expected=True) | |
1181 | ||
1182 | # Extract the video ids from the playlist pages | |
1183 | ids = [] | |
1184 | ||
1185 | for page_num in itertools.count(1): | |
1186 | matches = re.finditer(self._VIDEO_RE, content_html) | |
1187 | # We remove the duplicates and the link with index 0 | |
1188 | # (it's not the first video of the playlist) | |
1189 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
1190 | ids.extend(new_ids) | |
1191 | ||
1192 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) | |
1193 | if not mobj: | |
1194 | break | |
1195 | ||
1196 | more = self._download_json( | |
1197 | 'https://youtube.com/%s' % mobj.group('more'), playlist_id, | |
1198 | 'Downloading page #%s' % page_num, | |
1199 | transform_source=uppercase_escape) | |
1200 | content_html = more['content_html'] | |
1201 | more_widget_html = more['load_more_widget_html'] | |
1202 | ||
1203 | playlist_title = self._html_search_regex( | |
1204 | r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>', | |
1205 | page, 'title') | |
1206 | ||
1207 | url_results = self._ids_to_results(ids) | |
1208 | return self.playlist_result(url_results, playlist_id, playlist_title) | |
1209 | ||
1210 | ||
1211 | class YoutubeTopListIE(YoutubePlaylistIE): | |
1212 | IE_NAME = 'youtube:toplist' | |
1213 | IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
1214 | ' (Example: "yttoplist:music:Top Tracks")') | |
1215 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
1216 | _TESTS = [{ | |
1217 | 'url': 'yttoplist:music:Trending', | |
1218 | 'playlist_mincount': 5, | |
1219 | 'skip': 'Only works for logged-in users', | |
1220 | }] | |
1221 | ||
1222 | def _real_extract(self, url): | |
1223 | mobj = re.match(self._VALID_URL, url) | |
1224 | channel = mobj.group('chann') | |
1225 | title = mobj.group('title') | |
1226 | query = compat_urllib_parse.urlencode({'title': title}) | |
1227 | channel_page = self._download_webpage( | |
1228 | 'https://www.youtube.com/%s' % channel, title) | |
1229 | link = self._html_search_regex( | |
1230 | r'''(?x) | |
1231 | <a\s+href="([^"]+)".*?>\s* | |
1232 | <span\s+class="branded-page-module-title-text">\s* | |
1233 | <span[^>]*>.*?%s.*?</span>''' % re.escape(query), | |
1234 | channel_page, 'list') | |
1235 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
1236 | ||
1237 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
1238 | ids = [] | |
1239 | # sometimes the webpage doesn't contain the videos | |
1240 | # retry until we get them | |
1241 | for i in itertools.count(0): | |
1242 | msg = 'Downloading Youtube mix' | |
1243 | if i > 0: | |
1244 | msg += ', retry #%d' % i | |
1245 | ||
1246 | webpage = self._download_webpage(url, title, msg) | |
1247 | ids = orderedSet(re.findall(video_re, webpage)) | |
1248 | if ids: | |
1249 | break | |
1250 | url_results = self._ids_to_results(ids) | |
1251 | return self.playlist_result(url_results, playlist_title=title) | |
1252 | ||
1253 | ||
1254 | class YoutubeChannelIE(InfoExtractor): | |
1255 | IE_DESC = 'YouTube.com channels' | |
1256 | _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' | |
1257 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' | |
1258 | _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' | |
1259 | IE_NAME = 'youtube:channel' | |
1260 | _TESTS = [{ | |
1261 | 'note': 'paginated channel', | |
1262 | 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', | |
1263 | 'playlist_mincount': 91, | |
1264 | }] | |
1265 | ||
1266 | def extract_videos_from_page(self, page): | |
1267 | ids_in_page = [] | |
1268 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
1269 | if mobj.group(1) not in ids_in_page: | |
1270 | ids_in_page.append(mobj.group(1)) | |
1271 | return ids_in_page | |
1272 | ||
1273 | def _real_extract(self, url): | |
1274 | channel_id = self._match_id(url) | |
1275 | ||
1276 | video_ids = [] | |
1277 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id | |
1278 | channel_page = self._download_webpage(url, channel_id) | |
1279 | autogenerated = re.search(r'''(?x) | |
1280 | class="[^"]*?(?: | |
1281 | channel-header-autogenerated-label| | |
1282 | yt-channel-title-autogenerated | |
1283 | )[^"]*"''', channel_page) is not None | |
1284 | ||
1285 | if autogenerated: | |
1286 | # The videos are contained in a single page | |
1287 | # the ajax pages can't be used, they are empty | |
1288 | video_ids = self.extract_videos_from_page(channel_page) | |
1289 | entries = [ | |
1290 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1291 | for video_id in video_ids] | |
1292 | return self.playlist_result(entries, channel_id) | |
1293 | ||
1294 | def _entries(): | |
1295 | for pagenum in itertools.count(1): | |
1296 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
1297 | page = self._download_json( | |
1298 | url, channel_id, note='Downloading page #%s' % pagenum, | |
1299 | transform_source=uppercase_escape) | |
1300 | ||
1301 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
1302 | for video_id in ids_in_page: | |
1303 | yield self.url_result( | |
1304 | video_id, 'Youtube', video_id=video_id) | |
1305 | ||
1306 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
1307 | break | |
1308 | ||
1309 | return self.playlist_result(_entries(), channel_id) | |
1310 | ||
1311 | ||
1312 | class YoutubeUserIE(InfoExtractor): | |
1313 | IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' | |
1314 | _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_-]+)' | |
1315 | _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' | |
1316 | _GDATA_PAGE_SIZE = 50 | |
1317 | _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' | |
1318 | IE_NAME = 'youtube:user' | |
1319 | ||
1320 | _TESTS = [{ | |
1321 | 'url': 'https://www.youtube.com/user/TheLinuxFoundation', | |
1322 | 'playlist_mincount': 320, | |
1323 | 'info_dict': { | |
1324 | 'title': 'TheLinuxFoundation', | |
1325 | } | |
1326 | }, { | |
1327 | 'url': 'ytuser:phihag', | |
1328 | 'only_matching': True, | |
1329 | }] | |
1330 | ||
1331 | @classmethod | |
1332 | def suitable(cls, url): | |
1333 | # Don't return True if the url can be extracted with other youtube | |
1334 | # extractor, the regex would is too permissive and it would match. | |
1335 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1336 | if any(ie.suitable(url) for ie in other_ies): | |
1337 | return False | |
1338 | else: | |
1339 | return super(YoutubeUserIE, cls).suitable(url) | |
1340 | ||
1341 | def _real_extract(self, url): | |
1342 | username = self._match_id(url) | |
1343 | ||
1344 | # Download video ids using YouTube Data API. Result size per | |
1345 | # query is limited (currently to 50 videos) so we need to query | |
1346 | # page by page until there are no video ids - it means we got | |
1347 | # all of them. | |
1348 | ||
1349 | def download_page(pagenum): | |
1350 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 | |
1351 | ||
1352 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
1353 | page = self._download_webpage( | |
1354 | gdata_url, username, | |
1355 | 'Downloading video ids from %d to %d' % ( | |
1356 | start_index, start_index + self._GDATA_PAGE_SIZE)) | |
1357 | ||
1358 | try: | |
1359 | response = json.loads(page) | |
1360 | except ValueError as err: | |
1361 | raise ExtractorError('Invalid JSON in API response: ' + compat_str(err)) | |
1362 | if 'entry' not in response['feed']: | |
1363 | return | |
1364 | ||
1365 | # Extract video identifiers | |
1366 | entries = response['feed']['entry'] | |
1367 | for entry in entries: | |
1368 | title = entry['title']['$t'] | |
1369 | video_id = entry['id']['$t'].split('/')[-1] | |
1370 | yield { | |
1371 | '_type': 'url', | |
1372 | 'url': video_id, | |
1373 | 'ie_key': 'Youtube', | |
1374 | 'id': video_id, | |
1375 | 'title': title, | |
1376 | } | |
1377 | url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE) | |
1378 | ||
1379 | return self.playlist_result(url_results, playlist_title=username) | |
1380 | ||
1381 | ||
1382 | class YoutubeSearchIE(SearchInfoExtractor): | |
1383 | IE_DESC = 'YouTube.com searches' | |
1384 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' | |
1385 | _MAX_RESULTS = 1000 | |
1386 | IE_NAME = 'youtube:search' | |
1387 | _SEARCH_KEY = 'ytsearch' | |
1388 | ||
1389 | def _get_n_results(self, query, n): | |
1390 | """Get a specified number of results for a query""" | |
1391 | ||
1392 | video_ids = [] | |
1393 | pagenum = 0 | |
1394 | limit = n | |
1395 | PAGE_SIZE = 50 | |
1396 | ||
1397 | while (PAGE_SIZE * pagenum) < limit: | |
1398 | result_url = self._API_URL % ( | |
1399 | compat_urllib_parse.quote_plus(query.encode('utf-8')), | |
1400 | (PAGE_SIZE * pagenum) + 1) | |
1401 | data_json = self._download_webpage( | |
1402 | result_url, video_id='query "%s"' % query, | |
1403 | note='Downloading page %s' % (pagenum + 1), | |
1404 | errnote='Unable to download API page') | |
1405 | data = json.loads(data_json) | |
1406 | api_response = data['data'] | |
1407 | ||
1408 | if 'items' not in api_response: | |
1409 | raise ExtractorError( | |
1410 | '[youtube] No video results', expected=True) | |
1411 | ||
1412 | new_ids = list(video['id'] for video in api_response['items']) | |
1413 | video_ids += new_ids | |
1414 | ||
1415 | limit = min(n, api_response['totalItems']) | |
1416 | pagenum += 1 | |
1417 | ||
1418 | if len(video_ids) > n: | |
1419 | video_ids = video_ids[:n] | |
1420 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) | |
1421 | for video_id in video_ids] | |
1422 | return self.playlist_result(videos, query) | |
1423 | ||
1424 | ||
1425 | class YoutubeSearchDateIE(YoutubeSearchIE): | |
1426 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' | |
1427 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' | |
1428 | _SEARCH_KEY = 'ytsearchdate' | |
1429 | IE_DESC = 'YouTube.com searches, newest videos first' | |
1430 | ||
1431 | ||
1432 | class YoutubeSearchURLIE(InfoExtractor): | |
1433 | IE_DESC = 'YouTube.com search URLs' | |
1434 | IE_NAME = 'youtube:search_url' | |
1435 | _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)' | |
1436 | _TESTS = [{ | |
1437 | 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', | |
1438 | 'playlist_mincount': 5, | |
1439 | 'info_dict': { | |
1440 | 'title': 'youtube-dl test video', | |
1441 | } | |
1442 | }] | |
1443 | ||
1444 | def _real_extract(self, url): | |
1445 | mobj = re.match(self._VALID_URL, url) | |
1446 | query = compat_urllib_parse.unquote_plus(mobj.group('query')) | |
1447 | ||
1448 | webpage = self._download_webpage(url, query) | |
1449 | result_code = self._search_regex( | |
1450 | r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML') | |
1451 | ||
1452 | part_codes = re.findall( | |
1453 | r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code) | |
1454 | entries = [] | |
1455 | for part_code in part_codes: | |
1456 | part_title = self._html_search_regex( | |
1457 | [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) | |
1458 | part_url_snippet = self._html_search_regex( | |
1459 | r'(?s)href="([^"]+)"', part_code, 'item URL') | |
1460 | part_url = compat_urlparse.urljoin( | |
1461 | 'https://www.youtube.com/', part_url_snippet) | |
1462 | entries.append({ | |
1463 | '_type': 'url', | |
1464 | 'url': part_url, | |
1465 | 'title': part_title, | |
1466 | }) | |
1467 | ||
1468 | return { | |
1469 | '_type': 'playlist', | |
1470 | 'entries': entries, | |
1471 | 'title': query, | |
1472 | } | |
1473 | ||
1474 | ||
1475 | class YoutubeShowIE(InfoExtractor): | |
1476 | IE_DESC = 'YouTube.com (multi-season) shows' | |
1477 | _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)' | |
1478 | IE_NAME = 'youtube:show' | |
1479 | _TESTS = [{ | |
1480 | 'url': 'http://www.youtube.com/show/airdisasters', | |
1481 | 'playlist_mincount': 3, | |
1482 | 'info_dict': { | |
1483 | 'id': 'airdisasters', | |
1484 | 'title': 'Air Disasters', | |
1485 | } | |
1486 | }] | |
1487 | ||
1488 | def _real_extract(self, url): | |
1489 | mobj = re.match(self._VALID_URL, url) | |
1490 | playlist_id = mobj.group('id') | |
1491 | webpage = self._download_webpage( | |
1492 | url, playlist_id, 'Downloading show webpage') | |
1493 | # There's one playlist for each season of the show | |
1494 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1495 | self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) | |
1496 | entries = [ | |
1497 | self.url_result( | |
1498 | 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') | |
1499 | for season in m_seasons | |
1500 | ] | |
1501 | title = self._og_search_title(webpage, fatal=False) | |
1502 | ||
1503 | return { | |
1504 | '_type': 'playlist', | |
1505 | 'id': playlist_id, | |
1506 | 'title': title, | |
1507 | 'entries': entries, | |
1508 | } | |
1509 | ||
1510 | ||
1511 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): | |
1512 | """ | |
1513 | Base class for extractors that fetch info from | |
1514 | http://www.youtube.com/feed_ajax | |
1515 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1516 | """ | |
1517 | _LOGIN_REQUIRED = True | |
1518 | # use action_load_personal_feed instead of action_load_system_feed | |
1519 | _PERSONAL_FEED = False | |
1520 | ||
1521 | @property | |
1522 | def _FEED_TEMPLATE(self): | |
1523 | action = 'action_load_system_feed' | |
1524 | if self._PERSONAL_FEED: | |
1525 | action = 'action_load_personal_feed' | |
1526 | return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
1527 | ||
1528 | @property | |
1529 | def IE_NAME(self): | |
1530 | return 'youtube:%s' % self._FEED_NAME | |
1531 | ||
1532 | def _real_initialize(self): | |
1533 | self._login() | |
1534 | ||
1535 | def _real_extract(self, url): | |
1536 | feed_entries = [] | |
1537 | paging = 0 | |
1538 | for i in itertools.count(1): | |
1539 | info = self._download_json(self._FEED_TEMPLATE % paging, | |
1540 | '%s feed' % self._FEED_NAME, | |
1541 | 'Downloading page %s' % i) | |
1542 | feed_html = info.get('feed_html') or info.get('content_html') | |
1543 | load_more_widget_html = info.get('load_more_widget_html') or feed_html | |
1544 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) | |
1545 | ids = orderedSet(m.group(1) for m in m_ids) | |
1546 | feed_entries.extend( | |
1547 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1548 | for video_id in ids) | |
1549 | mobj = re.search( | |
1550 | r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)', | |
1551 | load_more_widget_html) | |
1552 | if mobj is None: | |
1553 | break | |
1554 | paging = mobj.group('paging') | |
1555 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) | |
1556 | ||
1557 | ||
1558 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1559 | IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)' | |
1560 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1561 | _FEED_NAME = 'recommended' | |
1562 | _PLAYLIST_TITLE = 'Youtube Recommended videos' | |
1563 | ||
1564 | ||
1565 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): | |
1566 | IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)' | |
1567 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1568 | _FEED_NAME = 'watch_later' | |
1569 | _PLAYLIST_TITLE = 'Youtube Watch Later' | |
1570 | _PERSONAL_FEED = True | |
1571 | ||
1572 | ||
1573 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): | |
1574 | IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)' | |
1575 | _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory' | |
1576 | _FEED_NAME = 'history' | |
1577 | _PERSONAL_FEED = True | |
1578 | _PLAYLIST_TITLE = 'Youtube Watch History' | |
1579 | ||
1580 | ||
1581 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): | |
1582 | IE_NAME = 'youtube:favorites' | |
1583 | IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)' | |
1584 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' | |
1585 | _LOGIN_REQUIRED = True | |
1586 | ||
1587 | def _real_extract(self, url): | |
1588 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1589 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id') | |
1590 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
1591 | ||
1592 | ||
1593 | class YoutubeSubscriptionsIE(YoutubePlaylistIE): | |
1594 | IE_NAME = 'youtube:subscriptions' | |
1595 | IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' | |
1596 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1597 | _TESTS = [] | |
1598 | ||
1599 | def _real_extract(self, url): | |
1600 | title = 'Youtube Subscriptions' | |
1601 | page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title) | |
1602 | ||
1603 | # The extraction process is the same as for playlists, but the regex | |
1604 | # for the video ids doesn't contain an index | |
1605 | ids = [] | |
1606 | more_widget_html = content_html = page | |
1607 | ||
1608 | for page_num in itertools.count(1): | |
1609 | matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html) | |
1610 | new_ids = orderedSet(matches) | |
1611 | ids.extend(new_ids) | |
1612 | ||
1613 | mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) | |
1614 | if not mobj: | |
1615 | break | |
1616 | ||
1617 | more = self._download_json( | |
1618 | 'https://youtube.com/%s' % mobj.group('more'), title, | |
1619 | 'Downloading page #%s' % page_num, | |
1620 | transform_source=uppercase_escape) | |
1621 | content_html = more['content_html'] | |
1622 | more_widget_html = more['load_more_widget_html'] | |
1623 | ||
1624 | return { | |
1625 | '_type': 'playlist', | |
1626 | 'title': title, | |
1627 | 'entries': self._ids_to_results(ids), | |
1628 | } | |
1629 | ||
1630 | ||
1631 | class YoutubeTruncatedURLIE(InfoExtractor): | |
1632 | IE_NAME = 'youtube:truncated_url' | |
1633 | IE_DESC = False # Do not list | |
1634 | _VALID_URL = r'''(?x) | |
1635 | (?:https?://)?[^/]+/watch\?(?: | |
1636 | feature=[a-z_]+| | |
1637 | annotation_id=annotation_[^&]+ | |
1638 | )?$| | |
1639 | (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$ | |
1640 | ''' | |
1641 | ||
1642 | _TESTS = [{ | |
1643 | 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041', | |
1644 | 'only_matching': True, | |
1645 | }, { | |
1646 | 'url': 'http://www.youtube.com/watch?', | |
1647 | 'only_matching': True, | |
1648 | }] | |
1649 | ||
1650 | def _real_extract(self, url): | |
1651 | raise ExtractorError( | |
1652 | 'Did you forget to quote the URL? Remember that & is a meta ' | |
1653 | 'character in most shells, so you want to put the URL in quotes, ' | |
1654 | 'like youtube-dl ' | |
1655 | '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' | |
1656 | ' or simply youtube-dl BaW_jenozKc .', | |
1657 | expected=True) |