]>
Commit | Line | Data |
---|---|---|
c5e8d7af | 1 | # coding: utf-8 |
c5e8d7af | 2 | |
e0df6211 | 3 | import collections |
edf3e38e | 4 | import errno |
e0df6211 | 5 | import io |
0ca96d48 | 6 | import itertools |
c5e8d7af | 7 | import json |
c4417ddb | 8 | import os.path |
c5e8d7af | 9 | import re |
e0df6211 PH |
10 | import string |
11 | import struct | |
12 | import traceback | |
13 | import zlib | |
c5e8d7af | 14 | |
b05654f0 | 15 | from .common import InfoExtractor, SearchInfoExtractor |
54d39d8b | 16 | from .subtitles import SubtitlesInfoExtractor |
c5e8d7af | 17 | from ..utils import ( |
edf3e38e | 18 | compat_chr, |
c5e8d7af | 19 | compat_parse_qs, |
c5e8d7af PH |
20 | compat_urllib_parse, |
21 | compat_urllib_request, | |
7c61bd36 | 22 | compat_urlparse, |
c5e8d7af PH |
23 | compat_str, |
24 | ||
25 | clean_html, | |
c38b1e77 | 26 | get_cachedir, |
c5e8d7af | 27 | get_element_by_id, |
652cdaa2 | 28 | get_element_by_attribute, |
c5e8d7af PH |
29 | ExtractorError, |
30 | unescapeHTML, | |
31 | unified_strdate, | |
04cc9617 | 32 | orderedSet, |
edf3e38e | 33 | write_json_file, |
c5e8d7af PH |
34 | ) |
35 | ||
de7f3446 | 36 | class YoutubeBaseInfoExtractor(InfoExtractor): |
b2e8bc1b JMF |
37 | """Provide base functions for Youtube extractors""" |
38 | _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' | |
39 | _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' | |
40 | _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' | |
41 | _NETRC_MACHINE = 'youtube' | |
42 | # If True it will raise an error if no login info is provided | |
43 | _LOGIN_REQUIRED = False | |
44 | ||
b2e8bc1b | 45 | def _set_language(self): |
7cc3570e PH |
46 | return bool(self._download_webpage( |
47 | self._LANG_URL, None, | |
48 | note=u'Setting language', errnote='unable to set language', | |
49 | fatal=False)) | |
b2e8bc1b JMF |
50 | |
51 | def _login(self): | |
52 | (username, password) = self._get_login_info() | |
53 | # No authentication to be performed | |
54 | if username is None: | |
55 | if self._LOGIN_REQUIRED: | |
56 | raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) | |
57 | return False | |
58 | ||
7cc3570e PH |
59 | login_page = self._download_webpage( |
60 | self._LOGIN_URL, None, | |
61 | note=u'Downloading login page', | |
62 | errnote=u'unable to fetch login page', fatal=False) | |
63 | if login_page is False: | |
64 | return | |
b2e8bc1b | 65 | |
795f28f8 PH |
66 | galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', |
67 | login_page, u'Login GALX parameter') | |
c5e8d7af | 68 | |
b2e8bc1b JMF |
69 | # Log in |
70 | login_form_strs = { | |
71 | u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', | |
72 | u'Email': username, | |
73 | u'GALX': galx, | |
74 | u'Passwd': password, | |
75 | u'PersistentCookie': u'yes', | |
76 | u'_utf8': u'霱', | |
77 | u'bgresponse': u'js_disabled', | |
78 | u'checkConnection': u'', | |
79 | u'checkedDomains': u'youtube', | |
80 | u'dnConn': u'', | |
b2e8bc1b JMF |
81 | u'pstMsg': u'0', |
82 | u'rmShown': u'1', | |
83 | u'secTok': u'', | |
84 | u'signIn': u'Sign in', | |
85 | u'timeStmp': u'', | |
86 | u'service': u'youtube', | |
87 | u'uilel': u'3', | |
88 | u'hl': u'en_US', | |
89 | } | |
90 | # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode | |
91 | # chokes on unicode | |
92 | login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) | |
93 | login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') | |
7cc3570e PH |
94 | |
95 | req = compat_urllib_request.Request(self._LOGIN_URL, login_data) | |
96 | login_results = self._download_webpage( | |
97 | req, None, | |
98 | note=u'Logging in', errnote=u'unable to log in', fatal=False) | |
99 | if login_results is False: | |
100 | return False | |
101 | if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: | |
102 | self._downloader.report_warning(u'unable to log in: bad username or password') | |
b2e8bc1b JMF |
103 | return False |
104 | return True | |
105 | ||
106 | def _confirm_age(self): | |
107 | age_form = { | |
7cc3570e PH |
108 | 'next_url': '/', |
109 | 'action_confirm': 'Confirm', | |
110 | } | |
111 | req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) | |
112 | ||
113 | self._download_webpage( | |
114 | req, None, | |
115 | note=u'Confirming age', errnote=u'Unable to confirm age') | |
b2e8bc1b JMF |
116 | return True |
117 | ||
118 | def _real_initialize(self): | |
119 | if self._downloader is None: | |
120 | return | |
121 | if not self._set_language(): | |
122 | return | |
123 | if not self._login(): | |
124 | return | |
125 | self._confirm_age() | |
c5e8d7af | 126 | |
8377574c | 127 | |
de7f3446 | 128 | class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): |
0f818663 | 129 | IE_DESC = u'YouTube.com' |
cb7dfeea | 130 | _VALID_URL = r"""(?x)^ |
c5e8d7af | 131 | ( |
83aa5293 | 132 | (?:https?://|//)? # http(s):// or protocol-independent URL (optional) |
cb7dfeea | 133 | (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| |
e69ae5b9 JMF |
134 | tube\.majestyc\.net/| |
135 | youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains | |
c5e8d7af PH |
136 | (?:.*?\#/)? # handle anchor (#/) redirect urls |
137 | (?: # the various things that can precede the ID: | |
138 | (?:(?:v|embed|e)/) # v/ or embed/ or e/ | |
139 | |(?: # or the v= param in all its forms | |
d741e55a | 140 | (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) |
c5e8d7af PH |
141 | (?:\?|\#!?) # the params delimiter ? or # or #! |
142 | (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) | |
143 | v= | |
144 | ) | |
f4b05232 JMF |
145 | )) |
146 | |youtu\.be/ # just youtu.be/xxxx | |
147 | ) | |
c5e8d7af | 148 | )? # all until now is optional -> you can pass the naked ID |
8963d9c2 | 149 | ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID |
c5e8d7af PH |
150 | (?(1).+)? # if we found the ID, everything can follow |
151 | $""" | |
c5e8d7af | 152 | _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' |
c5e8d7af | 153 | # Listed in order of quality |
bdc6b3fc | 154 | _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13', |
96fb5605 | 155 | # Apple HTTP Live Streaming |
bdc6b3fc | 156 | '96', '95', '94', '93', '92', '132', '151', |
939fbd26 JMF |
157 | # 3D |
158 | '85', '84', '102', '83', '101', '82', '100', | |
159 | # Dash video | |
160 | '138', '137', '248', '136', '247', '135', '246', | |
161 | '245', '244', '134', '243', '133', '242', '160', | |
162 | # Dash audio | |
163 | '141', '172', '140', '171', '139', | |
1d043b93 | 164 | ] |
bdc6b3fc | 165 | _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13', |
96fb5605 | 166 | # Apple HTTP Live Streaming |
bdc6b3fc AZ |
167 | '96', '95', '94', '93', '92', '132', '151', |
168 | # 3D | |
86fe61c8 | 169 | '85', '102', '84', '101', '83', '100', '82', |
939fbd26 JMF |
170 | # Dash video |
171 | '138', '248', '137', '247', '136', '246', '245', | |
172 | '244', '135', '243', '134', '242', '133', '160', | |
173 | # Dash audio | |
174 | '172', '141', '171', '140', '139', | |
1d043b93 | 175 | ] |
bdc6b3fc AZ |
176 | _video_formats_map = { |
177 | 'flv': ['35', '34', '6', '5'], | |
178 | '3gp': ['36', '17', '13'], | |
179 | 'mp4': ['38', '37', '22', '18'], | |
180 | 'webm': ['46', '45', '44', '43'], | |
181 | } | |
c5e8d7af PH |
182 | _video_extensions = { |
183 | '13': '3gp', | |
bdc6b3fc | 184 | '17': '3gp', |
c5e8d7af PH |
185 | '18': 'mp4', |
186 | '22': 'mp4', | |
bdc6b3fc | 187 | '36': '3gp', |
c5e8d7af | 188 | '37': 'mp4', |
d69cf69a | 189 | '38': 'mp4', |
c5e8d7af PH |
190 | '43': 'webm', |
191 | '44': 'webm', | |
192 | '45': 'webm', | |
193 | '46': 'webm', | |
1d043b93 | 194 | |
86fe61c8 AZ |
195 | # 3d videos |
196 | '82': 'mp4', | |
197 | '83': 'mp4', | |
198 | '84': 'mp4', | |
199 | '85': 'mp4', | |
200 | '100': 'webm', | |
201 | '101': 'webm', | |
202 | '102': 'webm', | |
836a086c | 203 | |
96fb5605 | 204 | # Apple HTTP Live Streaming |
1d043b93 JMF |
205 | '92': 'mp4', |
206 | '93': 'mp4', | |
207 | '94': 'mp4', | |
208 | '95': 'mp4', | |
209 | '96': 'mp4', | |
210 | '132': 'mp4', | |
211 | '151': 'mp4', | |
836a086c AZ |
212 | |
213 | # Dash mp4 | |
214 | '133': 'mp4', | |
215 | '134': 'mp4', | |
216 | '135': 'mp4', | |
217 | '136': 'mp4', | |
218 | '137': 'mp4', | |
219 | '138': 'mp4', | |
836a086c AZ |
220 | '160': 'mp4', |
221 | ||
f6f1fc92 RB |
222 | # Dash mp4 audio |
223 | '139': 'm4a', | |
16f36a6f RB |
224 | '140': 'm4a', |
225 | '141': 'm4a', | |
836a086c AZ |
226 | |
227 | # Dash webm | |
228 | '171': 'webm', | |
229 | '172': 'webm', | |
230 | '242': 'webm', | |
231 | '243': 'webm', | |
232 | '244': 'webm', | |
233 | '245': 'webm', | |
234 | '246': 'webm', | |
235 | '247': 'webm', | |
236 | '248': 'webm', | |
c5e8d7af PH |
237 | } |
238 | _video_dimensions = { | |
d5a9bb4e | 239 | '5': '400x240', |
c5e8d7af PH |
240 | '6': '???', |
241 | '13': '???', | |
d5a9bb4e RB |
242 | '17': '176x144', |
243 | '18': '640x360', | |
244 | '22': '1280x720', | |
245 | '34': '640x360', | |
246 | '35': '854x480', | |
247 | '36': '320x240', | |
248 | '37': '1920x1080', | |
249 | '38': '4096x3072', | |
250 | '43': '640x360', | |
251 | '44': '854x480', | |
252 | '45': '1280x720', | |
253 | '46': '1920x1080', | |
86fe61c8 AZ |
254 | '82': '360p', |
255 | '83': '480p', | |
256 | '84': '720p', | |
257 | '85': '1080p', | |
1d043b93 JMF |
258 | '92': '240p', |
259 | '93': '360p', | |
260 | '94': '480p', | |
261 | '95': '720p', | |
262 | '96': '1080p', | |
86fe61c8 AZ |
263 | '100': '360p', |
264 | '101': '480p', | |
836a086c | 265 | '102': '720p', |
1d043b93 JMF |
266 | '132': '240p', |
267 | '151': '72p', | |
836a086c AZ |
268 | '133': '240p', |
269 | '134': '360p', | |
270 | '135': '480p', | |
271 | '136': '720p', | |
272 | '137': '1080p', | |
273 | '138': '>1080p', | |
274 | '139': '48k', | |
275 | '140': '128k', | |
276 | '141': '256k', | |
277 | '160': '192p', | |
278 | '171': '128k', | |
279 | '172': '256k', | |
280 | '242': '240p', | |
281 | '243': '360p', | |
282 | '244': '480p', | |
283 | '245': '480p', | |
284 | '246': '480p', | |
285 | '247': '720p', | |
286 | '248': '1080p', | |
c5e8d7af | 287 | } |
836a086c AZ |
288 | _special_itags = { |
289 | '82': '3D', | |
290 | '83': '3D', | |
291 | '84': '3D', | |
292 | '85': '3D', | |
293 | '100': '3D', | |
294 | '101': '3D', | |
295 | '102': '3D', | |
296 | '133': 'DASH Video', | |
297 | '134': 'DASH Video', | |
298 | '135': 'DASH Video', | |
299 | '136': 'DASH Video', | |
300 | '137': 'DASH Video', | |
301 | '138': 'DASH Video', | |
302 | '139': 'DASH Audio', | |
303 | '140': 'DASH Audio', | |
304 | '141': 'DASH Audio', | |
305 | '160': 'DASH Video', | |
306 | '171': 'DASH Audio', | |
307 | '172': 'DASH Audio', | |
308 | '242': 'DASH Video', | |
309 | '243': 'DASH Video', | |
310 | '244': 'DASH Video', | |
311 | '245': 'DASH Video', | |
312 | '246': 'DASH Video', | |
313 | '247': 'DASH Video', | |
314 | '248': 'DASH Video', | |
c5e8d7af | 315 | } |
836a086c | 316 | |
c5e8d7af | 317 | IE_NAME = u'youtube' |
2eb88d95 PH |
318 | _TESTS = [ |
319 | { | |
0e853ca4 PH |
320 | u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", |
321 | u"file": u"BaW_jenozKc.mp4", | |
322 | u"info_dict": { | |
323 | u"title": u"youtube-dl test video \"'/\\ä↭𝕐", | |
324 | u"uploader": u"Philipp Hagemeister", | |
325 | u"uploader_id": u"phihag", | |
326 | u"upload_date": u"20121002", | |
27dcce19 | 327 | u"description": u"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 ." |
2eb88d95 | 328 | } |
0e853ca4 | 329 | }, |
0e853ca4 PH |
330 | { |
331 | u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", | |
332 | u"file": u"UxxajLWwzqY.mp4", | |
333 | u"note": u"Test generic use_cipher_signature video (#897)", | |
334 | u"info_dict": { | |
335 | u"upload_date": u"20120506", | |
336 | u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", | |
2dc59299 | 337 | u"description": u"md5:5b292926389560516e384ac437c0ec07", |
45ed795c | 338 | u"uploader": u"Icona Pop", |
0e853ca4 | 339 | u"uploader_id": u"IconaPop" |
2eb88d95 | 340 | } |
c108eb73 JMF |
341 | }, |
342 | { | |
343 | u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", | |
344 | u"file": u"07FYdnEawAQ.mp4", | |
345 | u"note": u"Test VEVO video with age protection (#956)", | |
346 | u"info_dict": { | |
347 | u"upload_date": u"20130703", | |
348 | u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", | |
349 | u"description": u"md5:64249768eec3bc4276236606ea996373", | |
350 | u"uploader": u"justintimberlakeVEVO", | |
351 | u"uploader_id": u"justintimberlakeVEVO" | |
352 | } | |
353 | }, | |
fccd3771 | 354 | { |
83aa5293 | 355 | u"url": u"//www.YouTube.com/watch?v=yZIXLfi8CZQ", |
fccd3771 PH |
356 | u"file": u"yZIXLfi8CZQ.mp4", |
357 | u"note": u"Embed-only video (#1746)", | |
358 | u"info_dict": { | |
359 | u"upload_date": u"20120608", | |
360 | u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", | |
361 | u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", | |
362 | u"uploader": u"SET India", | |
363 | u"uploader_id": u"setindia" | |
364 | } | |
365 | }, | |
2eb88d95 PH |
366 | ] |
367 | ||
c5e8d7af PH |
368 | |
369 | @classmethod | |
370 | def suitable(cls, url): | |
371 | """Receives a URL and returns True if suitable for this IE.""" | |
e3ea4790 | 372 | if YoutubePlaylistIE.suitable(url): return False |
fccd3771 | 373 | return re.match(cls._VALID_URL, url) is not None |
c5e8d7af | 374 | |
e0df6211 PH |
375 | def __init__(self, *args, **kwargs): |
376 | super(YoutubeIE, self).__init__(*args, **kwargs) | |
83799698 | 377 | self._player_cache = {} |
e0df6211 | 378 | |
c5e8d7af PH |
379 | def report_video_info_webpage_download(self, video_id): |
380 | """Report attempt to download video info webpage.""" | |
381 | self.to_screen(u'%s: Downloading video info webpage' % video_id) | |
382 | ||
c5e8d7af PH |
383 | def report_information_extraction(self, video_id): |
384 | """Report attempt to extract video information.""" | |
385 | self.to_screen(u'%s: Extracting video information' % video_id) | |
386 | ||
387 | def report_unavailable_format(self, video_id, format): | |
388 | """Report extracted video URL.""" | |
389 | self.to_screen(u'%s: Format %s not available' % (video_id, format)) | |
390 | ||
391 | def report_rtmp_download(self): | |
392 | """Indicate the download will use the RTMP protocol.""" | |
393 | self.to_screen(u'RTMP download detected') | |
394 | ||
c4417ddb PH |
395 | def _extract_signature_function(self, video_id, player_url, slen): |
396 | id_m = re.match(r'.*-(?P<id>[a-zA-Z0-9_-]+)\.(?P<ext>[a-z]+)$', | |
83799698 | 397 | player_url) |
e0df6211 PH |
398 | player_type = id_m.group('ext') |
399 | player_id = id_m.group('id') | |
400 | ||
c4417ddb PH |
401 | # Read from filesystem cache |
402 | func_id = '%s_%s_%d' % (player_type, player_id, slen) | |
403 | assert os.path.basename(func_id) == func_id | |
c38b1e77 | 404 | cache_dir = get_cachedir(self._downloader.params) |
c4417ddb | 405 | |
c3c88a26 | 406 | cache_enabled = cache_dir is not None |
f8061589 | 407 | if cache_enabled: |
c4417ddb PH |
408 | cache_fn = os.path.join(os.path.expanduser(cache_dir), |
409 | u'youtube-sigfuncs', | |
410 | func_id + '.json') | |
411 | try: | |
edf3e38e | 412 | with io.open(cache_fn, 'r', encoding='utf-8') as cachef: |
c4417ddb PH |
413 | cache_spec = json.load(cachef) |
414 | return lambda s: u''.join(s[i] for i in cache_spec) | |
edf3e38e | 415 | except IOError: |
c4417ddb | 416 | pass # No cache available |
83799698 | 417 | |
e0df6211 PH |
418 | if player_type == 'js': |
419 | code = self._download_webpage( | |
420 | player_url, video_id, | |
83799698 | 421 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 | 422 | errnote=u'Download of %s failed' % player_url) |
83799698 | 423 | res = self._parse_sig_js(code) |
c4417ddb | 424 | elif player_type == 'swf': |
e0df6211 PH |
425 | urlh = self._request_webpage( |
426 | player_url, video_id, | |
83799698 | 427 | note=u'Downloading %s player %s' % (player_type, player_id), |
e0df6211 PH |
428 | errnote=u'Download of %s failed' % player_url) |
429 | code = urlh.read() | |
83799698 | 430 | res = self._parse_sig_swf(code) |
e0df6211 PH |
431 | else: |
432 | assert False, 'Invalid player type %r' % player_type | |
433 | ||
f8061589 | 434 | if cache_enabled: |
edf3e38e | 435 | try: |
c705320f PH |
436 | test_string = u''.join(map(compat_chr, range(slen))) |
437 | cache_res = res(test_string) | |
edf3e38e PH |
438 | cache_spec = [ord(c) for c in cache_res] |
439 | try: | |
440 | os.makedirs(os.path.dirname(cache_fn)) | |
441 | except OSError as ose: | |
442 | if ose.errno != errno.EEXIST: | |
443 | raise | |
444 | write_json_file(cache_spec, cache_fn) | |
0ca96d48 | 445 | except Exception: |
edf3e38e PH |
446 | tb = traceback.format_exc() |
447 | self._downloader.report_warning( | |
448 | u'Writing cache to %r failed: %s' % (cache_fn, tb)) | |
83799698 PH |
449 | |
450 | return res | |
451 | ||
edf3e38e PH |
452 | def _print_sig_code(self, func, slen): |
453 | def gen_sig_code(idxs): | |
454 | def _genslice(start, end, step): | |
455 | starts = u'' if start == 0 else str(start) | |
e35e4ddc PH |
456 | ends = (u':%d' % (end+step)) if end + step >= 0 else u':' |
457 | steps = u'' if step == 1 else (u':%d' % step) | |
edf3e38e PH |
458 | return u's[%s%s%s]' % (starts, ends, steps) |
459 | ||
460 | step = None | |
0ca96d48 PH |
461 | start = '(Never used)' # Quelch pyflakes warnings - start will be |
462 | # set as soon as step is set | |
edf3e38e PH |
463 | for i, prev in zip(idxs[1:], idxs[:-1]): |
464 | if step is not None: | |
465 | if i - prev == step: | |
466 | continue | |
467 | yield _genslice(start, prev, step) | |
468 | step = None | |
469 | continue | |
470 | if i - prev in [-1, 1]: | |
471 | step = i - prev | |
472 | start = prev | |
473 | continue | |
474 | else: | |
475 | yield u's[%d]' % prev | |
476 | if step is None: | |
477 | yield u's[%d]' % i | |
478 | else: | |
479 | yield _genslice(start, i, step) | |
480 | ||
c705320f PH |
481 | test_string = u''.join(map(compat_chr, range(slen))) |
482 | cache_res = func(test_string) | |
edf3e38e PH |
483 | cache_spec = [ord(c) for c in cache_res] |
484 | expr_code = u' + '.join(gen_sig_code(cache_spec)) | |
485 | code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) | |
f8061589 | 486 | self.to_screen(u'Extracted signature function:\n' + code) |
edf3e38e | 487 | |
e0df6211 PH |
488 | def _parse_sig_js(self, jscode): |
489 | funcname = self._search_regex( | |
490 | r'signature=([a-zA-Z]+)', jscode, | |
491 | u'Initial JS player signature function name') | |
492 | ||
493 | functions = {} | |
494 | ||
495 | def argidx(varname): | |
496 | return string.lowercase.index(varname) | |
497 | ||
498 | def interpret_statement(stmt, local_vars, allow_recursion=20): | |
499 | if allow_recursion < 0: | |
0ca96d48 | 500 | raise ExtractorError(u'Recursion limit reached') |
e0df6211 PH |
501 | |
502 | if stmt.startswith(u'var '): | |
503 | stmt = stmt[len(u'var '):] | |
504 | ass_m = re.match(r'^(?P<out>[a-z]+)(?:\[(?P<index>[^\]]+)\])?' + | |
505 | r'=(?P<expr>.*)$', stmt) | |
506 | if ass_m: | |
507 | if ass_m.groupdict().get('index'): | |
508 | def assign(val): | |
509 | lvar = local_vars[ass_m.group('out')] | |
510 | idx = interpret_expression(ass_m.group('index'), | |
511 | local_vars, allow_recursion) | |
512 | assert isinstance(idx, int) | |
513 | lvar[idx] = val | |
514 | return val | |
515 | expr = ass_m.group('expr') | |
516 | else: | |
517 | def assign(val): | |
518 | local_vars[ass_m.group('out')] = val | |
519 | return val | |
520 | expr = ass_m.group('expr') | |
521 | elif stmt.startswith(u'return '): | |
522 | assign = lambda v: v | |
523 | expr = stmt[len(u'return '):] | |
524 | else: | |
525 | raise ExtractorError( | |
526 | u'Cannot determine left side of statement in %r' % stmt) | |
527 | ||
528 | v = interpret_expression(expr, local_vars, allow_recursion) | |
529 | return assign(v) | |
530 | ||
531 | def interpret_expression(expr, local_vars, allow_recursion): | |
532 | if expr.isdigit(): | |
533 | return int(expr) | |
534 | ||
535 | if expr.isalpha(): | |
536 | return local_vars[expr] | |
537 | ||
538 | m = re.match(r'^(?P<in>[a-z]+)\.(?P<member>.*)$', expr) | |
539 | if m: | |
540 | member = m.group('member') | |
541 | val = local_vars[m.group('in')] | |
542 | if member == 'split("")': | |
543 | return list(val) | |
544 | if member == 'join("")': | |
545 | return u''.join(val) | |
546 | if member == 'length': | |
547 | return len(val) | |
548 | if member == 'reverse()': | |
549 | return val[::-1] | |
550 | slice_m = re.match(r'slice\((?P<idx>.*)\)', member) | |
551 | if slice_m: | |
552 | idx = interpret_expression( | |
553 | slice_m.group('idx'), local_vars, allow_recursion-1) | |
554 | return val[idx:] | |
555 | ||
556 | m = re.match( | |
557 | r'^(?P<in>[a-z]+)\[(?P<idx>.+)\]$', expr) | |
558 | if m: | |
559 | val = local_vars[m.group('in')] | |
560 | idx = interpret_expression(m.group('idx'), local_vars, | |
561 | allow_recursion-1) | |
562 | return val[idx] | |
563 | ||
564 | m = re.match(r'^(?P<a>.+?)(?P<op>[%])(?P<b>.+?)$', expr) | |
565 | if m: | |
566 | a = interpret_expression(m.group('a'), | |
567 | local_vars, allow_recursion) | |
568 | b = interpret_expression(m.group('b'), | |
569 | local_vars, allow_recursion) | |
570 | return a % b | |
571 | ||
572 | m = re.match( | |
573 | r'^(?P<func>[a-zA-Z]+)\((?P<args>[a-z0-9,]+)\)$', expr) | |
574 | if m: | |
575 | fname = m.group('func') | |
576 | if fname not in functions: | |
577 | functions[fname] = extract_function(fname) | |
578 | argvals = [int(v) if v.isdigit() else local_vars[v] | |
579 | for v in m.group('args').split(',')] | |
580 | return functions[fname](argvals) | |
581 | raise ExtractorError(u'Unsupported JS expression %r' % expr) | |
582 | ||
583 | def extract_function(funcname): | |
584 | func_m = re.search( | |
585 | r'function ' + re.escape(funcname) + | |
586 | r'\((?P<args>[a-z,]+)\){(?P<code>[^}]+)}', | |
587 | jscode) | |
588 | argnames = func_m.group('args').split(',') | |
589 | ||
590 | def resf(args): | |
591 | local_vars = dict(zip(argnames, args)) | |
592 | for stmt in func_m.group('code').split(';'): | |
593 | res = interpret_statement(stmt, local_vars) | |
594 | return res | |
595 | return resf | |
596 | ||
597 | initial_function = extract_function(funcname) | |
598 | return lambda s: initial_function([s]) | |
599 | ||
600 | def _parse_sig_swf(self, file_contents): | |
601 | if file_contents[1:3] != b'WS': | |
602 | raise ExtractorError( | |
603 | u'Not an SWF file; header is %r' % file_contents[:3]) | |
604 | if file_contents[:1] == b'C': | |
605 | content = zlib.decompress(file_contents[8:]) | |
606 | else: | |
607 | raise NotImplementedError(u'Unsupported compression format %r' % | |
608 | file_contents[:1]) | |
609 | ||
610 | def extract_tags(content): | |
611 | pos = 0 | |
612 | while pos < len(content): | |
613 | header16 = struct.unpack('<H', content[pos:pos+2])[0] | |
614 | pos += 2 | |
615 | tag_code = header16 >> 6 | |
616 | tag_len = header16 & 0x3f | |
617 | if tag_len == 0x3f: | |
618 | tag_len = struct.unpack('<I', content[pos:pos+4])[0] | |
619 | pos += 4 | |
620 | assert pos+tag_len <= len(content) | |
621 | yield (tag_code, content[pos:pos+tag_len]) | |
622 | pos += tag_len | |
623 | ||
624 | code_tag = next(tag | |
625 | for tag_code, tag in extract_tags(content) | |
626 | if tag_code == 82) | |
627 | p = code_tag.index(b'\0', 4) + 1 | |
ba552f54 | 628 | code_reader = io.BytesIO(code_tag[p:]) |
e0df6211 PH |
629 | |
630 | # Parse ABC (AVM2 ByteCode) | |
ba552f54 PH |
631 | def read_int(reader=None): |
632 | if reader is None: | |
633 | reader = code_reader | |
e0df6211 PH |
634 | res = 0 |
635 | shift = 0 | |
636 | for _ in range(5): | |
ba552f54 PH |
637 | buf = reader.read(1) |
638 | assert len(buf) == 1 | |
639 | b = struct.unpack('<B', buf)[0] | |
e0df6211 PH |
640 | res = res | ((b & 0x7f) << shift) |
641 | if b & 0x80 == 0: | |
642 | break | |
643 | shift += 7 | |
ba552f54 PH |
644 | return res |
645 | ||
646 | def u30(reader=None): | |
647 | res = read_int(reader) | |
648 | assert res & 0xf0000000 == 0 | |
e0df6211 PH |
649 | return res |
650 | u32 = read_int | |
651 | ||
ba552f54 PH |
652 | def s32(reader=None): |
653 | v = read_int(reader) | |
e0df6211 PH |
654 | if v & 0x80000000 != 0: |
655 | v = - ((v ^ 0xffffffff) + 1) | |
ba552f54 PH |
656 | return v |
657 | ||
0ca96d48 | 658 | def read_string(reader=None): |
ba552f54 PH |
659 | if reader is None: |
660 | reader = code_reader | |
661 | slen = u30(reader) | |
662 | resb = reader.read(slen) | |
663 | assert len(resb) == slen | |
664 | return resb.decode('utf-8') | |
665 | ||
666 | def read_bytes(count, reader=None): | |
667 | if reader is None: | |
668 | reader = code_reader | |
669 | resb = reader.read(count) | |
670 | assert len(resb) == count | |
671 | return resb | |
672 | ||
673 | def read_byte(reader=None): | |
674 | resb = read_bytes(1, reader=reader) | |
675 | res = struct.unpack('<B', resb)[0] | |
676 | return res | |
e0df6211 PH |
677 | |
678 | # minor_version + major_version | |
0ca96d48 | 679 | read_bytes(2 + 2) |
e0df6211 PH |
680 | |
681 | # Constant pool | |
ba552f54 | 682 | int_count = u30() |
e0df6211 | 683 | for _c in range(1, int_count): |
0ca96d48 | 684 | s32() |
ba552f54 | 685 | uint_count = u30() |
e0df6211 | 686 | for _c in range(1, uint_count): |
0ca96d48 | 687 | u32() |
ba552f54 | 688 | double_count = u30() |
0ca96d48 | 689 | read_bytes((double_count-1) * 8) |
ba552f54 | 690 | string_count = u30() |
e0df6211 PH |
691 | constant_strings = [u''] |
692 | for _c in range(1, string_count): | |
0ca96d48 | 693 | s = read_string() |
e0df6211 | 694 | constant_strings.append(s) |
ba552f54 | 695 | namespace_count = u30() |
e0df6211 | 696 | for _c in range(1, namespace_count): |
0ca96d48 PH |
697 | read_bytes(1) # kind |
698 | u30() # name | |
ba552f54 | 699 | ns_set_count = u30() |
e0df6211 | 700 | for _c in range(1, ns_set_count): |
ba552f54 | 701 | count = u30() |
e0df6211 | 702 | for _c2 in range(count): |
0ca96d48 | 703 | u30() |
ba552f54 | 704 | multiname_count = u30() |
e0df6211 PH |
705 | MULTINAME_SIZES = { |
706 | 0x07: 2, # QName | |
707 | 0x0d: 2, # QNameA | |
708 | 0x0f: 1, # RTQName | |
709 | 0x10: 1, # RTQNameA | |
710 | 0x11: 0, # RTQNameL | |
711 | 0x12: 0, # RTQNameLA | |
712 | 0x09: 2, # Multiname | |
713 | 0x0e: 2, # MultinameA | |
714 | 0x1b: 1, # MultinameL | |
715 | 0x1c: 1, # MultinameLA | |
716 | } | |
717 | multinames = [u''] | |
718 | for _c in range(1, multiname_count): | |
ba552f54 | 719 | kind = u30() |
e0df6211 PH |
720 | assert kind in MULTINAME_SIZES, u'Invalid multiname kind %r' % kind |
721 | if kind == 0x07: | |
0ca96d48 | 722 | u30() # namespace_idx |
ba552f54 | 723 | name_idx = u30() |
e0df6211 PH |
724 | multinames.append(constant_strings[name_idx]) |
725 | else: | |
726 | multinames.append('[MULTINAME kind: %d]' % kind) | |
727 | for _c2 in range(MULTINAME_SIZES[kind]): | |
0ca96d48 | 728 | u30() |
e0df6211 PH |
729 | |
730 | # Methods | |
ba552f54 | 731 | method_count = u30() |
e0df6211 PH |
732 | MethodInfo = collections.namedtuple( |
733 | 'MethodInfo', | |
734 | ['NEED_ARGUMENTS', 'NEED_REST']) | |
735 | method_infos = [] | |
736 | for method_id in range(method_count): | |
ba552f54 | 737 | param_count = u30() |
0ca96d48 | 738 | u30() # return type |
e0df6211 | 739 | for _ in range(param_count): |
0ca96d48 PH |
740 | u30() # param type |
741 | u30() # name index (always 0 for youtube) | |
ba552f54 | 742 | flags = read_byte() |
e0df6211 PH |
743 | if flags & 0x08 != 0: |
744 | # Options present | |
ba552f54 | 745 | option_count = u30() |
e0df6211 | 746 | for c in range(option_count): |
0ca96d48 PH |
747 | u30() # val |
748 | read_bytes(1) # kind | |
e0df6211 PH |
749 | if flags & 0x80 != 0: |
750 | # Param names present | |
751 | for _ in range(param_count): | |
0ca96d48 | 752 | u30() # param name |
e0df6211 PH |
753 | mi = MethodInfo(flags & 0x01 != 0, flags & 0x04 != 0) |
754 | method_infos.append(mi) | |
755 | ||
756 | # Metadata | |
ba552f54 | 757 | metadata_count = u30() |
e0df6211 | 758 | for _c in range(metadata_count): |
0ca96d48 | 759 | u30() # name |
ba552f54 | 760 | item_count = u30() |
e0df6211 | 761 | for _c2 in range(item_count): |
0ca96d48 PH |
762 | u30() # key |
763 | u30() # value | |
ba552f54 PH |
764 | |
765 | def parse_traits_info(): | |
766 | trait_name_idx = u30() | |
767 | kind_full = read_byte() | |
e0df6211 PH |
768 | kind = kind_full & 0x0f |
769 | attrs = kind_full >> 4 | |
770 | methods = {} | |
771 | if kind in [0x00, 0x06]: # Slot or Const | |
0ca96d48 PH |
772 | u30() # Slot id |
773 | u30() # type_name_idx | |
ba552f54 | 774 | vindex = u30() |
e0df6211 | 775 | if vindex != 0: |
0ca96d48 | 776 | read_byte() # vkind |
e0df6211 | 777 | elif kind in [0x01, 0x02, 0x03]: # Method / Getter / Setter |
0ca96d48 | 778 | u30() # disp_id |
ba552f54 | 779 | method_idx = u30() |
e0df6211 PH |
780 | methods[multinames[trait_name_idx]] = method_idx |
781 | elif kind == 0x04: # Class | |
0ca96d48 PH |
782 | u30() # slot_id |
783 | u30() # classi | |
e0df6211 | 784 | elif kind == 0x05: # Function |
0ca96d48 | 785 | u30() # slot_id |
ba552f54 | 786 | function_idx = u30() |
e0df6211 PH |
787 | methods[function_idx] = multinames[trait_name_idx] |
788 | else: | |
789 | raise ExtractorError(u'Unsupported trait kind %d' % kind) | |
790 | ||
791 | if attrs & 0x4 != 0: # Metadata present | |
ba552f54 | 792 | metadata_count = u30() |
e0df6211 | 793 | for _c3 in range(metadata_count): |
0ca96d48 | 794 | u30() # metadata index |
e0df6211 | 795 | |
ba552f54 | 796 | return methods |
e0df6211 PH |
797 | |
798 | # Classes | |
799 | TARGET_CLASSNAME = u'SignatureDecipher' | |
800 | searched_idx = multinames.index(TARGET_CLASSNAME) | |
801 | searched_class_id = None | |
ba552f54 | 802 | class_count = u30() |
e0df6211 | 803 | for class_id in range(class_count): |
ba552f54 | 804 | name_idx = u30() |
e0df6211 PH |
805 | if name_idx == searched_idx: |
806 | # We found the class we're looking for! | |
807 | searched_class_id = class_id | |
0ca96d48 | 808 | u30() # super_name idx |
ba552f54 | 809 | flags = read_byte() |
e0df6211 | 810 | if flags & 0x08 != 0: # Protected namespace is present |
0ca96d48 | 811 | u30() # protected_ns_idx |
ba552f54 | 812 | intrf_count = u30() |
e0df6211 | 813 | for _c2 in range(intrf_count): |
0ca96d48 PH |
814 | u30() |
815 | u30() # iinit | |
ba552f54 | 816 | trait_count = u30() |
e0df6211 | 817 | for _c2 in range(trait_count): |
0ca96d48 | 818 | parse_traits_info() |
e0df6211 PH |
819 | |
820 | if searched_class_id is None: | |
821 | raise ExtractorError(u'Target class %r not found' % | |
822 | TARGET_CLASSNAME) | |
823 | ||
824 | method_names = {} | |
825 | method_idxs = {} | |
826 | for class_id in range(class_count): | |
0ca96d48 | 827 | u30() # cinit |
ba552f54 | 828 | trait_count = u30() |
e0df6211 | 829 | for _c2 in range(trait_count): |
ba552f54 | 830 | trait_methods = parse_traits_info() |
e0df6211 PH |
831 | if class_id == searched_class_id: |
832 | method_names.update(trait_methods.items()) | |
833 | method_idxs.update(dict( | |
834 | (idx, name) | |
835 | for name, idx in trait_methods.items())) | |
836 | ||
837 | # Scripts | |
ba552f54 | 838 | script_count = u30() |
e0df6211 | 839 | for _c in range(script_count): |
0ca96d48 | 840 | u30() # init |
ba552f54 | 841 | trait_count = u30() |
e0df6211 | 842 | for _c2 in range(trait_count): |
0ca96d48 | 843 | parse_traits_info() |
e0df6211 PH |
844 | |
845 | # Method bodies | |
ba552f54 | 846 | method_body_count = u30() |
e0df6211 PH |
847 | Method = collections.namedtuple('Method', ['code', 'local_count']) |
848 | methods = {} | |
849 | for _c in range(method_body_count): | |
ba552f54 | 850 | method_idx = u30() |
0ca96d48 | 851 | u30() # max_stack |
ba552f54 | 852 | local_count = u30() |
0ca96d48 PH |
853 | u30() # init_scope_depth |
854 | u30() # max_scope_depth | |
ba552f54 PH |
855 | code_length = u30() |
856 | code = read_bytes(code_length) | |
e0df6211 | 857 | if method_idx in method_idxs: |
ba552f54 | 858 | m = Method(code, local_count) |
e0df6211 | 859 | methods[method_idxs[method_idx]] = m |
ba552f54 | 860 | exception_count = u30() |
e0df6211 | 861 | for _c2 in range(exception_count): |
0ca96d48 PH |
862 | u30() # from |
863 | u30() # to | |
864 | u30() # target | |
865 | u30() # exc_type | |
866 | u30() # var_name | |
ba552f54 | 867 | trait_count = u30() |
e0df6211 | 868 | for _c2 in range(trait_count): |
0ca96d48 | 869 | parse_traits_info() |
e0df6211 | 870 | |
ba552f54 | 871 | assert p + code_reader.tell() == len(code_tag) |
e0df6211 PH |
872 | assert len(methods) == len(method_idxs) |
873 | ||
874 | method_pyfunctions = {} | |
875 | ||
876 | def extract_function(func_name): | |
877 | if func_name in method_pyfunctions: | |
878 | return method_pyfunctions[func_name] | |
879 | if func_name not in methods: | |
880 | raise ExtractorError(u'Cannot find function %r' % func_name) | |
881 | m = methods[func_name] | |
882 | ||
883 | def resfunc(args): | |
e0df6211 PH |
884 | registers = ['(this)'] + list(args) + [None] * m.local_count |
885 | stack = [] | |
886 | coder = io.BytesIO(m.code) | |
887 | while True: | |
888 | opcode = struct.unpack('!B', coder.read(1))[0] | |
a7177865 | 889 | if opcode == 36: # pushbyte |
e0df6211 PH |
890 | v = struct.unpack('!B', coder.read(1))[0] |
891 | stack.append(v) | |
892 | elif opcode == 44: # pushstring | |
893 | idx = u30(coder) | |
894 | stack.append(constant_strings[idx]) | |
895 | elif opcode == 48: # pushscope | |
896 | # We don't implement the scope register, so we'll just | |
897 | # ignore the popped value | |
898 | stack.pop() | |
899 | elif opcode == 70: # callproperty | |
900 | index = u30(coder) | |
901 | mname = multinames[index] | |
902 | arg_count = u30(coder) | |
903 | args = list(reversed( | |
904 | [stack.pop() for _ in range(arg_count)])) | |
905 | obj = stack.pop() | |
906 | if mname == u'split': | |
907 | assert len(args) == 1 | |
908 | assert isinstance(args[0], compat_str) | |
909 | assert isinstance(obj, compat_str) | |
910 | if args[0] == u'': | |
911 | res = list(obj) | |
912 | else: | |
913 | res = obj.split(args[0]) | |
914 | stack.append(res) | |
a7177865 PH |
915 | elif mname == u'slice': |
916 | assert len(args) == 1 | |
917 | assert isinstance(args[0], int) | |
918 | assert isinstance(obj, list) | |
919 | res = obj[args[0]:] | |
920 | stack.append(res) | |
921 | elif mname == u'join': | |
922 | assert len(args) == 1 | |
923 | assert isinstance(args[0], compat_str) | |
924 | assert isinstance(obj, list) | |
925 | res = args[0].join(obj) | |
926 | stack.append(res) | |
e0df6211 PH |
927 | elif mname in method_pyfunctions: |
928 | stack.append(method_pyfunctions[mname](args)) | |
929 | else: | |
930 | raise NotImplementedError( | |
931 | u'Unsupported property %r on %r' | |
932 | % (mname, obj)) | |
a7177865 PH |
933 | elif opcode == 72: # returnvalue |
934 | res = stack.pop() | |
935 | return res | |
936 | elif opcode == 79: # callpropvoid | |
937 | index = u30(coder) | |
938 | mname = multinames[index] | |
939 | arg_count = u30(coder) | |
940 | args = list(reversed( | |
941 | [stack.pop() for _ in range(arg_count)])) | |
942 | obj = stack.pop() | |
943 | if mname == u'reverse': | |
944 | assert isinstance(obj, list) | |
945 | obj.reverse() | |
946 | else: | |
947 | raise NotImplementedError( | |
948 | u'Unsupported (void) property %r on %r' | |
949 | % (mname, obj)) | |
e0df6211 PH |
950 | elif opcode == 93: # findpropstrict |
951 | index = u30(coder) | |
952 | mname = multinames[index] | |
953 | res = extract_function(mname) | |
954 | stack.append(res) | |
955 | elif opcode == 97: # setproperty | |
956 | index = u30(coder) | |
957 | value = stack.pop() | |
958 | idx = stack.pop() | |
959 | obj = stack.pop() | |
960 | assert isinstance(obj, list) | |
961 | assert isinstance(idx, int) | |
962 | obj[idx] = value | |
963 | elif opcode == 98: # getlocal | |
964 | index = u30(coder) | |
965 | stack.append(registers[index]) | |
966 | elif opcode == 99: # setlocal | |
967 | index = u30(coder) | |
968 | value = stack.pop() | |
969 | registers[index] = value | |
970 | elif opcode == 102: # getproperty | |
971 | index = u30(coder) | |
972 | pname = multinames[index] | |
973 | if pname == u'length': | |
974 | obj = stack.pop() | |
975 | assert isinstance(obj, list) | |
976 | stack.append(len(obj)) | |
977 | else: # Assume attribute access | |
978 | idx = stack.pop() | |
979 | assert isinstance(idx, int) | |
980 | obj = stack.pop() | |
981 | assert isinstance(obj, list) | |
982 | stack.append(obj[idx]) | |
983 | elif opcode == 128: # coerce | |
0ca96d48 | 984 | u30(coder) |
e0df6211 PH |
985 | elif opcode == 133: # coerce_s |
986 | assert isinstance(stack[-1], (type(None), compat_str)) | |
987 | elif opcode == 164: # modulo | |
988 | value2 = stack.pop() | |
989 | value1 = stack.pop() | |
990 | res = value1 % value2 | |
991 | stack.append(res) | |
a7177865 PH |
992 | elif opcode == 208: # getlocal_0 |
993 | stack.append(registers[0]) | |
994 | elif opcode == 209: # getlocal_1 | |
995 | stack.append(registers[1]) | |
996 | elif opcode == 210: # getlocal_2 | |
997 | stack.append(registers[2]) | |
998 | elif opcode == 211: # getlocal_3 | |
999 | stack.append(registers[3]) | |
e0df6211 PH |
1000 | elif opcode == 214: # setlocal_2 |
1001 | registers[2] = stack.pop() | |
1002 | elif opcode == 215: # setlocal_3 | |
1003 | registers[3] = stack.pop() | |
1004 | else: | |
1005 | raise NotImplementedError( | |
1006 | u'Unsupported opcode %d' % opcode) | |
1007 | ||
1008 | method_pyfunctions[func_name] = resfunc | |
1009 | return resfunc | |
1010 | ||
1011 | initial_function = extract_function(u'decipher') | |
1012 | return lambda s: initial_function([s]) | |
1013 | ||
83799698 | 1014 | def _decrypt_signature(self, s, video_id, player_url, age_gate=False): |
257a2501 | 1015 | """Turn the encrypted s field into a working signature""" |
6b37f0be | 1016 | |
83799698 | 1017 | if player_url is not None: |
9f9be844 PH |
1018 | if player_url.startswith(u'//'): |
1019 | player_url = u'https:' + player_url | |
e0df6211 | 1020 | try: |
7f8ae73a PH |
1021 | player_id = (player_url, len(s)) |
1022 | if player_id not in self._player_cache: | |
83799698 | 1023 | func = self._extract_signature_function( |
c4417ddb | 1024 | video_id, player_url, len(s) |
e0df6211 | 1025 | ) |
7f8ae73a PH |
1026 | self._player_cache[player_id] = func |
1027 | func = self._player_cache[player_id] | |
edf3e38e PH |
1028 | if self._downloader.params.get('youtube_print_sig_code'): |
1029 | self._print_sig_code(func, len(s)) | |
1030 | return func(s) | |
0ca96d48 | 1031 | except Exception: |
e0df6211 | 1032 | tb = traceback.format_exc() |
83799698 PH |
1033 | self._downloader.report_warning( |
1034 | u'Automatic signature extraction failed: ' + tb) | |
e0df6211 | 1035 | |
d2d8f895 PH |
1036 | self._downloader.report_warning( |
1037 | u'Warning: Falling back to static signature algorithm') | |
920de7a2 | 1038 | |
2f2ffea9 PH |
1039 | return self._static_decrypt_signature( |
1040 | s, video_id, player_url, age_gate) | |
e0df6211 | 1041 | |
2f2ffea9 | 1042 | def _static_decrypt_signature(self, s, video_id, player_url, age_gate): |
e0df6211 PH |
1043 | if age_gate: |
1044 | # The videos with age protection use another player, so the | |
1045 | # algorithms can be different. | |
1046 | if len(s) == 86: | |
1047 | return s[2:63] + s[82] + s[64:82] + s[63] | |
1048 | ||
bc4b9008 | 1049 | if len(s) == 93: |
1050 | return s[86:29:-1] + s[88] + s[28:5:-1] | |
1051 | elif len(s) == 92: | |
444b1165 | 1052 | return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] |
38d025b3 JMF |
1053 | elif len(s) == 91: |
1054 | return s[84:27:-1] + s[86] + s[26:5:-1] | |
444b1165 JMF |
1055 | elif len(s) == 90: |
1056 | return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] | |
8a9d86a2 | 1057 | elif len(s) == 89: |
1058 | return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1] | |
444b1165 | 1059 | elif len(s) == 88: |
3e223834 | 1060 | return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28] |
be547e1d | 1061 | elif len(s) == 87: |
3a725669 | 1062 | return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:] |
be547e1d | 1063 | elif len(s) == 86: |
f2c327fd | 1064 | return s[80:72:-1] + s[16] + s[71:39:-1] + s[72] + s[38:16:-1] + s[82] + s[15::-1] |
be547e1d | 1065 | elif len(s) == 85: |
6ae8ee3f | 1066 | return s[3:11] + s[0] + s[12:55] + s[84] + s[56:84] |
be547e1d | 1067 | elif len(s) == 84: |
6f56389b | 1068 | return s[78:70:-1] + s[14] + s[69:37:-1] + s[70] + s[36:14:-1] + s[80] + s[:14][::-1] |
be547e1d | 1069 | elif len(s) == 83: |
920de7a2 | 1070 | return s[80:63:-1] + s[0] + s[62:0:-1] + s[63] |
be547e1d | 1071 | elif len(s) == 82: |
c21315f2 | 1072 | return s[80:37:-1] + s[7] + s[36:7:-1] + s[0] + s[6:0:-1] + s[37] |
be547e1d | 1073 | elif len(s) == 81: |
aedd6bb9 | 1074 | return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] |
066090dd JMF |
1075 | elif len(s) == 80: |
1076 | return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80] | |
5c468ca8 JMF |
1077 | elif len(s) == 79: |
1078 | return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] | |
be547e1d PH |
1079 | |
1080 | else: | |
1081 | raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) | |
c5e8d7af | 1082 | |
1f343eaa | 1083 | def _get_available_subtitles(self, video_id, webpage): |
de7f3446 | 1084 | try: |
7fad1c63 JMF |
1085 | sub_list = self._download_webpage( |
1086 | 'http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, | |
1087 | video_id, note=False) | |
1088 | except ExtractorError as err: | |
de7f3446 JMF |
1089 | self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err)) |
1090 | return {} | |
1091 | lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) | |
1092 | ||
1093 | sub_lang_list = {} | |
1094 | for l in lang_list: | |
1095 | lang = l[1] | |
1096 | params = compat_urllib_parse.urlencode({ | |
1097 | 'lang': lang, | |
1098 | 'v': video_id, | |
ca715127 | 1099 | 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), |
8eddf3e9 | 1100 | 'name': l[0].encode('utf-8'), |
de7f3446 JMF |
1101 | }) |
1102 | url = u'http://www.youtube.com/api/timedtext?' + params | |
1103 | sub_lang_list[lang] = url | |
1104 | if not sub_lang_list: | |
1105 | self._downloader.report_warning(u'video doesn\'t have subtitles') | |
1106 | return {} | |
1107 | return sub_lang_list | |
1108 | ||
055e6f36 | 1109 | def _get_available_automatic_caption(self, video_id, webpage): |
de7f3446 JMF |
1110 | """We need the webpage for getting the captions url, pass it as an |
1111 | argument to speed up the process.""" | |
ca715127 | 1112 | sub_format = self._downloader.params.get('subtitlesformat', 'srt') |
de7f3446 JMF |
1113 | self.to_screen(u'%s: Looking for automatic captions' % video_id) |
1114 | mobj = re.search(r';ytplayer.config = ({.*?});', webpage) | |
055e6f36 | 1115 | err_msg = u'Couldn\'t find automatic captions for %s' % video_id |
de7f3446 JMF |
1116 | if mobj is None: |
1117 | self._downloader.report_warning(err_msg) | |
1118 | return {} | |
1119 | player_config = json.loads(mobj.group(1)) | |
1120 | try: | |
1121 | args = player_config[u'args'] | |
1122 | caption_url = args[u'ttsurl'] | |
1123 | timestamp = args[u'timestamp'] | |
055e6f36 JMF |
1124 | # We get the available subtitles |
1125 | list_params = compat_urllib_parse.urlencode({ | |
1126 | 'type': 'list', | |
1127 | 'tlangs': 1, | |
1128 | 'asrs': 1, | |
de7f3446 | 1129 | }) |
055e6f36 | 1130 | list_url = caption_url + '&' + list_params |
e26f8712 | 1131 | caption_list = self._download_xml(list_url, video_id) |
e3dc22ca | 1132 | original_lang_node = caption_list.find('track') |
f6a54188 | 1133 | if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : |
e3dc22ca JMF |
1134 | self._downloader.report_warning(u'Video doesn\'t have automatic captions') |
1135 | return {} | |
1136 | original_lang = original_lang_node.attrib['lang_code'] | |
055e6f36 JMF |
1137 | |
1138 | sub_lang_list = {} | |
1139 | for lang_node in caption_list.findall('target'): | |
1140 | sub_lang = lang_node.attrib['lang_code'] | |
1141 | params = compat_urllib_parse.urlencode({ | |
1142 | 'lang': original_lang, | |
1143 | 'tlang': sub_lang, | |
1144 | 'fmt': sub_format, | |
1145 | 'ts': timestamp, | |
1146 | 'kind': 'asr', | |
1147 | }) | |
1148 | sub_lang_list[sub_lang] = caption_url + '&' + params | |
1149 | return sub_lang_list | |
de7f3446 JMF |
1150 | # An extractor error can be raise by the download process if there are |
1151 | # no automatic captions but there are subtitles | |
1152 | except (KeyError, ExtractorError): | |
1153 | self._downloader.report_warning(err_msg) | |
1154 | return {} | |
1155 | ||
c5e8d7af PH |
1156 | def _print_formats(self, formats): |
1157 | print('Available formats:') | |
1158 | for x in formats: | |
03cc7c20 JMF |
1159 | print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), |
1160 | self._video_dimensions.get(x, '???'), | |
836a086c | 1161 | ' ('+self._special_itags[x]+')' if x in self._special_itags else '')) |
c5e8d7af PH |
1162 | |
1163 | def _extract_id(self, url): | |
1164 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
1165 | if mobj is None: | |
1166 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1167 | video_id = mobj.group(2) | |
1168 | return video_id | |
1169 | ||
1d043b93 JMF |
1170 | def _get_video_url_list(self, url_map): |
1171 | """ | |
1172 | Transform a dictionary in the format {itag:url} to a list of (itag, url) | |
1173 | with the requested formats. | |
1174 | """ | |
1175 | req_format = self._downloader.params.get('format', None) | |
1176 | format_limit = self._downloader.params.get('format_limit', None) | |
1177 | available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats | |
1178 | if format_limit is not None and format_limit in available_formats: | |
1179 | format_list = available_formats[available_formats.index(format_limit):] | |
1180 | else: | |
1181 | format_list = available_formats | |
1182 | existing_formats = [x for x in format_list if x in url_map] | |
1183 | if len(existing_formats) == 0: | |
1184 | raise ExtractorError(u'no known formats available for video') | |
1185 | if self._downloader.params.get('listformats', None): | |
1186 | self._print_formats(existing_formats) | |
1187 | return | |
1188 | if req_format is None or req_format == 'best': | |
1189 | video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality | |
1190 | elif req_format == 'worst': | |
1191 | video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality | |
1192 | elif req_format in ('-1', 'all'): | |
1193 | video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats | |
1194 | else: | |
1195 | # Specific formats. We pick the first in a slash-delimeted sequence. | |
bdc6b3fc AZ |
1196 | # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality |
1197 | # available in the specified format. For example, | |
1198 | # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. | |
1199 | # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'. | |
1200 | # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'. | |
1d043b93 JMF |
1201 | req_formats = req_format.split('/') |
1202 | video_url_list = None | |
1203 | for rf in req_formats: | |
1204 | if rf in url_map: | |
1205 | video_url_list = [(rf, url_map[rf])] | |
1206 | break | |
bdc6b3fc AZ |
1207 | if rf in self._video_formats_map: |
1208 | for srf in self._video_formats_map[rf]: | |
1209 | if srf in url_map: | |
1210 | video_url_list = [(srf, url_map[srf])] | |
1211 | break | |
1212 | else: | |
1213 | continue | |
1214 | break | |
1d043b93 JMF |
1215 | if video_url_list is None: |
1216 | raise ExtractorError(u'requested format not available') | |
1217 | return video_url_list | |
1218 | ||
1219 | def _extract_from_m3u8(self, manifest_url, video_id): | |
1220 | url_map = {} | |
1221 | def _get_urls(_manifest): | |
1222 | lines = _manifest.split('\n') | |
1223 | urls = filter(lambda l: l and not l.startswith('#'), | |
1224 | lines) | |
1225 | return urls | |
1226 | manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') | |
1227 | formats_urls = _get_urls(manifest) | |
1228 | for format_url in formats_urls: | |
890f62e8 | 1229 | itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') |
1d043b93 JMF |
1230 | url_map[itag] = format_url |
1231 | return url_map | |
1232 | ||
1fb07d10 JG |
1233 | def _extract_annotations(self, video_id): |
1234 | url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id | |
1235 | return self._download_webpage(url, video_id, note=u'Searching for annotations.', errnote=u'Unable to download video annotations.') | |
1236 | ||
c5e8d7af PH |
1237 | def _real_extract(self, url): |
1238 | # Extract original video URL from URL with redirection, like age verification, using next_url parameter | |
1239 | mobj = re.search(self._NEXT_URL_RE, url) | |
1240 | if mobj: | |
1241 | url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') | |
1242 | video_id = self._extract_id(url) | |
1243 | ||
1244 | # Get video webpage | |
c5e8d7af | 1245 | url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id |
336c3a69 | 1246 | video_webpage = self._download_webpage(url, video_id) |
c5e8d7af PH |
1247 | |
1248 | # Attempt to extract SWF player URL | |
e0df6211 | 1249 | mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) |
c5e8d7af PH |
1250 | if mobj is not None: |
1251 | player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) | |
1252 | else: | |
1253 | player_url = None | |
1254 | ||
1255 | # Get video info | |
1256 | self.report_video_info_webpage_download(video_id) | |
c108eb73 JMF |
1257 | if re.search(r'player-age-gate-content">', video_webpage) is not None: |
1258 | self.report_age_confirmation() | |
1259 | age_gate = True | |
1260 | # We simulate the access to the video from www.youtube.com/v/{video_id} | |
1261 | # this can be viewed without login into Youtube | |
1262 | data = compat_urllib_parse.urlencode({'video_id': video_id, | |
fccd3771 | 1263 | 'el': 'player_embedded', |
c108eb73 JMF |
1264 | 'gl': 'US', |
1265 | 'hl': 'en', | |
1266 | 'eurl': 'https://youtube.googleapis.com/v/' + video_id, | |
1267 | 'asv': 3, | |
1268 | 'sts':'1588', | |
1269 | }) | |
1270 | video_info_url = 'https://www.youtube.com/get_video_info?' + data | |
c5e8d7af PH |
1271 | video_info_webpage = self._download_webpage(video_info_url, video_id, |
1272 | note=False, | |
1273 | errnote='unable to download video info webpage') | |
1274 | video_info = compat_parse_qs(video_info_webpage) | |
c108eb73 JMF |
1275 | else: |
1276 | age_gate = False | |
1277 | for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: | |
1278 | video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' | |
1279 | % (video_id, el_type)) | |
1280 | video_info_webpage = self._download_webpage(video_info_url, video_id, | |
1281 | note=False, | |
1282 | errnote='unable to download video info webpage') | |
1283 | video_info = compat_parse_qs(video_info_webpage) | |
1284 | if 'token' in video_info: | |
1285 | break | |
c5e8d7af PH |
1286 | if 'token' not in video_info: |
1287 | if 'reason' in video_info: | |
9a82b238 | 1288 | raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) |
c5e8d7af PH |
1289 | else: |
1290 | raise ExtractorError(u'"token" parameter not in video info for unknown reason') | |
1291 | ||
1d699755 PH |
1292 | if 'view_count' in video_info: |
1293 | view_count = int(video_info['view_count'][0]) | |
1294 | else: | |
1295 | view_count = None | |
1296 | ||
c5e8d7af PH |
1297 | # Check for "rental" videos |
1298 | if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: | |
1299 | raise ExtractorError(u'"rental" videos not supported') | |
1300 | ||
1301 | # Start extracting information | |
1302 | self.report_information_extraction(video_id) | |
1303 | ||
1304 | # uploader | |
1305 | if 'author' not in video_info: | |
1306 | raise ExtractorError(u'Unable to extract uploader name') | |
1307 | video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) | |
1308 | ||
1309 | # uploader_id | |
1310 | video_uploader_id = None | |
1311 | mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage) | |
1312 | if mobj is not None: | |
1313 | video_uploader_id = mobj.group(1) | |
1314 | else: | |
1315 | self._downloader.report_warning(u'unable to extract uploader nickname') | |
1316 | ||
1317 | # title | |
a8c6b241 PH |
1318 | if 'title' in video_info: |
1319 | video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) | |
1320 | else: | |
1321 | self._downloader.report_warning(u'Unable to extract video title') | |
1322 | video_title = u'_' | |
c5e8d7af PH |
1323 | |
1324 | # thumbnail image | |
7763b04e JMF |
1325 | # We try first to get a high quality image: |
1326 | m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', | |
1327 | video_webpage, re.DOTALL) | |
1328 | if m_thumb is not None: | |
1329 | video_thumbnail = m_thumb.group(1) | |
1330 | elif 'thumbnail_url' not in video_info: | |
c5e8d7af | 1331 | self._downloader.report_warning(u'unable to extract video thumbnail') |
f490e77e | 1332 | video_thumbnail = None |
c5e8d7af PH |
1333 | else: # don't panic if we can't find it |
1334 | video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) | |
1335 | ||
1336 | # upload date | |
1337 | upload_date = None | |
1338 | mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL) | |
1339 | if mobj is not None: | |
1340 | upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) | |
1341 | upload_date = unified_strdate(upload_date) | |
1342 | ||
1343 | # description | |
1344 | video_description = get_element_by_id("eow-description", video_webpage) | |
1345 | if video_description: | |
27dcce19 PH |
1346 | video_description = re.sub(r'''(?x) |
1347 | <a\s+ | |
1348 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1349 | title="([^"]+)"\s+ | |
1350 | (?:[a-zA-Z-]+="[^"]+"\s+)*? | |
1351 | class="yt-uix-redirect-link"\s*> | |
1352 | [^<]+ | |
1353 | </a> | |
1354 | ''', r'\1', video_description) | |
c5e8d7af PH |
1355 | video_description = clean_html(video_description) |
1356 | else: | |
1357 | fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) | |
1358 | if fd_mobj: | |
1359 | video_description = unescapeHTML(fd_mobj.group(1)) | |
1360 | else: | |
1361 | video_description = u'' | |
1362 | ||
336c3a69 | 1363 | def _extract_count(klass): |
46374a56 PH |
1364 | count = self._search_regex( |
1365 | r'class="%s">([\d,]+)</span>' % re.escape(klass), | |
1366 | video_webpage, klass, default=None) | |
336c3a69 JMF |
1367 | if count is not None: |
1368 | return int(count.replace(',', '')) | |
1369 | return None | |
1370 | like_count = _extract_count(u'likes-count') | |
1371 | dislike_count = _extract_count(u'dislikes-count') | |
1372 | ||
c5e8d7af | 1373 | # subtitles |
d82134c3 | 1374 | video_subtitles = self.extract_subtitles(video_id, video_webpage) |
c5e8d7af | 1375 | |
c5e8d7af | 1376 | if self._downloader.params.get('listsubtitles', False): |
d665f8d3 | 1377 | self._list_available_subtitles(video_id, video_webpage) |
c5e8d7af PH |
1378 | return |
1379 | ||
1380 | if 'length_seconds' not in video_info: | |
1381 | self._downloader.report_warning(u'unable to extract video duration') | |
b466b702 | 1382 | video_duration = None |
c5e8d7af | 1383 | else: |
b466b702 | 1384 | video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) |
c5e8d7af | 1385 | |
1fb07d10 JG |
1386 | # annotations |
1387 | video_annotations = None | |
1388 | if self._downloader.params.get('writeannotations', False): | |
1389 | video_annotations = self._extract_annotations(video_id) | |
1390 | ||
c5e8d7af | 1391 | # Decide which formats to download |
c5e8d7af PH |
1392 | |
1393 | try: | |
1394 | mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) | |
50be92c1 PH |
1395 | if not mobj: |
1396 | raise ValueError('Could not find vevo ID') | |
c5e8d7af PH |
1397 | info = json.loads(mobj.group(1)) |
1398 | args = info['args'] | |
7ce7e394 JMF |
1399 | # Easy way to know if the 's' value is in url_encoded_fmt_stream_map |
1400 | # this signatures are encrypted | |
44d46655 | 1401 | if 'url_encoded_fmt_stream_map' not in args: |
f10503db | 1402 | raise ValueError(u'No stream_map present') # caught below |
00fe14fc JMF |
1403 | re_signature = re.compile(r'[&,]s=') |
1404 | m_s = re_signature.search(args['url_encoded_fmt_stream_map']) | |
7ce7e394 JMF |
1405 | if m_s is not None: |
1406 | self.to_screen(u'%s: Encrypted signatures detected.' % video_id) | |
c5e8d7af | 1407 | video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] |
00fe14fc | 1408 | m_s = re_signature.search(args.get('adaptive_fmts', u'')) |
b7a68384 | 1409 | if m_s is not None: |
00fe14fc JMF |
1410 | if 'adaptive_fmts' in video_info: |
1411 | video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] | |
37b6d5f6 | 1412 | else: |
00fe14fc | 1413 | video_info['adaptive_fmts'] = [args['adaptive_fmts']] |
c5e8d7af PH |
1414 | except ValueError: |
1415 | pass | |
1416 | ||
1417 | if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): | |
1418 | self.report_rtmp_download() | |
1419 | video_url_list = [(None, video_info['conn'][0])] | |
00fe14fc JMF |
1420 | elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1: |
1421 | encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts',[''])[0] | |
1422 | if 'rtmpe%3Dyes' in encoded_url_map: | |
a7055eb9 | 1423 | raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) |
c5e8d7af | 1424 | url_map = {} |
00fe14fc | 1425 | for url_data_str in encoded_url_map.split(','): |
c5e8d7af PH |
1426 | url_data = compat_parse_qs(url_data_str) |
1427 | if 'itag' in url_data and 'url' in url_data: | |
1428 | url = url_data['url'][0] | |
1429 | if 'sig' in url_data: | |
1430 | url += '&signature=' + url_data['sig'][0] | |
1431 | elif 's' in url_data: | |
e0df6211 | 1432 | encrypted_sig = url_data['s'][0] |
769fda3c | 1433 | if self._downloader.params.get('verbose'): |
c108eb73 | 1434 | if age_gate: |
bdde940e PH |
1435 | if player_url is None: |
1436 | player_version = 'unknown' | |
1437 | else: | |
1438 | player_version = self._search_regex( | |
1439 | r'-(.+)\.swf$', player_url, | |
1440 | u'flash player', fatal=False) | |
e0df6211 | 1441 | player_desc = 'flash player %s' % player_version |
c108eb73 | 1442 | else: |
83799698 PH |
1443 | player_version = self._search_regex( |
1444 | r'html5player-(.+?)\.js', video_webpage, | |
c108eb73 | 1445 | 'html5 player', fatal=False) |
e0df6211 PH |
1446 | player_desc = u'html5 player %s' % player_version |
1447 | ||
1448 | parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) | |
5a76c651 | 1449 | self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % |
e0df6211 PH |
1450 | (len(encrypted_sig), parts_sizes, url_data['itag'][0], player_desc)) |
1451 | ||
83799698 | 1452 | if not age_gate: |
e0df6211 PH |
1453 | jsplayer_url_json = self._search_regex( |
1454 | r'"assets":.+?"js":\s*("[^"]+")', | |
1455 | video_webpage, u'JS player URL') | |
83799698 | 1456 | player_url = json.loads(jsplayer_url_json) |
e0df6211 | 1457 | |
83799698 PH |
1458 | signature = self._decrypt_signature( |
1459 | encrypted_sig, video_id, player_url, age_gate) | |
c5e8d7af PH |
1460 | url += '&signature=' + signature |
1461 | if 'ratebypass' not in url: | |
1462 | url += '&ratebypass=yes' | |
1463 | url_map[url_data['itag'][0]] = url | |
1d043b93 JMF |
1464 | video_url_list = self._get_video_url_list(url_map) |
1465 | if not video_url_list: | |
c5e8d7af | 1466 | return |
1d043b93 JMF |
1467 | elif video_info.get('hlsvp'): |
1468 | manifest_url = video_info['hlsvp'][0] | |
1469 | url_map = self._extract_from_m3u8(manifest_url, video_id) | |
1470 | video_url_list = self._get_video_url_list(url_map) | |
1471 | if not video_url_list: | |
1472 | return | |
1473 | ||
c5e8d7af | 1474 | else: |
9abb3204 | 1475 | raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') |
c5e8d7af PH |
1476 | |
1477 | results = [] | |
600cc1a4 | 1478 | for itag, video_real_url in video_url_list: |
c5e8d7af | 1479 | # Extension |
600cc1a4 | 1480 | video_extension = self._video_extensions.get(itag, 'flv') |
c5e8d7af | 1481 | |
600cc1a4 JMF |
1482 | video_format = '{0} - {1}{2}'.format(itag if itag else video_extension, |
1483 | self._video_dimensions.get(itag, '???'), | |
1484 | ' ('+self._special_itags[itag]+')' if itag in self._special_itags else '') | |
c5e8d7af PH |
1485 | |
1486 | results.append({ | |
1487 | 'id': video_id, | |
1488 | 'url': video_real_url, | |
1489 | 'uploader': video_uploader, | |
1490 | 'uploader_id': video_uploader_id, | |
1491 | 'upload_date': upload_date, | |
1492 | 'title': video_title, | |
1493 | 'ext': video_extension, | |
1494 | 'format': video_format, | |
600cc1a4 | 1495 | 'format_id': itag, |
c5e8d7af PH |
1496 | 'thumbnail': video_thumbnail, |
1497 | 'description': video_description, | |
1498 | 'player_url': player_url, | |
1499 | 'subtitles': video_subtitles, | |
8dbe9899 | 1500 | 'duration': video_duration, |
cfadd183 | 1501 | 'age_limit': 18 if age_gate else 0, |
9103bbc5 JMF |
1502 | 'annotations': video_annotations, |
1503 | 'webpage_url': 'https://www.youtube.com/watch?v=%s' % video_id, | |
1d699755 | 1504 | 'view_count': view_count, |
336c3a69 JMF |
1505 | 'like_count': like_count, |
1506 | 'dislike_count': dislike_count, | |
c5e8d7af PH |
1507 | }) |
1508 | return results | |
1509 | ||
880e1c52 | 1510 | class YoutubePlaylistIE(YoutubeBaseInfoExtractor): |
0f818663 | 1511 | IE_DESC = u'YouTube.com playlists' |
c5e8d7af PH |
1512 | _VALID_URL = r"""(?: |
1513 | (?:https?://)? | |
1514 | (?:\w+\.)? | |
1515 | youtube\.com/ | |
1516 | (?: | |
1517 | (?:course|view_play_list|my_playlists|artist|playlist|watch) | |
1518 | \? (?:.*?&)*? (?:p|a|list)= | |
1519 | | p/ | |
1520 | ) | |
715c8e7b | 1521 | ((?:PL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}) |
c5e8d7af PH |
1522 | .* |
1523 | | | |
715c8e7b | 1524 | ((?:PL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,}) |
c5e8d7af | 1525 | )""" |
dcbb4580 JMF |
1526 | _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&page=%s' |
1527 | _MORE_PAGES_INDICATOR = r'data-link-type="next"' | |
6e47b51e | 1528 | _VIDEO_RE = r'href="/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)' |
c5e8d7af PH |
1529 | IE_NAME = u'youtube:playlist' |
1530 | ||
1531 | @classmethod | |
1532 | def suitable(cls, url): | |
1533 | """Receives a URL and returns True if suitable for this IE.""" | |
1534 | return re.match(cls._VALID_URL, url, re.VERBOSE) is not None | |
1535 | ||
880e1c52 JMF |
1536 | def _real_initialize(self): |
1537 | self._login() | |
1538 | ||
652cdaa2 JMF |
1539 | def _ids_to_results(self, ids): |
1540 | return [self.url_result(vid_id, 'Youtube', video_id=vid_id) | |
1541 | for vid_id in ids] | |
1542 | ||
1543 | def _extract_mix(self, playlist_id): | |
1544 | # The mixes are generated from a a single video | |
1545 | # the id of the playlist is just 'RD' + video_id | |
7d4afc55 | 1546 | url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) |
652cdaa2 | 1547 | webpage = self._download_webpage(url, playlist_id, u'Downloading Youtube mix') |
76d1700b JMF |
1548 | title_span = (get_element_by_attribute('class', 'title long-title', webpage) or |
1549 | get_element_by_attribute('class', 'title ', webpage)) | |
1550 | title = clean_html(title_span) | |
652cdaa2 JMF |
1551 | video_re = r'data-index="\d+".*?href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s' % re.escape(playlist_id) |
1552 | ids = orderedSet(re.findall(video_re, webpage)) | |
1553 | url_results = self._ids_to_results(ids) | |
1554 | ||
1555 | return self.playlist_result(url_results, playlist_id, title) | |
1556 | ||
c5e8d7af PH |
1557 | def _real_extract(self, url): |
1558 | # Extract playlist id | |
1559 | mobj = re.match(self._VALID_URL, url, re.VERBOSE) | |
1560 | if mobj is None: | |
1561 | raise ExtractorError(u'Invalid URL: %s' % url) | |
47192f92 FV |
1562 | playlist_id = mobj.group(1) or mobj.group(2) |
1563 | ||
1564 | # Check if it's a video-specific URL | |
7c61bd36 | 1565 | query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) |
47192f92 FV |
1566 | if 'v' in query_dict: |
1567 | video_id = query_dict['v'][0] | |
1568 | if self._downloader.params.get('noplaylist'): | |
1569 | self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) | |
7012b23c | 1570 | return self.url_result(video_id, 'Youtube', video_id=video_id) |
47192f92 FV |
1571 | else: |
1572 | self.to_screen(u'Downloading playlist PL%s - add --no-playlist to just download video %s' % (playlist_id, video_id)) | |
c5e8d7af | 1573 | |
7d4afc55 | 1574 | if playlist_id.startswith('RD'): |
652cdaa2 JMF |
1575 | # Mixes require a custom extraction process |
1576 | return self._extract_mix(playlist_id) | |
0a688bc0 JMF |
1577 | if playlist_id.startswith('TL'): |
1578 | raise ExtractorError(u'For downloading YouTube.com top lists, use ' | |
1579 | u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) | |
652cdaa2 | 1580 | |
dcbb4580 JMF |
1581 | # Extract the video ids from the playlist pages |
1582 | ids = [] | |
c5e8d7af | 1583 | |
755eb032 | 1584 | for page_num in itertools.count(1): |
dcbb4580 | 1585 | url = self._TEMPLATE_URL % (playlist_id, page_num) |
c5e8d7af | 1586 | page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) |
6e47b51e JMF |
1587 | matches = re.finditer(self._VIDEO_RE, page) |
1588 | # We remove the duplicates and the link with index 0 | |
1589 | # (it's not the first video of the playlist) | |
1590 | new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') | |
dcbb4580 | 1591 | ids.extend(new_ids) |
c5e8d7af | 1592 | |
dcbb4580 | 1593 | if re.search(self._MORE_PAGES_INDICATOR, page) is None: |
c5e8d7af PH |
1594 | break |
1595 | ||
dcbb4580 | 1596 | playlist_title = self._og_search_title(page) |
c5e8d7af | 1597 | |
652cdaa2 | 1598 | url_results = self._ids_to_results(ids) |
dcbb4580 | 1599 | return self.playlist_result(url_results, playlist_id, playlist_title) |
c5e8d7af PH |
1600 | |
1601 | ||
0a688bc0 JMF |
1602 | class YoutubeTopListIE(YoutubePlaylistIE): |
1603 | IE_NAME = u'youtube:toplist' | |
1604 | IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' | |
1605 | u' (Example: "yttoplist:music:Top Tracks")') | |
1606 | _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$' | |
1607 | ||
1608 | def _real_extract(self, url): | |
1609 | mobj = re.match(self._VALID_URL, url) | |
1610 | channel = mobj.group('chann') | |
1611 | title = mobj.group('title') | |
1612 | query = compat_urllib_parse.urlencode({'title': title}) | |
1613 | playlist_re = 'href="([^"]+?%s[^"]+?)"' % re.escape(query) | |
1614 | channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) | |
1615 | link = self._html_search_regex(playlist_re, channel_page, u'list') | |
1616 | url = compat_urlparse.urljoin('https://www.youtube.com/', link) | |
1617 | ||
1618 | video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' | |
1619 | ids = [] | |
1620 | # sometimes the webpage doesn't contain the videos | |
1621 | # retry until we get them | |
1622 | for i in itertools.count(0): | |
1623 | msg = u'Downloading Youtube mix' | |
1624 | if i > 0: | |
1625 | msg += ', retry #%d' % i | |
1626 | webpage = self._download_webpage(url, title, msg) | |
1627 | ids = orderedSet(re.findall(video_re, webpage)) | |
1628 | if ids: | |
1629 | break | |
1630 | url_results = self._ids_to_results(ids) | |
1631 | return self.playlist_result(url_results, playlist_title=title) | |
1632 | ||
1633 | ||
c5e8d7af | 1634 | class YoutubeChannelIE(InfoExtractor): |
0f818663 | 1635 | IE_DESC = u'YouTube.com channels' |
c5e8d7af | 1636 | _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" |
c5e8d7af | 1637 | _MORE_PAGES_INDICATOR = 'yt-uix-load-more' |
252580c5 | 1638 | _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' |
c5e8d7af PH |
1639 | IE_NAME = u'youtube:channel' |
1640 | ||
1641 | def extract_videos_from_page(self, page): | |
1642 | ids_in_page = [] | |
1643 | for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): | |
1644 | if mobj.group(1) not in ids_in_page: | |
1645 | ids_in_page.append(mobj.group(1)) | |
1646 | return ids_in_page | |
1647 | ||
1648 | def _real_extract(self, url): | |
1649 | # Extract channel id | |
1650 | mobj = re.match(self._VALID_URL, url) | |
1651 | if mobj is None: | |
1652 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1653 | ||
1654 | # Download channel page | |
1655 | channel_id = mobj.group(1) | |
1656 | video_ids = [] | |
b9643eed JMF |
1657 | url = 'https://www.youtube.com/channel/%s/videos' % channel_id |
1658 | channel_page = self._download_webpage(url, channel_id) | |
31812a9e PH |
1659 | autogenerated = re.search(r'''(?x) |
1660 | class="[^"]*?(?: | |
1661 | channel-header-autogenerated-label| | |
1662 | yt-channel-title-autogenerated | |
1663 | )[^"]*"''', channel_page) is not None | |
c5e8d7af | 1664 | |
b9643eed JMF |
1665 | if autogenerated: |
1666 | # The videos are contained in a single page | |
1667 | # the ajax pages can't be used, they are empty | |
1668 | video_ids = self.extract_videos_from_page(channel_page) | |
1669 | else: | |
1670 | # Download all channel pages using the json-based channel_ajax query | |
1671 | for pagenum in itertools.count(1): | |
1672 | url = self._MORE_PAGES_URL % (pagenum, channel_id) | |
1673 | page = self._download_webpage(url, channel_id, | |
1674 | u'Downloading page #%s' % pagenum) | |
1675 | ||
1676 | page = json.loads(page) | |
1677 | ||
1678 | ids_in_page = self.extract_videos_from_page(page['content_html']) | |
1679 | video_ids.extend(ids_in_page) | |
1680 | ||
1681 | if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: | |
1682 | break | |
c5e8d7af PH |
1683 | |
1684 | self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) | |
1685 | ||
7012b23c PH |
1686 | url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1687 | for video_id in video_ids] | |
1688 | return self.playlist_result(url_entries, channel_id) | |
c5e8d7af PH |
1689 | |
1690 | ||
1691 | class YoutubeUserIE(InfoExtractor): | |
0f818663 | 1692 | IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' |
57da92b7 | 1693 | _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)' |
c5e8d7af PH |
1694 | _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' |
1695 | _GDATA_PAGE_SIZE = 50 | |
fd9cf738 | 1696 | _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' |
c5e8d7af PH |
1697 | IE_NAME = u'youtube:user' |
1698 | ||
e3ea4790 | 1699 | @classmethod |
f4b05232 | 1700 | def suitable(cls, url): |
e3ea4790 JMF |
1701 | # Don't return True if the url can be extracted with other youtube |
1702 | # extractor, the regex would is too permissive and it would match. | |
1703 | other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) | |
1704 | if any(ie.suitable(url) for ie in other_ies): return False | |
f4b05232 JMF |
1705 | else: return super(YoutubeUserIE, cls).suitable(url) |
1706 | ||
c5e8d7af PH |
1707 | def _real_extract(self, url): |
1708 | # Extract username | |
1709 | mobj = re.match(self._VALID_URL, url) | |
1710 | if mobj is None: | |
1711 | raise ExtractorError(u'Invalid URL: %s' % url) | |
1712 | ||
1713 | username = mobj.group(1) | |
1714 | ||
1715 | # Download video ids using YouTube Data API. Result size per | |
1716 | # query is limited (currently to 50 videos) so we need to query | |
1717 | # page by page until there are no video ids - it means we got | |
1718 | # all of them. | |
1719 | ||
1720 | video_ids = [] | |
c5e8d7af | 1721 | |
755eb032 | 1722 | for pagenum in itertools.count(0): |
c5e8d7af PH |
1723 | start_index = pagenum * self._GDATA_PAGE_SIZE + 1 |
1724 | ||
1725 | gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) | |
1726 | page = self._download_webpage(gdata_url, username, | |
1727 | u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) | |
1728 | ||
fd9cf738 JMF |
1729 | try: |
1730 | response = json.loads(page) | |
1731 | except ValueError as err: | |
1732 | raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) | |
71c82637 JMF |
1733 | if 'entry' not in response['feed']: |
1734 | # Number of videos is a multiple of self._MAX_RESULTS | |
1735 | break | |
fd9cf738 | 1736 | |
c5e8d7af PH |
1737 | # Extract video identifiers |
1738 | ids_in_page = [] | |
fd9cf738 JMF |
1739 | for entry in response['feed']['entry']: |
1740 | ids_in_page.append(entry['id']['$t'].split('/')[-1]) | |
c5e8d7af PH |
1741 | video_ids.extend(ids_in_page) |
1742 | ||
1743 | # A little optimization - if current page is not | |
1744 | # "full", ie. does not contain PAGE_SIZE video ids then | |
1745 | # we can assume that this page is the last one - there | |
1746 | # are no more ids on further pages - no need to query | |
1747 | # again. | |
1748 | ||
1749 | if len(ids_in_page) < self._GDATA_PAGE_SIZE: | |
1750 | break | |
1751 | ||
7012b23c PH |
1752 | url_results = [ |
1753 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1754 | for video_id in video_ids] | |
1755 | return self.playlist_result(url_results, playlist_title=username) | |
1756 | ||
b05654f0 PH |
1757 | |
1758 | class YoutubeSearchIE(SearchInfoExtractor): | |
0f818663 | 1759 | IE_DESC = u'YouTube.com searches' |
b05654f0 PH |
1760 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' |
1761 | _MAX_RESULTS = 1000 | |
1762 | IE_NAME = u'youtube:search' | |
1763 | _SEARCH_KEY = 'ytsearch' | |
1764 | ||
b05654f0 PH |
1765 | def _get_n_results(self, query, n): |
1766 | """Get a specified number of results for a query""" | |
1767 | ||
1768 | video_ids = [] | |
1769 | pagenum = 0 | |
1770 | limit = n | |
1771 | ||
1772 | while (50 * pagenum) < limit: | |
b05654f0 | 1773 | result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) |
7cc3570e PH |
1774 | data_json = self._download_webpage( |
1775 | result_url, video_id=u'query "%s"' % query, | |
1776 | note=u'Downloading page %s' % (pagenum + 1), | |
1777 | errnote=u'Unable to download API page') | |
1778 | data = json.loads(data_json) | |
1779 | api_response = data['data'] | |
1780 | ||
1781 | if 'items' not in api_response: | |
b05654f0 PH |
1782 | raise ExtractorError(u'[youtube] No video results') |
1783 | ||
1784 | new_ids = list(video['id'] for video in api_response['items']) | |
1785 | video_ids += new_ids | |
1786 | ||
1787 | limit = min(n, api_response['totalItems']) | |
1788 | pagenum += 1 | |
1789 | ||
1790 | if len(video_ids) > n: | |
1791 | video_ids = video_ids[:n] | |
7012b23c PH |
1792 | videos = [self.url_result(video_id, 'Youtube', video_id=video_id) |
1793 | for video_id in video_ids] | |
b05654f0 | 1794 | return self.playlist_result(videos, query) |
75dff0ee | 1795 | |
a3dd9248 | 1796 | class YoutubeSearchDateIE(YoutubeSearchIE): |
cb7fb546 | 1797 | IE_NAME = YoutubeSearchIE.IE_NAME + ':date' |
a3dd9248 CM |
1798 | _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' |
1799 | _SEARCH_KEY = 'ytsearchdate' | |
08fb86c4 | 1800 | IE_DESC = u'YouTube.com searches, newest videos first' |
75dff0ee JMF |
1801 | |
1802 | class YoutubeShowIE(InfoExtractor): | |
0f818663 | 1803 | IE_DESC = u'YouTube.com (multi-season) shows' |
75dff0ee JMF |
1804 | _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' |
1805 | IE_NAME = u'youtube:show' | |
1806 | ||
1807 | def _real_extract(self, url): | |
1808 | mobj = re.match(self._VALID_URL, url) | |
1809 | show_name = mobj.group(1) | |
1810 | webpage = self._download_webpage(url, show_name, u'Downloading show webpage') | |
1811 | # There's one playlist for each season of the show | |
1812 | m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) | |
1813 | self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) | |
1814 | return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] | |
04cc9617 JMF |
1815 | |
1816 | ||
b2e8bc1b | 1817 | class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): |
d7ae0639 JMF |
1818 | """ |
1819 | Base class for extractors that fetch info from | |
1820 | http://www.youtube.com/feed_ajax | |
1821 | Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. | |
1822 | """ | |
b2e8bc1b | 1823 | _LOGIN_REQUIRED = True |
43ba5456 JMF |
1824 | # use action_load_personal_feed instead of action_load_system_feed |
1825 | _PERSONAL_FEED = False | |
04cc9617 | 1826 | |
d7ae0639 JMF |
1827 | @property |
1828 | def _FEED_TEMPLATE(self): | |
43ba5456 JMF |
1829 | action = 'action_load_system_feed' |
1830 | if self._PERSONAL_FEED: | |
1831 | action = 'action_load_personal_feed' | |
1832 | return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) | |
d7ae0639 JMF |
1833 | |
1834 | @property | |
1835 | def IE_NAME(self): | |
1836 | return u'youtube:%s' % self._FEED_NAME | |
04cc9617 | 1837 | |
81f0259b | 1838 | def _real_initialize(self): |
b2e8bc1b | 1839 | self._login() |
81f0259b | 1840 | |
04cc9617 JMF |
1841 | def _real_extract(self, url): |
1842 | feed_entries = [] | |
0e44d838 JMF |
1843 | paging = 0 |
1844 | for i in itertools.count(1): | |
d7ae0639 JMF |
1845 | info = self._download_webpage(self._FEED_TEMPLATE % paging, |
1846 | u'%s feed' % self._FEED_NAME, | |
04cc9617 JMF |
1847 | u'Downloading page %s' % i) |
1848 | info = json.loads(info) | |
1849 | feed_html = info['feed_html'] | |
43ba5456 | 1850 | m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) |
04cc9617 | 1851 | ids = orderedSet(m.group(1) for m in m_ids) |
7012b23c PH |
1852 | feed_entries.extend( |
1853 | self.url_result(video_id, 'Youtube', video_id=video_id) | |
1854 | for video_id in ids) | |
04cc9617 JMF |
1855 | if info['paging'] is None: |
1856 | break | |
0e44d838 | 1857 | paging = info['paging'] |
d7ae0639 JMF |
1858 | return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) |
1859 | ||
1860 | class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): | |
1861 | IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' | |
1862 | _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' | |
1863 | _FEED_NAME = 'subscriptions' | |
1864 | _PLAYLIST_TITLE = u'Youtube Subscriptions' | |
1865 | ||
1866 | class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): | |
1867 | IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' | |
1868 | _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' | |
1869 | _FEED_NAME = 'recommended' | |
1870 | _PLAYLIST_TITLE = u'Youtube Recommended videos' | |
c626a3d9 | 1871 | |
43ba5456 JMF |
1872 | class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): |
1873 | IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' | |
1874 | _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' | |
1875 | _FEED_NAME = 'watch_later' | |
1876 | _PLAYLIST_TITLE = u'Youtube Watch Later' | |
43ba5456 | 1877 | _PERSONAL_FEED = True |
c626a3d9 | 1878 | |
f459d170 JMF |
1879 | class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): |
1880 | IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' | |
1881 | _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' | |
1882 | _FEED_NAME = 'history' | |
1883 | _PERSONAL_FEED = True | |
1884 | _PLAYLIST_TITLE = u'Youtube Watch History' | |
1885 | ||
c626a3d9 JMF |
1886 | class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): |
1887 | IE_NAME = u'youtube:favorites' | |
1888 | IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' | |
c7a7750d | 1889 | _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?' |
c626a3d9 JMF |
1890 | _LOGIN_REQUIRED = True |
1891 | ||
1892 | def _real_extract(self, url): | |
1893 | webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') | |
1894 | playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') | |
1895 | return self.url_result(playlist_id, 'YoutubePlaylist') | |
15870e90 PH |
1896 | |
1897 | ||
1898 | class YoutubeTruncatedURLIE(InfoExtractor): | |
1899 | IE_NAME = 'youtube:truncated_url' | |
1900 | IE_DESC = False # Do not list | |
1901 | _VALID_URL = r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$' | |
1902 | ||
1903 | def _real_extract(self, url): | |
1904 | raise ExtractorError( | |
1905 | u'Did you forget to quote the URL? Remember that & is a meta ' | |
1906 | u'character in most shells, so you want to put the URL in quotes, ' | |
1907 | u'like youtube-dl ' | |
1908 | u'\'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\'' | |
1909 | u' (or simply youtube-dl BaW_jenozKc ).', | |
1910 | expected=True) |