]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
projects / yt-dlp.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
[youtube] add algo for length 80 and update player info
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 import json
4 import netrc
5 import re
6 import socket
7 import itertools
8
9 from .common import InfoExtractor, SearchInfoExtractor
10 from ..utils import (
11 compat_http_client,
12 compat_parse_qs,
13 compat_urllib_error,
14 compat_urllib_parse,
15 compat_urllib_request,
16 compat_str,
17
18 clean_html,
19 get_element_by_id,
20 ExtractorError,
21 unescapeHTML,
22 unified_strdate,
23 orderedSet,
24 )
25
26 class YoutubeBaseInfoExtractor(InfoExtractor):
27 """Provide base functions for Youtube extractors"""
28 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
29 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
30 _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
31 _NETRC_MACHINE = 'youtube'
32 # If True it will raise an error if no login info is provided
33 _LOGIN_REQUIRED = False
34
35 def report_lang(self):
36 """Report attempt to set language."""
37 self.to_screen(u'Setting language')
38
39 def _set_language(self):
40 request = compat_urllib_request.Request(self._LANG_URL)
41 try:
42 self.report_lang()
43 compat_urllib_request.urlopen(request).read()
44 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
45 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
46 return False
47 return True
48
49 def _login(self):
50 (username, password) = self._get_login_info()
51 # No authentication to be performed
52 if username is None:
53 if self._LOGIN_REQUIRED:
54 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
55 return False
56
57 request = compat_urllib_request.Request(self._LOGIN_URL)
58 try:
59 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
60 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
61 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
62 return False
63
64 galx = None
65 dsh = None
66 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
67 if match:
68 galx = match.group(1)
69 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
70 if match:
71 dsh = match.group(1)
72
73 # Log in
74 login_form_strs = {
75 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
76 u'Email': username,
77 u'GALX': galx,
78 u'Passwd': password,
79 u'PersistentCookie': u'yes',
80 u'_utf8': u'霱',
81 u'bgresponse': u'js_disabled',
82 u'checkConnection': u'',
83 u'checkedDomains': u'youtube',
84 u'dnConn': u'',
85 u'dsh': dsh,
86 u'pstMsg': u'0',
87 u'rmShown': u'1',
88 u'secTok': u'',
89 u'signIn': u'Sign in',
90 u'timeStmp': u'',
91 u'service': u'youtube',
92 u'uilel': u'3',
93 u'hl': u'en_US',
94 }
95 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
96 # chokes on unicode
97 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
98 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
99 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
100 try:
101 self.report_login()
102 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
103 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
104 self._downloader.report_warning(u'unable to log in: bad username or password')
105 return False
106 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
107 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
108 return False
109 return True
110
111 def _confirm_age(self):
112 age_form = {
113 'next_url': '/',
114 'action_confirm': 'Confirm',
115 }
116 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
117 try:
118 self.report_age_confirmation()
119 compat_urllib_request.urlopen(request).read().decode('utf-8')
120 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
121 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
122 return True
123
124 def _real_initialize(self):
125 if self._downloader is None:
126 return
127 if not self._set_language():
128 return
129 if not self._login():
130 return
131 self._confirm_age()
132
133 class YoutubeIE(YoutubeBaseInfoExtractor):
134 IE_DESC = u'YouTube.com'
135 _VALID_URL = r"""^
136 (
137 (?:https?://)? # http(s):// (optional)
138 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
139 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
140 (?:.*?\#/)? # handle anchor (#/) redirect urls
141 (?: # the various things that can precede the ID:
142 (?:(?:v|embed|e)/) # v/ or embed/ or e/
143 |(?: # or the v= param in all its forms
144 (?:(?:watch|movie)(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
145 (?:\?|\#!?) # the params delimiter ? or # or #!
146 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
147 v=
148 )
149 )? # optional -> youtube.com/xxxx is OK
150 )? # all until now is optional -> you can pass the naked ID
151 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
152 (?(1).+)? # if we found the ID, everything can follow
153 $"""
154 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
155 # Listed in order of quality
156 _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13',
157 '95', '94', '93', '92', '132', '151',
158 # 3D
159 '85', '84', '102', '83', '101', '82', '100',
160 # Dash video
161 '138', '137', '248', '136', '247', '135', '246',
162 '245', '244', '134', '243', '133', '242', '160',
163 # Dash audio
164 '141', '172', '140', '171', '139',
165 ]
166 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
167 '95', '94', '93', '92', '132', '151',
168 '85', '102', '84', '101', '83', '100', '82',
169 # Dash video
170 '138', '248', '137', '247', '136', '246', '245',
171 '244', '135', '243', '134', '242', '133', '160',
172 # Dash audio
173 '172', '141', '171', '140', '139',
174 ]
175 _video_extensions = {
176 '13': '3gp',
177 '17': 'mp4',
178 '18': 'mp4',
179 '22': 'mp4',
180 '37': 'mp4',
181 '38': 'mp4',
182 '43': 'webm',
183 '44': 'webm',
184 '45': 'webm',
185 '46': 'webm',
186
187 # 3d videos
188 '82': 'mp4',
189 '83': 'mp4',
190 '84': 'mp4',
191 '85': 'mp4',
192 '100': 'webm',
193 '101': 'webm',
194 '102': 'webm',
195
196 # videos that use m3u8
197 '92': 'mp4',
198 '93': 'mp4',
199 '94': 'mp4',
200 '95': 'mp4',
201 '96': 'mp4',
202 '132': 'mp4',
203 '151': 'mp4',
204
205 # Dash mp4
206 '133': 'mp4',
207 '134': 'mp4',
208 '135': 'mp4',
209 '136': 'mp4',
210 '137': 'mp4',
211 '138': 'mp4',
212 '139': 'mp4',
213 '140': 'mp4',
214 '141': 'mp4',
215 '160': 'mp4',
216
217 # Dash webm
218 '171': 'webm',
219 '172': 'webm',
220 '242': 'webm',
221 '243': 'webm',
222 '244': 'webm',
223 '245': 'webm',
224 '246': 'webm',
225 '247': 'webm',
226 '248': 'webm',
227 }
228 _video_dimensions = {
229 '5': '240x400',
230 '6': '???',
231 '13': '???',
232 '17': '144x176',
233 '18': '360x640',
234 '22': '720x1280',
235 '34': '360x640',
236 '35': '480x854',
237 '37': '1080x1920',
238 '38': '3072x4096',
239 '43': '360x640',
240 '44': '480x854',
241 '45': '720x1280',
242 '46': '1080x1920',
243 '82': '360p',
244 '83': '480p',
245 '84': '720p',
246 '85': '1080p',
247 '92': '240p',
248 '93': '360p',
249 '94': '480p',
250 '95': '720p',
251 '96': '1080p',
252 '100': '360p',
253 '101': '480p',
254 '102': '720p',
255 '132': '240p',
256 '151': '72p',
257 '133': '240p',
258 '134': '360p',
259 '135': '480p',
260 '136': '720p',
261 '137': '1080p',
262 '138': '>1080p',
263 '139': '48k',
264 '140': '128k',
265 '141': '256k',
266 '160': '192p',
267 '171': '128k',
268 '172': '256k',
269 '242': '240p',
270 '243': '360p',
271 '244': '480p',
272 '245': '480p',
273 '246': '480p',
274 '247': '720p',
275 '248': '1080p',
276 }
277 _special_itags = {
278 '82': '3D',
279 '83': '3D',
280 '84': '3D',
281 '85': '3D',
282 '100': '3D',
283 '101': '3D',
284 '102': '3D',
285 '133': 'DASH Video',
286 '134': 'DASH Video',
287 '135': 'DASH Video',
288 '136': 'DASH Video',
289 '137': 'DASH Video',
290 '138': 'DASH Video',
291 '139': 'DASH Audio',
292 '140': 'DASH Audio',
293 '141': 'DASH Audio',
294 '160': 'DASH Video',
295 '171': 'DASH Audio',
296 '172': 'DASH Audio',
297 '242': 'DASH Video',
298 '243': 'DASH Video',
299 '244': 'DASH Video',
300 '245': 'DASH Video',
301 '246': 'DASH Video',
302 '247': 'DASH Video',
303 '248': 'DASH Video',
304 }
305
306 IE_NAME = u'youtube'
307 _TESTS = [
308 {
309 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
310 u"file": u"BaW_jenozKc.mp4",
311 u"info_dict": {
312 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
313 u"uploader": u"Philipp Hagemeister",
314 u"uploader_id": u"phihag",
315 u"upload_date": u"20121002",
316 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
317 }
318 },
319 {
320 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
321 u"file": u"1ltcDfZMA3U.flv",
322 u"note": u"Test VEVO video (#897)",
323 u"info_dict": {
324 u"upload_date": u"20070518",
325 u"title": u"Maps - It Will Find You",
326 u"description": u"Music video by Maps performing It Will Find You.",
327 u"uploader": u"MuteUSA",
328 u"uploader_id": u"MuteUSA"
329 }
330 },
331 {
332 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
333 u"file": u"UxxajLWwzqY.mp4",
334 u"note": u"Test generic use_cipher_signature video (#897)",
335 u"info_dict": {
336 u"upload_date": u"20120506",
337 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
338 u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
339 u"uploader": u"Icona Pop",
340 u"uploader_id": u"IconaPop"
341 }
342 },
343 {
344 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
345 u"file": u"07FYdnEawAQ.mp4",
346 u"note": u"Test VEVO video with age protection (#956)",
347 u"info_dict": {
348 u"upload_date": u"20130703",
349 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
350 u"description": u"md5:64249768eec3bc4276236606ea996373",
351 u"uploader": u"justintimberlakeVEVO",
352 u"uploader_id": u"justintimberlakeVEVO"
353 }
354 },
355 {
356 u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE',
357 u'file': u'TGi3HqYrWHE.mp4',
358 u'note': u'm3u8 video',
359 u'info_dict': {
360 u'title': u'Triathlon - Men - London 2012 Olympic Games',
361 u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
362 u'uploader': u'olympic',
363 u'upload_date': u'20120807',
364 u'uploader_id': u'olympic',
365 },
366 u'params': {
367 u'skip_download': True,
368 },
369 },
370 ]
371
372
373 @classmethod
374 def suitable(cls, url):
375 """Receives a URL and returns True if suitable for this IE."""
376 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
377 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
378
379 def report_video_webpage_download(self, video_id):
380 """Report attempt to download video webpage."""
381 self.to_screen(u'%s: Downloading video webpage' % video_id)
382
383 def report_video_info_webpage_download(self, video_id):
384 """Report attempt to download video info webpage."""
385 self.to_screen(u'%s: Downloading video info webpage' % video_id)
386
387 def report_video_subtitles_download(self, video_id):
388 """Report attempt to download video info webpage."""
389 self.to_screen(u'%s: Checking available subtitles' % video_id)
390
391 def report_video_subtitles_request(self, video_id, sub_lang, format):
392 """Report attempt to download video info webpage."""
393 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
394
395 def report_video_subtitles_available(self, video_id, sub_lang_list):
396 """Report available subtitles."""
397 sub_lang = ",".join(list(sub_lang_list.keys()))
398 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
399
400 def report_information_extraction(self, video_id):
401 """Report attempt to extract video information."""
402 self.to_screen(u'%s: Extracting video information' % video_id)
403
404 def report_unavailable_format(self, video_id, format):
405 """Report extracted video URL."""
406 self.to_screen(u'%s: Format %s not available' % (video_id, format))
407
408 def report_rtmp_download(self):
409 """Indicate the download will use the RTMP protocol."""
410 self.to_screen(u'RTMP download detected')
411
412 def _decrypt_signature(self, s):
413 """Turn the encrypted s field into a working signature"""
414
415 if len(s) == 92:
416 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
417 elif len(s) == 90:
418 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
419 elif len(s) == 89:
420 return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1]
421 elif len(s) == 88:
422 return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12]
423 elif len(s) == 87:
424 return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
425 elif len(s) == 86:
426 return s[5:20] + s[2] + s[21:]
427 elif len(s) == 85:
428 return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27]
429 elif len(s) == 84:
430 return s[83:27:-1] + s[0] + s[26:5:-1] + s[2:0:-1] + s[27]
431 elif len(s) == 83:
432 return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0]
433 elif len(s) == 82:
434 return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82]
435 elif len(s) == 81:
436 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]
437 elif len(s) == 80:
438 return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80]
439 elif len(s) == 79:
440 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]
441
442 else:
443 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
444
445 def _decrypt_signature_age_gate(self, s):
446 # The videos with age protection use another player, so the algorithms
447 # can be different.
448 if len(s) == 86:
449 return s[2:63] + s[82] + s[64:82] + s[63]
450 else:
451 # Fallback to the other algortihms
452 return self._decrypt_signature(s)
453
454
455 def _get_available_subtitles(self, video_id):
456 self.report_video_subtitles_download(video_id)
457 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
458 try:
459 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
460 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
461 return (u'unable to download video subtitles: %s' % compat_str(err), None)
462 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
463 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
464 if not sub_lang_list:
465 return (u'video doesn\'t have subtitles', None)
466 return sub_lang_list
467
468 def _list_available_subtitles(self, video_id):
469 sub_lang_list = self._get_available_subtitles(video_id)
470 self.report_video_subtitles_available(video_id, sub_lang_list)
471
472 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
473 """
474 Return tuple:
475 (error_message, sub_lang, sub)
476 """
477 self.report_video_subtitles_request(video_id, sub_lang, format)
478 params = compat_urllib_parse.urlencode({
479 'lang': sub_lang,
480 'name': sub_name,
481 'v': video_id,
482 'fmt': format,
483 })
484 url = 'http://www.youtube.com/api/timedtext?' + params
485 try:
486 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
487 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
488 return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
489 if not sub:
490 return (u'Did not fetch video subtitles', None, None)
491 return (None, sub_lang, sub)
492
493 def _request_automatic_caption(self, video_id, webpage):
494 """We need the webpage for getting the captions url, pass it as an
495 argument to speed up the process."""
496 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
497 sub_format = self._downloader.params.get('subtitlesformat')
498 self.to_screen(u'%s: Looking for automatic captions' % video_id)
499 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
500 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
501 if mobj is None:
502 return [(err_msg, None, None)]
503 player_config = json.loads(mobj.group(1))
504 try:
505 args = player_config[u'args']
506 caption_url = args[u'ttsurl']
507 timestamp = args[u'timestamp']
508 params = compat_urllib_parse.urlencode({
509 'lang': 'en',
510 'tlang': sub_lang,
511 'fmt': sub_format,
512 'ts': timestamp,
513 'kind': 'asr',
514 })
515 subtitles_url = caption_url + '&' + params
516 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
517 return [(None, sub_lang, sub)]
518 except KeyError:
519 return [(err_msg, None, None)]
520
521 def _extract_subtitle(self, video_id):
522 """
523 Return a list with a tuple:
524 [(error_message, sub_lang, sub)]
525 """
526 sub_lang_list = self._get_available_subtitles(video_id)
527 sub_format = self._downloader.params.get('subtitlesformat')
528 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
529 return [(sub_lang_list[0], None, None)]
530 if self._downloader.params.get('subtitleslang', False):
531 sub_lang = self._downloader.params.get('subtitleslang')
532 elif 'en' in sub_lang_list:
533 sub_lang = 'en'
534 else:
535 sub_lang = list(sub_lang_list.keys())[0]
536 if not sub_lang in sub_lang_list:
537 return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
538
539 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
540 return [subtitle]
541
542 def _extract_all_subtitles(self, video_id):
543 sub_lang_list = self._get_available_subtitles(video_id)
544 sub_format = self._downloader.params.get('subtitlesformat')
545 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
546 return [(sub_lang_list[0], None, None)]
547 subtitles = []
548 for sub_lang in sub_lang_list:
549 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
550 subtitles.append(subtitle)
551 return subtitles
552
553 def _print_formats(self, formats):
554 print('Available formats:')
555 for x in formats:
556 print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
557 self._video_dimensions.get(x, '???'),
558 ' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
559
560 def _extract_id(self, url):
561 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
562 if mobj is None:
563 raise ExtractorError(u'Invalid URL: %s' % url)
564 video_id = mobj.group(2)
565 return video_id
566
567 def _get_video_url_list(self, url_map):
568 """
569 Transform a dictionary in the format {itag:url} to a list of (itag, url)
570 with the requested formats.
571 """
572 req_format = self._downloader.params.get('format', None)
573 format_limit = self._downloader.params.get('format_limit', None)
574 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
575 if format_limit is not None and format_limit in available_formats:
576 format_list = available_formats[available_formats.index(format_limit):]
577 else:
578 format_list = available_formats
579 existing_formats = [x for x in format_list if x in url_map]
580 if len(existing_formats) == 0:
581 raise ExtractorError(u'no known formats available for video')
582 if self._downloader.params.get('listformats', None):
583 self._print_formats(existing_formats)
584 return
585 if req_format is None or req_format == 'best':
586 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
587 elif req_format == 'worst':
588 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
589 elif req_format in ('-1', 'all'):
590 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
591 else:
592 # Specific formats. We pick the first in a slash-delimeted sequence.
593 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
594 req_formats = req_format.split('/')
595 video_url_list = None
596 for rf in req_formats:
597 if rf in url_map:
598 video_url_list = [(rf, url_map[rf])]
599 break
600 if video_url_list is None:
601 raise ExtractorError(u'requested format not available')
602 return video_url_list
603
604 def _extract_from_m3u8(self, manifest_url, video_id):
605 url_map = {}
606 def _get_urls(_manifest):
607 lines = _manifest.split('\n')
608 urls = filter(lambda l: l and not l.startswith('#'),
609 lines)
610 return urls
611 manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
612 formats_urls = _get_urls(manifest)
613 for format_url in formats_urls:
614 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
615 url_map[itag] = format_url
616 return url_map
617
618 def _real_extract(self, url):
619 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
620 self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).')
621
622 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
623 mobj = re.search(self._NEXT_URL_RE, url)
624 if mobj:
625 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
626 video_id = self._extract_id(url)
627
628 # Get video webpage
629 self.report_video_webpage_download(video_id)
630 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
631 request = compat_urllib_request.Request(url)
632 try:
633 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
634 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
635 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
636
637 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
638
639 # Attempt to extract SWF player URL
640 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
641 if mobj is not None:
642 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
643 else:
644 player_url = None
645
646 # Get video info
647 self.report_video_info_webpage_download(video_id)
648 if re.search(r'player-age-gate-content">', video_webpage) is not None:
649 self.report_age_confirmation()
650 age_gate = True
651 # We simulate the access to the video from www.youtube.com/v/{video_id}
652 # this can be viewed without login into Youtube
653 data = compat_urllib_parse.urlencode({'video_id': video_id,
654 'el': 'embedded',
655 'gl': 'US',
656 'hl': 'en',
657 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
658 'asv': 3,
659 'sts':'1588',
660 })
661 video_info_url = 'https://www.youtube.com/get_video_info?' + data
662 video_info_webpage = self._download_webpage(video_info_url, video_id,
663 note=False,
664 errnote='unable to download video info webpage')
665 video_info = compat_parse_qs(video_info_webpage)
666 else:
667 age_gate = False
668 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
669 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
670 % (video_id, el_type))
671 video_info_webpage = self._download_webpage(video_info_url, video_id,
672 note=False,
673 errnote='unable to download video info webpage')
674 video_info = compat_parse_qs(video_info_webpage)
675 if 'token' in video_info:
676 break
677 if 'token' not in video_info:
678 if 'reason' in video_info:
679 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
680 else:
681 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
682
683 # Check for "rental" videos
684 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
685 raise ExtractorError(u'"rental" videos not supported')
686
687 # Start extracting information
688 self.report_information_extraction(video_id)
689
690 # uploader
691 if 'author' not in video_info:
692 raise ExtractorError(u'Unable to extract uploader name')
693 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
694
695 # uploader_id
696 video_uploader_id = None
697 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
698 if mobj is not None:
699 video_uploader_id = mobj.group(1)
700 else:
701 self._downloader.report_warning(u'unable to extract uploader nickname')
702
703 # title
704 if 'title' not in video_info:
705 raise ExtractorError(u'Unable to extract video title')
706 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
707
708 # thumbnail image
709 # We try first to get a high quality image:
710 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
711 video_webpage, re.DOTALL)
712 if m_thumb is not None:
713 video_thumbnail = m_thumb.group(1)
714 elif 'thumbnail_url' not in video_info:
715 self._downloader.report_warning(u'unable to extract video thumbnail')
716 video_thumbnail = ''
717 else: # don't panic if we can't find it
718 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
719
720 # upload date
721 upload_date = None
722 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
723 if mobj is not None:
724 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
725 upload_date = unified_strdate(upload_date)
726
727 # description
728 video_description = get_element_by_id("eow-description", video_webpage)
729 if video_description:
730 video_description = clean_html(video_description)
731 else:
732 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
733 if fd_mobj:
734 video_description = unescapeHTML(fd_mobj.group(1))
735 else:
736 video_description = u''
737
738 # subtitles
739 video_subtitles = None
740
741 if self._downloader.params.get('writesubtitles', False):
742 video_subtitles = self._extract_subtitle(video_id)
743 if video_subtitles:
744 (sub_error, sub_lang, sub) = video_subtitles[0]
745 if sub_error:
746 self._downloader.report_warning(sub_error)
747
748 if self._downloader.params.get('writeautomaticsub', False):
749 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
750 (sub_error, sub_lang, sub) = video_subtitles[0]
751 if sub_error:
752 self._downloader.report_warning(sub_error)
753
754 if self._downloader.params.get('allsubtitles', False):
755 video_subtitles = self._extract_all_subtitles(video_id)
756 for video_subtitle in video_subtitles:
757 (sub_error, sub_lang, sub) = video_subtitle
758 if sub_error:
759 self._downloader.report_warning(sub_error)
760
761 if self._downloader.params.get('listsubtitles', False):
762 self._list_available_subtitles(video_id)
763 return
764
765 if 'length_seconds' not in video_info:
766 self._downloader.report_warning(u'unable to extract video duration')
767 video_duration = ''
768 else:
769 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
770
771 # Decide which formats to download
772
773 try:
774 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
775 if not mobj:
776 raise ValueError('Could not find vevo ID')
777 info = json.loads(mobj.group(1))
778 args = info['args']
779 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
780 # this signatures are encrypted
781 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
782 if m_s is not None:
783 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
784 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
785 m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u''))
786 if m_s is not None:
787 if 'url_encoded_fmt_stream_map' in video_info:
788 video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
789 else:
790 video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
791 elif 'adaptive_fmts' in video_info:
792 if 'url_encoded_fmt_stream_map' in video_info:
793 video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
794 else:
795 video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
796 except ValueError:
797 pass
798
799 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
800 self.report_rtmp_download()
801 video_url_list = [(None, video_info['conn'][0])]
802 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
803 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
804 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
805 url_map = {}
806 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
807 url_data = compat_parse_qs(url_data_str)
808 if 'itag' in url_data and 'url' in url_data:
809 url = url_data['url'][0]
810 if 'sig' in url_data:
811 url += '&signature=' + url_data['sig'][0]
812 elif 's' in url_data:
813 if self._downloader.params.get('verbose'):
814 s = url_data['s'][0]
815 if age_gate:
816 player_version = self._search_regex(r'ad3-(.+?)\.swf',
817 video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND',
818 'flash player', fatal=False)
819 player = 'flash player %s' % player_version
820 else:
821 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
822 'html5 player', fatal=False)
823 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
824 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
825 (len(s), parts_sizes, url_data['itag'][0], player))
826 encrypted_sig = url_data['s'][0]
827 if age_gate:
828 signature = self._decrypt_signature_age_gate(encrypted_sig)
829 else:
830 signature = self._decrypt_signature(encrypted_sig)
831 url += '&signature=' + signature
832 if 'ratebypass' not in url:
833 url += '&ratebypass=yes'
834 url_map[url_data['itag'][0]] = url
835 video_url_list = self._get_video_url_list(url_map)
836 if not video_url_list:
837 return
838 elif video_info.get('hlsvp'):
839 manifest_url = video_info['hlsvp'][0]
840 url_map = self._extract_from_m3u8(manifest_url, video_id)
841 video_url_list = self._get_video_url_list(url_map)
842 if not video_url_list:
843 return
844
845 else:
846 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
847
848 results = []
849 for format_param, video_real_url in video_url_list:
850 # Extension
851 video_extension = self._video_extensions.get(format_param, 'flv')
852
853 video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
854 self._video_dimensions.get(format_param, '???'),
855 ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
856
857 results.append({
858 'id': video_id,
859 'url': video_real_url,
860 'uploader': video_uploader,
861 'uploader_id': video_uploader_id,
862 'upload_date': upload_date,
863 'title': video_title,
864 'ext': video_extension,
865 'format': video_format,
866 'thumbnail': video_thumbnail,
867 'description': video_description,
868 'player_url': player_url,
869 'subtitles': video_subtitles,
870 'duration': video_duration
871 })
872 return results
873
874 class YoutubePlaylistIE(InfoExtractor):
875 IE_DESC = u'YouTube.com playlists'
876 _VALID_URL = r"""(?:
877 (?:https?://)?
878 (?:\w+\.)?
879 youtube\.com/
880 (?:
881 (?:course|view_play_list|my_playlists|artist|playlist|watch)
882 \? (?:.*?&)*? (?:p|a|list)=
883 | p/
884 )
885 ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
886 .*
887 |
888 ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
889 )"""
890 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
891 _MAX_RESULTS = 50
892 IE_NAME = u'youtube:playlist'
893
894 @classmethod
895 def suitable(cls, url):
896 """Receives a URL and returns True if suitable for this IE."""
897 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
898
899 def _real_extract(self, url):
900 # Extract playlist id
901 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
902 if mobj is None:
903 raise ExtractorError(u'Invalid URL: %s' % url)
904
905 # Download playlist videos from API
906 playlist_id = mobj.group(1) or mobj.group(2)
907 videos = []
908
909 for page_num in itertools.count(1):
910 start_index = self._MAX_RESULTS * (page_num - 1) + 1
911 if start_index >= 1000:
912 self._downloader.report_warning(u'Max number of results reached')
913 break
914 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
915 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
916
917 try:
918 response = json.loads(page)
919 except ValueError as err:
920 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
921
922 if 'feed' not in response:
923 raise ExtractorError(u'Got a malformed response from YouTube API')
924 playlist_title = response['feed']['title']['$t']
925 if 'entry' not in response['feed']:
926 # Number of videos is a multiple of self._MAX_RESULTS
927 break
928
929 for entry in response['feed']['entry']:
930 index = entry['yt$position']['$t']
931 if 'media$group' in entry and 'media$player' in entry['media$group']:
932 videos.append((index, entry['media$group']['media$player']['url']))
933
934 videos = [v[1] for v in sorted(videos)]
935
936 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
937 return [self.playlist_result(url_results, playlist_id, playlist_title)]
938
939
940 class YoutubeChannelIE(InfoExtractor):
941 IE_DESC = u'YouTube.com channels'
942 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
943 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
944 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
945 _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'
946 IE_NAME = u'youtube:channel'
947
948 def extract_videos_from_page(self, page):
949 ids_in_page = []
950 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
951 if mobj.group(1) not in ids_in_page:
952 ids_in_page.append(mobj.group(1))
953 return ids_in_page
954
955 def _real_extract(self, url):
956 # Extract channel id
957 mobj = re.match(self._VALID_URL, url)
958 if mobj is None:
959 raise ExtractorError(u'Invalid URL: %s' % url)
960
961 # Download channel page
962 channel_id = mobj.group(1)
963 video_ids = []
964 pagenum = 1
965
966 url = self._TEMPLATE_URL % (channel_id, pagenum)
967 page = self._download_webpage(url, channel_id,
968 u'Downloading page #%s' % pagenum)
969
970 # Extract video identifiers
971 ids_in_page = self.extract_videos_from_page(page)
972 video_ids.extend(ids_in_page)
973
974 # Download any subsequent channel pages using the json-based channel_ajax query
975 if self._MORE_PAGES_INDICATOR in page:
976 for pagenum in itertools.count(1):
977 url = self._MORE_PAGES_URL % (pagenum, channel_id)
978 page = self._download_webpage(url, channel_id,
979 u'Downloading page #%s' % pagenum)
980
981 page = json.loads(page)
982
983 ids_in_page = self.extract_videos_from_page(page['content_html'])
984 video_ids.extend(ids_in_page)
985
986 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
987 break
988
989 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
990
991 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
992 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
993 return [self.playlist_result(url_entries, channel_id)]
994
995
996 class YoutubeUserIE(InfoExtractor):
997 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
998 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
999 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
1000 _GDATA_PAGE_SIZE = 50
1001 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
1002 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
1003 IE_NAME = u'youtube:user'
1004
1005 def _real_extract(self, url):
1006 # Extract username
1007 mobj = re.match(self._VALID_URL, url)
1008 if mobj is None:
1009 raise ExtractorError(u'Invalid URL: %s' % url)
1010
1011 username = mobj.group(1)
1012
1013 # Download video ids using YouTube Data API. Result size per
1014 # query is limited (currently to 50 videos) so we need to query
1015 # page by page until there are no video ids - it means we got
1016 # all of them.
1017
1018 video_ids = []
1019
1020 for pagenum in itertools.count(0):
1021 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1022
1023 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1024 page = self._download_webpage(gdata_url, username,
1025 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
1026
1027 # Extract video identifiers
1028 ids_in_page = []
1029
1030 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
1031 if mobj.group(1) not in ids_in_page:
1032 ids_in_page.append(mobj.group(1))
1033
1034 video_ids.extend(ids_in_page)
1035
1036 # A little optimization - if current page is not
1037 # "full", ie. does not contain PAGE_SIZE video ids then
1038 # we can assume that this page is the last one - there
1039 # are no more ids on further pages - no need to query
1040 # again.
1041
1042 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
1043 break
1044
1045 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
1046 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
1047 return [self.playlist_result(url_results, playlist_title = username)]
1048
1049 class YoutubeSearchIE(SearchInfoExtractor):
1050 IE_DESC = u'YouTube.com searches'
1051 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1052 _MAX_RESULTS = 1000
1053 IE_NAME = u'youtube:search'
1054 _SEARCH_KEY = 'ytsearch'
1055
1056 def report_download_page(self, query, pagenum):
1057 """Report attempt to download search page with given number."""
1058 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
1059
1060 def _get_n_results(self, query, n):
1061 """Get a specified number of results for a query"""
1062
1063 video_ids = []
1064 pagenum = 0
1065 limit = n
1066
1067 while (50 * pagenum) < limit:
1068 self.report_download_page(query, pagenum+1)
1069 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
1070 request = compat_urllib_request.Request(result_url)
1071 try:
1072 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
1073 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1074 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
1075 api_response = json.loads(data)['data']
1076
1077 if not 'items' in api_response:
1078 raise ExtractorError(u'[youtube] No video results')
1079
1080 new_ids = list(video['id'] for video in api_response['items'])
1081 video_ids += new_ids
1082
1083 limit = min(n, api_response['totalItems'])
1084 pagenum += 1
1085
1086 if len(video_ids) > n:
1087 video_ids = video_ids[:n]
1088 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
1089 return self.playlist_result(videos, query)
1090
1091
1092 class YoutubeShowIE(InfoExtractor):
1093 IE_DESC = u'YouTube.com (multi-season) shows'
1094 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
1095 IE_NAME = u'youtube:show'
1096
1097 def _real_extract(self, url):
1098 mobj = re.match(self._VALID_URL, url)
1099 show_name = mobj.group(1)
1100 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
1101 # There's one playlist for each season of the show
1102 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1103 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
1104 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
1105
1106
1107 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1108 """
1109 Base class for extractors that fetch info from
1110 http://www.youtube.com/feed_ajax
1111 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1112 """
1113 _LOGIN_REQUIRED = True
1114 _PAGING_STEP = 30
1115 # use action_load_personal_feed instead of action_load_system_feed
1116 _PERSONAL_FEED = False
1117
1118 @property
1119 def _FEED_TEMPLATE(self):
1120 action = 'action_load_system_feed'
1121 if self._PERSONAL_FEED:
1122 action = 'action_load_personal_feed'
1123 return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1124
1125 @property
1126 def IE_NAME(self):
1127 return u'youtube:%s' % self._FEED_NAME
1128
1129 def _real_initialize(self):
1130 self._login()
1131
1132 def _real_extract(self, url):
1133 feed_entries = []
1134 # The step argument is available only in 2.7 or higher
1135 for i in itertools.count(0):
1136 paging = i*self._PAGING_STEP
1137 info = self._download_webpage(self._FEED_TEMPLATE % paging,
1138 u'%s feed' % self._FEED_NAME,
1139 u'Downloading page %s' % i)
1140 info = json.loads(info)
1141 feed_html = info['feed_html']
1142 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1143 ids = orderedSet(m.group(1) for m in m_ids)
1144 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
1145 if info['paging'] is None:
1146 break
1147 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1148
1149 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1150 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
1151 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1152 _FEED_NAME = 'subscriptions'
1153 _PLAYLIST_TITLE = u'Youtube Subscriptions'
1154
1155 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1156 IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1157 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1158 _FEED_NAME = 'recommended'
1159 _PLAYLIST_TITLE = u'Youtube Recommended videos'
1160
1161 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1162 IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1163 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1164 _FEED_NAME = 'watch_later'
1165 _PLAYLIST_TITLE = u'Youtube Watch Later'
1166 _PAGING_STEP = 100
1167 _PERSONAL_FEED = True
1168
1169 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1170 IE_NAME = u'youtube:favorites'
1171 IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1172 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
1173 _LOGIN_REQUIRED = True
1174
1175 def _real_extract(self, url):
1176 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1177 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
1178 return self.url_result(playlist_id, 'YoutubePlaylist')
fork of https://github.com/yt-dlp/yt-dlp/