]> 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] Fix some issues with the detection of playlist/channel urls (reported in...
[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 (?:(?:(?:(?:\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 ))
150 |youtu\.be/ # just youtu.be/xxxx
151 )
152 )? # all until now is optional -> you can pass the naked ID
153 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
154 (?(1).+)? # if we found the ID, everything can follow
155 $"""
156 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
157 # Listed in order of quality
158 _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13',
159 # Apple HTTP Live Streaming
160 '96', '95', '94', '93', '92', '132', '151',
161 # 3D
162 '85', '84', '102', '83', '101', '82', '100',
163 # Dash video
164 '138', '137', '248', '136', '247', '135', '246',
165 '245', '244', '134', '243', '133', '242', '160',
166 # Dash audio
167 '141', '172', '140', '171', '139',
168 ]
169 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13',
170 # Apple HTTP Live Streaming
171 '96', '95', '94', '93', '92', '132', '151',
172 # 3D
173 '85', '102', '84', '101', '83', '100', '82',
174 # Dash video
175 '138', '248', '137', '247', '136', '246', '245',
176 '244', '135', '243', '134', '242', '133', '160',
177 # Dash audio
178 '172', '141', '171', '140', '139',
179 ]
180 _video_formats_map = {
181 'flv': ['35', '34', '6', '5'],
182 '3gp': ['36', '17', '13'],
183 'mp4': ['38', '37', '22', '18'],
184 'webm': ['46', '45', '44', '43'],
185 }
186 _video_extensions = {
187 '13': '3gp',
188 '17': '3gp',
189 '18': 'mp4',
190 '22': 'mp4',
191 '36': '3gp',
192 '37': 'mp4',
193 '38': 'mp4',
194 '43': 'webm',
195 '44': 'webm',
196 '45': 'webm',
197 '46': 'webm',
198
199 # 3d videos
200 '82': 'mp4',
201 '83': 'mp4',
202 '84': 'mp4',
203 '85': 'mp4',
204 '100': 'webm',
205 '101': 'webm',
206 '102': 'webm',
207
208 # Apple HTTP Live Streaming
209 '92': 'mp4',
210 '93': 'mp4',
211 '94': 'mp4',
212 '95': 'mp4',
213 '96': 'mp4',
214 '132': 'mp4',
215 '151': 'mp4',
216
217 # Dash mp4
218 '133': 'mp4',
219 '134': 'mp4',
220 '135': 'mp4',
221 '136': 'mp4',
222 '137': 'mp4',
223 '138': 'mp4',
224 '139': 'mp4',
225 '140': 'mp4',
226 '141': 'mp4',
227 '160': 'mp4',
228
229 # Dash webm
230 '171': 'webm',
231 '172': 'webm',
232 '242': 'webm',
233 '243': 'webm',
234 '244': 'webm',
235 '245': 'webm',
236 '246': 'webm',
237 '247': 'webm',
238 '248': 'webm',
239 }
240 _video_dimensions = {
241 '5': '240x400',
242 '6': '???',
243 '13': '???',
244 '17': '144x176',
245 '18': '360x640',
246 '22': '720x1280',
247 '34': '360x640',
248 '35': '480x854',
249 '36': '240x320',
250 '37': '1080x1920',
251 '38': '3072x4096',
252 '43': '360x640',
253 '44': '480x854',
254 '45': '720x1280',
255 '46': '1080x1920',
256 '82': '360p',
257 '83': '480p',
258 '84': '720p',
259 '85': '1080p',
260 '92': '240p',
261 '93': '360p',
262 '94': '480p',
263 '95': '720p',
264 '96': '1080p',
265 '100': '360p',
266 '101': '480p',
267 '102': '720p',
268 '132': '240p',
269 '151': '72p',
270 '133': '240p',
271 '134': '360p',
272 '135': '480p',
273 '136': '720p',
274 '137': '1080p',
275 '138': '>1080p',
276 '139': '48k',
277 '140': '128k',
278 '141': '256k',
279 '160': '192p',
280 '171': '128k',
281 '172': '256k',
282 '242': '240p',
283 '243': '360p',
284 '244': '480p',
285 '245': '480p',
286 '246': '480p',
287 '247': '720p',
288 '248': '1080p',
289 }
290 _special_itags = {
291 '82': '3D',
292 '83': '3D',
293 '84': '3D',
294 '85': '3D',
295 '100': '3D',
296 '101': '3D',
297 '102': '3D',
298 '133': 'DASH Video',
299 '134': 'DASH Video',
300 '135': 'DASH Video',
301 '136': 'DASH Video',
302 '137': 'DASH Video',
303 '138': 'DASH Video',
304 '139': 'DASH Audio',
305 '140': 'DASH Audio',
306 '141': 'DASH Audio',
307 '160': 'DASH Video',
308 '171': 'DASH Audio',
309 '172': 'DASH Audio',
310 '242': 'DASH Video',
311 '243': 'DASH Video',
312 '244': 'DASH Video',
313 '245': 'DASH Video',
314 '246': 'DASH Video',
315 '247': 'DASH Video',
316 '248': 'DASH Video',
317 }
318
319 IE_NAME = u'youtube'
320 _TESTS = [
321 {
322 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
323 u"file": u"BaW_jenozKc.mp4",
324 u"info_dict": {
325 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
326 u"uploader": u"Philipp Hagemeister",
327 u"uploader_id": u"phihag",
328 u"upload_date": u"20121002",
329 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
330 }
331 },
332 {
333 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
334 u"file": u"1ltcDfZMA3U.flv",
335 u"note": u"Test VEVO video (#897)",
336 u"info_dict": {
337 u"upload_date": u"20070518",
338 u"title": u"Maps - It Will Find You",
339 u"description": u"Music video by Maps performing It Will Find You.",
340 u"uploader": u"MuteUSA",
341 u"uploader_id": u"MuteUSA"
342 }
343 },
344 {
345 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
346 u"file": u"UxxajLWwzqY.mp4",
347 u"note": u"Test generic use_cipher_signature video (#897)",
348 u"info_dict": {
349 u"upload_date": u"20120506",
350 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
351 u"description": u"md5:3e2666e0a55044490499ea45fe9037b7",
352 u"uploader": u"Icona Pop",
353 u"uploader_id": u"IconaPop"
354 }
355 },
356 {
357 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
358 u"file": u"07FYdnEawAQ.mp4",
359 u"note": u"Test VEVO video with age protection (#956)",
360 u"info_dict": {
361 u"upload_date": u"20130703",
362 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
363 u"description": u"md5:64249768eec3bc4276236606ea996373",
364 u"uploader": u"justintimberlakeVEVO",
365 u"uploader_id": u"justintimberlakeVEVO"
366 }
367 },
368 {
369 u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE',
370 u'file': u'TGi3HqYrWHE.mp4',
371 u'note': u'm3u8 video',
372 u'info_dict': {
373 u'title': u'Triathlon - Men - London 2012 Olympic Games',
374 u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
375 u'uploader': u'olympic',
376 u'upload_date': u'20120807',
377 u'uploader_id': u'olympic',
378 },
379 u'params': {
380 u'skip_download': True,
381 },
382 },
383 ]
384
385
386 @classmethod
387 def suitable(cls, url):
388 """Receives a URL and returns True if suitable for this IE."""
389 if YoutubePlaylistIE.suitable(url): return False
390 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
391
392 def report_video_webpage_download(self, video_id):
393 """Report attempt to download video webpage."""
394 self.to_screen(u'%s: Downloading video webpage' % video_id)
395
396 def report_video_info_webpage_download(self, video_id):
397 """Report attempt to download video info webpage."""
398 self.to_screen(u'%s: Downloading video info webpage' % video_id)
399
400 def report_video_subtitles_download(self, video_id):
401 """Report attempt to download video info webpage."""
402 self.to_screen(u'%s: Checking available subtitles' % video_id)
403
404 def report_video_subtitles_request(self, video_id, sub_lang, format):
405 """Report attempt to download video info webpage."""
406 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
407
408 def report_video_subtitles_available(self, video_id, sub_lang_list):
409 """Report available subtitles."""
410 sub_lang = ",".join(list(sub_lang_list.keys()))
411 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
412
413 def report_information_extraction(self, video_id):
414 """Report attempt to extract video information."""
415 self.to_screen(u'%s: Extracting video information' % video_id)
416
417 def report_unavailable_format(self, video_id, format):
418 """Report extracted video URL."""
419 self.to_screen(u'%s: Format %s not available' % (video_id, format))
420
421 def report_rtmp_download(self):
422 """Indicate the download will use the RTMP protocol."""
423 self.to_screen(u'RTMP download detected')
424
425 def _decrypt_signature(self, s):
426 """Turn the encrypted s field into a working signature"""
427
428 if len(s) == 92:
429 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
430 elif len(s) == 90:
431 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
432 elif len(s) == 89:
433 return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1]
434 elif len(s) == 88:
435 return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28]
436 elif len(s) == 87:
437 return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
438 elif len(s) == 86:
439 return s[5:34] + s[0] + s[35:38] + s[3] + s[39:45] + s[38] + s[46:53] + s[73] + s[54:73] + s[85] + s[74:85] + s[53]
440 elif len(s) == 85:
441 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]
442 elif len(s) == 84:
443 return s[81:36:-1] + s[0] + s[35:2:-1]
444 elif len(s) == 83:
445 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]
446 elif len(s) == 82:
447 return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82]
448 elif len(s) == 81:
449 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]
450 elif len(s) == 80:
451 return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80]
452 elif len(s) == 79:
453 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]
454
455 else:
456 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
457
458 def _decrypt_signature_age_gate(self, s):
459 # The videos with age protection use another player, so the algorithms
460 # can be different.
461 if len(s) == 86:
462 return s[2:63] + s[82] + s[64:82] + s[63]
463 else:
464 # Fallback to the other algortihms
465 return self._decrypt_signature(s)
466
467
468 def _get_available_subtitles(self, video_id):
469 self.report_video_subtitles_download(video_id)
470 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
471 try:
472 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
473 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
474 self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
475 return {}
476 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
477 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
478 if not sub_lang_list:
479 self._downloader.report_warning(u'video doesn\'t have subtitles')
480 return {}
481 return sub_lang_list
482
483 def _list_available_subtitles(self, video_id):
484 sub_lang_list = self._get_available_subtitles(video_id)
485 self.report_video_subtitles_available(video_id, sub_lang_list)
486
487 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
488 """
489 Return the subtitle as a string or None if they are not found
490 """
491 self.report_video_subtitles_request(video_id, sub_lang, format)
492 params = compat_urllib_parse.urlencode({
493 'lang': sub_lang,
494 'name': sub_name,
495 'v': video_id,
496 'fmt': format,
497 })
498 url = 'http://www.youtube.com/api/timedtext?' + params
499 try:
500 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
501 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
502 self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
503 return
504 if not sub:
505 self._downloader.report_warning(u'Did not fetch video subtitles')
506 return
507 return sub
508
509 def _request_automatic_caption(self, video_id, webpage):
510 """We need the webpage for getting the captions url, pass it as an
511 argument to speed up the process."""
512 sub_lang = (self._downloader.params.get('subtitleslangs') or ['en'])[0]
513 sub_format = self._downloader.params.get('subtitlesformat')
514 self.to_screen(u'%s: Looking for automatic captions' % video_id)
515 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
516 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
517 if mobj is None:
518 self._downloader.report_warning(err_msg)
519 return {}
520 player_config = json.loads(mobj.group(1))
521 try:
522 args = player_config[u'args']
523 caption_url = args[u'ttsurl']
524 timestamp = args[u'timestamp']
525 params = compat_urllib_parse.urlencode({
526 'lang': 'en',
527 'tlang': sub_lang,
528 'fmt': sub_format,
529 'ts': timestamp,
530 'kind': 'asr',
531 })
532 subtitles_url = caption_url + '&' + params
533 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
534 return {sub_lang: sub}
535 # An extractor error can be raise by the download process if there are
536 # no automatic captions but there are subtitles
537 except (KeyError, ExtractorError):
538 self._downloader.report_warning(err_msg)
539 return {}
540
541 def _extract_subtitles(self, video_id):
542 """
543 Return a dictionary: {language: subtitles} or {} if the subtitles
544 couldn't be found
545 """
546 available_subs_list = self._get_available_subtitles(video_id)
547 sub_format = self._downloader.params.get('subtitlesformat')
548 if not available_subs_list: #There was some error, it didn't get the available subtitles
549 return {}
550 if self._downloader.params.get('allsubtitles', False):
551 sub_lang_list = available_subs_list
552 else:
553 if self._downloader.params.get('subtitleslangs', False):
554 reqested_langs = self._downloader.params.get('subtitleslangs')
555 elif 'en' in available_subs_list:
556 reqested_langs = ['en']
557 else:
558 reqested_langs = [list(available_subs_list.keys())[0]]
559
560 sub_lang_list = {}
561 for sub_lang in reqested_langs:
562 if not sub_lang in available_subs_list:
563 self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
564 continue
565 sub_lang_list[sub_lang] = available_subs_list[sub_lang]
566 subtitles = {}
567 for sub_lang in sub_lang_list:
568 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
569 if subtitle:
570 subtitles[sub_lang] = subtitle
571 return subtitles
572
573 def _print_formats(self, formats):
574 print('Available formats:')
575 for x in formats:
576 print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
577 self._video_dimensions.get(x, '???'),
578 ' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
579
580 def _extract_id(self, url):
581 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
582 if mobj is None:
583 raise ExtractorError(u'Invalid URL: %s' % url)
584 video_id = mobj.group(2)
585 return video_id
586
587 def _get_video_url_list(self, url_map):
588 """
589 Transform a dictionary in the format {itag:url} to a list of (itag, url)
590 with the requested formats.
591 """
592 req_format = self._downloader.params.get('format', None)
593 format_limit = self._downloader.params.get('format_limit', None)
594 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
595 if format_limit is not None and format_limit in available_formats:
596 format_list = available_formats[available_formats.index(format_limit):]
597 else:
598 format_list = available_formats
599 existing_formats = [x for x in format_list if x in url_map]
600 if len(existing_formats) == 0:
601 raise ExtractorError(u'no known formats available for video')
602 if self._downloader.params.get('listformats', None):
603 self._print_formats(existing_formats)
604 return
605 if req_format is None or req_format == 'best':
606 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
607 elif req_format == 'worst':
608 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
609 elif req_format in ('-1', 'all'):
610 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
611 else:
612 # Specific formats. We pick the first in a slash-delimeted sequence.
613 # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality
614 # available in the specified format. For example,
615 # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
616 # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'.
617 # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'.
618 req_formats = req_format.split('/')
619 video_url_list = None
620 for rf in req_formats:
621 if rf in url_map:
622 video_url_list = [(rf, url_map[rf])]
623 break
624 if rf in self._video_formats_map:
625 for srf in self._video_formats_map[rf]:
626 if srf in url_map:
627 video_url_list = [(srf, url_map[srf])]
628 break
629 else:
630 continue
631 break
632 if video_url_list is None:
633 raise ExtractorError(u'requested format not available')
634 return video_url_list
635
636 def _extract_from_m3u8(self, manifest_url, video_id):
637 url_map = {}
638 def _get_urls(_manifest):
639 lines = _manifest.split('\n')
640 urls = filter(lambda l: l and not l.startswith('#'),
641 lines)
642 return urls
643 manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
644 formats_urls = _get_urls(manifest)
645 for format_url in formats_urls:
646 itag = self._search_regex(r'itag%3D(\d+?)/', format_url, 'itag')
647 url_map[itag] = format_url
648 return url_map
649
650 def _real_extract(self, url):
651 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
652 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 ).')
653
654 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
655 mobj = re.search(self._NEXT_URL_RE, url)
656 if mobj:
657 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
658 video_id = self._extract_id(url)
659
660 # Get video webpage
661 self.report_video_webpage_download(video_id)
662 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
663 request = compat_urllib_request.Request(url)
664 try:
665 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
666 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
667 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
668
669 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
670
671 # Attempt to extract SWF player URL
672 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
673 if mobj is not None:
674 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
675 else:
676 player_url = None
677
678 # Get video info
679 self.report_video_info_webpage_download(video_id)
680 if re.search(r'player-age-gate-content">', video_webpage) is not None:
681 self.report_age_confirmation()
682 age_gate = True
683 # We simulate the access to the video from www.youtube.com/v/{video_id}
684 # this can be viewed without login into Youtube
685 data = compat_urllib_parse.urlencode({'video_id': video_id,
686 'el': 'embedded',
687 'gl': 'US',
688 'hl': 'en',
689 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
690 'asv': 3,
691 'sts':'1588',
692 })
693 video_info_url = 'https://www.youtube.com/get_video_info?' + data
694 video_info_webpage = self._download_webpage(video_info_url, video_id,
695 note=False,
696 errnote='unable to download video info webpage')
697 video_info = compat_parse_qs(video_info_webpage)
698 else:
699 age_gate = False
700 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
701 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
702 % (video_id, el_type))
703 video_info_webpage = self._download_webpage(video_info_url, video_id,
704 note=False,
705 errnote='unable to download video info webpage')
706 video_info = compat_parse_qs(video_info_webpage)
707 if 'token' in video_info:
708 break
709 if 'token' not in video_info:
710 if 'reason' in video_info:
711 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
712 else:
713 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
714
715 # Check for "rental" videos
716 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
717 raise ExtractorError(u'"rental" videos not supported')
718
719 # Start extracting information
720 self.report_information_extraction(video_id)
721
722 # uploader
723 if 'author' not in video_info:
724 raise ExtractorError(u'Unable to extract uploader name')
725 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
726
727 # uploader_id
728 video_uploader_id = None
729 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
730 if mobj is not None:
731 video_uploader_id = mobj.group(1)
732 else:
733 self._downloader.report_warning(u'unable to extract uploader nickname')
734
735 # title
736 if 'title' not in video_info:
737 raise ExtractorError(u'Unable to extract video title')
738 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
739
740 # thumbnail image
741 # We try first to get a high quality image:
742 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
743 video_webpage, re.DOTALL)
744 if m_thumb is not None:
745 video_thumbnail = m_thumb.group(1)
746 elif 'thumbnail_url' not in video_info:
747 self._downloader.report_warning(u'unable to extract video thumbnail')
748 video_thumbnail = ''
749 else: # don't panic if we can't find it
750 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
751
752 # upload date
753 upload_date = None
754 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
755 if mobj is not None:
756 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
757 upload_date = unified_strdate(upload_date)
758
759 # description
760 video_description = get_element_by_id("eow-description", video_webpage)
761 if video_description:
762 video_description = clean_html(video_description)
763 else:
764 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
765 if fd_mobj:
766 video_description = unescapeHTML(fd_mobj.group(1))
767 else:
768 video_description = u''
769
770 # subtitles
771 video_subtitles = None
772
773 if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
774 video_subtitles = self._extract_subtitles(video_id)
775 elif self._downloader.params.get('writeautomaticsub', False):
776 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
777
778 if self._downloader.params.get('listsubtitles', False):
779 self._list_available_subtitles(video_id)
780 return
781
782 if 'length_seconds' not in video_info:
783 self._downloader.report_warning(u'unable to extract video duration')
784 video_duration = ''
785 else:
786 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
787
788 # Decide which formats to download
789
790 try:
791 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
792 if not mobj:
793 raise ValueError('Could not find vevo ID')
794 info = json.loads(mobj.group(1))
795 args = info['args']
796 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
797 # this signatures are encrypted
798 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
799 if m_s is not None:
800 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
801 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
802 m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u''))
803 if m_s is not None:
804 if 'url_encoded_fmt_stream_map' in video_info:
805 video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
806 else:
807 video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
808 elif 'adaptive_fmts' in video_info:
809 if 'url_encoded_fmt_stream_map' in video_info:
810 video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
811 else:
812 video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
813 except ValueError:
814 pass
815
816 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
817 self.report_rtmp_download()
818 video_url_list = [(None, video_info['conn'][0])]
819 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
820 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
821 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
822 url_map = {}
823 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
824 url_data = compat_parse_qs(url_data_str)
825 if 'itag' in url_data and 'url' in url_data:
826 url = url_data['url'][0]
827 if 'sig' in url_data:
828 url += '&signature=' + url_data['sig'][0]
829 elif 's' in url_data:
830 if self._downloader.params.get('verbose'):
831 s = url_data['s'][0]
832 if age_gate:
833 player_version = self._search_regex(r'ad3-(.+?)\.swf',
834 video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND',
835 'flash player', fatal=False)
836 player = 'flash player %s' % player_version
837 else:
838 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
839 'html5 player', fatal=False)
840 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
841 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
842 (len(s), parts_sizes, url_data['itag'][0], player))
843 encrypted_sig = url_data['s'][0]
844 if age_gate:
845 signature = self._decrypt_signature_age_gate(encrypted_sig)
846 else:
847 signature = self._decrypt_signature(encrypted_sig)
848 url += '&signature=' + signature
849 if 'ratebypass' not in url:
850 url += '&ratebypass=yes'
851 url_map[url_data['itag'][0]] = url
852 video_url_list = self._get_video_url_list(url_map)
853 if not video_url_list:
854 return
855 elif video_info.get('hlsvp'):
856 manifest_url = video_info['hlsvp'][0]
857 url_map = self._extract_from_m3u8(manifest_url, video_id)
858 video_url_list = self._get_video_url_list(url_map)
859 if not video_url_list:
860 return
861
862 else:
863 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
864
865 results = []
866 for format_param, video_real_url in video_url_list:
867 # Extension
868 video_extension = self._video_extensions.get(format_param, 'flv')
869
870 video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
871 self._video_dimensions.get(format_param, '???'),
872 ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
873
874 results.append({
875 'id': video_id,
876 'url': video_real_url,
877 'uploader': video_uploader,
878 'uploader_id': video_uploader_id,
879 'upload_date': upload_date,
880 'title': video_title,
881 'ext': video_extension,
882 'format': video_format,
883 'thumbnail': video_thumbnail,
884 'description': video_description,
885 'player_url': player_url,
886 'subtitles': video_subtitles,
887 'duration': video_duration
888 })
889 return results
890
891 class YoutubePlaylistIE(InfoExtractor):
892 IE_DESC = u'YouTube.com playlists'
893 _VALID_URL = r"""(?:
894 (?:https?://)?
895 (?:\w+\.)?
896 youtube\.com/
897 (?:
898 (?:course|view_play_list|my_playlists|artist|playlist|watch)
899 \? (?:.*?&)*? (?:p|a|list)=
900 | p/
901 )
902 ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
903 .*
904 |
905 ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
906 )"""
907 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
908 _MAX_RESULTS = 50
909 IE_NAME = u'youtube:playlist'
910
911 @classmethod
912 def suitable(cls, url):
913 """Receives a URL and returns True if suitable for this IE."""
914 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
915
916 def _real_extract(self, url):
917 # Extract playlist id
918 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
919 if mobj is None:
920 raise ExtractorError(u'Invalid URL: %s' % url)
921
922 # Download playlist videos from API
923 playlist_id = mobj.group(1) or mobj.group(2)
924 videos = []
925
926 for page_num in itertools.count(1):
927 start_index = self._MAX_RESULTS * (page_num - 1) + 1
928 if start_index >= 1000:
929 self._downloader.report_warning(u'Max number of results reached')
930 break
931 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
932 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
933
934 try:
935 response = json.loads(page)
936 except ValueError as err:
937 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
938
939 if 'feed' not in response:
940 raise ExtractorError(u'Got a malformed response from YouTube API')
941 playlist_title = response['feed']['title']['$t']
942 if 'entry' not in response['feed']:
943 # Number of videos is a multiple of self._MAX_RESULTS
944 break
945
946 for entry in response['feed']['entry']:
947 index = entry['yt$position']['$t']
948 if 'media$group' in entry and 'yt$videoid' in entry['media$group']:
949 videos.append((
950 index,
951 'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t']
952 ))
953
954 videos = [v[1] for v in sorted(videos)]
955
956 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
957 return [self.playlist_result(url_results, playlist_id, playlist_title)]
958
959
960 class YoutubeChannelIE(InfoExtractor):
961 IE_DESC = u'YouTube.com channels'
962 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
963 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
964 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
965 _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'
966 IE_NAME = u'youtube:channel'
967
968 def extract_videos_from_page(self, page):
969 ids_in_page = []
970 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
971 if mobj.group(1) not in ids_in_page:
972 ids_in_page.append(mobj.group(1))
973 return ids_in_page
974
975 def _real_extract(self, url):
976 # Extract channel id
977 mobj = re.match(self._VALID_URL, url)
978 if mobj is None:
979 raise ExtractorError(u'Invalid URL: %s' % url)
980
981 # Download channel page
982 channel_id = mobj.group(1)
983 video_ids = []
984 pagenum = 1
985
986 url = self._TEMPLATE_URL % (channel_id, pagenum)
987 page = self._download_webpage(url, channel_id,
988 u'Downloading page #%s' % pagenum)
989
990 # Extract video identifiers
991 ids_in_page = self.extract_videos_from_page(page)
992 video_ids.extend(ids_in_page)
993
994 # Download any subsequent channel pages using the json-based channel_ajax query
995 if self._MORE_PAGES_INDICATOR in page:
996 for pagenum in itertools.count(1):
997 url = self._MORE_PAGES_URL % (pagenum, channel_id)
998 page = self._download_webpage(url, channel_id,
999 u'Downloading page #%s' % pagenum)
1000
1001 page = json.loads(page)
1002
1003 ids_in_page = self.extract_videos_from_page(page['content_html'])
1004 video_ids.extend(ids_in_page)
1005
1006 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1007 break
1008
1009 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1010
1011 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
1012 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
1013 return [self.playlist_result(url_entries, channel_id)]
1014
1015
1016 class YoutubeUserIE(InfoExtractor):
1017 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
1018 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?)|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1019 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
1020 _GDATA_PAGE_SIZE = 50
1021 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1022 IE_NAME = u'youtube:user'
1023
1024 @classmethod
1025 def suitable(cls, url):
1026 # Don't return True if the url can be extracted with other youtube
1027 # extractor, the regex would is too permissive and it would match.
1028 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1029 if any(ie.suitable(url) for ie in other_ies): return False
1030 else: return super(YoutubeUserIE, cls).suitable(url)
1031
1032 def _real_extract(self, url):
1033 # Extract username
1034 mobj = re.match(self._VALID_URL, url)
1035 if mobj is None:
1036 raise ExtractorError(u'Invalid URL: %s' % url)
1037
1038 username = mobj.group(1)
1039
1040 # Download video ids using YouTube Data API. Result size per
1041 # query is limited (currently to 50 videos) so we need to query
1042 # page by page until there are no video ids - it means we got
1043 # all of them.
1044
1045 video_ids = []
1046
1047 for pagenum in itertools.count(0):
1048 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1049
1050 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1051 page = self._download_webpage(gdata_url, username,
1052 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
1053
1054 try:
1055 response = json.loads(page)
1056 except ValueError as err:
1057 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
1058
1059 # Extract video identifiers
1060 ids_in_page = []
1061 for entry in response['feed']['entry']:
1062 ids_in_page.append(entry['id']['$t'].split('/')[-1])
1063 video_ids.extend(ids_in_page)
1064
1065 # A little optimization - if current page is not
1066 # "full", ie. does not contain PAGE_SIZE video ids then
1067 # we can assume that this page is the last one - there
1068 # are no more ids on further pages - no need to query
1069 # again.
1070
1071 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
1072 break
1073
1074 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
1075 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
1076 return [self.playlist_result(url_results, playlist_title = username)]
1077
1078 class YoutubeSearchIE(SearchInfoExtractor):
1079 IE_DESC = u'YouTube.com searches'
1080 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1081 _MAX_RESULTS = 1000
1082 IE_NAME = u'youtube:search'
1083 _SEARCH_KEY = 'ytsearch'
1084
1085 def report_download_page(self, query, pagenum):
1086 """Report attempt to download search page with given number."""
1087 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
1088
1089 def _get_n_results(self, query, n):
1090 """Get a specified number of results for a query"""
1091
1092 video_ids = []
1093 pagenum = 0
1094 limit = n
1095
1096 while (50 * pagenum) < limit:
1097 self.report_download_page(query, pagenum+1)
1098 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
1099 request = compat_urllib_request.Request(result_url)
1100 try:
1101 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
1102 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1103 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
1104 api_response = json.loads(data)['data']
1105
1106 if not 'items' in api_response:
1107 raise ExtractorError(u'[youtube] No video results')
1108
1109 new_ids = list(video['id'] for video in api_response['items'])
1110 video_ids += new_ids
1111
1112 limit = min(n, api_response['totalItems'])
1113 pagenum += 1
1114
1115 if len(video_ids) > n:
1116 video_ids = video_ids[:n]
1117 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
1118 return self.playlist_result(videos, query)
1119
1120
1121 class YoutubeShowIE(InfoExtractor):
1122 IE_DESC = u'YouTube.com (multi-season) shows'
1123 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
1124 IE_NAME = u'youtube:show'
1125
1126 def _real_extract(self, url):
1127 mobj = re.match(self._VALID_URL, url)
1128 show_name = mobj.group(1)
1129 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
1130 # There's one playlist for each season of the show
1131 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1132 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
1133 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
1134
1135
1136 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1137 """
1138 Base class for extractors that fetch info from
1139 http://www.youtube.com/feed_ajax
1140 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1141 """
1142 _LOGIN_REQUIRED = True
1143 _PAGING_STEP = 30
1144 # use action_load_personal_feed instead of action_load_system_feed
1145 _PERSONAL_FEED = False
1146
1147 @property
1148 def _FEED_TEMPLATE(self):
1149 action = 'action_load_system_feed'
1150 if self._PERSONAL_FEED:
1151 action = 'action_load_personal_feed'
1152 return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1153
1154 @property
1155 def IE_NAME(self):
1156 return u'youtube:%s' % self._FEED_NAME
1157
1158 def _real_initialize(self):
1159 self._login()
1160
1161 def _real_extract(self, url):
1162 feed_entries = []
1163 # The step argument is available only in 2.7 or higher
1164 for i in itertools.count(0):
1165 paging = i*self._PAGING_STEP
1166 info = self._download_webpage(self._FEED_TEMPLATE % paging,
1167 u'%s feed' % self._FEED_NAME,
1168 u'Downloading page %s' % i)
1169 info = json.loads(info)
1170 feed_html = info['feed_html']
1171 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1172 ids = orderedSet(m.group(1) for m in m_ids)
1173 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
1174 if info['paging'] is None:
1175 break
1176 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1177
1178 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1179 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
1180 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1181 _FEED_NAME = 'subscriptions'
1182 _PLAYLIST_TITLE = u'Youtube Subscriptions'
1183
1184 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1185 IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1186 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1187 _FEED_NAME = 'recommended'
1188 _PLAYLIST_TITLE = u'Youtube Recommended videos'
1189
1190 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1191 IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1192 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1193 _FEED_NAME = 'watch_later'
1194 _PLAYLIST_TITLE = u'Youtube Watch Later'
1195 _PAGING_STEP = 100
1196 _PERSONAL_FEED = True
1197
1198 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1199 IE_NAME = u'youtube:favorites'
1200 IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1201 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1202 _LOGIN_REQUIRED = True
1203
1204 def _real_extract(self, url):
1205 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1206 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
1207 return self.url_result(playlist_id, 'YoutubePlaylist')
fork of https://github.com/yt-dlp/yt-dlp/