9 from .common import InfoExtractor, SearchInfoExtractor
15 compat_urllib_request,
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
35 def report_lang(self):
36 """Report attempt to set language."""
37 self.to_screen(u'Setting language')
39 def _set_language(self):
40 request = compat_urllib_request.Request(self._LANG_URL)
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))
50 (username, password) = self._get_login_info()
51 # No authentication to be performed
53 if self._LOGIN_REQUIRED:
54 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
57 request = compat_urllib_request.Request(self._LOGIN_URL)
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))
66 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
69 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
75 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
79 u'PersistentCookie': u'yes',
81 u'bgresponse': u'js_disabled',
82 u'checkConnection': u'',
83 u'checkedDomains': u'youtube',
89 u'signIn': u'Sign in',
91 u'service': u'youtube',
95 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
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)
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')
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))
111 def _confirm_age(self):
114 'action_confirm': 'Confirm',
116 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
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))
124 def _real_initialize(self):
125 if self._downloader is None:
127 if not self._set_language():
129 if not self._login():
133 class YoutubeIE(YoutubeBaseInfoExtractor):
134 IE_DESC = u'YouTube.com'
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)
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
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 '85', '84', '102', '83', '101', '82', '100',
160 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13',
161 '95', '94', '93', '92', '132', '151',
162 '85', '102', '84', '101', '83', '100', '82',
164 _video_extensions = {
185 # videos that use m3u8
194 _video_dimensions = {
224 _3d_itags = ['85', '84', '102', '83', '101', '82', '100']
228 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
229 u"file": u"BaW_jenozKc.mp4",
231 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
232 u"uploader": u"Philipp Hagemeister",
233 u"uploader_id": u"phihag",
234 u"upload_date": u"20121002",
235 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
239 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
240 u"file": u"1ltcDfZMA3U.flv",
241 u"note": u"Test VEVO video (#897)",
243 u"upload_date": u"20070518",
244 u"title": u"Maps - It Will Find You",
245 u"description": u"Music video by Maps performing It Will Find You.",
246 u"uploader": u"MuteUSA",
247 u"uploader_id": u"MuteUSA"
251 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
252 u"file": u"UxxajLWwzqY.mp4",
253 u"note": u"Test generic use_cipher_signature video (#897)",
255 u"upload_date": u"20120506",
256 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
257 u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
258 u"uploader": u"IconaPop",
259 u"uploader_id": u"IconaPop"
263 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
264 u"file": u"07FYdnEawAQ.mp4",
265 u"note": u"Test VEVO video with age protection (#956)",
267 u"upload_date": u"20130703",
268 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
269 u"description": u"md5:64249768eec3bc4276236606ea996373",
270 u"uploader": u"justintimberlakeVEVO",
271 u"uploader_id": u"justintimberlakeVEVO"
275 u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE',
276 u'file': u'TGi3HqYrWHE.mp4',
277 u'note': u'm3u8 video',
279 u'title': u'Triathlon - Men - London 2012 Olympic Games',
280 u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
281 u'uploader': u'olympic',
282 u'upload_date': u'20120807',
283 u'uploader_id': u'olympic',
286 u'skip_download': True,
293 def suitable(cls, url):
294 """Receives a URL and returns True if suitable for this IE."""
295 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
296 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
298 def report_video_webpage_download(self, video_id):
299 """Report attempt to download video webpage."""
300 self.to_screen(u'%s: Downloading video webpage' % video_id)
302 def report_video_info_webpage_download(self, video_id):
303 """Report attempt to download video info webpage."""
304 self.to_screen(u'%s: Downloading video info webpage' % video_id)
306 def report_video_subtitles_download(self, video_id):
307 """Report attempt to download video info webpage."""
308 self.to_screen(u'%s: Checking available subtitles' % video_id)
310 def report_video_subtitles_request(self, video_id, sub_lang, format):
311 """Report attempt to download video info webpage."""
312 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
314 def report_video_subtitles_available(self, video_id, sub_lang_list):
315 """Report available subtitles."""
316 sub_lang = ",".join(list(sub_lang_list.keys()))
317 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
319 def report_information_extraction(self, video_id):
320 """Report attempt to extract video information."""
321 self.to_screen(u'%s: Extracting video information' % video_id)
323 def report_unavailable_format(self, video_id, format):
324 """Report extracted video URL."""
325 self.to_screen(u'%s: Format %s not available' % (video_id, format))
327 def report_rtmp_download(self):
328 """Indicate the download will use the RTMP protocol."""
329 self.to_screen(u'RTMP download detected')
331 def _decrypt_signature(self, s):
332 """Turn the encrypted s field into a working signature"""
335 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
337 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
339 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]
341 return s[83:53:-1] + s[3] + s[52:40:-1] + s[86] + s[39:10:-1] + s[0] + s[9:3:-1] + s[53]
343 return s[83:85] + s[26] + s[79:46:-1] + s[85] + s[45:36:-1] + s[30] + s[35:30:-1] + s[46] + s[29:26:-1] + s[82] + s[25:1:-1]
345 return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
347 return s[83:27:-1] + s[0] + s[26:5:-1] + s[2:0:-1] + s[27]
349 return s[:15] + s[80] + s[16:80] + s[15]
351 return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34]
353 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]
355 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]
358 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
360 def _decrypt_signature_age_gate(self, s):
361 # The videos with age protection use another player, so the algorithms
364 return s[2:63] + s[82] + s[64:82] + s[63]
366 # Fallback to the other algortihms
367 return self._decrypt_signature(s)
370 def _get_available_subtitles(self, video_id):
371 self.report_video_subtitles_download(video_id)
372 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
374 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
375 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
376 return (u'unable to download video subtitles: %s' % compat_str(err), None)
377 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
378 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
379 if not sub_lang_list:
380 return (u'video doesn\'t have subtitles', None)
383 def _list_available_subtitles(self, video_id):
384 sub_lang_list = self._get_available_subtitles(video_id)
385 self.report_video_subtitles_available(video_id, sub_lang_list)
387 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
390 (error_message, sub_lang, sub)
392 self.report_video_subtitles_request(video_id, sub_lang, format)
393 params = compat_urllib_parse.urlencode({
399 url = 'http://www.youtube.com/api/timedtext?' + params
401 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
402 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
403 return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
405 return (u'Did not fetch video subtitles', None, None)
406 return (None, sub_lang, sub)
408 def _request_automatic_caption(self, video_id, webpage):
409 """We need the webpage for getting the captions url, pass it as an
410 argument to speed up the process."""
411 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
412 sub_format = self._downloader.params.get('subtitlesformat')
413 self.to_screen(u'%s: Looking for automatic captions' % video_id)
414 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
415 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
417 return [(err_msg, None, None)]
418 player_config = json.loads(mobj.group(1))
420 args = player_config[u'args']
421 caption_url = args[u'ttsurl']
422 timestamp = args[u'timestamp']
423 params = compat_urllib_parse.urlencode({
430 subtitles_url = caption_url + '&' + params
431 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
432 return [(None, sub_lang, sub)]
434 return [(err_msg, None, None)]
436 def _extract_subtitle(self, video_id):
438 Return a list with a tuple:
439 [(error_message, sub_lang, sub)]
441 sub_lang_list = self._get_available_subtitles(video_id)
442 sub_format = self._downloader.params.get('subtitlesformat')
443 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
444 return [(sub_lang_list[0], None, None)]
445 if self._downloader.params.get('subtitleslang', False):
446 sub_lang = self._downloader.params.get('subtitleslang')
447 elif 'en' in sub_lang_list:
450 sub_lang = list(sub_lang_list.keys())[0]
451 if not sub_lang in sub_lang_list:
452 return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
454 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
457 def _extract_all_subtitles(self, video_id):
458 sub_lang_list = self._get_available_subtitles(video_id)
459 sub_format = self._downloader.params.get('subtitlesformat')
460 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
461 return [(sub_lang_list[0], None, None)]
463 for sub_lang in sub_lang_list:
464 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
465 subtitles.append(subtitle)
468 def _print_formats(self, formats):
469 print('Available formats:')
471 print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
472 self._video_dimensions.get(x, '???'),
473 ' (3D)' if x in self._3d_itags else ''))
475 def _extract_id(self, url):
476 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
478 raise ExtractorError(u'Invalid URL: %s' % url)
479 video_id = mobj.group(2)
482 def _get_video_url_list(self, url_map):
484 Transform a dictionary in the format {itag:url} to a list of (itag, url)
485 with the requested formats.
487 req_format = self._downloader.params.get('format', None)
488 format_limit = self._downloader.params.get('format_limit', None)
489 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
490 if format_limit is not None and format_limit in available_formats:
491 format_list = available_formats[available_formats.index(format_limit):]
493 format_list = available_formats
494 existing_formats = [x for x in format_list if x in url_map]
495 if len(existing_formats) == 0:
496 raise ExtractorError(u'no known formats available for video')
497 if self._downloader.params.get('listformats', None):
498 self._print_formats(existing_formats)
500 if req_format is None or req_format == 'best':
501 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
502 elif req_format == 'worst':
503 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
504 elif req_format in ('-1', 'all'):
505 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
507 # Specific formats. We pick the first in a slash-delimeted sequence.
508 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
509 req_formats = req_format.split('/')
510 video_url_list = None
511 for rf in req_formats:
513 video_url_list = [(rf, url_map[rf])]
515 if video_url_list is None:
516 raise ExtractorError(u'requested format not available')
517 return video_url_list
519 def _extract_from_m3u8(self, manifest_url, video_id):
521 def _get_urls(_manifest):
522 lines = _manifest.split('\n')
523 urls = filter(lambda l: l and not l.startswith('#'),
526 manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
527 formats_urls = _get_urls(manifest)
528 for format_url in formats_urls:
529 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
530 url_map[itag] = format_url
533 def _real_extract(self, url):
534 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
535 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 ).')
537 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
538 mobj = re.search(self._NEXT_URL_RE, url)
540 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
541 video_id = self._extract_id(url)
544 self.report_video_webpage_download(video_id)
545 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
546 request = compat_urllib_request.Request(url)
548 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
549 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
550 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
552 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
554 # Attempt to extract SWF player URL
555 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
557 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
562 self.report_video_info_webpage_download(video_id)
563 if re.search(r'player-age-gate-content">', video_webpage) is not None:
564 self.report_age_confirmation()
566 # We simulate the access to the video from www.youtube.com/v/{video_id}
567 # this can be viewed without login into Youtube
568 data = compat_urllib_parse.urlencode({'video_id': video_id,
572 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
576 video_info_url = 'https://www.youtube.com/get_video_info?' + data
577 video_info_webpage = self._download_webpage(video_info_url, video_id,
579 errnote='unable to download video info webpage')
580 video_info = compat_parse_qs(video_info_webpage)
583 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
584 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
585 % (video_id, el_type))
586 video_info_webpage = self._download_webpage(video_info_url, video_id,
588 errnote='unable to download video info webpage')
589 video_info = compat_parse_qs(video_info_webpage)
590 if 'token' in video_info:
592 if 'token' not in video_info:
593 if 'reason' in video_info:
594 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
596 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
598 # Check for "rental" videos
599 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
600 raise ExtractorError(u'"rental" videos not supported')
602 # Start extracting information
603 self.report_information_extraction(video_id)
606 if 'author' not in video_info:
607 raise ExtractorError(u'Unable to extract uploader name')
608 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
611 video_uploader_id = None
612 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
614 video_uploader_id = mobj.group(1)
616 self._downloader.report_warning(u'unable to extract uploader nickname')
619 if 'title' not in video_info:
620 raise ExtractorError(u'Unable to extract video title')
621 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
624 # We try first to get a high quality image:
625 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
626 video_webpage, re.DOTALL)
627 if m_thumb is not None:
628 video_thumbnail = m_thumb.group(1)
629 elif 'thumbnail_url' not in video_info:
630 self._downloader.report_warning(u'unable to extract video thumbnail')
632 else: # don't panic if we can't find it
633 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
637 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
639 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
640 upload_date = unified_strdate(upload_date)
643 video_description = get_element_by_id("eow-description", video_webpage)
644 if video_description:
645 video_description = clean_html(video_description)
647 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
649 video_description = unescapeHTML(fd_mobj.group(1))
651 video_description = u''
654 video_subtitles = None
656 if self._downloader.params.get('writesubtitles', False):
657 video_subtitles = self._extract_subtitle(video_id)
659 (sub_error, sub_lang, sub) = video_subtitles[0]
661 self._downloader.report_warning(sub_error)
663 if self._downloader.params.get('writeautomaticsub', False):
664 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
665 (sub_error, sub_lang, sub) = video_subtitles[0]
667 self._downloader.report_warning(sub_error)
669 if self._downloader.params.get('allsubtitles', False):
670 video_subtitles = self._extract_all_subtitles(video_id)
671 for video_subtitle in video_subtitles:
672 (sub_error, sub_lang, sub) = video_subtitle
674 self._downloader.report_warning(sub_error)
676 if self._downloader.params.get('listsubtitles', False):
677 self._list_available_subtitles(video_id)
680 if 'length_seconds' not in video_info:
681 self._downloader.report_warning(u'unable to extract video duration')
684 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
686 # Decide which formats to download
689 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
691 raise ValueError('Could not find vevo ID')
692 info = json.loads(mobj.group(1))
694 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
695 # this signatures are encrypted
696 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
698 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
699 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
703 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
704 self.report_rtmp_download()
705 video_url_list = [(None, video_info['conn'][0])]
706 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
707 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
708 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
710 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
711 url_data = compat_parse_qs(url_data_str)
712 if 'itag' in url_data and 'url' in url_data:
713 url = url_data['url'][0]
714 if 'sig' in url_data:
715 url += '&signature=' + url_data['sig'][0]
716 elif 's' in url_data:
717 if self._downloader.params.get('verbose'):
720 player_version = self._search_regex(r'ad3-(.+?)\.swf',
721 video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND',
722 'flash player', fatal=False)
723 player = 'flash player %s' % player_version
725 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
726 'html5 player', fatal=False)
727 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
728 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
729 (len(s), parts_sizes, url_data['itag'][0], player))
730 encrypted_sig = url_data['s'][0]
732 signature = self._decrypt_signature_age_gate(encrypted_sig)
734 signature = self._decrypt_signature(encrypted_sig)
735 url += '&signature=' + signature
736 if 'ratebypass' not in url:
737 url += '&ratebypass=yes'
738 url_map[url_data['itag'][0]] = url
739 video_url_list = self._get_video_url_list(url_map)
740 if not video_url_list:
742 elif video_info.get('hlsvp'):
743 manifest_url = video_info['hlsvp'][0]
744 url_map = self._extract_from_m3u8(manifest_url, video_id)
745 video_url_list = self._get_video_url_list(url_map)
746 if not video_url_list:
750 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
753 for format_param, video_real_url in video_url_list:
755 video_extension = self._video_extensions.get(format_param, 'flv')
757 video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
758 self._video_dimensions.get(format_param, '???'),
759 ' (3D)' if format_param in self._3d_itags else '')
763 'url': video_real_url,
764 'uploader': video_uploader,
765 'uploader_id': video_uploader_id,
766 'upload_date': upload_date,
767 'title': video_title,
768 'ext': video_extension,
769 'format': video_format,
770 'thumbnail': video_thumbnail,
771 'description': video_description,
772 'player_url': player_url,
773 'subtitles': video_subtitles,
774 'duration': video_duration
778 class YoutubePlaylistIE(InfoExtractor):
779 IE_DESC = u'YouTube.com playlists'
785 (?:course|view_play_list|my_playlists|artist|playlist|watch)
786 \? (?:.*?&)*? (?:p|a|list)=
789 ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
792 ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
794 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
796 IE_NAME = u'youtube:playlist'
799 def suitable(cls, url):
800 """Receives a URL and returns True if suitable for this IE."""
801 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
803 def _real_extract(self, url):
804 # Extract playlist id
805 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
807 raise ExtractorError(u'Invalid URL: %s' % url)
809 # Download playlist videos from API
810 playlist_id = mobj.group(1) or mobj.group(2)
813 for page_num in itertools.count(1):
814 start_index = self._MAX_RESULTS * (page_num - 1) + 1
815 if start_index >= 1000:
816 self._downloader.report_warning(u'Max number of results reached')
818 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
819 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
822 response = json.loads(page)
823 except ValueError as err:
824 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
826 if 'feed' not in response:
827 raise ExtractorError(u'Got a malformed response from YouTube API')
828 playlist_title = response['feed']['title']['$t']
829 if 'entry' not in response['feed']:
830 # Number of videos is a multiple of self._MAX_RESULTS
833 for entry in response['feed']['entry']:
834 index = entry['yt$position']['$t']
835 if 'media$group' in entry and 'media$player' in entry['media$group']:
836 videos.append((index, entry['media$group']['media$player']['url']))
838 videos = [v[1] for v in sorted(videos)]
840 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
841 return [self.playlist_result(url_results, playlist_id, playlist_title)]
844 class YoutubeChannelIE(InfoExtractor):
845 IE_DESC = u'YouTube.com channels'
846 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
847 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
848 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
849 _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'
850 IE_NAME = u'youtube:channel'
852 def extract_videos_from_page(self, page):
854 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
855 if mobj.group(1) not in ids_in_page:
856 ids_in_page.append(mobj.group(1))
859 def _real_extract(self, url):
861 mobj = re.match(self._VALID_URL, url)
863 raise ExtractorError(u'Invalid URL: %s' % url)
865 # Download channel page
866 channel_id = mobj.group(1)
870 url = self._TEMPLATE_URL % (channel_id, pagenum)
871 page = self._download_webpage(url, channel_id,
872 u'Downloading page #%s' % pagenum)
874 # Extract video identifiers
875 ids_in_page = self.extract_videos_from_page(page)
876 video_ids.extend(ids_in_page)
878 # Download any subsequent channel pages using the json-based channel_ajax query
879 if self._MORE_PAGES_INDICATOR in page:
880 for pagenum in itertools.count(1):
881 url = self._MORE_PAGES_URL % (pagenum, channel_id)
882 page = self._download_webpage(url, channel_id,
883 u'Downloading page #%s' % pagenum)
885 page = json.loads(page)
887 ids_in_page = self.extract_videos_from_page(page['content_html'])
888 video_ids.extend(ids_in_page)
890 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
893 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
895 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
896 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
897 return [self.playlist_result(url_entries, channel_id)]
900 class YoutubeUserIE(InfoExtractor):
901 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
902 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
903 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
904 _GDATA_PAGE_SIZE = 50
905 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
906 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
907 IE_NAME = u'youtube:user'
909 def _real_extract(self, url):
911 mobj = re.match(self._VALID_URL, url)
913 raise ExtractorError(u'Invalid URL: %s' % url)
915 username = mobj.group(1)
917 # Download video ids using YouTube Data API. Result size per
918 # query is limited (currently to 50 videos) so we need to query
919 # page by page until there are no video ids - it means we got
924 for pagenum in itertools.count(0):
925 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
927 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
928 page = self._download_webpage(gdata_url, username,
929 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
931 # Extract video identifiers
934 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
935 if mobj.group(1) not in ids_in_page:
936 ids_in_page.append(mobj.group(1))
938 video_ids.extend(ids_in_page)
940 # A little optimization - if current page is not
941 # "full", ie. does not contain PAGE_SIZE video ids then
942 # we can assume that this page is the last one - there
943 # are no more ids on further pages - no need to query
946 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
949 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
950 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
951 return [self.playlist_result(url_results, playlist_title = username)]
953 class YoutubeSearchIE(SearchInfoExtractor):
954 IE_DESC = u'YouTube.com searches'
955 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
957 IE_NAME = u'youtube:search'
958 _SEARCH_KEY = 'ytsearch'
960 def report_download_page(self, query, pagenum):
961 """Report attempt to download search page with given number."""
962 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
964 def _get_n_results(self, query, n):
965 """Get a specified number of results for a query"""
971 while (50 * pagenum) < limit:
972 self.report_download_page(query, pagenum+1)
973 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
974 request = compat_urllib_request.Request(result_url)
976 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
977 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
978 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
979 api_response = json.loads(data)['data']
981 if not 'items' in api_response:
982 raise ExtractorError(u'[youtube] No video results')
984 new_ids = list(video['id'] for video in api_response['items'])
987 limit = min(n, api_response['totalItems'])
990 if len(video_ids) > n:
991 video_ids = video_ids[:n]
992 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
993 return self.playlist_result(videos, query)
996 class YoutubeShowIE(InfoExtractor):
997 IE_DESC = u'YouTube.com (multi-season) shows'
998 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
999 IE_NAME = u'youtube:show'
1001 def _real_extract(self, url):
1002 mobj = re.match(self._VALID_URL, url)
1003 show_name = mobj.group(1)
1004 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
1005 # There's one playlist for each season of the show
1006 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1007 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
1008 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
1011 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1013 Base class for extractors that fetch info from
1014 http://www.youtube.com/feed_ajax
1015 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1017 _LOGIN_REQUIRED = True
1019 # use action_load_personal_feed instead of action_load_system_feed
1020 _PERSONAL_FEED = False
1023 def _FEED_TEMPLATE(self):
1024 action = 'action_load_system_feed'
1025 if self._PERSONAL_FEED:
1026 action = 'action_load_personal_feed'
1027 return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1031 return u'youtube:%s' % self._FEED_NAME
1033 def _real_initialize(self):
1036 def _real_extract(self, url):
1038 # The step argument is available only in 2.7 or higher
1039 for i in itertools.count(0):
1040 paging = i*self._PAGING_STEP
1041 info = self._download_webpage(self._FEED_TEMPLATE % paging,
1042 u'%s feed' % self._FEED_NAME,
1043 u'Downloading page %s' % i)
1044 info = json.loads(info)
1045 feed_html = info['feed_html']
1046 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1047 ids = orderedSet(m.group(1) for m in m_ids)
1048 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
1049 if info['paging'] is None:
1051 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1053 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1054 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
1055 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1056 _FEED_NAME = 'subscriptions'
1057 _PLAYLIST_TITLE = u'Youtube Subscriptions'
1059 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1060 IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1061 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1062 _FEED_NAME = 'recommended'
1063 _PLAYLIST_TITLE = u'Youtube Recommended videos'
1065 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1066 IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1067 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1068 _FEED_NAME = 'watch_later'
1069 _PLAYLIST_TITLE = u'Youtube Watch Later'
1071 _PERSONAL_FEED = True
1073 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1074 IE_NAME = u'youtube:favorites'
1075 IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1076 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
1077 _LOGIN_REQUIRED = True
1079 def _real_extract(self, url):
1080 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1081 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
1082 return self.url_result(playlist_id, 'YoutubePlaylist')