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 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
158 _video_extensions = {
170 _video_dimensions = {
189 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
190 u"file": u"BaW_jenozKc.mp4",
192 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
193 u"uploader": u"Philipp Hagemeister",
194 u"uploader_id": u"phihag",
195 u"upload_date": u"20121002",
196 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
200 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
201 u"file": u"1ltcDfZMA3U.flv",
202 u"note": u"Test VEVO video (#897)",
204 u"upload_date": u"20070518",
205 u"title": u"Maps - It Will Find You",
206 u"description": u"Music video by Maps performing It Will Find You.",
207 u"uploader": u"MuteUSA",
208 u"uploader_id": u"MuteUSA"
212 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
213 u"file": u"UxxajLWwzqY.mp4",
214 u"note": u"Test generic use_cipher_signature video (#897)",
216 u"upload_date": u"20120506",
217 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
218 u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
219 u"uploader": u"IconaPop",
220 u"uploader_id": u"IconaPop"
224 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
225 u"file": u"07FYdnEawAQ.mp4",
226 u"note": u"Test VEVO video with age protection (#956)",
228 u"upload_date": u"20130703",
229 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
230 u"description": u"md5:64249768eec3bc4276236606ea996373",
231 u"uploader": u"justintimberlakeVEVO",
232 u"uploader_id": u"justintimberlakeVEVO"
239 def suitable(cls, url):
240 """Receives a URL and returns True if suitable for this IE."""
241 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
242 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
244 def report_video_webpage_download(self, video_id):
245 """Report attempt to download video webpage."""
246 self.to_screen(u'%s: Downloading video webpage' % video_id)
248 def report_video_info_webpage_download(self, video_id):
249 """Report attempt to download video info webpage."""
250 self.to_screen(u'%s: Downloading video info webpage' % video_id)
252 def report_video_subtitles_download(self, video_id):
253 """Report attempt to download video info webpage."""
254 self.to_screen(u'%s: Checking available subtitles' % video_id)
256 def report_video_subtitles_request(self, video_id, sub_lang, format):
257 """Report attempt to download video info webpage."""
258 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
260 def report_video_subtitles_available(self, video_id, sub_lang_list):
261 """Report available subtitles."""
262 sub_lang = ",".join(list(sub_lang_list.keys()))
263 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
265 def report_information_extraction(self, video_id):
266 """Report attempt to extract video information."""
267 self.to_screen(u'%s: Extracting video information' % video_id)
269 def report_unavailable_format(self, video_id, format):
270 """Report extracted video URL."""
271 self.to_screen(u'%s: Format %s not available' % (video_id, format))
273 def report_rtmp_download(self):
274 """Indicate the download will use the RTMP protocol."""
275 self.to_screen(u'RTMP download detected')
277 def _decrypt_signature(self, s):
278 """Turn the encrypted s field into a working signature"""
281 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
283 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
285 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]
287 return s[4:23] + s[86] + s[24:85]
289 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]
291 return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
293 return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26]
295 return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[53] + s[34:53] + s[24] + s[54:]
297 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]
299 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]
301 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]
304 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
306 def _decrypt_signature_age_gate(self, s):
307 # The videos with age protection use another player, so the algorithms
310 return s[2:63] + s[82] + s[64:82] + s[63]
312 # Fallback to the other algortihms
313 self._decrypt_signature(s)
316 def _get_available_subtitles(self, video_id):
317 self.report_video_subtitles_download(video_id)
318 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
320 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
321 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
322 return (u'unable to download video subtitles: %s' % compat_str(err), None)
323 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
324 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
325 if not sub_lang_list:
326 return (u'video doesn\'t have subtitles', None)
329 def _list_available_subtitles(self, video_id):
330 sub_lang_list = self._get_available_subtitles(video_id)
331 self.report_video_subtitles_available(video_id, sub_lang_list)
333 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
336 (error_message, sub_lang, sub)
338 self.report_video_subtitles_request(video_id, sub_lang, format)
339 params = compat_urllib_parse.urlencode({
345 url = 'http://www.youtube.com/api/timedtext?' + params
347 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
348 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
349 return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
351 return (u'Did not fetch video subtitles', None, None)
352 return (None, sub_lang, sub)
354 def _request_automatic_caption(self, video_id, webpage):
355 """We need the webpage for getting the captions url, pass it as an
356 argument to speed up the process."""
357 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
358 sub_format = self._downloader.params.get('subtitlesformat')
359 self.to_screen(u'%s: Looking for automatic captions' % video_id)
360 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
361 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
363 return [(err_msg, None, None)]
364 player_config = json.loads(mobj.group(1))
366 args = player_config[u'args']
367 caption_url = args[u'ttsurl']
368 timestamp = args[u'timestamp']
369 params = compat_urllib_parse.urlencode({
376 subtitles_url = caption_url + '&' + params
377 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
378 return [(None, sub_lang, sub)]
380 return [(err_msg, None, None)]
382 def _extract_subtitle(self, video_id):
384 Return a list with a tuple:
385 [(error_message, sub_lang, sub)]
387 sub_lang_list = self._get_available_subtitles(video_id)
388 sub_format = self._downloader.params.get('subtitlesformat')
389 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
390 return [(sub_lang_list[0], None, None)]
391 if self._downloader.params.get('subtitleslang', False):
392 sub_lang = self._downloader.params.get('subtitleslang')
393 elif 'en' in sub_lang_list:
396 sub_lang = list(sub_lang_list.keys())[0]
397 if not sub_lang in sub_lang_list:
398 return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
400 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
403 def _extract_all_subtitles(self, video_id):
404 sub_lang_list = self._get_available_subtitles(video_id)
405 sub_format = self._downloader.params.get('subtitlesformat')
406 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
407 return [(sub_lang_list[0], None, None)]
409 for sub_lang in sub_lang_list:
410 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
411 subtitles.append(subtitle)
414 def _print_formats(self, formats):
415 print('Available formats:')
417 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
419 def _extract_id(self, url):
420 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
422 raise ExtractorError(u'Invalid URL: %s' % url)
423 video_id = mobj.group(2)
426 def _real_extract(self, url):
427 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
428 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 ).')
430 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
431 mobj = re.search(self._NEXT_URL_RE, url)
433 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
434 video_id = self._extract_id(url)
437 self.report_video_webpage_download(video_id)
438 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
439 request = compat_urllib_request.Request(url)
441 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
442 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
443 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
445 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
447 # Attempt to extract SWF player URL
448 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
450 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
455 self.report_video_info_webpage_download(video_id)
456 if re.search(r'player-age-gate-content">', video_webpage) is not None:
457 self.report_age_confirmation()
459 # We simulate the access to the video from www.youtube.com/v/{video_id}
460 # this can be viewed without login into Youtube
461 data = compat_urllib_parse.urlencode({'video_id': video_id,
465 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
469 video_info_url = 'https://www.youtube.com/get_video_info?' + data
470 video_info_webpage = self._download_webpage(video_info_url, video_id,
472 errnote='unable to download video info webpage')
473 video_info = compat_parse_qs(video_info_webpage)
476 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
477 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
478 % (video_id, el_type))
479 video_info_webpage = self._download_webpage(video_info_url, video_id,
481 errnote='unable to download video info webpage')
482 video_info = compat_parse_qs(video_info_webpage)
483 if 'token' in video_info:
485 if 'token' not in video_info:
486 if 'reason' in video_info:
487 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
489 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
491 # Check for "rental" videos
492 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
493 raise ExtractorError(u'"rental" videos not supported')
495 # Start extracting information
496 self.report_information_extraction(video_id)
499 if 'author' not in video_info:
500 raise ExtractorError(u'Unable to extract uploader name')
501 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
504 video_uploader_id = None
505 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
507 video_uploader_id = mobj.group(1)
509 self._downloader.report_warning(u'unable to extract uploader nickname')
512 if 'title' not in video_info:
513 raise ExtractorError(u'Unable to extract video title')
514 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
517 # We try first to get a high quality image:
518 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
519 video_webpage, re.DOTALL)
520 if m_thumb is not None:
521 video_thumbnail = m_thumb.group(1)
522 elif 'thumbnail_url' not in video_info:
523 self._downloader.report_warning(u'unable to extract video thumbnail')
525 else: # don't panic if we can't find it
526 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
530 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
532 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
533 upload_date = unified_strdate(upload_date)
536 video_description = get_element_by_id("eow-description", video_webpage)
537 if video_description:
538 video_description = clean_html(video_description)
540 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
542 video_description = unescapeHTML(fd_mobj.group(1))
544 video_description = u''
547 video_subtitles = None
549 if self._downloader.params.get('writesubtitles', False):
550 video_subtitles = self._extract_subtitle(video_id)
552 (sub_error, sub_lang, sub) = video_subtitles[0]
554 self._downloader.report_warning(sub_error)
556 if self._downloader.params.get('writeautomaticsub', False):
557 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
558 (sub_error, sub_lang, sub) = video_subtitles[0]
560 self._downloader.report_warning(sub_error)
562 if self._downloader.params.get('allsubtitles', False):
563 video_subtitles = self._extract_all_subtitles(video_id)
564 for video_subtitle in video_subtitles:
565 (sub_error, sub_lang, sub) = video_subtitle
567 self._downloader.report_warning(sub_error)
569 if self._downloader.params.get('listsubtitles', False):
570 self._list_available_subtitles(video_id)
573 if 'length_seconds' not in video_info:
574 self._downloader.report_warning(u'unable to extract video duration')
577 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
579 # Decide which formats to download
580 req_format = self._downloader.params.get('format', None)
583 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
585 raise ValueError('Could not find vevo ID')
586 info = json.loads(mobj.group(1))
588 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
589 # this signatures are encrypted
590 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
592 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
593 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
597 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
598 self.report_rtmp_download()
599 video_url_list = [(None, video_info['conn'][0])]
600 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
601 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
602 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
604 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
605 url_data = compat_parse_qs(url_data_str)
606 if 'itag' in url_data and 'url' in url_data:
607 url = url_data['url'][0]
608 if 'sig' in url_data:
609 url += '&signature=' + url_data['sig'][0]
610 elif 's' in url_data:
611 if self._downloader.params.get('verbose'):
614 player_version = self._search_regex(r'ad3-(.+?)\.swf',
615 video_info['ad3_module'][0], 'flash player',
617 player = 'flash player %s' % player_version
619 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
620 'html5 player', fatal=False)
621 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
622 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
623 (len(s), parts_sizes, url_data['itag'][0], player))
624 encrypted_sig = url_data['s'][0]
626 signature = self._decrypt_signature_age_gate(encrypted_sig)
628 signature = self._decrypt_signature(encrypted_sig)
629 url += '&signature=' + signature
630 if 'ratebypass' not in url:
631 url += '&ratebypass=yes'
632 url_map[url_data['itag'][0]] = url
634 format_limit = self._downloader.params.get('format_limit', None)
635 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
636 if format_limit is not None and format_limit in available_formats:
637 format_list = available_formats[available_formats.index(format_limit):]
639 format_list = available_formats
640 existing_formats = [x for x in format_list if x in url_map]
641 if len(existing_formats) == 0:
642 raise ExtractorError(u'no known formats available for video')
643 if self._downloader.params.get('listformats', None):
644 self._print_formats(existing_formats)
646 if req_format is None or req_format == 'best':
647 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
648 elif req_format == 'worst':
649 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
650 elif req_format in ('-1', 'all'):
651 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
653 # Specific formats. We pick the first in a slash-delimeted sequence.
654 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
655 req_formats = req_format.split('/')
656 video_url_list = None
657 for rf in req_formats:
659 video_url_list = [(rf, url_map[rf])]
661 if video_url_list is None:
662 raise ExtractorError(u'requested format not available')
664 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
667 for format_param, video_real_url in video_url_list:
669 video_extension = self._video_extensions.get(format_param, 'flv')
671 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
672 self._video_dimensions.get(format_param, '???'))
676 'url': video_real_url,
677 'uploader': video_uploader,
678 'uploader_id': video_uploader_id,
679 'upload_date': upload_date,
680 'title': video_title,
681 'ext': video_extension,
682 'format': video_format,
683 'thumbnail': video_thumbnail,
684 'description': video_description,
685 'player_url': player_url,
686 'subtitles': video_subtitles,
687 'duration': video_duration
691 class YoutubePlaylistIE(InfoExtractor):
692 IE_DESC = u'YouTube.com playlists'
698 (?:course|view_play_list|my_playlists|artist|playlist|watch)
699 \? (?:.*?&)*? (?:p|a|list)=
702 ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
705 ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
707 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
709 IE_NAME = u'youtube:playlist'
712 def suitable(cls, url):
713 """Receives a URL and returns True if suitable for this IE."""
714 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
716 def _real_extract(self, url):
717 # Extract playlist id
718 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
720 raise ExtractorError(u'Invalid URL: %s' % url)
722 # Download playlist videos from API
723 playlist_id = mobj.group(1) or mobj.group(2)
726 for page_num in itertools.count(1):
727 start_index = self._MAX_RESULTS * (page_num - 1) + 1
728 if start_index >= 1000:
729 self._downloader.report_warning(u'Max number of results reached')
731 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
732 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
735 response = json.loads(page)
736 except ValueError as err:
737 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
739 if 'feed' not in response:
740 raise ExtractorError(u'Got a malformed response from YouTube API')
741 playlist_title = response['feed']['title']['$t']
742 if 'entry' not in response['feed']:
743 # Number of videos is a multiple of self._MAX_RESULTS
746 for entry in response['feed']['entry']:
747 index = entry['yt$position']['$t']
748 if 'media$group' in entry and 'media$player' in entry['media$group']:
749 videos.append((index, entry['media$group']['media$player']['url']))
751 videos = [v[1] for v in sorted(videos)]
753 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
754 return [self.playlist_result(url_results, playlist_id, playlist_title)]
757 class YoutubeChannelIE(InfoExtractor):
758 IE_DESC = u'YouTube.com channels'
759 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
760 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
761 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
762 _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'
763 IE_NAME = u'youtube:channel'
765 def extract_videos_from_page(self, page):
767 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
768 if mobj.group(1) not in ids_in_page:
769 ids_in_page.append(mobj.group(1))
772 def _real_extract(self, url):
774 mobj = re.match(self._VALID_URL, url)
776 raise ExtractorError(u'Invalid URL: %s' % url)
778 # Download channel page
779 channel_id = mobj.group(1)
783 url = self._TEMPLATE_URL % (channel_id, pagenum)
784 page = self._download_webpage(url, channel_id,
785 u'Downloading page #%s' % pagenum)
787 # Extract video identifiers
788 ids_in_page = self.extract_videos_from_page(page)
789 video_ids.extend(ids_in_page)
791 # Download any subsequent channel pages using the json-based channel_ajax query
792 if self._MORE_PAGES_INDICATOR in page:
793 for pagenum in itertools.count(1):
794 url = self._MORE_PAGES_URL % (pagenum, channel_id)
795 page = self._download_webpage(url, channel_id,
796 u'Downloading page #%s' % pagenum)
798 page = json.loads(page)
800 ids_in_page = self.extract_videos_from_page(page['content_html'])
801 video_ids.extend(ids_in_page)
803 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
806 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
808 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
809 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
810 return [self.playlist_result(url_entries, channel_id)]
813 class YoutubeUserIE(InfoExtractor):
814 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
815 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
816 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
817 _GDATA_PAGE_SIZE = 50
818 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
819 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
820 IE_NAME = u'youtube:user'
822 def _real_extract(self, url):
824 mobj = re.match(self._VALID_URL, url)
826 raise ExtractorError(u'Invalid URL: %s' % url)
828 username = mobj.group(1)
830 # Download video ids using YouTube Data API. Result size per
831 # query is limited (currently to 50 videos) so we need to query
832 # page by page until there are no video ids - it means we got
837 for pagenum in itertools.count(0):
838 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
840 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
841 page = self._download_webpage(gdata_url, username,
842 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
844 # Extract video identifiers
847 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
848 if mobj.group(1) not in ids_in_page:
849 ids_in_page.append(mobj.group(1))
851 video_ids.extend(ids_in_page)
853 # A little optimization - if current page is not
854 # "full", ie. does not contain PAGE_SIZE video ids then
855 # we can assume that this page is the last one - there
856 # are no more ids on further pages - no need to query
859 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
862 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
863 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
864 return [self.playlist_result(url_results, playlist_title = username)]
866 class YoutubeSearchIE(SearchInfoExtractor):
867 IE_DESC = u'YouTube.com searches'
868 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
870 IE_NAME = u'youtube:search'
871 _SEARCH_KEY = 'ytsearch'
873 def report_download_page(self, query, pagenum):
874 """Report attempt to download search page with given number."""
875 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
877 def _get_n_results(self, query, n):
878 """Get a specified number of results for a query"""
884 while (50 * pagenum) < limit:
885 self.report_download_page(query, pagenum+1)
886 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
887 request = compat_urllib_request.Request(result_url)
889 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
890 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
891 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
892 api_response = json.loads(data)['data']
894 if not 'items' in api_response:
895 raise ExtractorError(u'[youtube] No video results')
897 new_ids = list(video['id'] for video in api_response['items'])
900 limit = min(n, api_response['totalItems'])
903 if len(video_ids) > n:
904 video_ids = video_ids[:n]
905 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
906 return self.playlist_result(videos, query)
909 class YoutubeShowIE(InfoExtractor):
910 IE_DESC = u'YouTube.com (multi-season) shows'
911 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
912 IE_NAME = u'youtube:show'
914 def _real_extract(self, url):
915 mobj = re.match(self._VALID_URL, url)
916 show_name = mobj.group(1)
917 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
918 # There's one playlist for each season of the show
919 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
920 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
921 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
924 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
926 Base class for extractors that fetch info from
927 http://www.youtube.com/feed_ajax
928 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
930 _LOGIN_REQUIRED = True
932 # use action_load_personal_feed instead of action_load_system_feed
933 _PERSONAL_FEED = False
936 def _FEED_TEMPLATE(self):
937 action = 'action_load_system_feed'
938 if self._PERSONAL_FEED:
939 action = 'action_load_personal_feed'
940 return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
944 return u'youtube:%s' % self._FEED_NAME
946 def _real_initialize(self):
949 def _real_extract(self, url):
951 # The step argument is available only in 2.7 or higher
952 for i in itertools.count(0):
953 paging = i*self._PAGING_STEP
954 info = self._download_webpage(self._FEED_TEMPLATE % paging,
955 u'%s feed' % self._FEED_NAME,
956 u'Downloading page %s' % i)
957 info = json.loads(info)
958 feed_html = info['feed_html']
959 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
960 ids = orderedSet(m.group(1) for m in m_ids)
961 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
962 if info['paging'] is None:
964 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
966 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
967 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
968 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
969 _FEED_NAME = 'subscriptions'
970 _PLAYLIST_TITLE = u'Youtube Subscriptions'
972 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
973 IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
974 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
975 _FEED_NAME = 'recommended'
976 _PLAYLIST_TITLE = u'Youtube Recommended videos'
978 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
979 IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
980 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
981 _FEED_NAME = 'watch_later'
982 _PLAYLIST_TITLE = u'Youtube Watch Later'
984 _PERSONAL_FEED = True
986 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
987 IE_NAME = u'youtube:favorites'
988 IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
989 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?'
990 _LOGIN_REQUIRED = True
992 def _real_extract(self, url):
993 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
994 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
995 return self.url_result(playlist_id, 'YoutubePlaylist')