9 from .common import InfoExtractor, SearchInfoExtractor
15 compat_urllib_request,
27 class YoutubeIE(InfoExtractor):
28 IE_DESC = u'YouTube.com'
31 (?:https?://)? # http(s):// (optional)
32 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
33 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
34 (?:.*?\#/)? # handle anchor (#/) redirect urls
35 (?: # the various things that can precede the ID:
36 (?:(?:v|embed|e)/) # v/ or embed/ or e/
37 |(?: # or the v= param in all its forms
38 (?:watch|movie(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
39 (?:\?|\#!?) # the params delimiter ? or # or #!
40 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
43 )? # optional -> youtube.com/xxxx is OK
44 )? # all until now is optional -> you can pass the naked ID
45 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
46 (?(1).+)? # if we found the ID, everything can follow
48 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
49 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
50 _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
51 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
52 _NETRC_MACHINE = 'youtube'
53 # Listed in order of quality
54 _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13']
55 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
87 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
88 u"file": u"BaW_jenozKc.mp4",
90 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
91 u"uploader": u"Philipp Hagemeister",
92 u"uploader_id": u"phihag",
93 u"upload_date": u"20121002",
94 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
98 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
99 u"file": u"1ltcDfZMA3U.flv",
100 u"note": u"Test VEVO video (#897)",
102 u"upload_date": u"20070518",
103 u"title": u"Maps - It Will Find You",
104 u"description": u"Music video by Maps performing It Will Find You.",
105 u"uploader": u"MuteUSA",
106 u"uploader_id": u"MuteUSA"
110 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
111 u"file": u"UxxajLWwzqY.mp4",
112 u"note": u"Test generic use_cipher_signature video (#897)",
114 u"upload_date": u"20120506",
115 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
116 u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
117 u"uploader": u"IconaPop",
118 u"uploader_id": u"IconaPop"
122 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
123 u"file": u"07FYdnEawAQ.mp4",
124 u"note": u"Test VEVO video with age protection (#956)",
126 u"upload_date": u"20130703",
127 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
128 u"description": u"md5:64249768eec3bc4276236606ea996373",
129 u"uploader": u"justintimberlakeVEVO",
130 u"uploader_id": u"justintimberlakeVEVO"
137 def suitable(cls, url):
138 """Receives a URL and returns True if suitable for this IE."""
139 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
140 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
142 def report_lang(self):
143 """Report attempt to set language."""
144 self.to_screen(u'Setting language')
146 def report_video_webpage_download(self, video_id):
147 """Report attempt to download video webpage."""
148 self.to_screen(u'%s: Downloading video webpage' % video_id)
150 def report_video_info_webpage_download(self, video_id):
151 """Report attempt to download video info webpage."""
152 self.to_screen(u'%s: Downloading video info webpage' % video_id)
154 def report_video_subtitles_download(self, video_id):
155 """Report attempt to download video info webpage."""
156 self.to_screen(u'%s: Checking available subtitles' % video_id)
158 def report_video_subtitles_request(self, video_id, sub_lang, format):
159 """Report attempt to download video info webpage."""
160 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
162 def report_video_subtitles_available(self, video_id, sub_lang_list):
163 """Report available subtitles."""
164 sub_lang = ",".join(list(sub_lang_list.keys()))
165 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
167 def report_information_extraction(self, video_id):
168 """Report attempt to extract video information."""
169 self.to_screen(u'%s: Extracting video information' % video_id)
171 def report_unavailable_format(self, video_id, format):
172 """Report extracted video URL."""
173 self.to_screen(u'%s: Format %s not available' % (video_id, format))
175 def report_rtmp_download(self):
176 """Indicate the download will use the RTMP protocol."""
177 self.to_screen(u'RTMP download detected')
179 def _decrypt_signature(self, s):
180 """Turn the encrypted s field into a working signature"""
183 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
185 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
187 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]
189 return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1]
191 return s[2:63] + s[82] + s[64:82] + s[63]
193 return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
195 return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26]
197 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:]
199 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]
201 return s[6] + s[3:6] + s[33] + s[7:24] + s[0] + s[25:33] + s[2] + s[34:53] + s[24] + s[54:81]
204 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
206 def _get_available_subtitles(self, video_id):
207 self.report_video_subtitles_download(video_id)
208 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
210 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
211 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
212 self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
214 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
215 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
216 if not sub_lang_list:
217 self._downloader.report_warning(u'video doesn\'t have subtitles')
221 def _list_available_subtitles(self, video_id):
222 sub_lang_list = self._get_available_subtitles(video_id)
223 self.report_video_subtitles_available(video_id, sub_lang_list)
225 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
227 Return the subtitle as a string or None if they are not found
229 self.report_video_subtitles_request(video_id, sub_lang, format)
230 params = compat_urllib_parse.urlencode({
236 url = 'http://www.youtube.com/api/timedtext?' + params
238 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
239 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
240 self._downloader.report_warning(u'unable to download video subtitles for %s: %s' % (sub_lang, compat_str(err)))
243 self._downloader.report_warning(u'Did not fetch video subtitles')
247 def _request_automatic_caption(self, video_id, webpage):
248 """We need the webpage for getting the captions url, pass it as an
249 argument to speed up the process."""
250 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
251 sub_format = self._downloader.params.get('subtitlesformat')
252 self.to_screen(u'%s: Looking for automatic captions' % video_id)
253 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
254 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
256 self._downloader.report_warning(err_msg)
258 player_config = json.loads(mobj.group(1))
260 args = player_config[u'args']
261 caption_url = args[u'ttsurl']
262 timestamp = args[u'timestamp']
263 params = compat_urllib_parse.urlencode({
270 subtitles_url = caption_url + '&' + params
271 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
272 return {sub_lang: sub}
273 # An extractor error can be raise by the download process if there are
274 # no automatic captions but there are subtitles
275 except (KeyError, ExtractorError):
276 self._downloader.report_warning(err_msg)
279 def _extract_subtitles(self, video_id):
281 Return a dictionary: {language: subtitles} or {} if the subtitles
284 sub_lang_list = self._get_available_subtitles(video_id)
285 sub_format = self._downloader.params.get('subtitlesformat')
286 if not sub_lang_list: #There was some error, it didn't get the available subtitles
288 if self._downloader.params.get('writesubtitles', False):
289 if self._downloader.params.get('subtitleslang', False):
290 sub_lang = self._downloader.params.get('subtitleslang')
291 elif 'en' in sub_lang_list:
294 sub_lang = list(sub_lang_list.keys())[0]
295 if not sub_lang in sub_lang_list:
296 self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
298 sub_lang_list = {sub_lang: sub_lang_list[sub_lang]}
300 for sub_lang in sub_lang_list:
301 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
303 subtitles[sub_lang] = subtitle
306 def _print_formats(self, formats):
307 print('Available formats:')
309 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
311 def _real_initialize(self):
312 if self._downloader is None:
316 request = compat_urllib_request.Request(self._LANG_URL)
319 compat_urllib_request.urlopen(request).read()
320 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
321 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
324 (username, password) = self._get_login_info()
326 # No authentication to be performed
330 request = compat_urllib_request.Request(self._LOGIN_URL)
332 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
333 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
334 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
339 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
341 galx = match.group(1)
343 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
349 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
353 u'PersistentCookie': u'yes',
355 u'bgresponse': u'js_disabled',
356 u'checkConnection': u'',
357 u'checkedDomains': u'youtube',
363 u'signIn': u'Sign in',
365 u'service': u'youtube',
369 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
371 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
372 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
373 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
376 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
377 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
378 self._downloader.report_warning(u'unable to log in: bad username or password')
380 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
381 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
387 'action_confirm': 'Confirm',
389 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
391 self.report_age_confirmation()
392 compat_urllib_request.urlopen(request).read().decode('utf-8')
393 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
394 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
396 def _extract_id(self, url):
397 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
399 raise ExtractorError(u'Invalid URL: %s' % url)
400 video_id = mobj.group(2)
403 def _real_extract(self, url):
404 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
405 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 ).')
407 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
408 mobj = re.search(self._NEXT_URL_RE, url)
410 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
411 video_id = self._extract_id(url)
414 self.report_video_webpage_download(video_id)
415 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
416 request = compat_urllib_request.Request(url)
418 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
419 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
420 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
422 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
424 # Attempt to extract SWF player URL
425 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
427 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
432 self.report_video_info_webpage_download(video_id)
433 if re.search(r'player-age-gate-content">', video_webpage) is not None:
434 self.report_age_confirmation()
436 # We simulate the access to the video from www.youtube.com/v/{video_id}
437 # this can be viewed without login into Youtube
438 data = compat_urllib_parse.urlencode({'video_id': video_id,
442 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
446 video_info_url = 'https://www.youtube.com/get_video_info?' + data
447 video_info_webpage = self._download_webpage(video_info_url, video_id,
449 errnote='unable to download video info webpage')
450 video_info = compat_parse_qs(video_info_webpage)
453 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
454 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
455 % (video_id, el_type))
456 video_info_webpage = self._download_webpage(video_info_url, video_id,
458 errnote='unable to download video info webpage')
459 video_info = compat_parse_qs(video_info_webpage)
460 if 'token' in video_info:
462 if 'token' not in video_info:
463 if 'reason' in video_info:
464 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
466 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
468 # Check for "rental" videos
469 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
470 raise ExtractorError(u'"rental" videos not supported')
472 # Start extracting information
473 self.report_information_extraction(video_id)
476 if 'author' not in video_info:
477 raise ExtractorError(u'Unable to extract uploader name')
478 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
481 video_uploader_id = None
482 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
484 video_uploader_id = mobj.group(1)
486 self._downloader.report_warning(u'unable to extract uploader nickname')
489 if 'title' not in video_info:
490 raise ExtractorError(u'Unable to extract video title')
491 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
494 # We try first to get a high quality image:
495 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
496 video_webpage, re.DOTALL)
497 if m_thumb is not None:
498 video_thumbnail = m_thumb.group(1)
499 elif 'thumbnail_url' not in video_info:
500 self._downloader.report_warning(u'unable to extract video thumbnail')
502 else: # don't panic if we can't find it
503 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
507 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
509 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
510 upload_date = unified_strdate(upload_date)
513 video_description = get_element_by_id("eow-description", video_webpage)
514 if video_description:
515 video_description = clean_html(video_description)
517 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
519 video_description = unescapeHTML(fd_mobj.group(1))
521 video_description = u''
524 video_subtitles = None
526 if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
527 video_subtitles = self._extract_subtitles(video_id)
528 elif self._downloader.params.get('writeautomaticsub', False):
529 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
531 if self._downloader.params.get('listsubtitles', False):
532 self._list_available_subtitles(video_id)
535 if 'length_seconds' not in video_info:
536 self._downloader.report_warning(u'unable to extract video duration')
539 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
541 # Decide which formats to download
542 req_format = self._downloader.params.get('format', None)
545 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
547 raise ValueError('Could not find vevo ID')
548 info = json.loads(mobj.group(1))
550 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
551 # this signatures are encrypted
552 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
554 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
555 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
559 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
560 self.report_rtmp_download()
561 video_url_list = [(None, video_info['conn'][0])]
562 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
563 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
564 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
566 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
567 url_data = compat_parse_qs(url_data_str)
568 if 'itag' in url_data and 'url' in url_data:
569 url = url_data['url'][0]
570 if 'sig' in url_data:
571 url += '&signature=' + url_data['sig'][0]
572 elif 's' in url_data:
573 if self._downloader.params.get('verbose'):
576 player_version = self._search_regex(r'ad3-(.+?)\.swf',
577 video_info['ad3_module'][0], 'flash player',
579 player = 'flash player %s' % player_version
581 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
582 'html5 player', fatal=False)
583 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
584 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
585 (len(s), parts_sizes, url_data['itag'][0], player))
586 signature = self._decrypt_signature(url_data['s'][0])
587 url += '&signature=' + signature
588 if 'ratebypass' not in url:
589 url += '&ratebypass=yes'
590 url_map[url_data['itag'][0]] = url
592 format_limit = self._downloader.params.get('format_limit', None)
593 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
594 if format_limit is not None and format_limit in available_formats:
595 format_list = available_formats[available_formats.index(format_limit):]
597 format_list = available_formats
598 existing_formats = [x for x in format_list if x in url_map]
599 if len(existing_formats) == 0:
600 raise ExtractorError(u'no known formats available for video')
601 if self._downloader.params.get('listformats', None):
602 self._print_formats(existing_formats)
604 if req_format is None or req_format == 'best':
605 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
606 elif req_format == 'worst':
607 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
608 elif req_format in ('-1', 'all'):
609 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
611 # Specific formats. We pick the first in a slash-delimeted sequence.
612 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
613 req_formats = req_format.split('/')
614 video_url_list = None
615 for rf in req_formats:
617 video_url_list = [(rf, url_map[rf])]
619 if video_url_list is None:
620 raise ExtractorError(u'requested format not available')
622 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
625 for format_param, video_real_url in video_url_list:
627 video_extension = self._video_extensions.get(format_param, 'flv')
629 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
630 self._video_dimensions.get(format_param, '???'))
634 'url': video_real_url,
635 'uploader': video_uploader,
636 'uploader_id': video_uploader_id,
637 'upload_date': upload_date,
638 'title': video_title,
639 'ext': video_extension,
640 'format': video_format,
641 'thumbnail': video_thumbnail,
642 'description': video_description,
643 'player_url': player_url,
644 'subtitles': video_subtitles,
645 'duration': video_duration
649 class YoutubePlaylistIE(InfoExtractor):
650 IE_DESC = u'YouTube.com playlists'
656 (?:course|view_play_list|my_playlists|artist|playlist|watch)
657 \? (?:.*?&)*? (?:p|a|list)=
660 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
663 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
665 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
667 IE_NAME = u'youtube:playlist'
670 def suitable(cls, url):
671 """Receives a URL and returns True if suitable for this IE."""
672 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
674 def _real_extract(self, url):
675 # Extract playlist id
676 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
678 raise ExtractorError(u'Invalid URL: %s' % url)
680 # Download playlist videos from API
681 playlist_id = mobj.group(1) or mobj.group(2)
686 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
687 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
690 response = json.loads(page)
691 except ValueError as err:
692 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
694 if 'feed' not in response:
695 raise ExtractorError(u'Got a malformed response from YouTube API')
696 playlist_title = response['feed']['title']['$t']
697 if 'entry' not in response['feed']:
698 # Number of videos is a multiple of self._MAX_RESULTS
701 for entry in response['feed']['entry']:
702 index = entry['yt$position']['$t']
703 if 'media$group' in entry and 'media$player' in entry['media$group']:
704 videos.append((index, entry['media$group']['media$player']['url']))
706 if len(response['feed']['entry']) < self._MAX_RESULTS:
710 videos = [v[1] for v in sorted(videos)]
712 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
713 return [self.playlist_result(url_results, playlist_id, playlist_title)]
716 class YoutubeChannelIE(InfoExtractor):
717 IE_DESC = u'YouTube.com channels'
718 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
719 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
720 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
721 _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
722 IE_NAME = u'youtube:channel'
724 def extract_videos_from_page(self, page):
726 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
727 if mobj.group(1) not in ids_in_page:
728 ids_in_page.append(mobj.group(1))
731 def _real_extract(self, url):
733 mobj = re.match(self._VALID_URL, url)
735 raise ExtractorError(u'Invalid URL: %s' % url)
737 # Download channel page
738 channel_id = mobj.group(1)
742 url = self._TEMPLATE_URL % (channel_id, pagenum)
743 page = self._download_webpage(url, channel_id,
744 u'Downloading page #%s' % pagenum)
746 # Extract video identifiers
747 ids_in_page = self.extract_videos_from_page(page)
748 video_ids.extend(ids_in_page)
750 # Download any subsequent channel pages using the json-based channel_ajax query
751 if self._MORE_PAGES_INDICATOR in page:
753 pagenum = pagenum + 1
755 url = self._MORE_PAGES_URL % (pagenum, channel_id)
756 page = self._download_webpage(url, channel_id,
757 u'Downloading page #%s' % pagenum)
759 page = json.loads(page)
761 ids_in_page = self.extract_videos_from_page(page['content_html'])
762 video_ids.extend(ids_in_page)
764 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
767 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
769 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
770 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
771 return [self.playlist_result(url_entries, channel_id)]
774 class YoutubeUserIE(InfoExtractor):
775 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
776 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
777 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
778 _GDATA_PAGE_SIZE = 50
779 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
780 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
781 IE_NAME = u'youtube:user'
783 def _real_extract(self, url):
785 mobj = re.match(self._VALID_URL, url)
787 raise ExtractorError(u'Invalid URL: %s' % url)
789 username = mobj.group(1)
791 # Download video ids using YouTube Data API. Result size per
792 # query is limited (currently to 50 videos) so we need to query
793 # page by page until there are no video ids - it means we got
800 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
802 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
803 page = self._download_webpage(gdata_url, username,
804 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
806 # Extract video identifiers
809 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
810 if mobj.group(1) not in ids_in_page:
811 ids_in_page.append(mobj.group(1))
813 video_ids.extend(ids_in_page)
815 # A little optimization - if current page is not
816 # "full", ie. does not contain PAGE_SIZE video ids then
817 # we can assume that this page is the last one - there
818 # are no more ids on further pages - no need to query
821 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
826 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
827 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
828 return [self.playlist_result(url_results, playlist_title = username)]
830 class YoutubeSearchIE(SearchInfoExtractor):
831 IE_DESC = u'YouTube.com searches'
832 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
834 IE_NAME = u'youtube:search'
835 _SEARCH_KEY = 'ytsearch'
837 def report_download_page(self, query, pagenum):
838 """Report attempt to download search page with given number."""
839 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
841 def _get_n_results(self, query, n):
842 """Get a specified number of results for a query"""
848 while (50 * pagenum) < limit:
849 self.report_download_page(query, pagenum+1)
850 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
851 request = compat_urllib_request.Request(result_url)
853 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
854 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
855 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
856 api_response = json.loads(data)['data']
858 if not 'items' in api_response:
859 raise ExtractorError(u'[youtube] No video results')
861 new_ids = list(video['id'] for video in api_response['items'])
864 limit = min(n, api_response['totalItems'])
867 if len(video_ids) > n:
868 video_ids = video_ids[:n]
869 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
870 return self.playlist_result(videos, query)
873 class YoutubeShowIE(InfoExtractor):
874 IE_DESC = u'YouTube.com (multi-season) shows'
875 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
876 IE_NAME = u'youtube:show'
878 def _real_extract(self, url):
879 mobj = re.match(self._VALID_URL, url)
880 show_name = mobj.group(1)
881 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
882 # There's one playlist for each season of the show
883 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
884 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
885 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
888 class YoutubeSubscriptionsIE(YoutubeIE):
889 """It's a subclass of YoutubeIE because we need to login"""
890 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
891 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
892 IE_NAME = u'youtube:subscriptions'
893 _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
896 # Overwrite YoutubeIE properties we don't want
899 def suitable(cls, url):
900 return re.match(cls._VALID_URL, url) is not None
902 def _real_initialize(self):
903 (username, password) = self._get_login_info()
905 raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True)
906 super(YoutubeSubscriptionsIE, self)._real_initialize()
908 def _real_extract(self, url):
910 # The step argument is available only in 2.7 or higher
911 for i in itertools.count(0):
912 paging = i*self._PAGING_STEP
913 info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
914 u'Downloading page %s' % i)
915 info = json.loads(info)
916 feed_html = info['feed_html']
917 m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html)
918 ids = orderedSet(m.group(1) for m in m_ids)
919 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
920 if info['paging'] is None:
922 return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')