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}
274 self._downloader.report_warning(err_msg)
277 def _extract_subtitles(self, video_id):
279 Return a dictionary: {language: subtitles} or {} if the subtitles
282 sub_lang_list = self._get_available_subtitles(video_id)
283 sub_format = self._downloader.params.get('subtitlesformat')
284 if not sub_lang_list: #There was some error, it didn't get the available subtitles
286 if self._downloader.params.get('writesubtitles', False):
287 if self._downloader.params.get('subtitleslang', False):
288 sub_lang = self._downloader.params.get('subtitleslang')
289 elif 'en' in sub_lang_list:
292 sub_lang = list(sub_lang_list.keys())[0]
293 if not sub_lang in sub_lang_list:
294 self._downloader.report_warning(u'no closed captions found in the specified language "%s"' % sub_lang)
296 sub_lang_list = {sub_lang: sub_lang_list[sub_lang]}
298 for sub_lang in sub_lang_list:
299 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
301 subtitles[sub_lang] = subtitle
304 def _print_formats(self, formats):
305 print('Available formats:')
307 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
309 def _real_initialize(self):
310 if self._downloader is None:
314 request = compat_urllib_request.Request(self._LANG_URL)
317 compat_urllib_request.urlopen(request).read()
318 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
319 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
322 (username, password) = self._get_login_info()
324 # No authentication to be performed
328 request = compat_urllib_request.Request(self._LOGIN_URL)
330 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
331 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
332 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
337 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
339 galx = match.group(1)
341 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
347 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
351 u'PersistentCookie': u'yes',
353 u'bgresponse': u'js_disabled',
354 u'checkConnection': u'',
355 u'checkedDomains': u'youtube',
361 u'signIn': u'Sign in',
363 u'service': u'youtube',
367 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
369 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
370 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
371 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
374 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
375 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
376 self._downloader.report_warning(u'unable to log in: bad username or password')
378 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
379 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
385 'action_confirm': 'Confirm',
387 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
389 self.report_age_confirmation()
390 compat_urllib_request.urlopen(request).read().decode('utf-8')
391 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
392 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
394 def _extract_id(self, url):
395 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
397 raise ExtractorError(u'Invalid URL: %s' % url)
398 video_id = mobj.group(2)
401 def _real_extract(self, url):
402 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
403 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 ).')
405 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
406 mobj = re.search(self._NEXT_URL_RE, url)
408 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
409 video_id = self._extract_id(url)
412 self.report_video_webpage_download(video_id)
413 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
414 request = compat_urllib_request.Request(url)
416 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
417 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
418 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
420 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
422 # Attempt to extract SWF player URL
423 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
425 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
430 self.report_video_info_webpage_download(video_id)
431 if re.search(r'player-age-gate-content">', video_webpage) is not None:
432 self.report_age_confirmation()
434 # We simulate the access to the video from www.youtube.com/v/{video_id}
435 # this can be viewed without login into Youtube
436 data = compat_urllib_parse.urlencode({'video_id': video_id,
440 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
444 video_info_url = 'https://www.youtube.com/get_video_info?' + data
445 video_info_webpage = self._download_webpage(video_info_url, video_id,
447 errnote='unable to download video info webpage')
448 video_info = compat_parse_qs(video_info_webpage)
451 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
452 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
453 % (video_id, el_type))
454 video_info_webpage = self._download_webpage(video_info_url, video_id,
456 errnote='unable to download video info webpage')
457 video_info = compat_parse_qs(video_info_webpage)
458 if 'token' in video_info:
460 if 'token' not in video_info:
461 if 'reason' in video_info:
462 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
464 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
466 # Check for "rental" videos
467 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
468 raise ExtractorError(u'"rental" videos not supported')
470 # Start extracting information
471 self.report_information_extraction(video_id)
474 if 'author' not in video_info:
475 raise ExtractorError(u'Unable to extract uploader name')
476 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
479 video_uploader_id = None
480 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
482 video_uploader_id = mobj.group(1)
484 self._downloader.report_warning(u'unable to extract uploader nickname')
487 if 'title' not in video_info:
488 raise ExtractorError(u'Unable to extract video title')
489 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
492 # We try first to get a high quality image:
493 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
494 video_webpage, re.DOTALL)
495 if m_thumb is not None:
496 video_thumbnail = m_thumb.group(1)
497 elif 'thumbnail_url' not in video_info:
498 self._downloader.report_warning(u'unable to extract video thumbnail')
500 else: # don't panic if we can't find it
501 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
505 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
507 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
508 upload_date = unified_strdate(upload_date)
511 video_description = get_element_by_id("eow-description", video_webpage)
512 if video_description:
513 video_description = clean_html(video_description)
515 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
517 video_description = unescapeHTML(fd_mobj.group(1))
519 video_description = u''
522 video_subtitles = None
524 if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
525 video_subtitles = self._extract_subtitles(video_id)
526 elif self._downloader.params.get('writeautomaticsub', False):
527 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
529 if self._downloader.params.get('listsubtitles', False):
530 self._list_available_subtitles(video_id)
533 if 'length_seconds' not in video_info:
534 self._downloader.report_warning(u'unable to extract video duration')
537 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
539 # Decide which formats to download
540 req_format = self._downloader.params.get('format', None)
543 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
545 raise ValueError('Could not find vevo ID')
546 info = json.loads(mobj.group(1))
548 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
549 # this signatures are encrypted
550 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
552 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
553 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
557 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
558 self.report_rtmp_download()
559 video_url_list = [(None, video_info['conn'][0])]
560 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
561 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
562 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
564 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
565 url_data = compat_parse_qs(url_data_str)
566 if 'itag' in url_data and 'url' in url_data:
567 url = url_data['url'][0]
568 if 'sig' in url_data:
569 url += '&signature=' + url_data['sig'][0]
570 elif 's' in url_data:
571 if self._downloader.params.get('verbose'):
574 player_version = self._search_regex(r'ad3-(.+?)\.swf',
575 video_info['ad3_module'][0], 'flash player',
577 player = 'flash player %s' % player_version
579 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
580 'html5 player', fatal=False)
581 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
582 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
583 (len(s), parts_sizes, url_data['itag'][0], player))
584 signature = self._decrypt_signature(url_data['s'][0])
585 url += '&signature=' + signature
586 if 'ratebypass' not in url:
587 url += '&ratebypass=yes'
588 url_map[url_data['itag'][0]] = url
590 format_limit = self._downloader.params.get('format_limit', None)
591 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
592 if format_limit is not None and format_limit in available_formats:
593 format_list = available_formats[available_formats.index(format_limit):]
595 format_list = available_formats
596 existing_formats = [x for x in format_list if x in url_map]
597 if len(existing_formats) == 0:
598 raise ExtractorError(u'no known formats available for video')
599 if self._downloader.params.get('listformats', None):
600 self._print_formats(existing_formats)
602 if req_format is None or req_format == 'best':
603 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
604 elif req_format == 'worst':
605 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
606 elif req_format in ('-1', 'all'):
607 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
609 # Specific formats. We pick the first in a slash-delimeted sequence.
610 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
611 req_formats = req_format.split('/')
612 video_url_list = None
613 for rf in req_formats:
615 video_url_list = [(rf, url_map[rf])]
617 if video_url_list is None:
618 raise ExtractorError(u'requested format not available')
620 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
623 for format_param, video_real_url in video_url_list:
625 video_extension = self._video_extensions.get(format_param, 'flv')
627 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
628 self._video_dimensions.get(format_param, '???'))
632 'url': video_real_url,
633 'uploader': video_uploader,
634 'uploader_id': video_uploader_id,
635 'upload_date': upload_date,
636 'title': video_title,
637 'ext': video_extension,
638 'format': video_format,
639 'thumbnail': video_thumbnail,
640 'description': video_description,
641 'player_url': player_url,
642 'subtitles': video_subtitles,
643 'duration': video_duration
647 class YoutubePlaylistIE(InfoExtractor):
648 IE_DESC = u'YouTube.com playlists'
654 (?:course|view_play_list|my_playlists|artist|playlist|watch)
655 \? (?:.*?&)*? (?:p|a|list)=
658 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
661 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
663 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
665 IE_NAME = u'youtube:playlist'
668 def suitable(cls, url):
669 """Receives a URL and returns True if suitable for this IE."""
670 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
672 def _real_extract(self, url):
673 # Extract playlist id
674 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
676 raise ExtractorError(u'Invalid URL: %s' % url)
678 # Download playlist videos from API
679 playlist_id = mobj.group(1) or mobj.group(2)
684 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
685 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
688 response = json.loads(page)
689 except ValueError as err:
690 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
692 if 'feed' not in response:
693 raise ExtractorError(u'Got a malformed response from YouTube API')
694 playlist_title = response['feed']['title']['$t']
695 if 'entry' not in response['feed']:
696 # Number of videos is a multiple of self._MAX_RESULTS
699 for entry in response['feed']['entry']:
700 index = entry['yt$position']['$t']
701 if 'media$group' in entry and 'media$player' in entry['media$group']:
702 videos.append((index, entry['media$group']['media$player']['url']))
704 if len(response['feed']['entry']) < self._MAX_RESULTS:
708 videos = [v[1] for v in sorted(videos)]
710 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
711 return [self.playlist_result(url_results, playlist_id, playlist_title)]
714 class YoutubeChannelIE(InfoExtractor):
715 IE_DESC = u'YouTube.com channels'
716 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
717 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
718 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
719 _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'
720 IE_NAME = u'youtube:channel'
722 def extract_videos_from_page(self, page):
724 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
725 if mobj.group(1) not in ids_in_page:
726 ids_in_page.append(mobj.group(1))
729 def _real_extract(self, url):
731 mobj = re.match(self._VALID_URL, url)
733 raise ExtractorError(u'Invalid URL: %s' % url)
735 # Download channel page
736 channel_id = mobj.group(1)
740 url = self._TEMPLATE_URL % (channel_id, pagenum)
741 page = self._download_webpage(url, channel_id,
742 u'Downloading page #%s' % pagenum)
744 # Extract video identifiers
745 ids_in_page = self.extract_videos_from_page(page)
746 video_ids.extend(ids_in_page)
748 # Download any subsequent channel pages using the json-based channel_ajax query
749 if self._MORE_PAGES_INDICATOR in page:
751 pagenum = pagenum + 1
753 url = self._MORE_PAGES_URL % (pagenum, channel_id)
754 page = self._download_webpage(url, channel_id,
755 u'Downloading page #%s' % pagenum)
757 page = json.loads(page)
759 ids_in_page = self.extract_videos_from_page(page['content_html'])
760 video_ids.extend(ids_in_page)
762 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
765 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
767 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
768 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
769 return [self.playlist_result(url_entries, channel_id)]
772 class YoutubeUserIE(InfoExtractor):
773 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
774 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
775 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
776 _GDATA_PAGE_SIZE = 50
777 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
778 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
779 IE_NAME = u'youtube:user'
781 def _real_extract(self, url):
783 mobj = re.match(self._VALID_URL, url)
785 raise ExtractorError(u'Invalid URL: %s' % url)
787 username = mobj.group(1)
789 # Download video ids using YouTube Data API. Result size per
790 # query is limited (currently to 50 videos) so we need to query
791 # page by page until there are no video ids - it means we got
798 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
800 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
801 page = self._download_webpage(gdata_url, username,
802 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
804 # Extract video identifiers
807 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
808 if mobj.group(1) not in ids_in_page:
809 ids_in_page.append(mobj.group(1))
811 video_ids.extend(ids_in_page)
813 # A little optimization - if current page is not
814 # "full", ie. does not contain PAGE_SIZE video ids then
815 # we can assume that this page is the last one - there
816 # are no more ids on further pages - no need to query
819 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
824 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
825 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
826 return [self.playlist_result(url_results, playlist_title = username)]
828 class YoutubeSearchIE(SearchInfoExtractor):
829 IE_DESC = u'YouTube.com searches'
830 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
832 IE_NAME = u'youtube:search'
833 _SEARCH_KEY = 'ytsearch'
835 def report_download_page(self, query, pagenum):
836 """Report attempt to download search page with given number."""
837 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
839 def _get_n_results(self, query, n):
840 """Get a specified number of results for a query"""
846 while (50 * pagenum) < limit:
847 self.report_download_page(query, pagenum+1)
848 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
849 request = compat_urllib_request.Request(result_url)
851 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
852 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
853 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
854 api_response = json.loads(data)['data']
856 if not 'items' in api_response:
857 raise ExtractorError(u'[youtube] No video results')
859 new_ids = list(video['id'] for video in api_response['items'])
862 limit = min(n, api_response['totalItems'])
865 if len(video_ids) > n:
866 video_ids = video_ids[:n]
867 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
868 return self.playlist_result(videos, query)
871 class YoutubeShowIE(InfoExtractor):
872 IE_DESC = u'YouTube.com (multi-season) shows'
873 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
874 IE_NAME = u'youtube:show'
876 def _real_extract(self, url):
877 mobj = re.match(self._VALID_URL, url)
878 show_name = mobj.group(1)
879 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
880 # There's one playlist for each season of the show
881 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
882 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
883 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
886 class YoutubeSubscriptionsIE(YoutubeIE):
887 """It's a subclass of YoutubeIE because we need to login"""
888 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
889 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
890 IE_NAME = u'youtube:subscriptions'
891 _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
894 # Overwrite YoutubeIE properties we don't want
897 def suitable(cls, url):
898 return re.match(cls._VALID_URL, url) is not None
900 def _real_initialize(self):
901 (username, password) = self._get_login_info()
903 raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True)
904 super(YoutubeSubscriptionsIE, self)._real_initialize()
906 def _real_extract(self, url):
908 # The step argument is available only in 2.7 or higher
909 for i in itertools.count(0):
910 paging = i*self._PAGING_STEP
911 info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
912 u'Downloading page %s' % i)
913 info = json.loads(info)
914 feed_html = info['feed_html']
915 m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html)
916 ids = orderedSet(m.group(1) for m in m_ids)
917 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
918 if info['paging'] is None:
920 return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')