9 from .common import InfoExtractor, SearchInfoExtractor
10 from .subtitles import SubtitlesIE
16 compat_urllib_request,
28 class YoutubeSubtitlesIE(SubtitlesIE):
30 def _get_available_subtitles(self, video_id):
31 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
33 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
34 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
35 self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
37 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
42 params = compat_urllib_parse.urlencode({
45 'fmt': self._downloader.params.get('subtitlesformat'),
47 url = u'http://www.youtube.com/api/timedtext?' + params
48 sub_lang_list[lang] = url
50 self._downloader.report_warning(u'video doesn\'t have subtitles')
54 def _request_automatic_caption(self, video_id, webpage):
55 """We need the webpage for getting the captions url, pass it as an
56 argument to speed up the process."""
57 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
58 sub_format = self._downloader.params.get('subtitlesformat')
59 self.to_screen(u'%s: Looking for automatic captions' % video_id)
60 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
61 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
63 self._downloader.report_warning(err_msg)
65 player_config = json.loads(mobj.group(1))
67 args = player_config[u'args']
68 caption_url = args[u'ttsurl']
69 timestamp = args[u'timestamp']
70 params = compat_urllib_parse.urlencode({
77 subtitles_url = caption_url + '&' + params
78 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
79 return {sub_lang: sub}
80 # An extractor error can be raise by the download process if there are
81 # no automatic captions but there are subtitles
82 except (KeyError, ExtractorError):
83 self._downloader.report_warning(err_msg)
87 class YoutubeIE(YoutubeSubtitlesIE):
88 IE_DESC = u'YouTube.com'
91 (?:https?://)? # http(s):// (optional)
92 (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
93 tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
94 (?:.*?\#/)? # handle anchor (#/) redirect urls
95 (?: # the various things that can precede the ID:
96 (?:(?:v|embed|e)/) # v/ or embed/ or e/
97 |(?: # or the v= param in all its forms
98 (?:watch|movie(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
99 (?:\?|\#!?) # the params delimiter ? or # or #!
100 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
103 )? # optional -> youtube.com/xxxx is OK
104 )? # all until now is optional -> you can pass the naked ID
105 ([0-9A-Za-z_-]+) # here is it! the YouTube video ID
106 (?(1).+)? # if we found the ID, everything can follow
108 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
109 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
110 _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
111 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
112 _NETRC_MACHINE = 'youtube'
113 # Listed in order of quality
114 _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13']
115 _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13']
116 _video_extensions = {
128 _video_dimensions = {
147 u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc",
148 u"file": u"BaW_jenozKc.mp4",
150 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
151 u"uploader": u"Philipp Hagemeister",
152 u"uploader_id": u"phihag",
153 u"upload_date": u"20121002",
154 u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
158 u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
159 u"file": u"1ltcDfZMA3U.flv",
160 u"note": u"Test VEVO video (#897)",
162 u"upload_date": u"20070518",
163 u"title": u"Maps - It Will Find You",
164 u"description": u"Music video by Maps performing It Will Find You.",
165 u"uploader": u"MuteUSA",
166 u"uploader_id": u"MuteUSA"
170 u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY",
171 u"file": u"UxxajLWwzqY.mp4",
172 u"note": u"Test generic use_cipher_signature video (#897)",
174 u"upload_date": u"20120506",
175 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
176 u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c",
177 u"uploader": u"IconaPop",
178 u"uploader_id": u"IconaPop"
182 u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ",
183 u"file": u"07FYdnEawAQ.mp4",
184 u"note": u"Test VEVO video with age protection (#956)",
186 u"upload_date": u"20130703",
187 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
188 u"description": u"md5:64249768eec3bc4276236606ea996373",
189 u"uploader": u"justintimberlakeVEVO",
190 u"uploader_id": u"justintimberlakeVEVO"
197 def suitable(cls, url):
198 """Receives a URL and returns True if suitable for this IE."""
199 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
200 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
202 def report_lang(self):
203 """Report attempt to set language."""
204 self.to_screen(u'Setting language')
206 def report_video_webpage_download(self, video_id):
207 """Report attempt to download video webpage."""
208 self.to_screen(u'%s: Downloading video webpage' % video_id)
210 def report_video_info_webpage_download(self, video_id):
211 """Report attempt to download video info webpage."""
212 self.to_screen(u'%s: Downloading video info webpage' % video_id)
214 def report_information_extraction(self, video_id):
215 """Report attempt to extract video information."""
216 self.to_screen(u'%s: Extracting video information' % video_id)
218 def report_unavailable_format(self, video_id, format):
219 """Report extracted video URL."""
220 self.to_screen(u'%s: Format %s not available' % (video_id, format))
222 def report_rtmp_download(self):
223 """Indicate the download will use the RTMP protocol."""
224 self.to_screen(u'RTMP download detected')
226 def _decrypt_signature(self, s):
227 """Turn the encrypted s field into a working signature"""
230 return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
232 return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
234 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]
236 return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1]
238 return s[2:63] + s[82] + s[64:82] + s[63]
240 return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21]
242 return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26]
244 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:]
246 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]
248 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]
251 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
253 def _print_formats(self, formats):
254 print('Available formats:')
256 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
258 def _real_initialize(self):
259 if self._downloader is None:
263 request = compat_urllib_request.Request(self._LANG_URL)
266 compat_urllib_request.urlopen(request).read()
267 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
268 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
271 (username, password) = self._get_login_info()
273 # No authentication to be performed
277 request = compat_urllib_request.Request(self._LOGIN_URL)
279 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
280 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
281 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
286 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
288 galx = match.group(1)
290 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
296 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
300 u'PersistentCookie': u'yes',
302 u'bgresponse': u'js_disabled',
303 u'checkConnection': u'',
304 u'checkedDomains': u'youtube',
310 u'signIn': u'Sign in',
312 u'service': u'youtube',
316 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
318 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
319 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
320 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
323 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
324 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
325 self._downloader.report_warning(u'unable to log in: bad username or password')
327 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
328 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
334 'action_confirm': 'Confirm',
336 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
338 self.report_age_confirmation()
339 compat_urllib_request.urlopen(request).read().decode('utf-8')
340 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
341 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
343 def _extract_id(self, url):
344 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
346 raise ExtractorError(u'Invalid URL: %s' % url)
347 video_id = mobj.group(2)
350 def _real_extract(self, url):
351 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
352 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 ).')
354 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
355 mobj = re.search(self._NEXT_URL_RE, url)
357 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
358 video_id = self._extract_id(url)
361 self.report_video_webpage_download(video_id)
362 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
363 request = compat_urllib_request.Request(url)
365 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
366 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
367 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
369 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
371 # Attempt to extract SWF player URL
372 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
374 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
379 self.report_video_info_webpage_download(video_id)
380 if re.search(r'player-age-gate-content">', video_webpage) is not None:
381 self.report_age_confirmation()
383 # We simulate the access to the video from www.youtube.com/v/{video_id}
384 # this can be viewed without login into Youtube
385 data = compat_urllib_parse.urlencode({'video_id': video_id,
389 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
393 video_info_url = 'https://www.youtube.com/get_video_info?' + data
394 video_info_webpage = self._download_webpage(video_info_url, video_id,
396 errnote='unable to download video info webpage')
397 video_info = compat_parse_qs(video_info_webpage)
400 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
401 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
402 % (video_id, el_type))
403 video_info_webpage = self._download_webpage(video_info_url, video_id,
405 errnote='unable to download video info webpage')
406 video_info = compat_parse_qs(video_info_webpage)
407 if 'token' in video_info:
409 if 'token' not in video_info:
410 if 'reason' in video_info:
411 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
413 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
415 # Check for "rental" videos
416 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
417 raise ExtractorError(u'"rental" videos not supported')
419 # Start extracting information
420 self.report_information_extraction(video_id)
423 if 'author' not in video_info:
424 raise ExtractorError(u'Unable to extract uploader name')
425 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
428 video_uploader_id = None
429 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
431 video_uploader_id = mobj.group(1)
433 self._downloader.report_warning(u'unable to extract uploader nickname')
436 if 'title' not in video_info:
437 raise ExtractorError(u'Unable to extract video title')
438 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
441 # We try first to get a high quality image:
442 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
443 video_webpage, re.DOTALL)
444 if m_thumb is not None:
445 video_thumbnail = m_thumb.group(1)
446 elif 'thumbnail_url' not in video_info:
447 self._downloader.report_warning(u'unable to extract video thumbnail')
449 else: # don't panic if we can't find it
450 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
454 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
456 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
457 upload_date = unified_strdate(upload_date)
460 video_description = get_element_by_id("eow-description", video_webpage)
461 if video_description:
462 video_description = clean_html(video_description)
464 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
466 video_description = unescapeHTML(fd_mobj.group(1))
468 video_description = u''
471 video_subtitles = None
473 if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
474 video_subtitles = self._extract_subtitles(video_id)
475 elif self._downloader.params.get('writeautomaticsub', False):
476 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
478 if self._downloader.params.get('listsubtitles', False):
479 self._list_available_subtitles(video_id)
482 if 'length_seconds' not in video_info:
483 self._downloader.report_warning(u'unable to extract video duration')
486 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
488 # Decide which formats to download
489 req_format = self._downloader.params.get('format', None)
492 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
494 raise ValueError('Could not find vevo ID')
495 info = json.loads(mobj.group(1))
497 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
498 # this signatures are encrypted
499 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
501 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
502 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
506 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
507 self.report_rtmp_download()
508 video_url_list = [(None, video_info['conn'][0])]
509 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
510 if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
511 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
513 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
514 url_data = compat_parse_qs(url_data_str)
515 if 'itag' in url_data and 'url' in url_data:
516 url = url_data['url'][0]
517 if 'sig' in url_data:
518 url += '&signature=' + url_data['sig'][0]
519 elif 's' in url_data:
520 if self._downloader.params.get('verbose'):
523 player_version = self._search_regex(r'ad3-(.+?)\.swf',
524 video_info['ad3_module'][0], 'flash player',
526 player = 'flash player %s' % player_version
528 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
529 'html5 player', fatal=False)
530 parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
531 self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
532 (len(s), parts_sizes, url_data['itag'][0], player))
533 signature = self._decrypt_signature(url_data['s'][0])
534 url += '&signature=' + signature
535 if 'ratebypass' not in url:
536 url += '&ratebypass=yes'
537 url_map[url_data['itag'][0]] = url
539 format_limit = self._downloader.params.get('format_limit', None)
540 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
541 if format_limit is not None and format_limit in available_formats:
542 format_list = available_formats[available_formats.index(format_limit):]
544 format_list = available_formats
545 existing_formats = [x for x in format_list if x in url_map]
546 if len(existing_formats) == 0:
547 raise ExtractorError(u'no known formats available for video')
548 if self._downloader.params.get('listformats', None):
549 self._print_formats(existing_formats)
551 if req_format is None or req_format == 'best':
552 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
553 elif req_format == 'worst':
554 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
555 elif req_format in ('-1', 'all'):
556 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
558 # Specific formats. We pick the first in a slash-delimeted sequence.
559 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
560 req_formats = req_format.split('/')
561 video_url_list = None
562 for rf in req_formats:
564 video_url_list = [(rf, url_map[rf])]
566 if video_url_list is None:
567 raise ExtractorError(u'requested format not available')
569 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
572 for format_param, video_real_url in video_url_list:
574 video_extension = self._video_extensions.get(format_param, 'flv')
576 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
577 self._video_dimensions.get(format_param, '???'))
581 'url': video_real_url,
582 'uploader': video_uploader,
583 'uploader_id': video_uploader_id,
584 'upload_date': upload_date,
585 'title': video_title,
586 'ext': video_extension,
587 'format': video_format,
588 'thumbnail': video_thumbnail,
589 'description': video_description,
590 'player_url': player_url,
591 'subtitles': video_subtitles,
592 'duration': video_duration
596 class YoutubePlaylistIE(InfoExtractor):
597 IE_DESC = u'YouTube.com playlists'
603 (?:course|view_play_list|my_playlists|artist|playlist|watch)
604 \? (?:.*?&)*? (?:p|a|list)=
607 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
610 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
612 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
614 IE_NAME = u'youtube:playlist'
617 def suitable(cls, url):
618 """Receives a URL and returns True if suitable for this IE."""
619 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
621 def _real_extract(self, url):
622 # Extract playlist id
623 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
625 raise ExtractorError(u'Invalid URL: %s' % url)
627 # Download playlist videos from API
628 playlist_id = mobj.group(1) or mobj.group(2)
633 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
634 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
637 response = json.loads(page)
638 except ValueError as err:
639 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
641 if 'feed' not in response:
642 raise ExtractorError(u'Got a malformed response from YouTube API')
643 playlist_title = response['feed']['title']['$t']
644 if 'entry' not in response['feed']:
645 # Number of videos is a multiple of self._MAX_RESULTS
648 for entry in response['feed']['entry']:
649 index = entry['yt$position']['$t']
650 if 'media$group' in entry and 'media$player' in entry['media$group']:
651 videos.append((index, entry['media$group']['media$player']['url']))
653 if len(response['feed']['entry']) < self._MAX_RESULTS:
657 videos = [v[1] for v in sorted(videos)]
659 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
660 return [self.playlist_result(url_results, playlist_id, playlist_title)]
663 class YoutubeChannelIE(InfoExtractor):
664 IE_DESC = u'YouTube.com channels'
665 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
666 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
667 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
668 _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'
669 IE_NAME = u'youtube:channel'
671 def extract_videos_from_page(self, page):
673 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
674 if mobj.group(1) not in ids_in_page:
675 ids_in_page.append(mobj.group(1))
678 def _real_extract(self, url):
680 mobj = re.match(self._VALID_URL, url)
682 raise ExtractorError(u'Invalid URL: %s' % url)
684 # Download channel page
685 channel_id = mobj.group(1)
689 url = self._TEMPLATE_URL % (channel_id, pagenum)
690 page = self._download_webpage(url, channel_id,
691 u'Downloading page #%s' % pagenum)
693 # Extract video identifiers
694 ids_in_page = self.extract_videos_from_page(page)
695 video_ids.extend(ids_in_page)
697 # Download any subsequent channel pages using the json-based channel_ajax query
698 if self._MORE_PAGES_INDICATOR in page:
700 pagenum = pagenum + 1
702 url = self._MORE_PAGES_URL % (pagenum, channel_id)
703 page = self._download_webpage(url, channel_id,
704 u'Downloading page #%s' % pagenum)
706 page = json.loads(page)
708 ids_in_page = self.extract_videos_from_page(page['content_html'])
709 video_ids.extend(ids_in_page)
711 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
714 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
716 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
717 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
718 return [self.playlist_result(url_entries, channel_id)]
721 class YoutubeUserIE(InfoExtractor):
722 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
723 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
724 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
725 _GDATA_PAGE_SIZE = 50
726 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
727 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
728 IE_NAME = u'youtube:user'
730 def _real_extract(self, url):
732 mobj = re.match(self._VALID_URL, url)
734 raise ExtractorError(u'Invalid URL: %s' % url)
736 username = mobj.group(1)
738 # Download video ids using YouTube Data API. Result size per
739 # query is limited (currently to 50 videos) so we need to query
740 # page by page until there are no video ids - it means we got
747 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
749 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
750 page = self._download_webpage(gdata_url, username,
751 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
753 # Extract video identifiers
756 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
757 if mobj.group(1) not in ids_in_page:
758 ids_in_page.append(mobj.group(1))
760 video_ids.extend(ids_in_page)
762 # A little optimization - if current page is not
763 # "full", ie. does not contain PAGE_SIZE video ids then
764 # we can assume that this page is the last one - there
765 # are no more ids on further pages - no need to query
768 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
773 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
774 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
775 return [self.playlist_result(url_results, playlist_title = username)]
777 class YoutubeSearchIE(SearchInfoExtractor):
778 IE_DESC = u'YouTube.com searches'
779 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
781 IE_NAME = u'youtube:search'
782 _SEARCH_KEY = 'ytsearch'
784 def report_download_page(self, query, pagenum):
785 """Report attempt to download search page with given number."""
786 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
788 def _get_n_results(self, query, n):
789 """Get a specified number of results for a query"""
795 while (50 * pagenum) < limit:
796 self.report_download_page(query, pagenum+1)
797 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
798 request = compat_urllib_request.Request(result_url)
800 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
801 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
802 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
803 api_response = json.loads(data)['data']
805 if not 'items' in api_response:
806 raise ExtractorError(u'[youtube] No video results')
808 new_ids = list(video['id'] for video in api_response['items'])
811 limit = min(n, api_response['totalItems'])
814 if len(video_ids) > n:
815 video_ids = video_ids[:n]
816 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
817 return self.playlist_result(videos, query)
820 class YoutubeShowIE(InfoExtractor):
821 IE_DESC = u'YouTube.com (multi-season) shows'
822 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
823 IE_NAME = u'youtube:show'
825 def _real_extract(self, url):
826 mobj = re.match(self._VALID_URL, url)
827 show_name = mobj.group(1)
828 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
829 # There's one playlist for each season of the show
830 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
831 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
832 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
835 class YoutubeSubscriptionsIE(YoutubeIE):
836 """It's a subclass of YoutubeIE because we need to login"""
837 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
838 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
839 IE_NAME = u'youtube:subscriptions'
840 _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
843 # Overwrite YoutubeIE properties we don't want
846 def suitable(cls, url):
847 return re.match(cls._VALID_URL, url) is not None
849 def _real_initialize(self):
850 (username, password) = self._get_login_info()
852 raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True)
853 super(YoutubeSubscriptionsIE, self)._real_initialize()
855 def _real_extract(self, url):
857 # The step argument is available only in 2.7 or higher
858 for i in itertools.count(0):
859 paging = i*self._PAGING_STEP
860 info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
861 u'Downloading page %s' % i)
862 info = json.loads(info)
863 feed_html = info['feed_html']
864 m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html)
865 ids = orderedSet(m.group(1) for m in m_ids)
866 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
867 if info['paging'] is None:
869 return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')