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"
125 def suitable(cls, url):
126 """Receives a URL and returns True if suitable for this IE."""
127 if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False
128 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
130 def report_lang(self):
131 """Report attempt to set language."""
132 self.to_screen(u'Setting language')
134 def report_login(self):
135 """Report attempt to log in."""
136 self.to_screen(u'Logging in')
138 def report_video_webpage_download(self, video_id):
139 """Report attempt to download video webpage."""
140 self.to_screen(u'%s: Downloading video webpage' % video_id)
142 def report_video_info_webpage_download(self, video_id):
143 """Report attempt to download video info webpage."""
144 self.to_screen(u'%s: Downloading video info webpage' % video_id)
146 def report_video_subtitles_download(self, video_id):
147 """Report attempt to download video info webpage."""
148 self.to_screen(u'%s: Checking available subtitles' % video_id)
150 def report_video_subtitles_request(self, video_id, sub_lang, format):
151 """Report attempt to download video info webpage."""
152 self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format))
154 def report_video_subtitles_available(self, video_id, sub_lang_list):
155 """Report available subtitles."""
156 sub_lang = ",".join(list(sub_lang_list.keys()))
157 self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang))
159 def report_information_extraction(self, video_id):
160 """Report attempt to extract video information."""
161 self.to_screen(u'%s: Extracting video information' % video_id)
163 def report_unavailable_format(self, video_id, format):
164 """Report extracted video URL."""
165 self.to_screen(u'%s: Format %s not available' % (video_id, format))
167 def report_rtmp_download(self):
168 """Indicate the download will use the RTMP protocol."""
169 self.to_screen(u'RTMP download detected')
171 def _decrypt_signature(self, s):
172 """Turn the encrypted s field into a working signature"""
175 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]
177 return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1]
179 return s[2:63] + s[82] + s[64:82] + s[63]
181 return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1]
183 return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26]
185 return s[52] + s[81:55:-1] + s[2] + s[54:52:-1] + s[82] + s[51:36:-1] + s[55] + s[35:2:-1] + s[36]
187 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]
190 raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
192 def _get_available_subtitles(self, video_id):
193 self.report_video_subtitles_download(video_id)
194 request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
196 sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
197 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
198 return (u'unable to download video subtitles: %s' % compat_str(err), None)
199 sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
200 sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list)
201 if not sub_lang_list:
202 return (u'video doesn\'t have subtitles', None)
205 def _list_available_subtitles(self, video_id):
206 sub_lang_list = self._get_available_subtitles(video_id)
207 self.report_video_subtitles_available(video_id, sub_lang_list)
209 def _request_subtitle(self, sub_lang, sub_name, video_id, format):
212 (error_message, sub_lang, sub)
214 self.report_video_subtitles_request(video_id, sub_lang, format)
215 params = compat_urllib_parse.urlencode({
221 url = 'http://www.youtube.com/api/timedtext?' + params
223 sub = compat_urllib_request.urlopen(url).read().decode('utf-8')
224 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
225 return (u'unable to download video subtitles: %s' % compat_str(err), None, None)
227 return (u'Did not fetch video subtitles', None, None)
228 return (None, sub_lang, sub)
230 def _request_automatic_caption(self, video_id, webpage):
231 """We need the webpage for getting the captions url, pass it as an
232 argument to speed up the process."""
233 sub_lang = self._downloader.params.get('subtitleslang') or 'en'
234 sub_format = self._downloader.params.get('subtitlesformat')
235 self.to_screen(u'%s: Looking for automatic captions' % video_id)
236 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
237 err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
239 return [(err_msg, None, None)]
240 player_config = json.loads(mobj.group(1))
242 args = player_config[u'args']
243 caption_url = args[u'ttsurl']
244 timestamp = args[u'timestamp']
245 params = compat_urllib_parse.urlencode({
252 subtitles_url = caption_url + '&' + params
253 sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
254 return [(None, sub_lang, sub)]
256 return [(err_msg, None, None)]
258 def _extract_subtitle(self, video_id):
260 Return a list with a tuple:
261 [(error_message, sub_lang, sub)]
263 sub_lang_list = self._get_available_subtitles(video_id)
264 sub_format = self._downloader.params.get('subtitlesformat')
265 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
266 return [(sub_lang_list[0], None, None)]
267 if self._downloader.params.get('subtitleslang', False):
268 sub_lang = self._downloader.params.get('subtitleslang')
269 elif 'en' in sub_lang_list:
272 sub_lang = list(sub_lang_list.keys())[0]
273 if not sub_lang in sub_lang_list:
274 return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)]
276 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
279 def _extract_all_subtitles(self, video_id):
280 sub_lang_list = self._get_available_subtitles(video_id)
281 sub_format = self._downloader.params.get('subtitlesformat')
282 if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles
283 return [(sub_lang_list[0], None, None)]
285 for sub_lang in sub_lang_list:
286 subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format)
287 subtitles.append(subtitle)
290 def _print_formats(self, formats):
291 print('Available formats:')
293 print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???')))
295 def _real_initialize(self):
296 if self._downloader is None:
301 downloader_params = self._downloader.params
303 # Attempt to use provided username and password or .netrc data
304 if downloader_params.get('username', None) is not None:
305 username = downloader_params['username']
306 password = downloader_params['password']
307 elif downloader_params.get('usenetrc', False):
309 info = netrc.netrc().authenticators(self._NETRC_MACHINE)
314 raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
315 except (IOError, netrc.NetrcParseError) as err:
316 self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err))
320 request = compat_urllib_request.Request(self._LANG_URL)
323 compat_urllib_request.urlopen(request).read()
324 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
325 self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
328 # No authentication to be performed
332 request = compat_urllib_request.Request(self._LOGIN_URL)
334 login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
335 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
336 self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
341 match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
343 galx = match.group(1)
345 match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
351 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
355 u'PersistentCookie': u'yes',
357 u'bgresponse': u'js_disabled',
358 u'checkConnection': u'',
359 u'checkedDomains': u'youtube',
365 u'signIn': u'Sign in',
367 u'service': u'youtube',
371 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
373 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
374 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
375 request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
378 login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
379 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
380 self._downloader.report_warning(u'unable to log in: bad username or password')
382 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
383 self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
389 'action_confirm': 'Confirm',
391 request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
393 self.report_age_confirmation()
394 compat_urllib_request.urlopen(request).read().decode('utf-8')
395 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
396 raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
398 def _extract_id(self, url):
399 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
401 raise ExtractorError(u'Invalid URL: %s' % url)
402 video_id = mobj.group(2)
405 def _real_extract(self, url):
406 if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
407 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 ).')
409 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
410 mobj = re.search(self._NEXT_URL_RE, url)
412 url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
413 video_id = self._extract_id(url)
416 self.report_video_webpage_download(video_id)
417 url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
418 request = compat_urllib_request.Request(url)
420 video_webpage_bytes = compat_urllib_request.urlopen(request).read()
421 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
422 raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
424 video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
426 # Attempt to extract SWF player URL
427 mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
429 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
434 self.report_video_info_webpage_download(video_id)
435 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
436 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
437 % (video_id, el_type))
438 video_info_webpage = self._download_webpage(video_info_url, video_id,
440 errnote='unable to download video info webpage')
441 video_info = compat_parse_qs(video_info_webpage)
442 if 'token' in video_info:
444 if 'token' not in video_info:
445 if 'reason' in video_info:
446 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
448 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
450 # Check for "rental" videos
451 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
452 raise ExtractorError(u'"rental" videos not supported')
454 # Start extracting information
455 self.report_information_extraction(video_id)
458 if 'author' not in video_info:
459 raise ExtractorError(u'Unable to extract uploader name')
460 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
463 video_uploader_id = None
464 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
466 video_uploader_id = mobj.group(1)
468 self._downloader.report_warning(u'unable to extract uploader nickname')
471 if 'title' not in video_info:
472 raise ExtractorError(u'Unable to extract video title')
473 video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
476 # We try first to get a high quality image:
477 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
478 video_webpage, re.DOTALL)
479 if m_thumb is not None:
480 video_thumbnail = m_thumb.group(1)
481 elif 'thumbnail_url' not in video_info:
482 self._downloader.report_warning(u'unable to extract video thumbnail')
484 else: # don't panic if we can't find it
485 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
489 mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
491 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
492 upload_date = unified_strdate(upload_date)
495 video_description = get_element_by_id("eow-description", video_webpage)
496 if video_description:
497 video_description = clean_html(video_description)
499 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
501 video_description = unescapeHTML(fd_mobj.group(1))
503 video_description = u''
506 video_subtitles = None
508 if self._downloader.params.get('writesubtitles', False):
509 video_subtitles = self._extract_subtitle(video_id)
511 (sub_error, sub_lang, sub) = video_subtitles[0]
513 self._downloader.report_warning(sub_error)
515 if self._downloader.params.get('writeautomaticsub', False):
516 video_subtitles = self._request_automatic_caption(video_id, video_webpage)
517 (sub_error, sub_lang, sub) = video_subtitles[0]
519 self._downloader.report_warning(sub_error)
521 if self._downloader.params.get('allsubtitles', False):
522 video_subtitles = self._extract_all_subtitles(video_id)
523 for video_subtitle in video_subtitles:
524 (sub_error, sub_lang, sub) = video_subtitle
526 self._downloader.report_warning(sub_error)
528 if self._downloader.params.get('listsubtitles', False):
529 self._list_available_subtitles(video_id)
532 if 'length_seconds' not in video_info:
533 self._downloader.report_warning(u'unable to extract video duration')
536 video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
538 # Decide which formats to download
539 req_format = self._downloader.params.get('format', None)
542 mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
544 raise ValueError('Could not find vevo ID')
545 info = json.loads(mobj.group(1))
547 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
548 # this signatures are encrypted
549 m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
551 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
552 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
556 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
557 self.report_rtmp_download()
558 video_url_list = [(None, video_info['conn'][0])]
559 elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
561 for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
562 url_data = compat_parse_qs(url_data_str)
563 if 'itag' in url_data and 'url' in url_data:
564 url = url_data['url'][0]
565 if 'sig' in url_data:
566 url += '&signature=' + url_data['sig'][0]
567 elif 's' in url_data:
568 if self._downloader.params.get('verbose'):
570 player = self._search_regex(r'html5player-(.+?)\.js', video_webpage,
571 'html5 player', fatal=False)
572 self.to_screen('encrypted signature length %d (%d.%d), itag %s, html5 player %s' %
573 (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player))
574 signature = self._decrypt_signature(url_data['s'][0])
575 url += '&signature=' + signature
576 if 'ratebypass' not in url:
577 url += '&ratebypass=yes'
578 url_map[url_data['itag'][0]] = url
580 format_limit = self._downloader.params.get('format_limit', None)
581 available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
582 if format_limit is not None and format_limit in available_formats:
583 format_list = available_formats[available_formats.index(format_limit):]
585 format_list = available_formats
586 existing_formats = [x for x in format_list if x in url_map]
587 if len(existing_formats) == 0:
588 raise ExtractorError(u'no known formats available for video')
589 if self._downloader.params.get('listformats', None):
590 self._print_formats(existing_formats)
592 if req_format is None or req_format == 'best':
593 video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
594 elif req_format == 'worst':
595 video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
596 elif req_format in ('-1', 'all'):
597 video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
599 # Specific formats. We pick the first in a slash-delimeted sequence.
600 # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
601 req_formats = req_format.split('/')
602 video_url_list = None
603 for rf in req_formats:
605 video_url_list = [(rf, url_map[rf])]
607 if video_url_list is None:
608 raise ExtractorError(u'requested format not available')
610 raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
613 for format_param, video_real_url in video_url_list:
615 video_extension = self._video_extensions.get(format_param, 'flv')
617 video_format = '{0} - {1}'.format(format_param if format_param else video_extension,
618 self._video_dimensions.get(format_param, '???'))
622 'url': video_real_url,
623 'uploader': video_uploader,
624 'uploader_id': video_uploader_id,
625 'upload_date': upload_date,
626 'title': video_title,
627 'ext': video_extension,
628 'format': video_format,
629 'thumbnail': video_thumbnail,
630 'description': video_description,
631 'player_url': player_url,
632 'subtitles': video_subtitles,
633 'duration': video_duration
637 class YoutubePlaylistIE(InfoExtractor):
638 IE_DESC = u'YouTube.com playlists'
644 (?:course|view_play_list|my_playlists|artist|playlist|watch)
645 \? (?:.*?&)*? (?:p|a|list)=
648 ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,})
651 ((?:PL|EC|UU)[0-9A-Za-z-_]{10,})
653 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
655 IE_NAME = u'youtube:playlist'
658 def suitable(cls, url):
659 """Receives a URL and returns True if suitable for this IE."""
660 return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
662 def _real_extract(self, url):
663 # Extract playlist id
664 mobj = re.match(self._VALID_URL, url, re.VERBOSE)
666 raise ExtractorError(u'Invalid URL: %s' % url)
668 # Download playlist videos from API
669 playlist_id = mobj.group(1) or mobj.group(2)
674 url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1)
675 page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
678 response = json.loads(page)
679 except ValueError as err:
680 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
682 if 'feed' not in response:
683 raise ExtractorError(u'Got a malformed response from YouTube API')
684 playlist_title = response['feed']['title']['$t']
685 if 'entry' not in response['feed']:
686 # Number of videos is a multiple of self._MAX_RESULTS
689 for entry in response['feed']['entry']:
690 index = entry['yt$position']['$t']
691 if 'media$group' in entry and 'media$player' in entry['media$group']:
692 videos.append((index, entry['media$group']['media$player']['url']))
694 if len(response['feed']['entry']) < self._MAX_RESULTS:
698 videos = [v[1] for v in sorted(videos)]
700 url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
701 return [self.playlist_result(url_results, playlist_id, playlist_title)]
704 class YoutubeChannelIE(InfoExtractor):
705 IE_DESC = u'YouTube.com channels'
706 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
707 _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
708 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
709 _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'
710 IE_NAME = u'youtube:channel'
712 def extract_videos_from_page(self, page):
714 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
715 if mobj.group(1) not in ids_in_page:
716 ids_in_page.append(mobj.group(1))
719 def _real_extract(self, url):
721 mobj = re.match(self._VALID_URL, url)
723 raise ExtractorError(u'Invalid URL: %s' % url)
725 # Download channel page
726 channel_id = mobj.group(1)
730 url = self._TEMPLATE_URL % (channel_id, pagenum)
731 page = self._download_webpage(url, channel_id,
732 u'Downloading page #%s' % pagenum)
734 # Extract video identifiers
735 ids_in_page = self.extract_videos_from_page(page)
736 video_ids.extend(ids_in_page)
738 # Download any subsequent channel pages using the json-based channel_ajax query
739 if self._MORE_PAGES_INDICATOR in page:
741 pagenum = pagenum + 1
743 url = self._MORE_PAGES_URL % (pagenum, channel_id)
744 page = self._download_webpage(url, channel_id,
745 u'Downloading page #%s' % pagenum)
747 page = json.loads(page)
749 ids_in_page = self.extract_videos_from_page(page['content_html'])
750 video_ids.extend(ids_in_page)
752 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
755 self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
757 urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
758 url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
759 return [self.playlist_result(url_entries, channel_id)]
762 class YoutubeUserIE(InfoExtractor):
763 IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
764 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)'
765 _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
766 _GDATA_PAGE_SIZE = 50
767 _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d'
768 _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]'
769 IE_NAME = u'youtube:user'
771 def _real_extract(self, url):
773 mobj = re.match(self._VALID_URL, url)
775 raise ExtractorError(u'Invalid URL: %s' % url)
777 username = mobj.group(1)
779 # Download video ids using YouTube Data API. Result size per
780 # query is limited (currently to 50 videos) so we need to query
781 # page by page until there are no video ids - it means we got
788 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
790 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
791 page = self._download_webpage(gdata_url, username,
792 u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
794 # Extract video identifiers
797 for mobj in re.finditer(self._VIDEO_INDICATOR, page):
798 if mobj.group(1) not in ids_in_page:
799 ids_in_page.append(mobj.group(1))
801 video_ids.extend(ids_in_page)
803 # A little optimization - if current page is not
804 # "full", ie. does not contain PAGE_SIZE video ids then
805 # we can assume that this page is the last one - there
806 # are no more ids on further pages - no need to query
809 if len(ids_in_page) < self._GDATA_PAGE_SIZE:
814 urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
815 url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
816 return [self.playlist_result(url_results, playlist_title = username)]
818 class YoutubeSearchIE(SearchInfoExtractor):
819 IE_DESC = u'YouTube.com searches'
820 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
822 IE_NAME = u'youtube:search'
823 _SEARCH_KEY = 'ytsearch'
825 def report_download_page(self, query, pagenum):
826 """Report attempt to download search page with given number."""
827 self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
829 def _get_n_results(self, query, n):
830 """Get a specified number of results for a query"""
836 while (50 * pagenum) < limit:
837 self.report_download_page(query, pagenum+1)
838 result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
839 request = compat_urllib_request.Request(result_url)
841 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
842 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
843 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
844 api_response = json.loads(data)['data']
846 if not 'items' in api_response:
847 raise ExtractorError(u'[youtube] No video results')
849 new_ids = list(video['id'] for video in api_response['items'])
852 limit = min(n, api_response['totalItems'])
855 if len(video_ids) > n:
856 video_ids = video_ids[:n]
857 videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
858 return self.playlist_result(videos, query)
861 class YoutubeShowIE(InfoExtractor):
862 IE_DESC = u'YouTube.com (multi-season) shows'
863 _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
864 IE_NAME = u'youtube:show'
866 def _real_extract(self, url):
867 mobj = re.match(self._VALID_URL, url)
868 show_name = mobj.group(1)
869 webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
870 # There's one playlist for each season of the show
871 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
872 self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
873 return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
876 class YoutubeSubscriptionsIE(YoutubeIE):
877 """It's a subclass of YoutubeIE because we need to login"""
878 IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
879 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
880 IE_NAME = u'youtube:subscriptions'
881 _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s'
884 # Overwrite YoutubeIE properties we don't want
887 def suitable(cls, url):
888 return re.match(cls._VALID_URL, url) is not None
890 def _real_extract(self, url):
892 # The step argument is available only in 2.7 or higher
893 for i in itertools.count(0):
894 paging = i*self._PAGING_STEP
895 info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed',
896 u'Downloading page %s' % i)
897 info = json.loads(info)
898 feed_html = info['feed_html']
899 m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html)
900 ids = orderedSet(m.group(1) for m in m_ids)
901 feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
902 if info['paging'] is None:
904 return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions')