X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9815bb0a551468e4939cacfffbc2d5cb8dd12431..33cd347759d6d999325ebf3c69b7ed5692c343b2:/youtube_dl/extractor/youtube.py diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index fbe8d63..bc89a14 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -4,6 +4,7 @@ import json import netrc import re import socket +import itertools from .common import InfoExtractor, SearchInfoExtractor from ..utils import ( @@ -19,12 +20,118 @@ from ..utils import ( ExtractorError, unescapeHTML, unified_strdate, + orderedSet, ) +class YoutubeBaseInfoExtractor(InfoExtractor): + """Provide base functions for Youtube extractors""" + _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' + _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' + _NETRC_MACHINE = 'youtube' + # If True it will raise an error if no login info is provided + _LOGIN_REQUIRED = False + + def report_lang(self): + """Report attempt to set language.""" + self.to_screen(u'Setting language') + + def _set_language(self): + request = compat_urllib_request.Request(self._LANG_URL) + try: + self.report_lang() + compat_urllib_request.urlopen(request).read() + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) + return False + return True + + def _login(self): + (username, password) = self._get_login_info() + # No authentication to be performed + if username is None: + if self._LOGIN_REQUIRED: + raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) + return False + + request = compat_urllib_request.Request(self._LOGIN_URL) + try: + login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) + return False + + galx = None + dsh = None + match = re.search(re.compile(r']* id="gaia_loginform"', login_results) is not None: + self._downloader.report_warning(u'unable to log in: bad username or password') + return False + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) + return False + return True -class YoutubeIE(InfoExtractor): - """Information extractor for youtube.com.""" + def _confirm_age(self): + age_form = { + 'next_url': '/', + 'action_confirm': 'Confirm', + } + request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) + try: + self.report_age_confirmation() + compat_urllib_request.urlopen(request).read().decode('utf-8') + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + return True + + def _real_initialize(self): + if self._downloader is None: + return + if not self._set_language(): + return + if not self._login(): + return + self._confirm_age() +class YoutubeIE(YoutubeBaseInfoExtractor): + IE_DESC = u'YouTube.com' _VALID_URL = r"""^ ( (?:https?://)? # http(s):// (optional) @@ -34,7 +141,7 @@ class YoutubeIE(InfoExtractor): (?: # the various things that can precede the ID: (?:(?:v|embed|e)/) # v/ or embed/ or e/ |(?: # or the v= param in all its forms - (?:watch(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) + (?:watch|movie(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) v= @@ -44,25 +151,45 @@ class YoutubeIE(InfoExtractor): ([0-9A-Za-z_-]+) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow $""" - _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' - _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' - _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' - _NETRC_MACHINE = 'youtube' # Listed in order of quality - _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13'] - _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13'] + _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '17', '13', + '95', '94', '93', '92', '132', '151', + '85', '84', '102', '83', '101', '82', '100', + ] + _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '17', '13', + '95', '94', '93', '92', '132', '151', + '85', '102', '84', '101', '83', '100', '82', + ] _video_extensions = { '13': '3gp', '17': 'mp4', '18': 'mp4', '22': 'mp4', '37': 'mp4', - '38': 'video', # You actually don't know if this will be MOV, AVI or whatever + '38': 'mp4', '43': 'webm', '44': 'webm', '45': 'webm', '46': 'webm', + + # 3d videos + '82': 'mp4', + '83': 'mp4', + '84': 'mp4', + '85': 'mp4', + '100': 'webm', + '101': 'webm', + '102': 'webm', + + # videos that use m3u8 + '92': 'mp4', + '93': 'mp4', + '94': 'mp4', + '95': 'mp4', + '96': 'mp4', + '132': 'mp4', + '151': 'mp4', } _video_dimensions = { '5': '240x400', @@ -79,23 +206,95 @@ class YoutubeIE(InfoExtractor): '44': '480x854', '45': '720x1280', '46': '1080x1920', + '82': '360p', + '83': '480p', + '84': '720p', + '85': '1080p', + '92': '240p', + '93': '360p', + '94': '480p', + '95': '720p', + '96': '1080p', + '100': '360p', + '101': '480p', + '102': '720p', + '132': '240p', + '151': '72p', } + _3d_itags = ['85', '84', '102', '83', '101', '82', '100'] IE_NAME = u'youtube' + _TESTS = [ + { + u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", + u"file": u"BaW_jenozKc.mp4", + u"info_dict": { + u"title": u"youtube-dl test video \"'/\\ä↭𝕐", + u"uploader": u"Philipp Hagemeister", + u"uploader_id": u"phihag", + u"upload_date": u"20121002", + u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." + } + }, + { + u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", + u"file": u"1ltcDfZMA3U.flv", + u"note": u"Test VEVO video (#897)", + u"info_dict": { + u"upload_date": u"20070518", + u"title": u"Maps - It Will Find You", + u"description": u"Music video by Maps performing It Will Find You.", + u"uploader": u"MuteUSA", + u"uploader_id": u"MuteUSA" + } + }, + { + u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", + u"file": u"UxxajLWwzqY.mp4", + u"note": u"Test generic use_cipher_signature video (#897)", + u"info_dict": { + u"upload_date": u"20120506", + u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", + u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c", + u"uploader": u"IconaPop", + u"uploader_id": u"IconaPop" + } + }, + { + u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", + u"file": u"07FYdnEawAQ.mp4", + u"note": u"Test VEVO video with age protection (#956)", + u"info_dict": { + u"upload_date": u"20130703", + u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", + u"description": u"md5:64249768eec3bc4276236606ea996373", + u"uploader": u"justintimberlakeVEVO", + u"uploader_id": u"justintimberlakeVEVO" + } + }, + { + u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE', + u'file': u'TGi3HqYrWHE.mp4', + u'note': u'm3u8 video', + u'info_dict': { + u'title': u'Triathlon - Men - London 2012 Olympic Games', + u'description': u'- Men - TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games', + u'uploader': u'olympic', + u'upload_date': u'20120807', + u'uploader_id': u'olympic', + }, + u'params': { + u'skip_download': True, + }, + }, + ] + @classmethod def suitable(cls, url): """Receives a URL and returns True if suitable for this IE.""" - if YoutubePlaylistIE.suitable(url): return False + if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - def report_lang(self): - """Report attempt to set language.""" - self.to_screen(u'Setting language') - - def report_login(self): - """Report attempt to log in.""" - self.to_screen(u'Logging in') - def report_video_webpage_download(self, video_id): """Report attempt to download video webpage.""" self.to_screen(u'%s: Downloading video webpage' % video_id) @@ -129,16 +328,44 @@ class YoutubeIE(InfoExtractor): """Indicate the download will use the RTMP protocol.""" self.to_screen(u'RTMP download detected') - @staticmethod - def _decrypt_signature(s): - """Decrypt the key the two subkeys must have a length of 43""" - (a,b) = s.split('.') - if len(a) != 43 or len(b) != 43: - raise ExtractorError(u'Unable to decrypt signature, subkeys lengths not valid') - b = ''.join([b[:8],a[0],b[9:18],b[-4],b[19:39], b[18]])[0:40] - a = a[-40:] - s_dec = '.'.join((a,b))[::-1] - return s_dec + def _decrypt_signature(self, s): + """Turn the encrypted s field into a working signature""" + + if len(s) == 92: + return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83] + elif len(s) == 90: + return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81] + elif len(s) == 88: + 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] + elif len(s) == 87: + return s[4:23] + s[86] + s[24:85] + elif len(s) == 86: + return s[83:85] + s[26] + s[79:46:-1] + s[85] + s[45:36:-1] + s[30] + s[35:30:-1] + s[46] + s[29:26:-1] + s[82] + s[25:1:-1] + elif len(s) == 85: + return s[2:8] + s[0] + s[9:21] + s[65] + s[22:65] + s[84] + s[66:82] + s[21] + elif len(s) == 84: + return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26] + elif len(s) == 83: + return s[:15] + s[80] + s[16:80] + s[15] + elif len(s) == 82: + 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] + elif len(s) == 81: + return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] + elif len(s) == 79: + return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9] + + else: + raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) + + def _decrypt_signature_age_gate(self, s): + # The videos with age protection use another player, so the algorithms + # can be different. + if len(s) == 86: + return s[2:63] + s[82] + s[64:82] + s[63] + else: + # Fallback to the other algortihms + return self._decrypt_signature(s) + def _get_available_subtitles(self, video_id): self.report_video_subtitles_download(video_id) @@ -241,110 +468,9 @@ class YoutubeIE(InfoExtractor): def _print_formats(self, formats): print('Available formats:') for x in formats: - print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'flv'), self._video_dimensions.get(x, '???'))) - - def _real_initialize(self): - if self._downloader is None: - return - - username = None - password = None - downloader_params = self._downloader.params - - # Attempt to use provided username and password or .netrc data - if downloader_params.get('username', None) is not None: - username = downloader_params['username'] - password = downloader_params['password'] - elif downloader_params.get('usenetrc', False): - try: - info = netrc.netrc().authenticators(self._NETRC_MACHINE) - if info is not None: - username = info[0] - password = info[2] - else: - raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE) - except (IOError, netrc.NetrcParseError) as err: - self._downloader.report_warning(u'parsing .netrc: %s' % compat_str(err)) - return - - # Set language - request = compat_urllib_request.Request(self._LANG_URL) - try: - self.report_lang() - compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to set language: %s' % compat_str(err)) - return - - # No authentication to be performed - if username is None: - return - - request = compat_urllib_request.Request(self._LOGIN_URL) - try: - login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) - return - - galx = None - dsh = None - match = re.search(re.compile(r']* id="gaia_loginform"', login_results) is not None: - self._downloader.report_warning(u'unable to log in: bad username or password') - return - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) - return - - # Confirm age - age_form = { - 'next_url': '/', - 'action_confirm': 'Confirm', - } - request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) - try: - self.report_age_confirmation() - compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) + print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'), + self._video_dimensions.get(x, '???'), + ' (3D)' if x in self._3d_itags else '')) def _extract_id(self, url): mobj = re.match(self._VALID_URL, url, re.VERBOSE) @@ -353,7 +479,61 @@ class YoutubeIE(InfoExtractor): video_id = mobj.group(2) return video_id + def _get_video_url_list(self, url_map): + """ + Transform a dictionary in the format {itag:url} to a list of (itag, url) + with the requested formats. + """ + req_format = self._downloader.params.get('format', None) + format_limit = self._downloader.params.get('format_limit', None) + available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats + if format_limit is not None and format_limit in available_formats: + format_list = available_formats[available_formats.index(format_limit):] + else: + format_list = available_formats + existing_formats = [x for x in format_list if x in url_map] + if len(existing_formats) == 0: + raise ExtractorError(u'no known formats available for video') + if self._downloader.params.get('listformats', None): + self._print_formats(existing_formats) + return + if req_format is None or req_format == 'best': + video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality + elif req_format == 'worst': + video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality + elif req_format in ('-1', 'all'): + video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats + else: + # Specific formats. We pick the first in a slash-delimeted sequence. + # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. + req_formats = req_format.split('/') + video_url_list = None + for rf in req_formats: + if rf in url_map: + video_url_list = [(rf, url_map[rf])] + break + if video_url_list is None: + raise ExtractorError(u'requested format not available') + return video_url_list + + def _extract_from_m3u8(self, manifest_url, video_id): + url_map = {} + def _get_urls(_manifest): + lines = _manifest.split('\n') + urls = filter(lambda l: l and not l.startswith('#'), + lines) + return urls + manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') + formats_urls = _get_urls(manifest) + for format_url in formats_urls: + itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag') + url_map[itag] = format_url + return url_map + def _real_extract(self, url): + if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): + 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 ).') + # Extract original video URL from URL with redirection, like age verification, using next_url parameter mobj = re.search(self._NEXT_URL_RE, url) if mobj: @@ -380,18 +560,38 @@ class YoutubeIE(InfoExtractor): # Get video info self.report_video_info_webpage_download(video_id) - for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: - video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' - % (video_id, el_type)) + if re.search(r'player-age-gate-content">', video_webpage) is not None: + self.report_age_confirmation() + age_gate = True + # We simulate the access to the video from www.youtube.com/v/{video_id} + # this can be viewed without login into Youtube + data = compat_urllib_parse.urlencode({'video_id': video_id, + 'el': 'embedded', + 'gl': 'US', + 'hl': 'en', + 'eurl': 'https://youtube.googleapis.com/v/' + video_id, + 'asv': 3, + 'sts':'1588', + }) + video_info_url = 'https://www.youtube.com/get_video_info?' + data video_info_webpage = self._download_webpage(video_info_url, video_id, note=False, errnote='unable to download video info webpage') video_info = compat_parse_qs(video_info_webpage) - if 'token' in video_info: - break + else: + age_gate = False + for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: + video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' + % (video_id, el_type)) + video_info_webpage = self._download_webpage(video_info_url, video_id, + note=False, + errnote='unable to download video info webpage') + video_info = compat_parse_qs(video_info_webpage) + if 'token' in video_info: + break if 'token' not in video_info: if 'reason' in video_info: - raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0]) + raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True) else: raise ExtractorError(u'"token" parameter not in video info for unknown reason') @@ -421,7 +621,12 @@ class YoutubeIE(InfoExtractor): video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) # thumbnail image - if 'thumbnail_url' not in video_info: + # We try first to get a high quality image: + m_thumb = re.search(r'', + video_webpage, re.DOTALL) + if m_thumb is not None: + video_thumbnail = m_thumb.group(1) + elif 'thumbnail_url' not in video_info: self._downloader.report_warning(u'unable to extract video thumbnail') video_thumbnail = '' else: # don't panic if we can't find it @@ -453,14 +658,13 @@ class YoutubeIE(InfoExtractor): if video_subtitles: (sub_error, sub_lang, sub) = video_subtitles[0] if sub_error: - # We try with the automatic captions - video_subtitles = self._request_automatic_caption(video_id, video_webpage) - (sub_error_auto, sub_lang, sub) = video_subtitles[0] - if sub is not None: - pass - else: - # We report the original error - self._downloader.report_warning(sub_error) + self._downloader.report_warning(sub_error) + + if self._downloader.params.get('writeautomaticsub', False): + video_subtitles = self._request_automatic_caption(video_id, video_webpage) + (sub_error, sub_lang, sub) = video_subtitles[0] + if sub_error: + self._downloader.report_warning(sub_error) if self._downloader.params.get('allsubtitles', False): video_subtitles = self._extract_all_subtitles(video_id) @@ -480,15 +684,18 @@ class YoutubeIE(InfoExtractor): video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]) # Decide which formats to download - req_format = self._downloader.params.get('format', None) try: mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage) + if not mobj: + raise ValueError('Could not find vevo ID') info = json.loads(mobj.group(1)) args = info['args'] - if args.get('ptk','') == 'vevo' or 'dashmpd': - # Vevo videos with encrypted signatures - self.to_screen(u'%s: Vevo video detected.' % video_id) + # Easy way to know if the 's' value is in url_encoded_fmt_stream_map + # this signatures are encrypted + m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) + if m_s is not None: + self.to_screen(u'%s: Encrypted signatures detected.' % video_id) video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] except ValueError: pass @@ -497,6 +704,8 @@ class YoutubeIE(InfoExtractor): self.report_rtmp_download() video_url_list = [(None, video_info['conn'][0])] elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1: + if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]: + raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) url_map = {} for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): url_data = compat_parse_qs(url_data_str) @@ -505,41 +714,38 @@ class YoutubeIE(InfoExtractor): if 'sig' in url_data: url += '&signature=' + url_data['sig'][0] elif 's' in url_data: - signature = self._decrypt_signature(url_data['s'][0]) + if self._downloader.params.get('verbose'): + s = url_data['s'][0] + if age_gate: + player_version = self._search_regex(r'ad3-(.+?)\.swf', + video_info['ad3_module'][0], 'flash player', + fatal=False) + player = 'flash player %s' % player_version + else: + player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, + 'html5 player', fatal=False) + parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.')) + self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % + (len(s), parts_sizes, url_data['itag'][0], player)) + encrypted_sig = url_data['s'][0] + if age_gate: + signature = self._decrypt_signature_age_gate(encrypted_sig) + else: + signature = self._decrypt_signature(encrypted_sig) url += '&signature=' + signature if 'ratebypass' not in url: url += '&ratebypass=yes' url_map[url_data['itag'][0]] = url - - format_limit = self._downloader.params.get('format_limit', None) - available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats - if format_limit is not None and format_limit in available_formats: - format_list = available_formats[available_formats.index(format_limit):] - else: - format_list = available_formats - existing_formats = [x for x in format_list if x in url_map] - if len(existing_formats) == 0: - raise ExtractorError(u'no known formats available for video') - if self._downloader.params.get('listformats', None): - self._print_formats(existing_formats) + video_url_list = self._get_video_url_list(url_map) + if not video_url_list: return - if req_format is None or req_format == 'best': - video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality - elif req_format == 'worst': - video_url_list = [(existing_formats[len(existing_formats)-1], url_map[existing_formats[len(existing_formats)-1]])] # worst quality - elif req_format in ('-1', 'all'): - video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats - else: - # Specific formats. We pick the first in a slash-delimeted sequence. - # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. - req_formats = req_format.split('/') - video_url_list = None - for rf in req_formats: - if rf in url_map: - video_url_list = [(rf, url_map[rf])] - break - if video_url_list is None: - raise ExtractorError(u'requested format not available') + elif video_info.get('hlsvp'): + manifest_url = video_info['hlsvp'][0] + url_map = self._extract_from_m3u8(manifest_url, video_id) + video_url_list = self._get_video_url_list(url_map) + if not video_url_list: + return + else: raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') @@ -548,8 +754,9 @@ class YoutubeIE(InfoExtractor): # Extension video_extension = self._video_extensions.get(format_param, 'flv') - video_format = '{0} - {1}'.format(format_param if format_param else video_extension, - self._video_dimensions.get(format_param, '???')) + video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension, + self._video_dimensions.get(format_param, '???'), + ' (3D)' if format_param in self._3d_itags else '') results.append({ 'id': video_id, @@ -569,8 +776,7 @@ class YoutubeIE(InfoExtractor): return results class YoutubePlaylistIE(InfoExtractor): - """Information Extractor for YouTube playlists.""" - + IE_DESC = u'YouTube.com playlists' _VALID_URL = r"""(?: (?:https?://)? (?:\w+\.)? @@ -580,10 +786,10 @@ class YoutubePlaylistIE(InfoExtractor): \? (?:.*?&)*? (?:p|a|list)= | p/ ) - ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) + ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,}) .* | - ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) + ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,}) )""" _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none' _MAX_RESULTS = 50 @@ -602,11 +808,14 @@ class YoutubePlaylistIE(InfoExtractor): # Download playlist videos from API playlist_id = mobj.group(1) or mobj.group(2) - page_num = 1 videos = [] - while True: - url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) + for page_num in itertools.count(1): + start_index = self._MAX_RESULTS * (page_num - 1) + 1 + if start_index >= 1000: + self._downloader.report_warning(u'Max number of results reached') + break + url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index) page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) try: @@ -626,23 +835,18 @@ class YoutubePlaylistIE(InfoExtractor): if 'media$group' in entry and 'media$player' in entry['media$group']: videos.append((index, entry['media$group']['media$player']['url'])) - if len(response['feed']['entry']) < self._MAX_RESULTS: - break - page_num += 1 - videos = [v[1] for v in sorted(videos)] - url_results = [self.url_result(url, 'Youtube') for url in videos] + url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] return [self.playlist_result(url_results, playlist_id, playlist_title)] class YoutubeChannelIE(InfoExtractor): - """Information Extractor for YouTube channels.""" - + IE_DESC = u'YouTube.com channels' _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' _MORE_PAGES_INDICATOR = 'yt-uix-load-more' - _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' + _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' IE_NAME = u'youtube:channel' def extract_videos_from_page(self, page): @@ -673,9 +877,7 @@ class YoutubeChannelIE(InfoExtractor): # Download any subsequent channel pages using the json-based channel_ajax query if self._MORE_PAGES_INDICATOR in page: - while True: - pagenum = pagenum + 1 - + for pagenum in itertools.count(1): url = self._MORE_PAGES_URL % (pagenum, channel_id) page = self._download_webpage(url, channel_id, u'Downloading page #%s' % pagenum) @@ -691,13 +893,12 @@ class YoutubeChannelIE(InfoExtractor): self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] - url_entries = [self.url_result(url, 'Youtube') for url in urls] + url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] return [self.playlist_result(url_entries, channel_id)] class YoutubeUserIE(InfoExtractor): - """Information Extractor for YouTube users.""" - + IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' _GDATA_PAGE_SIZE = 50 @@ -719,9 +920,8 @@ class YoutubeUserIE(InfoExtractor): # all of them. video_ids = [] - pagenum = 0 - while True: + for pagenum in itertools.count(0): start_index = pagenum * self._GDATA_PAGE_SIZE + 1 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) @@ -746,14 +946,12 @@ class YoutubeUserIE(InfoExtractor): if len(ids_in_page) < self._GDATA_PAGE_SIZE: break - pagenum += 1 - urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] - url_results = [self.url_result(url, 'Youtube') for url in urls] + url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] return [self.playlist_result(url_results, playlist_title = username)] class YoutubeSearchIE(SearchInfoExtractor): - """Information Extractor for YouTube search queries.""" + IE_DESC = u'YouTube.com searches' _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' _MAX_RESULTS = 1000 IE_NAME = u'youtube:search' @@ -793,3 +991,92 @@ class YoutubeSearchIE(SearchInfoExtractor): video_ids = video_ids[:n] videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids] return self.playlist_result(videos, query) + + +class YoutubeShowIE(InfoExtractor): + IE_DESC = u'YouTube.com (multi-season) shows' + _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' + IE_NAME = u'youtube:show' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + show_name = mobj.group(1) + webpage = self._download_webpage(url, show_name, u'Downloading show webpage') + # There's one playlist for each season of the show + m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) + self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) + return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] + + +class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): + """ + Base class for extractors that fetch info from + http://www.youtube.com/feed_ajax + Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. + """ + _LOGIN_REQUIRED = True + _PAGING_STEP = 30 + # use action_load_personal_feed instead of action_load_system_feed + _PERSONAL_FEED = False + + @property + def _FEED_TEMPLATE(self): + action = 'action_load_system_feed' + if self._PERSONAL_FEED: + action = 'action_load_personal_feed' + return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) + + @property + def IE_NAME(self): + return u'youtube:%s' % self._FEED_NAME + + def _real_initialize(self): + self._login() + + def _real_extract(self, url): + feed_entries = [] + # The step argument is available only in 2.7 or higher + for i in itertools.count(0): + paging = i*self._PAGING_STEP + info = self._download_webpage(self._FEED_TEMPLATE % paging, + u'%s feed' % self._FEED_NAME, + u'Downloading page %s' % i) + info = json.loads(info) + feed_html = info['feed_html'] + m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) + ids = orderedSet(m.group(1) for m in m_ids) + feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) + if info['paging'] is None: + break + return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) + +class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' + _FEED_NAME = 'subscriptions' + _PLAYLIST_TITLE = u'Youtube Subscriptions' + +class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' + _FEED_NAME = 'recommended' + _PLAYLIST_TITLE = u'Youtube Recommended videos' + +class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): + IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' + _FEED_NAME = 'watch_later' + _PLAYLIST_TITLE = u'Youtube Watch Later' + _PAGING_STEP = 100 + _PERSONAL_FEED = True + +class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): + IE_NAME = u'youtube:favorites' + IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:o?rites)?' + _LOGIN_REQUIRED = True + + def _real_extract(self, url): + webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos') + playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id') + return self.url_result(playlist_id, 'YoutubePlaylist')