X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/bddc9fc577d16b1428924bf8a5c37ef1d9295f14..83c8b40add31fb501d587b64fb17a936ffd5dbc5:/youtube_dl/extractor/youtube.py?ds=sidebyside diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index fbe8d63..bd24a28 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1,795 +1,2517 @@ # coding: utf-8 +from __future__ import unicode_literals + + +import itertools import json -import netrc +import os.path +import random import re -import socket +import time +import traceback from .common import InfoExtractor, SearchInfoExtractor -from ..utils import ( - compat_http_client, +from ..jsinterp import JSInterpreter +from ..swfinterp import SWFInterpreter +from ..compat import ( + compat_chr, compat_parse_qs, - compat_urllib_error, - compat_urllib_parse, - compat_urllib_request, + compat_urllib_parse_unquote, + compat_urllib_parse_unquote_plus, + compat_urllib_parse_urlencode, + compat_urllib_parse_urlparse, + compat_urlparse, compat_str, - +) +from ..utils import ( clean_html, - get_element_by_id, + error_to_compat_str, ExtractorError, + float_or_none, + get_element_by_attribute, + get_element_by_id, + int_or_none, + mimetype2ext, + orderedSet, + parse_duration, + remove_quotes, + remove_start, + sanitized_Request, + smuggle_url, + str_to_int, unescapeHTML, unified_strdate, + unsmuggle_url, + uppercase_escape, + urlencode_postdata, + ISO3166Utils, ) -class YoutubeIE(InfoExtractor): - """Information extractor for youtube.com.""" +class YoutubeBaseInfoExtractor(InfoExtractor): + """Provide base functions for Youtube extractors""" + _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge' + _PASSWORD_CHALLENGE_URL = 'https://accounts.google.com/signin/challenge/sl/password' + _NETRC_MACHINE = 'youtube' + # If True it will raise an error if no login info is provided + _LOGIN_REQUIRED = False + + def _set_language(self): + self._set_cookie( + '.youtube.com', 'PREF', 'f1=50000000&hl=en', + # YouTube sets the expire time to about two months + expire_time=time.time() + 2 * 30 * 24 * 3600) + + def _ids_to_results(self, ids): + return [ + self.url_result(vid_id, 'Youtube', video_id=vid_id) + for vid_id in ids] + + def _login(self): + """ + Attempt to log in to YouTube. + True is returned if successful or skipped. + False is returned if login failed. + + If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. + """ + (username, password) = self._get_login_info() + # No authentication to be performed + if username is None: + if self._LOGIN_REQUIRED: + raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) + return True + + login_page = self._download_webpage( + self._LOGIN_URL, None, + note='Downloading login page', + errnote='unable to fetch login page', fatal=False) + if login_page is False: + return + + login_form = self._hidden_inputs(login_page) + + login_form.update({ + 'checkConnection': 'youtube', + 'Email': username, + 'Passwd': password, + }) + + login_results = self._download_webpage( + self._PASSWORD_CHALLENGE_URL, None, + note='Logging in', errnote='unable to log in', fatal=False, + data=urlencode_postdata(login_form)) + if login_results is False: + return False + + error_msg = self._html_search_regex( + r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<', + login_results, 'error message', default=None) + if error_msg: + raise ExtractorError('Unable to login: %s' % error_msg, expected=True) + + if re.search(r'id="errormsg_0_Passwd"', login_results) is not None: + raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True) + + # Two-Factor + # TODO add SMS and phone call support - these require making a request and then prompting the user + + if re.search(r'(?i)]+id="challenge"', login_results) is not None: + tfa_code = self._get_tfa_info('2-step verification code') + + if not tfa_code: + self._downloader.report_warning( + 'Two-factor authentication required. Provide it either interactively or with --twofactor ' + '(Note that only TOTP (Google Authenticator App) codes work at this time.)') + return False + + tfa_code = remove_start(tfa_code, 'G-') + + tfa_form_strs = self._form_hidden_inputs('challenge', login_results) - _VALID_URL = r"""^ + tfa_form_strs.update({ + 'Pin': tfa_code, + 'TrustDevice': 'on', + }) + + tfa_data = urlencode_postdata(tfa_form_strs) + + tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data) + tfa_results = self._download_webpage( + tfa_req, None, + note='Submitting TFA code', errnote='unable to submit tfa', fatal=False) + + if tfa_results is False: + return False + + if re.search(r'(?i)]+id="challenge"', tfa_results) is not None: + self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.') + return False + if re.search(r'(?i)]+id="gaia_loginform"', tfa_results) is not None: + self._downloader.report_warning('unable to log in - did the page structure change?') + return False + if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None: + self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.') + return False + + if re.search(r'(?i)]+id="gaia_loginform"', login_results) is not None: + self._downloader.report_warning('unable to log in: bad username or password') + return False + return True + + def _real_initialize(self): + if self._downloader is None: + return + self._set_language() + if not self._login(): + return + + +class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor): + # Extract entries from page with "Load more" button + def _entries(self, page, playlist_id): + more_widget_html = content_html = page + for page_num in itertools.count(1): + for entry in self._process_page(content_html): + yield entry + + mobj = re.search(r'data-uix-load-more-href="/?(?P[^"]+)"', more_widget_html) + if not mobj: + break + + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), playlist_id, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + if not content_html.strip(): + # Some webpages show a "Load more" button but they don't + # have more videos + break + more_widget_html = more['load_more_widget_html'] + + +class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for video_id, video_title in self.extract_videos_from_page(content): + yield self.url_result(video_id, 'Youtube', video_id, video_title) + + def extract_videos_from_page(self, page): + ids_in_page = [] + titles_in_page = [] + for mobj in re.finditer(self._VIDEO_RE, page): + # The link with index 0 is not the first video of the playlist (not sure if still actual) + if 'index' in mobj.groupdict() and mobj.group('id') == '0': + continue + video_id = mobj.group('id') + video_title = unescapeHTML(mobj.group('title')) + if video_title: + video_title = video_title.strip() + try: + idx = ids_in_page.index(video_id) + if video_title and not titles_in_page[idx]: + titles_in_page[idx] = video_title + except ValueError: + ids_in_page.append(video_id) + titles_in_page.append(video_title) + return zip(ids_in_page, titles_in_page) + + +class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): + def _process_page(self, content): + for playlist_id in orderedSet(re.findall( + r']+class="[^"]*yt-lockup-title[^"]*"[^>]*>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', + content)): + yield self.url_result( + 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist') + + def _real_extract(self, url): + playlist_id = self._match_id(url) + webpage = self._download_webpage(url, playlist_id) + title = self._og_search_title(webpage, fatal=False) + return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title) + + +class YoutubeIE(YoutubeBaseInfoExtractor): + IE_DESC = 'YouTube.com' + _VALID_URL = r"""(?x)^ ( - (?:https?://)? # http(s):// (optional) - (?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/| - tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains + (?:https?://|//) # http(s):// or protocol-independent URL + (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/| + (?:www\.)?deturl\.com/www\.youtube\.com/| + (?:www\.)?pwnyoutube\.com/| + (?:www\.)?yourepeat\.com/| + tube\.majestyc\.net/| + youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls (?: # the various things that can precede the ID: - (?:(?:v|embed|e)/) # v/ or embed/ or e/ + (?:(?:v|embed|e)/(?!videoseries)) # 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) + (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY) v= ) - )? # optional -> youtube.com/xxxx is OK + )) + |(?: + youtu\.be| # just youtu.be/xxxx + vid\.plus| # or vid.plus/xxxx + zwearz\.com/watch| # or zwearz.com/watch/xxxx + )/ + |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= + ) )? # all until now is optional -> you can pass the naked ID - ([0-9A-Za-z_-]+) # here is it! the YouTube video ID + ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID + (?!.*?\blist=) # combined list/video URLs are handled by the playlist IE (?(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'] - _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 - '43': 'webm', - '44': 'webm', - '45': 'webm', - '46': 'webm', - } - _video_dimensions = { - '5': '240x400', - '6': '???', - '13': '???', - '17': '144x176', - '18': '360x640', - '22': '720x1280', - '34': '360x640', - '35': '480x854', - '37': '1080x1920', - '38': '3072x4096', - '43': '360x640', - '44': '480x854', - '45': '720x1280', - '46': '1080x1920', + _formats = { + '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, + '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, + '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'}, + '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'}, + '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'}, + '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well + '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'}, + '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + + + # 3D videos + '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20}, + '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + + # Apple HTTP Live Streaming + '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, + + # DASH mp4 video + '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559) + '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, + '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40}, + '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, + + # Dash mp4 audio + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, + '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'}, + '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'}, + + # Dash webm + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40}, + '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) + '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40}, + + # Dash webm audio + '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, + '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, + + # Dash webm audio with opus inside + '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, + '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, + '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, + + # RTMP (unnamed) + '_rtmp': {'protocol': 'rtmp'}, } - IE_NAME = u'youtube' - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - if YoutubePlaylistIE.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') + _SUBTITLE_FORMATS = ('ttml', 'vtt') + + IE_NAME = 'youtube' + _TESTS = [ + { + 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9', + 'info_dict': { + 'id': 'BaW_jenozKc', + 'ext': 'mp4', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', + 'uploader': 'Philipp Hagemeister', + 'uploader_id': 'phihag', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag', + 'upload_date': '20121002', + 'license': 'Standard YouTube License', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], + 'like_count': int, + 'dislike_count': int, + 'start_time': 1, + 'end_time': 9, + } + }, + { + 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY', + 'note': 'Test generic use_cipher_signature video (#897)', + 'info_dict': { + 'id': 'UxxajLWwzqY', + 'ext': 'mp4', + 'upload_date': '20120506', + 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', + 'alt_title': 'I Love It (feat. Charli XCX)', + 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8', + 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli', + 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop', + 'iconic ep', 'iconic', 'love', 'it'], + 'uploader': 'Icona Pop', + 'uploader_id': 'IconaPop', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop', + 'license': 'Standard YouTube License', + 'creator': 'Icona Pop', + } + }, + { + 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ', + 'note': 'Test VEVO video with age protection (#956)', + 'info_dict': { + 'id': '07FYdnEawAQ', + 'ext': 'mp4', + 'upload_date': '20130703', + 'title': 'Justin Timberlake - Tunnel Vision (Explicit)', + 'alt_title': 'Tunnel Vision', + 'description': 'md5:64249768eec3bc4276236606ea996373', + 'uploader': 'justintimberlakeVEVO', + 'uploader_id': 'justintimberlakeVEVO', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO', + 'license': 'Standard YouTube License', + 'creator': 'Justin Timberlake', + 'age_limit': 18, + } + }, + { + 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ', + 'note': 'Embed-only video (#1746)', + 'info_dict': { + 'id': 'yZIXLfi8CZQ', + 'ext': 'mp4', + 'upload_date': '20120608', + 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012', + 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7', + 'uploader': 'SET India', + 'uploader_id': 'setindia', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia', + 'license': 'Standard YouTube License', + 'age_limit': 18, + } + }, + { + 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY', + 'note': 'Use the first video ID in the URL', + 'info_dict': { + 'id': 'BaW_jenozKc', + 'ext': 'mp4', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', + 'uploader': 'Philipp Hagemeister', + 'uploader_id': 'phihag', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag', + 'upload_date': '20121002', + 'license': 'Standard YouTube License', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'categories': ['Science & Technology'], + 'tags': ['youtube-dl'], + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, + { + 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I', + 'note': '256k DASH audio (format 141) via DASH manifest', + 'info_dict': { + 'id': 'a9LDPn-MO4I', + 'ext': 'm4a', + 'upload_date': '20121002', + 'uploader_id': '8KVIDEO', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO', + 'description': '', + 'uploader': '8KVIDEO', + 'license': 'Standard YouTube License', + 'title': 'UHDTV TEST 8K VIDEO.mp4' + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', + }, + 'skip': 'format 141 not served anymore', + }, + # DASH manifest with encrypted signature + { + 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA', + 'info_dict': { + 'id': 'IB3lcPjvWLA', + 'ext': 'm4a', + 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson', + 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d', + 'uploader': 'AfrojackVEVO', + 'uploader_id': 'AfrojackVEVO', + 'upload_date': '20131011', + 'license': 'Standard YouTube License', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141/bestaudio[ext=m4a]', + }, + }, + # JS player signature function name containing $ + { + 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM', + 'info_dict': { + 'id': 'nfWlot6h_JM', + 'ext': 'm4a', + 'title': 'Taylor Swift - Shake It Off', + 'alt_title': 'Shake It Off', + 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3', + 'uploader': 'TaylorSwiftVEVO', + 'uploader_id': 'TaylorSwiftVEVO', + 'upload_date': '20140818', + 'license': 'Standard YouTube License', + 'creator': 'Taylor Swift', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141/bestaudio[ext=m4a]', + }, + }, + # Controversy video + { + 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8', + 'info_dict': { + 'id': 'T4XJQO3qol8', + 'ext': 'mp4', + 'upload_date': '20100909', + 'uploader': 'The Amazing Atheist', + 'uploader_id': 'TheAmazingAtheist', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist', + 'license': 'Standard YouTube License', + 'title': 'Burning Everyone\'s Koran', + 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html', + } + }, + # Normal age-gate video (No vevo, embed allowed) + { + 'url': 'https://youtube.com/watch?v=HtVdAasjOgU', + 'info_dict': { + 'id': 'HtVdAasjOgU', + 'ext': 'mp4', + 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer', + 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}', + 'uploader': 'The Witcher', + 'uploader_id': 'WitcherGame', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame', + 'upload_date': '20140605', + 'license': 'Standard YouTube License', + 'age_limit': 18, + }, + }, + # Age-gate video with encrypted signature + { + 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU', + 'info_dict': { + 'id': '6kLq3WMV1nU', + 'ext': 'mp4', + 'title': 'Dedication To My Ex (Miss That) (Lyric Video)', + 'description': 'md5:33765bb339e1b47e7e72b5490139bb41', + 'uploader': 'LloydVEVO', + 'uploader_id': 'LloydVEVO', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO', + 'upload_date': '20110629', + 'license': 'Standard YouTube License', + 'age_limit': 18, + }, + }, + # video_info is None (https://github.com/rg3/youtube-dl/issues/4421) + { + 'url': '__2ABJjxzNo', + 'info_dict': { + 'id': '__2ABJjxzNo', + 'ext': 'mp4', + 'upload_date': '20100430', + 'uploader_id': 'deadmau5', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5', + 'creator': 'deadmau5', + 'description': 'md5:12c56784b8032162bb936a5f76d55360', + 'uploader': 'deadmau5', + 'license': 'Standard YouTube License', + 'title': 'Deadmau5 - Some Chords (HD)', + 'alt_title': 'Some Chords', + }, + 'expected_warnings': [ + 'DASH manifest missing', + ] + }, + # Olympics (https://github.com/rg3/youtube-dl/issues/4431) + { + 'url': 'lqQg6PlCWgI', + 'info_dict': { + 'id': 'lqQg6PlCWgI', + 'ext': 'mp4', + 'upload_date': '20150827', + 'uploader_id': 'olympic', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic', + 'license': 'Standard YouTube License', + 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', + 'uploader': 'Olympic', + 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', + }, + 'params': { + 'skip_download': 'requires avconv', + } + }, + # Non-square pixels + { + 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0', + 'info_dict': { + 'id': '_b-2C3KPAM0', + 'ext': 'mp4', + 'stretched_ratio': 16 / 9., + 'upload_date': '20110310', + 'uploader_id': 'AllenMeow', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow', + 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯', + 'uploader': '孫艾倫', + 'license': 'Standard YouTube License', + 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人', + }, + }, + # url_encoded_fmt_stream_map is empty string + { + 'url': 'qEJwOuvDf7I', + 'info_dict': { + 'id': 'qEJwOuvDf7I', + 'ext': 'webm', + 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге', + 'description': '', + 'upload_date': '20150404', + 'uploader_id': 'spbelect', + 'uploader': 'Наблюдатели Петербурга', + }, + 'params': { + 'skip_download': 'requires avconv', + }, + 'skip': 'This live event has ended.', + }, + # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097) + { + 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y', + 'info_dict': { + 'id': 'FIl7x6_3R5Y', + 'ext': 'mp4', + 'title': 'md5:7b81415841e02ecd4313668cde88737a', + 'description': 'md5:116377fd2963b81ec4ce64b542173306', + 'upload_date': '20150625', + 'uploader_id': 'dorappi2000', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000', + 'uploader': 'dorappi2000', + 'license': 'Standard YouTube License', + 'formats': 'mincount:32', + }, + }, + # DASH manifest with segment_list + { + 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8', + 'md5': '8ce563a1d667b599d21064e982ab9e31', + 'info_dict': { + 'id': 'CsmdDsKjzN8', + 'ext': 'mp4', + 'upload_date': '20150501', # According to '[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P[a-z]+)$', + player_url) + if not id_m: + raise ExtractorError('Cannot identify player %r' % player_url) + player_type = id_m.group('ext') + player_id = id_m.group('id') + + # Read from filesystem cache + func_id = '%s_%s_%s' % ( + player_type, player_id, self._signature_cache_id(example_sig)) + assert os.path.basename(func_id) == func_id + + cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id) + if cache_spec is not None: + return lambda s: ''.join(s[i] for i in cache_spec) + + download_note = ( + 'Downloading player %s' % player_url + if self._downloader.params.get('verbose') else + 'Downloading %s player %s' % (player_type, player_id) + ) + if player_type == 'js': + code = self._download_webpage( + player_url, video_id, + note=download_note, + errnote='Download of %s failed' % player_url) + res = self._parse_sig_js(code) + elif player_type == 'swf': + urlh = self._request_webpage( + player_url, video_id, + note=download_note, + errnote='Download of %s failed' % player_url) + code = urlh.read() + res = self._parse_sig_swf(code) + else: + assert False, 'Invalid player type %r' % player_type + + test_string = ''.join(map(compat_chr, range(len(example_sig)))) + cache_res = res(test_string) + cache_spec = [ord(c) for c in cache_res] + + self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) + return res + + def _print_sig_code(self, func, example_sig): + def gen_sig_code(idxs): + def _genslice(start, end, step): + starts = '' if start == 0 else str(start) + ends = (':%d' % (end + step)) if end + step >= 0 else ':' + steps = '' if step == 1 else (':%d' % step) + return 's[%s%s%s]' % (starts, ends, steps) + + step = None + # Quelch pyflakes warnings - start will be set when step is set + start = '(Never used)' + for i, prev in zip(idxs[1:], idxs[:-1]): + if step is not None: + if i - prev == step: + continue + yield _genslice(start, prev, step) + step = None + continue + if i - prev in [-1, 1]: + step = i - prev + start = prev + continue + else: + yield 's[%d]' % prev + if step is None: + yield 's[%d]' % i + else: + yield _genslice(start, i, step) + + test_string = ''.join(map(compat_chr, range(len(example_sig)))) + cache_res = func(test_string) + cache_spec = [ord(c) for c in cache_res] + expr_code = ' + '.join(gen_sig_code(cache_spec)) + signature_id_tuple = '(%s)' % ( + ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) + code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n' + ' return %s\n') % (signature_id_tuple, expr_code) + self.to_screen('Extracted signature function:\n' + code) + + def _parse_sig_js(self, jscode): + funcname = self._search_regex( + r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode, + 'Initial JS player signature function name') + + jsi = JSInterpreter(jscode) + initial_function = jsi.extract_function(funcname) + return lambda s: initial_function([s]) + + def _parse_sig_swf(self, file_contents): + swfi = SWFInterpreter(file_contents) + TARGET_CLASSNAME = 'SignatureDecipher' + searched_class = swfi.extract_class(TARGET_CLASSNAME) + initial_function = swfi.extract_function(searched_class, 'decipher') + return lambda s: initial_function([s]) + + def _decrypt_signature(self, s, video_id, player_url, age_gate=False): + """Turn the encrypted s field into a working signature""" + + if player_url is None: + raise ExtractorError('Cannot decrypt signature without player_url') + + if player_url.startswith('//'): + player_url = 'https:' + player_url + try: + player_id = (player_url, self._signature_cache_id(s)) + if player_id not in self._player_cache: + func = self._extract_signature_function( + video_id, player_url, s + ) + self._player_cache[player_id] = func + func = self._player_cache[player_id] + if self._downloader.params.get('youtube_print_sig_code'): + self._print_sig_code(func, s) + return func(s) + except Exception as e: + tb = traceback.format_exc() + raise ExtractorError( + 'Signature extraction failed: ' + tb, cause=e) + + def _get_subtitles(self, video_id, webpage): try: - sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None) - sub_lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) - sub_lang_list = dict((l[1], l[0]) for l in sub_lang_list) + subs_doc = self._download_xml( + 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, + video_id, note=False) + except ExtractorError as err: + self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err)) + return {} + + sub_lang_list = {} + for track in subs_doc.findall('track'): + lang = track.attrib['lang_code'] + if lang in sub_lang_list: + continue + sub_formats = [] + for ext in self._SUBTITLE_FORMATS: + params = compat_urllib_parse_urlencode({ + 'lang': lang, + 'v': video_id, + 'fmt': ext, + 'name': track.attrib['name'].encode('utf-8'), + }) + sub_formats.append({ + 'url': 'https://www.youtube.com/api/timedtext?' + params, + 'ext': ext, + }) + sub_lang_list[lang] = sub_formats if not sub_lang_list: - return (u'video doesn\'t have subtitles', None) + self._downloader.report_warning('video doesn\'t have subtitles') + return {} return sub_lang_list - def _list_available_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - self.report_video_subtitles_available(video_id, sub_lang_list) - - def _request_subtitle(self, sub_lang, sub_name, video_id, format): - """ - Return tuple: - (error_message, sub_lang, sub) - """ - self.report_video_subtitles_request(video_id, sub_lang, format) - params = compat_urllib_parse.urlencode({ - 'lang': sub_lang, - 'name': sub_name, - 'v': video_id, - 'fmt': format, - }) - url = 'http://www.youtube.com/api/timedtext?' + params - try: - sub = compat_urllib_request.urlopen(url).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - return (u'unable to download video subtitles: %s' % compat_str(err), None, None) - if not sub: - return (u'Did not fetch video subtitles', None, None) - return (None, sub_lang, sub) - - def _request_automatic_caption(self, video_id, webpage): + def _get_ytplayer_config(self, video_id, webpage): + patterns = ( + # User data may contain arbitrary character sequences that may affect + # JSON extraction with regex, e.g. when '};' is contained the second + # regex won't capture the whole JSON. Yet working around by trying more + # concrete regex first keeping in mind proper quoted string handling + # to be implemented in future that will replace this workaround (see + # https://github.com/rg3/youtube-dl/issues/7468, + # https://github.com/rg3/youtube-dl/pull/7599) + r';ytplayer\.config\s*=\s*({.+?});ytplayer', + r';ytplayer\.config\s*=\s*({.+?});', + ) + config = self._search_regex( + patterns, webpage, 'ytplayer.config', default=None) + if config: + return self._parse_json( + uppercase_escape(config), video_id, fatal=False) + + def _get_automatic_captions(self, video_id, webpage): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" - sub_lang = self._downloader.params.get('subtitleslang') or 'en' - sub_format = self._downloader.params.get('subtitlesformat') - self.to_screen(u'%s: Looking for automatic captions' % video_id) - mobj = re.search(r';ytplayer.config = ({.*?});', webpage) - err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang - if mobj is None: - return [(err_msg, None, None)] - player_config = json.loads(mobj.group(1)) + self.to_screen('%s: Looking for automatic captions' % video_id) + player_config = self._get_ytplayer_config(video_id, webpage) + err_msg = 'Couldn\'t find automatic captions for %s' % video_id + if not player_config: + self._downloader.report_warning(err_msg) + return {} try: - args = player_config[u'args'] - caption_url = args[u'ttsurl'] - timestamp = args[u'timestamp'] - params = compat_urllib_parse.urlencode({ - 'lang': 'en', - 'tlang': sub_lang, - 'fmt': sub_format, - 'ts': timestamp, - 'kind': 'asr', - }) - subtitles_url = caption_url + '&' + params - sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions') - return [(None, sub_lang, sub)] - except KeyError: - return [(err_msg, None, None)] - - def _extract_subtitle(self, video_id): - """ - Return a list with a tuple: - [(error_message, sub_lang, sub)] - """ - sub_lang_list = self._get_available_subtitles(video_id) - sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - if self._downloader.params.get('subtitleslang', False): - sub_lang = self._downloader.params.get('subtitleslang') - elif 'en' in sub_lang_list: - sub_lang = 'en' - else: - sub_lang = list(sub_lang_list.keys())[0] - if not sub_lang in sub_lang_list: - return [(u'no closed captions found in the specified language "%s"' % sub_lang, None, None)] - - subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - return [subtitle] - - def _extract_all_subtitles(self, video_id): - sub_lang_list = self._get_available_subtitles(video_id) - sub_format = self._downloader.params.get('subtitlesformat') - if isinstance(sub_lang_list,tuple): #There was some error, it didn't get the available subtitles - return [(sub_lang_list[0], None, None)] - subtitles = [] - for sub_lang in sub_lang_list: - subtitle = self._request_subtitle(sub_lang, sub_lang_list[sub_lang].encode('utf-8'), video_id, sub_format) - subtitles.append(subtitle) - return subtitles - - 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: + args = player_config['args'] + caption_url = args.get('ttsurl') + if caption_url: + timestamp = args['timestamp'] + # We get the available subtitles + list_params = compat_urllib_parse_urlencode({ + 'type': 'list', + 'tlangs': 1, + 'asrs': 1, + }) + list_url = caption_url + '&' + list_params + caption_list = self._download_xml(list_url, video_id) + original_lang_node = caption_list.find('track') + if original_lang_node is None: + self._downloader.report_warning('Video doesn\'t have automatic captions') + return {} + original_lang = original_lang_node.attrib['lang_code'] + caption_kind = original_lang_node.attrib.get('kind', '') + + sub_lang_list = {} + for lang_node in caption_list.findall('target'): + sub_lang = lang_node.attrib['lang_code'] + sub_formats = [] + for ext in self._SUBTITLE_FORMATS: + params = compat_urllib_parse_urlencode({ + 'lang': original_lang, + 'tlang': sub_lang, + 'fmt': ext, + 'ts': timestamp, + 'kind': caption_kind, + }) + sub_formats.append({ + 'url': caption_url + '&' + params, + 'ext': ext, + }) + sub_lang_list[sub_lang] = sub_formats + return sub_lang_list + + # Some videos don't provide ttsurl but rather caption_tracks and + # caption_translation_languages (e.g. 20LmZk1hakA) + caption_tracks = args['caption_tracks'] + caption_translation_languages = args['caption_translation_languages'] + caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0] + parsed_caption_url = compat_urllib_parse_urlparse(caption_url) + caption_qs = compat_parse_qs(parsed_caption_url.query) + + sub_lang_list = {} + for lang in caption_translation_languages.split(','): + lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang)) + sub_lang = lang_qs.get('lc', [None])[0] + if not sub_lang: + continue + sub_formats = [] + for ext in self._SUBTITLE_FORMATS: + caption_qs.update({ + 'tlang': [sub_lang], + 'fmt': [ext], + }) + sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace( + query=compat_urllib_parse_urlencode(caption_qs, True))) + sub_formats.append({ + 'url': sub_url, + 'ext': ext, + }) + sub_lang_list[sub_lang] = sub_formats + return sub_lang_list + # An extractor error can be raise by the download process if there are + # no automatic captions but there are subtitles + except (KeyError, ExtractorError): + self._downloader.report_warning(err_msg) + return {} + + def _mark_watched(self, video_id, video_info): + playback_url = video_info.get('videostats_playback_base_url', [None])[0] + if not playback_url: return + parsed_playback_url = compat_urlparse.urlparse(playback_url) + qs = compat_urlparse.parse_qs(parsed_playback_url.query) - username = None - password = None - downloader_params = self._downloader.params + # cpn generation algorithm is reverse engineered from base.js. + # In fact it works even with dummy cpn. + CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_' + cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))) - # 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 + qs.update({ + 'ver': ['2'], + 'cpn': [cpn], + }) + playback_url = compat_urlparse.urlunparse( + parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True))) - # 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)) + self._download_webpage( + playback_url, video_id, 'Marking watched', + 'Unable to mark watched', fatal=False) - def _extract_id(self, url): - mobj = re.match(self._VALID_URL, url, re.VERBOSE) + @classmethod + def extract_id(cls, url): + mobj = re.match(cls._VALID_URL, url, re.VERBOSE) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError('Invalid URL: %s' % url) video_id = mobj.group(2) return video_id + 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, '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 _extract_annotations(self, video_id): + url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id + return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') + def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + + proto = ( + 'http' if self._downloader.params.get('prefer_insecure', False) + else 'https') + + start_time = None + end_time = None + parsed_url = compat_urllib_parse_urlparse(url) + for component in [parsed_url.fragment, parsed_url.query]: + query = compat_parse_qs(component) + if start_time is None and 't' in query: + start_time = parse_duration(query['t'][0]) + if start_time is None and 'start' in query: + start_time = parse_duration(query['start'][0]) + if end_time is None and 'end' in query: + end_time = parse_duration(query['end'][0]) + # 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: - url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/') - video_id = self._extract_id(url) + url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/') + video_id = self.extract_id(url) # Get video webpage - self.report_video_webpage_download(video_id) - url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id - request = compat_urllib_request.Request(url) - try: - video_webpage_bytes = compat_urllib_request.urlopen(request).read() - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err)) - - video_webpage = video_webpage_bytes.decode('utf-8', 'ignore') + url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id + video_webpage = self._download_webpage(url, video_id) # Attempt to extract SWF player URL - mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) + mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) if mobj is not None: player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) else: player_url = None + dash_mpds = [] + + def add_dash_mpd(video_info): + dash_mpd = video_info.get('dashmpd') + if dash_mpd and dash_mpd[0] not in dash_mpds: + dash_mpds.append(dash_mpd[0]) + # 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)) - video_info_webpage = self._download_webpage(video_info_url, video_id, - note=False, - errnote='unable to download video info webpage') + embed_webpage = None + is_live = None + if re.search(r'player-age-gate-content">', video_webpage) is not None: + 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 + url = proto + '://www.youtube.com/embed/%s' % video_id + embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage') + data = compat_urllib_parse_urlencode({ + 'video_id': video_id, + 'eurl': 'https://youtube.googleapis.com/v/' + video_id, + 'sts': self._search_regex( + r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''), + }) + video_info_url = proto + '://www.youtube.com/get_video_info?' + data + video_info_webpage = self._download_webpage( + video_info_url, video_id, + note='Refetching age-gated info webpage', + errnote='unable to download video info webpage') video_info = compat_parse_qs(video_info_webpage) - if 'token' in video_info: - break + add_dash_mpd(video_info) + else: + age_gate = False + video_info = None + # Try looking directly into the video webpage + ytplayer_config = self._get_ytplayer_config(video_id, video_webpage) + if ytplayer_config: + args = ytplayer_config['args'] + if args.get('url_encoded_fmt_stream_map'): + # Convert to the same format returned by compat_parse_qs + video_info = dict((k, [v]) for k, v in args.items()) + add_dash_mpd(video_info) + # Rental video is not rented but preview is available (e.g. + # https://www.youtube.com/watch?v=yYr8q0y5Jfg, + # https://github.com/rg3/youtube-dl/issues/10532) + if not video_info and args.get('ypc_vid'): + return self.url_result( + args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid']) + if args.get('livestream') == '1' or args.get('live_playback') == 1: + is_live = True + if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): + # We also try looking in get_video_info since it may contain different dashmpd + # URL that points to a DASH manifest with possibly different itag set (some itags + # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH + # manifest pointed by get_video_info's dashmpd). + # The general idea is to take a union of itags of both DASH manifests (for example + # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093) + self.report_video_info_webpage_download(video_id) + for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']: + video_info_url = ( + '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' + % (proto, video_id, el_type)) + video_info_webpage = self._download_webpage( + video_info_url, + video_id, note=False, + errnote='unable to download video info webpage') + get_video_info = compat_parse_qs(video_info_webpage) + if get_video_info.get('use_cipher_signature') != ['True']: + add_dash_mpd(get_video_info) + if not video_info: + video_info = get_video_info + if 'token' in get_video_info: + # Different get_video_info requests may report different results, e.g. + # some may report video unavailability, but some may serve it without + # any complaint (see https://github.com/rg3/youtube-dl/issues/7362, + # the original webpage as well as el=info and el=embedded get_video_info + # requests report video unavailability due to geo restriction while + # el=detailpage succeeds and returns valid data). This is probably + # due to YouTube measures against IP ranges of hosting providers. + # Working around by preferring the first succeeded video_info containing + # the token if no such video_info yet was found. + if 'token' not in video_info: + video_info = get_video_info + break if 'token' not in video_info: if 'reason' in video_info: - raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0]) + if 'The uploader has not made this video available in your country.' in video_info['reason']: + regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None) + if regions_allowed: + raise ExtractorError('YouTube said: This video is available in %s only' % ( + ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))), + expected=True) + raise ExtractorError( + 'YouTube said: %s' % video_info['reason'][0], + expected=True, video_id=video_id) else: - raise ExtractorError(u'"token" parameter not in video info for unknown reason') + raise ExtractorError( + '"token" parameter not in video info for unknown reason', + video_id=video_id) + + # title + if 'title' in video_info: + video_title = video_info['title'][0] + else: + self._downloader.report_warning('Unable to extract video title') + video_title = '_' + + # description + video_description = get_element_by_id("eow-description", video_webpage) + if video_description: + video_description = re.sub(r'''(?x) + ]*> + [^<]+\.{3}\s* + + ''', r'\1', video_description) + video_description = clean_html(video_description) + else: + fd_mobj = re.search(r'', video_webpage) + video_uploader_url = None + mobj = re.search( + r'', + video_webpage) if mobj is not None: - video_uploader_id = mobj.group(1) + video_uploader_id = mobj.group('uploader_id') + video_uploader_url = mobj.group('uploader_url') else: - self._downloader.report_warning(u'unable to extract uploader nickname') - - # title - if 'title' not in video_info: - raise ExtractorError(u'Unable to extract video title') - video_title = compat_urllib_parse.unquote_plus(video_info['title'][0]) + self._downloader.report_warning('unable to extract uploader nickname') # thumbnail image - if 'thumbnail_url' not in video_info: - self._downloader.report_warning(u'unable to extract video thumbnail') - video_thumbnail = '' + # 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('unable to extract video thumbnail') + video_thumbnail = None else: # don't panic if we can't find it - video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0]) + video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0]) # upload date - upload_date = None - mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) - if mobj is not None: - upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) - upload_date = unified_strdate(upload_date) - - # description - video_description = get_element_by_id("eow-description", video_webpage) - if video_description: - video_description = clean_html(video_description) + upload_date = self._html_search_meta( + 'datePublished', video_webpage, 'upload date', default=None) + if not upload_date: + upload_date = self._search_regex( + [r'(?s)id="eow-date.*?>(.*?)', + r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)'], + video_webpage, 'upload date', default=None) + if upload_date: + upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) + upload_date = unified_strdate(upload_date) + + video_license = self._html_search_regex( + r']+class="title"[^>]*>\s*License\s*\s*]*>\s*
  • (.+?)]+class="title"[^>]*>\s*Music\s*\s*]*>\s*
  • (?P.+?) by (?P<creator>.+?)(?:\(.+?\))?</li', + video_webpage) + if m_music: + video_alt_title = remove_quotes(unescapeHTML(m_music.group('title'))) + video_creator = clean_html(m_music.group('creator')) else: - fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage) - if fd_mobj: - video_description = unescapeHTML(fd_mobj.group(1)) - else: - video_description = u'' + video_alt_title = video_creator = None + + m_cat_container = self._search_regex( + r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>', + video_webpage, 'categories', default=None) + if m_cat_container: + category = self._html_search_regex( + r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', + default=None) + video_categories = None if category is None else [category] + else: + video_categories = None + + video_tags = [ + unescapeHTML(m.group('content')) + for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)] + + def _extract_count(count_name): + return str_to_int(self._search_regex( + r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>' + % re.escape(count_name), + video_webpage, count_name, default=None)) + + like_count = _extract_count('like') + dislike_count = _extract_count('dislike') # subtitles - video_subtitles = None - - if self._downloader.params.get('writesubtitles', False): - video_subtitles = self._extract_subtitle(video_id) - 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) - - if self._downloader.params.get('allsubtitles', False): - video_subtitles = self._extract_all_subtitles(video_id) - for video_subtitle in video_subtitles: - (sub_error, sub_lang, sub) = video_subtitle - if sub_error: - self._downloader.report_warning(sub_error) - - if self._downloader.params.get('listsubtitles', False): - self._list_available_subtitles(video_id) - return + video_subtitles = self.extract_subtitles(video_id, video_webpage) + automatic_captions = self.extract_automatic_captions(video_id, video_webpage) if 'length_seconds' not in video_info: - self._downloader.report_warning(u'unable to extract video duration') - video_duration = '' + self._downloader.report_warning('unable to extract video duration') + video_duration = None else: - 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) - 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) - video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] - except ValueError: - pass + video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0])) + + # annotations + video_annotations = None + if self._downloader.params.get('writeannotations', False): + video_annotations = self._extract_annotations(video_id) + + def _map_to_format_list(urlmap): + formats = [] + for itag, video_real_url in urlmap.items(): + dct = { + 'format_id': itag, + 'url': video_real_url, + 'player_url': player_url, + } + if itag in self._formats: + dct.update(self._formats[itag]) + formats.append(dct) + return formats if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): 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: - url_map = {} - for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): + formats = [{ + 'format_id': '_rtmp', + 'protocol': 'rtmp', + 'url': video_info['conn'][0], + 'player_url': player_url, + }] + elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1: + encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0] + if 'rtmpe%3Dyes' in encoded_url_map: + raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True) + formats_spec = {} + fmt_list = video_info.get('fmt_list', [''])[0] + if fmt_list: + for fmt in fmt_list.split(','): + spec = fmt.split('/') + if len(spec) > 1: + width_height = spec[1].split('x') + if len(width_height) == 2: + formats_spec[spec[0]] = { + 'resolution': spec[1], + 'width': int_or_none(width_height[0]), + 'height': int_or_none(width_height[1]), + } + formats = [] + for url_data_str in encoded_url_map.split(','): url_data = compat_parse_qs(url_data_str) - if 'itag' in url_data and 'url' in url_data: - url = url_data['url'][0] - if 'sig' in url_data: - url += '&signature=' + url_data['sig'][0] - elif 's' in url_data: - signature = self._decrypt_signature(url_data['s'][0]) - 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) - 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') + if 'itag' not in url_data or 'url' not in url_data: + continue + format_id = url_data['itag'][0] + url = url_data['url'][0] + + if 'sig' in url_data: + url += '&signature=' + url_data['sig'][0] + elif 's' in url_data: + encrypted_sig = url_data['s'][0] + ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")' + + jsplayer_url_json = self._search_regex( + ASSETS_RE, + embed_webpage if age_gate else video_webpage, + 'JS player URL (1)', default=None) + if not jsplayer_url_json and not age_gate: + # We need the embed website after all + if embed_webpage is None: + embed_url = proto + '://www.youtube.com/embed/%s' % video_id + embed_webpage = self._download_webpage( + embed_url, video_id, 'Downloading embed webpage') + jsplayer_url_json = self._search_regex( + ASSETS_RE, embed_webpage, 'JS player URL') + + player_url = json.loads(jsplayer_url_json) + if player_url is None: + player_url_json = self._search_regex( + r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', + video_webpage, 'age gate player URL') + player_url = json.loads(player_url_json) + + if self._downloader.params.get('verbose'): + if player_url is None: + player_version = 'unknown' + player_desc = 'unknown' + else: + if player_url.endswith('swf'): + player_version = self._search_regex( + r'-(.+?)(?:/watch_as3)?\.swf$', player_url, + 'flash player', fatal=False) + player_desc = 'flash player %s' % player_version + else: + player_version = self._search_regex( + [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'], + player_url, + 'html5 player', fatal=False) + player_desc = 'html5 player %s' % player_version + + parts_sizes = self._signature_cache_id(encrypted_sig) + self.to_screen('{%s} signature length %s, %s' % + (format_id, parts_sizes, player_desc)) + + signature = self._decrypt_signature( + encrypted_sig, video_id, player_url, age_gate) + url += '&signature=' + signature + if 'ratebypass' not in url: + url += '&ratebypass=yes' + + dct = { + 'format_id': format_id, + 'url': url, + 'player_url': player_url, + } + if format_id in self._formats: + dct.update(self._formats[format_id]) + if format_id in formats_spec: + dct.update(formats_spec[format_id]) + + # Some itags are not included in DASH manifest thus corresponding formats will + # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993). + # Trying to extract metadata from url_encoded_fmt_stream_map entry. + mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0]) + width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) + + more_fields = { + 'filesize': int_or_none(url_data.get('clen', [None])[0]), + 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000), + 'width': width, + 'height': height, + 'fps': int_or_none(url_data.get('fps', [None])[0]), + 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0], + } + for key, value in more_fields.items(): + if value: + dct[key] = value + type_ = url_data.get('type', [None])[0] + if type_: + type_split = type_.split(';') + kind_ext = type_split[0].split('/') + if len(kind_ext) == 2: + kind, _ = kind_ext + dct['ext'] = mimetype2ext(type_split[0]) + if kind in ('audio', 'video'): + codecs = None + for mobj in re.finditer( + r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_): + if mobj.group('key') == 'codecs': + codecs = mobj.group('val') + break + if codecs: + codecs = codecs.split(',') + if len(codecs) == 2: + acodec, vcodec = codecs[1], codecs[0] + else: + acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0]) + dct.update({ + 'acodec': acodec, + 'vcodec': vcodec, + }) + formats.append(dct) + elif video_info.get('hlsvp'): + manifest_url = video_info['hlsvp'][0] + url_map = self._extract_from_m3u8(manifest_url, video_id) + formats = _map_to_format_list(url_map) + # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming + for a_format in formats: + a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True' else: - raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info') - - results = [] - for format_param, video_real_url in video_url_list: - # 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, '???')) - - results.append({ - 'id': video_id, - 'url': video_real_url, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'upload_date': upload_date, - 'title': video_title, - 'ext': video_extension, - 'format': video_format, - 'thumbnail': video_thumbnail, - 'description': video_description, - 'player_url': player_url, - 'subtitles': video_subtitles, - 'duration': video_duration - }) - return results + unavailable_message = self._html_search_regex( + r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>', + video_webpage, 'unavailable message', default=None) + if unavailable_message: + raise ExtractorError(unavailable_message, expected=True) + raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') + + # Look for the DASH manifest + if self._downloader.params.get('youtube_include_dash_manifest', True): + dash_mpd_fatal = True + for mpd_url in dash_mpds: + dash_formats = {} + try: + def decrypt_sig(mobj): + s = mobj.group(1) + dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) + return '/signature/%s' % dec_s + + mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url) + + for df in self._extract_mpd_formats( + mpd_url, video_id, fatal=dash_mpd_fatal, + formats_dict=self._formats): + # Do not overwrite DASH format found in some previous DASH manifest + if df['format_id'] not in dash_formats: + dash_formats[df['format_id']] = df + # Additional DASH manifests may end up in HTTP Error 403 therefore + # allow them to fail without bug report message if we already have + # some DASH manifest succeeded. This is temporary workaround to reduce + # burst of bug reports until we figure out the reason and whether it + # can be fixed at all. + dash_mpd_fatal = False + except (ExtractorError, KeyError) as e: + self.report_warning( + 'Skipping DASH manifest: %r' % e, video_id) + if dash_formats: + # Remove the formats we found through non-DASH, they + # contain less info and it can be wrong, because we use + # fixed values (for example the resolution). See + # https://github.com/rg3/youtube-dl/issues/5774 for an + # example. + formats = [f for f in formats if f['format_id'] not in dash_formats.keys()] + formats.extend(dash_formats.values()) + + # Check for malformed aspect ratio + stretched_m = re.search( + r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">', + video_webpage) + if stretched_m: + w = float(stretched_m.group('w')) + h = float(stretched_m.group('h')) + # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0). + # We will only process correct ratios. + if w > 0 and h > 0: + ratio = w / h + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio + + self._sort_formats(formats) + + self.mark_watched(video_id, video_info) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'uploader_id': video_uploader_id, + 'uploader_url': video_uploader_url, + 'upload_date': upload_date, + 'license': video_license, + 'creator': video_creator, + 'title': video_title, + 'alt_title': video_alt_title, + 'thumbnail': video_thumbnail, + 'description': video_description, + 'categories': video_categories, + 'tags': video_tags, + 'subtitles': video_subtitles, + 'automatic_captions': automatic_captions, + 'duration': video_duration, + 'age_limit': 18 if age_gate else 0, + 'annotations': video_annotations, + 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, + 'view_count': view_count, + 'like_count': like_count, + 'dislike_count': dislike_count, + 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]), + 'formats': formats, + 'is_live': is_live, + 'start_time': start_time, + 'end_time': end_time, + } + + +class YoutubeSharedVideoIE(InfoExtractor): + _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})' + IE_NAME = 'youtube:shared' + + _TEST = { + 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU', + 'info_dict': { + 'id': 'uPDB5I9wfp8', + 'ext': 'webm', + 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3', + 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d', + 'upload_date': '20160219', + 'uploader': 'Pocoyo - Português (BR)', + 'uploader_id': 'PocoyoBrazil', + }, + 'add_ie': ['Youtube'], + 'params': { + # There are already too many Youtube downloads + 'skip_download': True, + }, + } -class YoutubePlaylistIE(InfoExtractor): - """Information Extractor for YouTube playlists.""" + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + real_video_id = self._html_search_meta( + 'videoId', webpage, 'YouTube video id', fatal=True) + + return self.url_result(real_video_id, YoutubeIE.ie_key()) - _VALID_URL = r"""(?: + +class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): + IE_DESC = 'YouTube.com playlists' + _VALID_URL = r"""(?x)(?: (?:https?://)? (?:\w+\.)? - youtube\.com/ (?: - (?:course|view_play_list|my_playlists|artist|playlist|watch) - \? (?:.*?&)*? (?:p|a|list)= - | p/ + youtube\.com/ + (?: + (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) + \? (?:.*?[&;])*? (?:p|a|list)= + | p/ + )| + youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist= + ) + ( + (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,} + # Top tracks, they can also include dots + |(?:MC)[\w\.]* ) - ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) .* | - ((?:PL|EC|UU)[0-9A-Za-z-_]{10,}) + ((?:PL|LL|EC|UU|FL|RD|UL)[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 - IE_NAME = u'youtube:playlist' - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None + _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?' + IE_NAME = 'youtube:playlist' + _TESTS = [{ + 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', + 'info_dict': { + 'title': 'ytdl test PL', + 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', + }, + 'playlist_count': 3, + }, { + 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', + 'info_dict': { + 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', + 'title': 'YDL_Empty_List', + }, + 'playlist_count': 0, + }, { + 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', + 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', + 'info_dict': { + 'title': '29C3: Not my department', + 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', + }, + 'playlist_count': 95, + }, { + 'note': 'issue #673', + 'url': 'PLBB231211A4F62143', + 'info_dict': { + 'title': '[OLD]Team Fortress 2 (Class-based LP)', + 'id': 'PLBB231211A4F62143', + }, + 'playlist_mincount': 26, + }, { + 'note': 'Large playlist', + 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', + 'info_dict': { + 'title': 'Uploads from Cauchemar', + 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q', + }, + 'playlist_mincount': 799, + }, { + 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', + 'info_dict': { + 'title': 'YDL_safe_search', + 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', + }, + 'playlist_count': 2, + }, { + 'note': 'embedded', + 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + 'playlist_count': 4, + 'info_dict': { + 'title': 'JODA15', + 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + } + }, { + 'note': 'Embedded SWF player', + 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0', + 'playlist_count': 4, + 'info_dict': { + 'title': 'JODA7', + 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ', + } + }, { + 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos', + 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA', + 'info_dict': { + 'title': 'Uploads from Interstellar Movie', + 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', + }, + 'playlist_mincount': 21, + }, { + # Playlist URL that does not actually serve a playlist + 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4', + 'info_dict': { + 'id': 'FqZTN594JQw', + 'ext': 'webm', + 'title': "Smiley's People 01 detective, Adventure Series, Action", + 'uploader': 'STREEM', + 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng', + 'upload_date': '20150526', + 'license': 'Standard YouTube License', + 'description': 'md5:507cdcb5a49ac0da37a920ece610be80', + 'categories': ['People & Blogs'], + 'tags': list, + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [YoutubeIE.ie_key()], + }, { + 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5', + 'info_dict': { + 'id': 'yeWKywCrFtk', + 'ext': 'mp4', + 'title': 'Small Scale Baler and Braiding Rugs', + 'uploader': 'Backus-Page House Museum', + 'uploader_id': 'backuspagemuseum', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum', + 'upload_date': '20161008', + 'license': 'Standard YouTube License', + 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a', + 'categories': ['Nonprofits & Activism'], + 'tags': list, + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'noplaylist': True, + 'skip_download': True, + }, + }, { + 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21', + 'only_matching': True, + }] - def _real_extract(self, url): - # Extract playlist id - mobj = re.match(self._VALID_URL, url, re.VERBOSE) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + def _real_initialize(self): + self._login() + + def _extract_mix(self, playlist_id): + # The mixes are generated from a single video + # the id of the playlist is just 'RD' + video_id + ids = [] + last_id = playlist_id[-11:] + for n in itertools.count(1): + url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id) + webpage = self._download_webpage( + url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n)) + new_ids = orderedSet(re.findall( + r'''(?xs)data-video-username=".*?".*? + href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id), + webpage)) + # Fetch new pages until all the videos are repeated, it seems that + # there are always 51 unique videos. + new_ids = [_id for _id in new_ids if _id not in ids] + if not new_ids: + break + ids.extend(new_ids) + last_id = ids[-1] + + url_results = self._ids_to_results(ids) + + search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage) + title_span = ( + search_title('playlist-title') or + search_title('title long-title') or + search_title('title')) + title = clean_html(title_span) + + return self.playlist_result(url_results, playlist_id, title) + + def _extract_playlist(self, playlist_id): + url = self._TEMPLATE_URL % playlist_id + page = self._download_webpage(url, playlist_id) + + for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page): + match = match.strip() + # Check if the playlist exists or is private + if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match): + raise ExtractorError( + 'The playlist doesn\'t exist or is private, use --username or ' + '--netrc to access it.', + expected=True) + elif re.match(r'[^<]*Invalid parameters[^<]*', match): + raise ExtractorError( + 'Invalid parameters. Maybe URL is incorrect.', + expected=True) + elif re.match(r'[^<]*Choose your language[^<]*', match): + continue + else: + self.report_warning('Youtube gives an alert message: ' + match) - # Download playlist videos from API - playlist_id = mobj.group(1) or mobj.group(2) - page_num = 1 - videos = [] + playlist_title = self._html_search_regex( + r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>', + page, 'title', default=None) - while True: - url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) - page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) + has_videos = True + if not playlist_title: try: - response = json.loads(page) - except ValueError as err: - raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) - - if 'feed' not in response: - raise ExtractorError(u'Got a malformed response from YouTube API') - playlist_title = response['feed']['title']['$t'] - if 'entry' not in response['feed']: - # Number of videos is a multiple of self._MAX_RESULTS - break + # Some playlist URLs don't actually serve a playlist (e.g. + # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4) + next(self._entries(page, playlist_id)) + except StopIteration: + has_videos = False + + return has_videos, self.playlist_result( + self._entries(page, playlist_id), playlist_id, playlist_title) + + def _check_download_just_video(self, url, playlist_id): + # Check if it's a video-specific URL + query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + video_id = query_dict.get('v', [None])[0] or self._search_regex( + r'(?:^|//)youtu\.be/([0-9A-Za-z_-]{11})', url, + 'video id', default=None) + if video_id: + if self._downloader.params.get('noplaylist'): + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + return video_id, self.url_result(video_id, 'Youtube', video_id=video_id) + else: + self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) + return video_id, None + return None, None - for entry in response['feed']['entry']: - index = entry['yt$position']['$t'] - if 'media$group' in entry and 'media$player' in entry['media$group']: - videos.append((index, entry['media$group']['media$player']['url'])) + def _real_extract(self, url): + # Extract playlist id + mobj = re.match(self._VALID_URL, url) + if mobj is None: + raise ExtractorError('Invalid URL: %s' % url) + playlist_id = mobj.group(1) or mobj.group(2) - if len(response['feed']['entry']) < self._MAX_RESULTS: - break - page_num += 1 + video_id, video = self._check_download_just_video(url, playlist_id) + if video: + return video + + if playlist_id.startswith(('RD', 'UL', 'PU')): + # Mixes require a custom extraction process + return self._extract_mix(playlist_id) + + has_videos, playlist = self._extract_playlist(playlist_id) + if has_videos or not video_id: + return playlist + + # Some playlist URLs don't actually serve a playlist (see + # https://github.com/rg3/youtube-dl/issues/10537). + # Fallback to plain video extraction if there is a video id + # along with playlist id. + return self.url_result(video_id, 'Youtube', video_id=video_id) + + +class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): + IE_DESC = 'YouTube.com channels' + _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' + _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos' + _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?' + IE_NAME = 'youtube:channel' + _TESTS = [{ + 'note': 'paginated channel', + 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'playlist_mincount': 91, + 'info_dict': { + 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w', + 'title': 'Uploads from lex will', + } + }, { + 'note': 'Age restricted channel', + # from https://www.youtube.com/user/DeusExOfficial + 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w', + 'playlist_mincount': 64, + 'info_dict': { + 'id': 'UUs0ifCMCm1icqRbqhUINa0w', + 'title': 'Uploads from Deus Ex', + }, + }] - videos = [v[1] for v in sorted(videos)] + @classmethod + def suitable(cls, url): + return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url) + else super(YoutubeChannelIE, cls).suitable(url)) - url_results = [self.url_result(url, 'Youtube') for url in videos] - return [self.playlist_result(url_results, playlist_id, playlist_title)] + def _build_template_url(self, url, channel_id): + return self._TEMPLATE_URL % channel_id + def _real_extract(self, url): + channel_id = self._match_id(url) + + url = self._build_template_url(url, channel_id) + + # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778) + # Workaround by extracting as a playlist if managed to obtain channel playlist URL + # otherwise fallback on channel by page extraction + channel_page = self._download_webpage( + url + '?view=57', channel_id, + 'Downloading channel page', fatal=False) + if channel_page is False: + channel_playlist_id = False + else: + channel_playlist_id = self._html_search_meta( + 'channelId', channel_page, 'channel id', default=None) + if not channel_playlist_id: + channel_url = self._html_search_meta( + ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'), + channel_page, 'channel url', default=None) + if channel_url: + channel_playlist_id = self._search_regex( + r'vnd\.youtube://user/([0-9A-Za-z_-]+)', + channel_url, 'channel id', default=None) + if channel_playlist_id and channel_playlist_id.startswith('UC'): + playlist_id = 'UU' + channel_playlist_id[2:] + return self.url_result( + compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist') + + channel_page = self._download_webpage(url, channel_id, 'Downloading page #1') + autogenerated = re.search(r'''(?x) + class="[^"]*?(?: + channel-header-autogenerated-label| + yt-channel-title-autogenerated + )[^"]*"''', channel_page) is not None + + if autogenerated: + # The videos are contained in a single page + # the ajax pages can't be used, they are empty + entries = [ + self.url_result( + video_id, 'Youtube', video_id=video_id, + video_title=video_title) + for video_id, video_title in self.extract_videos_from_page(channel_page)] + return self.playlist_result(entries, channel_id) -class YoutubeChannelIE(InfoExtractor): - """Information Extractor for YouTube channels.""" + try: + next(self._entries(channel_page, channel_id)) + except StopIteration: + alert_message = self._html_search_regex( + r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>', + channel_page, 'alert', default=None, group='alert') + if alert_message: + raise ExtractorError('Youtube said: %s' % alert_message, expected=True) + + return self.playlist_result(self._entries(channel_page, channel_id), channel_id) + + +class YoutubeUserIE(YoutubeChannelIE): + IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' + _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos' + IE_NAME = 'youtube:user' + + _TESTS = [{ + 'url': 'https://www.youtube.com/user/TheLinuxFoundation', + 'playlist_mincount': 320, + 'info_dict': { + 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ', + 'title': 'Uploads from The Linux Foundation', + } + }, { + # Only available via https://www.youtube.com/c/12minuteathlete/videos + # but not https://www.youtube.com/user/12minuteathlete/videos + 'url': 'https://www.youtube.com/c/12minuteathlete/videos', + 'playlist_mincount': 249, + 'info_dict': { + 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ', + 'title': 'Uploads from 12 Minute Athlete', + } + }, { + 'url': 'ytuser:phihag', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/c/gametrailers', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/gametrailers', + 'only_matching': True, + }, { + # This channel is not available. + 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos', + 'only_matching': True, + }] - _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' - IE_NAME = u'youtube:channel' + @classmethod + def suitable(cls, url): + # Don't return True if the url can be extracted with other youtube + # extractor, the regex would is too permissive and it would match. + other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls) + if any(ie.suitable(url) for ie in other_yt_ies): + return False + else: + return super(YoutubeUserIE, cls).suitable(url) - def extract_videos_from_page(self, page): - ids_in_page = [] - for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page): - if mobj.group(1) not in ids_in_page: - ids_in_page.append(mobj.group(1)) - return ids_in_page + def _build_template_url(self, url, channel_id): + mobj = re.match(self._VALID_URL, url) + return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id')) + + +class YoutubeLiveIE(YoutubeBaseInfoExtractor): + IE_DESC = 'YouTube.com live streams' + _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live' + IE_NAME = 'youtube:live' + + _TESTS = [{ + 'url': 'https://www.youtube.com/user/TheYoungTurks/live', + 'info_dict': { + 'id': 'a48o2S1cPoo', + 'ext': 'mp4', + 'title': 'The Young Turks - Live Main Show', + 'uploader': 'The Young Turks', + 'uploader_id': 'TheYoungTurks', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks', + 'upload_date': '20150715', + 'license': 'Standard YouTube License', + 'description': 'md5:438179573adcdff3c97ebb1ee632b891', + 'categories': ['News & Politics'], + 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'], + 'like_count': int, + 'dislike_count': int, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/c/CommanderVideoHq/live', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/TheYoungTurks/live', + 'only_matching': True, + }] def _real_extract(self, url): - # Extract channel id mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - # Download channel page - channel_id = mobj.group(1) - video_ids = [] - pagenum = 1 - - url = self._TEMPLATE_URL % (channel_id, pagenum) - page = self._download_webpage(url, channel_id, - u'Downloading page #%s' % pagenum) - - # Extract video identifiers - ids_in_page = self.extract_videos_from_page(page) - video_ids.extend(ids_in_page) - - # Download any subsequent channel pages using the json-based channel_ajax query - if self._MORE_PAGES_INDICATOR in page: - while True: - pagenum = pagenum + 1 - - url = self._MORE_PAGES_URL % (pagenum, channel_id) - page = self._download_webpage(url, channel_id, - u'Downloading page #%s' % pagenum) + channel_id = mobj.group('id') + base_url = mobj.group('base_url') + webpage = self._download_webpage(url, channel_id, fatal=False) + if webpage: + page_type = self._og_search_property( + 'type', webpage, 'page type', default=None) + video_id = self._html_search_meta( + 'videoId', webpage, 'video id', default=None) + if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id): + return self.url_result(video_id, YoutubeIE.ie_key()) + return self.url_result(base_url) + + +class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor): + IE_DESC = 'YouTube.com user/channel playlists' + _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists' + IE_NAME = 'youtube:playlists' + + _TESTS = [{ + 'url': 'https://www.youtube.com/user/ThirstForScience/playlists', + 'playlist_mincount': 4, + 'info_dict': { + 'id': 'ThirstForScience', + 'title': 'Thirst for Science', + }, + }, { + # with "Load more" button + 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd', + 'playlist_mincount': 70, + 'info_dict': { + 'id': 'igorkle1', + 'title': 'Игорь Клейнер', + }, + }, { + 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists', + 'playlist_mincount': 17, + 'info_dict': { + 'id': 'UCiU1dHvZObB2iP6xkJ__Icw', + 'title': 'Chem Player', + }, + }] + + +class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): + IE_DESC = 'YouTube.com searches' + # there doesn't appear to be a real limit, for example if you search for + # 'python' you get more than 8.000.000 results + _MAX_RESULTS = float('inf') + IE_NAME = 'youtube:search' + _SEARCH_KEY = 'ytsearch' + _EXTRA_QUERY_ARGS = {} + _TESTS = [] - page = json.loads(page) + def _get_n_results(self, query, n): + """Get a specified number of results for a query""" - ids_in_page = self.extract_videos_from_page(page['content_html']) - video_ids.extend(ids_in_page) + videos = [] + limit = n - if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: - break + for pagenum in itertools.count(1): + url_query = { + 'search_query': query.encode('utf-8'), + 'page': pagenum, + 'spf': 'navigate', + } + url_query.update(self._EXTRA_QUERY_ARGS) + result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query) + data = self._download_json( + result_url, video_id='query "%s"' % query, + note='Downloading page %s' % pagenum, + errnote='Unable to download API page') + html_content = data[1]['body']['content'] + + if 'class="search-message' in html_content: + raise ExtractorError( + '[youtube] No video results', expected=True) + + new_videos = self._ids_to_results(orderedSet(re.findall( + r'href="/watch\?v=(.{11})', html_content))) + videos += new_videos + if not new_videos or len(videos) > limit: + break - self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) + if len(videos) > n: + videos = videos[:n] + return self.playlist_result(videos, query) - 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] - return [self.playlist_result(url_entries, channel_id)] +class YoutubeSearchDateIE(YoutubeSearchIE): + IE_NAME = YoutubeSearchIE.IE_NAME + ':date' + _SEARCH_KEY = 'ytsearchdate' + IE_DESC = 'YouTube.com searches, newest videos first' + _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'} -class YoutubeUserIE(InfoExtractor): - """Information Extractor for YouTube users.""" - _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 - _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d' - _VIDEO_INDICATOR = r'/watch\?v=(.+?)[\<&]' - IE_NAME = u'youtube:user' +class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor): + IE_DESC = 'YouTube.com search URLs' + IE_NAME = 'youtube:search_url' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)' + _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?' + _TESTS = [{ + 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', + 'playlist_mincount': 5, + 'info_dict': { + 'title': 'youtube-dl test video', + } + }, { + 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB', + 'only_matching': True, + }] def _real_extract(self, url): - # Extract username mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + query = compat_urllib_parse_unquote_plus(mobj.group('query')) + webpage = self._download_webpage(url, query) + return self.playlist_result(self._process_page(webpage), playlist_title=query) + + +class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor): + IE_DESC = 'YouTube.com (multi-season) shows' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)' + IE_NAME = 'youtube:show' + _TESTS = [{ + 'url': 'https://www.youtube.com/show/airdisasters', + 'playlist_mincount': 5, + 'info_dict': { + 'id': 'airdisasters', + 'title': 'Air Disasters', + } + }] - username = mobj.group(1) + def _real_extract(self, url): + playlist_id = self._match_id(url) + return super(YoutubeShowIE, self)._real_extract( + 'https://www.youtube.com/show/%s/playlists' % playlist_id) - # Download video ids using YouTube Data API. Result size per - # query is limited (currently to 50 videos) so we need to query - # page by page until there are no video ids - it means we got - # all of them. - video_ids = [] - pagenum = 0 +class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): + """ + Base class for feed extractors + Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. + """ + _LOGIN_REQUIRED = True - while True: - start_index = pagenum * self._GDATA_PAGE_SIZE + 1 + @property + def IE_NAME(self): + return 'youtube:%s' % self._FEED_NAME - gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index) - page = self._download_webpage(gdata_url, username, - u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE)) + def _real_initialize(self): + self._login() - # Extract video identifiers - ids_in_page = [] + def _real_extract(self, url): + page = self._download_webpage( + 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE) + + # The extraction process is the same as for playlists, but the regex + # for the video ids doesn't contain an index + ids = [] + more_widget_html = content_html = page + for page_num in itertools.count(1): + matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html) + + # 'recommended' feed has infinite 'load more' and each new portion spins + # the same videos in (sometimes) slightly different order, so we'll check + # for unicity and break when portion has no new videos + new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches)) + if not new_ids: + break - for mobj in re.finditer(self._VIDEO_INDICATOR, page): - if mobj.group(1) not in ids_in_page: - ids_in_page.append(mobj.group(1)) + ids.extend(new_ids) - video_ids.extend(ids_in_page) + mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html) + if not mobj: + break - # A little optimization - if current page is not - # "full", ie. does not contain PAGE_SIZE video ids then - # we can assume that this page is the last one - there - # are no more ids on further pages - no need to query - # again. + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + more_widget_html = more['load_more_widget_html'] - if len(ids_in_page) < self._GDATA_PAGE_SIZE: - break + return self.playlist_result( + self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE) - 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] - return [self.playlist_result(url_results, playlist_title = username)] +class YoutubeWatchLaterIE(YoutubePlaylistIE): + IE_NAME = 'youtube:watchlater' + IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater' -class YoutubeSearchIE(SearchInfoExtractor): - """Information Extractor for YouTube search queries.""" - _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' - _SEARCH_KEY = 'ytsearch' + _TESTS = [{ + 'url': 'https://www.youtube.com/playlist?list=WL', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL', + 'only_matching': True, + }] - def report_download_page(self, query, pagenum): - """Report attempt to download search page with given number.""" - self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum)) + def _real_extract(self, url): + _, video = self._check_download_just_video(url, 'WL') + if video: + return video + _, playlist = self._extract_playlist('WL') + return playlist - def _get_n_results(self, query, n): - """Get a specified number of results for a query""" - video_ids = [] - pagenum = 0 - limit = n +class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): + IE_NAME = 'youtube:favorites' + IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?' + _LOGIN_REQUIRED = True - while (50 * pagenum) < limit: - self.report_download_page(query, pagenum+1) - result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) - request = compat_urllib_request.Request(result_url) - try: - data = 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 download API page: %s' % compat_str(err)) - api_response = json.loads(data)['data'] + 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, 'favourites playlist id') + return self.url_result(playlist_id, 'YoutubePlaylist') + + +class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): + IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?' + _FEED_NAME = 'recommended' + _PLAYLIST_TITLE = 'Youtube Recommended videos' + + +class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): + IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' + _FEED_NAME = 'subscriptions' + _PLAYLIST_TITLE = 'Youtube Subscriptions' + + +class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): + IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory' + _FEED_NAME = 'history' + _PLAYLIST_TITLE = 'Youtube History' + + +class YoutubeTruncatedURLIE(InfoExtractor): + IE_NAME = 'youtube:truncated_url' + IE_DESC = False # Do not list + _VALID_URL = r'''(?x) + (?:https?://)? + (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/ + (?:watch\?(?: + feature=[a-z_]+| + annotation_id=annotation_[^&]+| + x-yt-cl=[0-9]+| + hl=[^&]*| + t=[0-9]+ + )? + | + attribution_link\?a=[^&]+ + ) + $ + ''' + + _TESTS = [{ + 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?feature=foo', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?hl=en-GB', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?t=2372', + 'only_matching': True, + }] - if not 'items' in api_response: - raise ExtractorError(u'[youtube] No video results') + def _real_extract(self, url): + raise ExtractorError( + '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 ' + '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' + ' or simply youtube-dl BaW_jenozKc .', + expected=True) - new_ids = list(video['id'] for video in api_response['items']) - video_ids += new_ids - limit = min(n, api_response['totalItems']) - pagenum += 1 +class YoutubeTruncatedIDIE(InfoExtractor): + IE_NAME = 'youtube:truncated_id' + IE_DESC = False # Do not list + _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$' - if len(video_ids) > n: - 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) + _TESTS = [{ + 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + raise ExtractorError( + 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url), + expected=True)