X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/03342304420e5daeb428ffdcc7bbd2bbfecfa61a..0ae75ad42cc6f25af12b54201dcbadfc116f923f:/youtube_dl/extractor/youtube.py diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index e24dd3e..268080b 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -6,6 +6,7 @@ from __future__ import unicode_literals import itertools import json import os.path +import random import re import time import traceback @@ -16,16 +17,15 @@ from ..swfinterp import SWFInterpreter from ..compat import ( compat_chr, compat_parse_qs, - compat_urllib_parse, 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, - encode_dict, error_to_compat_str, ExtractorError, float_or_none, @@ -44,6 +44,7 @@ from ..utils import ( unified_strdate, unsmuggle_url, uppercase_escape, + urlencode_postdata, ISO3166Utils, ) @@ -52,6 +53,7 @@ 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 @@ -115,22 +117,26 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 'hl': 'en_US', } - login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii') - - req = sanitized_Request(self._LOGIN_URL, login_data) login_results = self._download_webpage( - req, None, - note='Logging in', errnote='unable to log in', fatal=False) + self._PASSWORD_CHALLENGE_URL, None, + note='Logging in', errnote='unable to log in', fatal=False, + data=urlencode_postdata(login_form_strs)) 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: + 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: @@ -148,7 +154,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 'TrustDevice': 'on', }) - tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii') + tfa_data = urlencode_postdata(tfa_form_strs) tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data) tfa_results = self._download_webpage( @@ -158,17 +164,17 @@ class YoutubeBaseInfoExtractor(InfoExtractor): if tfa_results is False: return False - if re.search(r'(?i)]* id="challenge"', tfa_results) is not None: + 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: + 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: + 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 @@ -233,7 +239,9 @@ class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor): def _process_page(self, content): - for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', 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') @@ -267,7 +275,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): )) |(?: youtu\.be| # just youtu.be/xxxx - vid\.plus # or vid.plus/xxxx + vid\.plus| # or vid.plus/xxxx + zwearz\.com/watch| # or zwearz.com/watch/xxxx )/ |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= ) @@ -308,6 +317,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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}, @@ -333,6 +343,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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}, @@ -382,7 +394,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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'], @@ -401,12 +415,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20120506', 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]', 'alt_title': 'I Love It (feat. Charli XCX)', - 'description': 'md5:782e8651347686cba06e58f71ab51773', + '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', } }, @@ -422,6 +438,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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, } @@ -437,6 +455,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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, } }, @@ -449,7 +469,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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'], @@ -468,14 +490,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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 { @@ -488,10 +513,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'AfrojackVEVO', 'uploader_id': 'AfrojackVEVO', 'upload_date': '20131011', + 'license': 'Standard YouTube License', }, 'params': { 'youtube_include_dash_manifest': True, - 'format': '141', + 'format': '141/bestaudio[ext=m4a]', }, }, # JS player signature function name containing $ @@ -506,11 +532,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'TaylorSwiftVEVO', 'uploader_id': 'TaylorSwiftVEVO', 'upload_date': '20140818', + 'license': 'Standard YouTube License', 'creator': 'Taylor Swift', }, 'params': { 'youtube_include_dash_manifest': True, - 'format': '141', + 'format': '141/bestaudio[ext=m4a]', }, }, # Controversy video @@ -522,6 +549,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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', } @@ -536,7 +565,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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, }, }, @@ -550,7 +581,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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, }, }, @@ -562,9 +595,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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', }, @@ -580,8 +615,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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': 'Olympics', + 'uploader': 'Olympic', 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', }, 'params': { @@ -597,8 +634,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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', }, }, @@ -629,8 +668,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'md5:116377fd2963b81ec4ce64b542173306', 'upload_date': '20150625', 'uploader_id': 'dorappi2000', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000', 'uploader': 'dorappi2000', - 'formats': 'mincount:33', + 'license': 'Standard YouTube License', + 'formats': 'mincount:32', }, }, # DASH manifest with segment_list @@ -644,12 +685,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Airtek', 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.', 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ', + 'license': 'Standard YouTube License', 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015', }, 'params': { 'youtube_include_dash_manifest': True, 'format': '135', # bestvideo - } + }, + 'skip': 'This live event has ended.', }, { # Multifeed videos (multiple cameras), URL is for Main Camera @@ -668,6 +711,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20150721', 'uploader': 'Beer Games Beer', 'uploader_id': 'beergamesbeer', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', + 'license': 'Standard YouTube License', }, }, { 'info_dict': { @@ -678,6 +723,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20150721', 'uploader': 'Beer Games Beer', 'uploader_id': 'beergamesbeer', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', + 'license': 'Standard YouTube License', }, }, { 'info_dict': { @@ -688,6 +735,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20150721', 'uploader': 'Beer Games Beer', 'uploader_id': 'beergamesbeer', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', + 'license': 'Standard YouTube License', }, }, { 'info_dict': { @@ -698,6 +747,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20150721', 'uploader': 'Beer Games Beer', 'uploader_id': 'beergamesbeer', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', + 'license': 'Standard YouTube License', }, }], 'params': { @@ -712,11 +763,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30', }, 'playlist_count': 2, + 'skip': 'Not multifeed anymore', }, { 'url': 'http://vid.plus/FlRa-iH7PGw', 'only_matching': True, }, + { + 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html', + 'only_matching': True, + }, { # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468) # Also tests cut-off URL expansion in video description (see @@ -731,7 +787,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a', 'upload_date': '20151119', 'uploader_id': 'IronSoulElf', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf', 'uploader': 'IronSoulElf', + 'license': 'Standard YouTube License', 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan', }, 'params': { @@ -758,10 +816,52 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'This video does not exist.', + }, + { + # Video licensed under Creative Commons + 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA', + 'info_dict': { + 'id': 'M4gD1WSo5mA', + 'ext': 'mp4', + 'title': 'md5:e41008789470fc2533a3252216f1c1d1', + 'description': 'md5:a677553cf0840649b731a3024aeff4cc', + 'upload_date': '20150127', + 'uploader_id': 'BerkmanCenter', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter', + 'uploader': 'BerkmanCenter', + 'license': 'Creative Commons Attribution license (reuse allowed)', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # Channel-like uploader_url + 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg', + 'info_dict': { + 'id': 'eQcmzGIKrzg', + 'ext': 'mp4', + 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders', + 'description': 'md5:dda0d780d5a6e120758d1711d062a867', + 'upload_date': '20151119', + 'uploader': 'Bernie 2016', + 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg', + 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg', + 'license': 'Creative Commons Attribution license (reuse allowed)', + }, + 'params': { + 'skip_download': True, + }, }, { 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY', 'only_matching': True, + }, + { + # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059) + 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo', + 'only_matching': True, } ] @@ -930,7 +1030,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): continue sub_formats = [] for ext in self._SUBTITLE_FORMATS: - params = compat_urllib_parse.urlencode({ + params = compat_urllib_parse_urlencode({ 'lang': lang, 'v': video_id, 'fmt': ext, @@ -975,40 +1075,67 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return {} try: args = player_config['args'] - caption_url = args['ttsurl'] - if not caption_url: - self._downloader.report_warning(err_msg) - return {} - 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', '') + 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_node in caption_list.findall('target'): - sub_lang = lang_node.attrib['lang_code'] + 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: - params = compat_urllib_parse.urlencode({ - 'lang': original_lang, - 'tlang': sub_lang, - 'fmt': ext, - 'ts': timestamp, - 'kind': caption_kind, + 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': caption_url + '&' + params, + 'url': sub_url, 'ext': ext, }) sub_lang_list[sub_lang] = sub_formats @@ -1019,6 +1146,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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) + + # 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))) + + qs.update({ + 'ver': ['2'], + 'cpn': [cpn], + }) + playback_url = compat_urlparse.urlunparse( + parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True))) + + self._download_webpage( + playback_url, video_id, 'Marking watched', + 'Unable to mark watched', fatal=False) + @classmethod def extract_id(cls, url): mobj = re.match(cls._VALID_URL, url, re.VERBOSE) @@ -1098,7 +1248,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # 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({ + data = compat_urllib_parse_urlencode({ 'video_id': video_id, 'eurl': 'https://youtube.googleapis.com/v/' + video_id, 'sts': self._search_regex( @@ -1186,10 +1336,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if video_description: video_description = re.sub(r'''(?x) ]*> + (?:[a-zA-Z-]+="[^"]*"\s+)*? + class="[^"]*"[^>]*> [^<]+\.{3}\s* ''', r'\1', video_description) @@ -1245,9 +1395,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # uploader_id video_uploader_id = None - 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('unable to extract uploader nickname') @@ -1275,6 +1429,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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) @@ -1348,6 +1506,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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) @@ -1416,6 +1587,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } 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). @@ -1528,11 +1701,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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, @@ -1557,6 +1734,39 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } +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, + }, + } + + 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()) + + class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com playlists' _VALID_URL = r"""(?x)(?: @@ -1657,20 +1867,32 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): def _extract_mix(self, playlist_id): # The mixes are generated from a single video # the id of the playlist is just 'RD' + video_id - url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id) - webpage = self._download_webpage( - url, playlist_id, 'Downloading Youtube mix') + 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) - ids = orderedSet(re.findall( - r'''(?xs)data-video-username=".*?".*? - href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id), - webpage)) - url_results = self._ids_to_results(ids) return self.playlist_result(url_results, playlist_id, title) @@ -1723,7 +1945,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor): if video: return video - if playlist_id.startswith('RD') or playlist_id.startswith('UL'): + if playlist_id.startswith(('RD', 'UL', 'PU')): # Mixes require a custom extraction process return self._extract_mix(playlist_id) @@ -1757,12 +1979,16 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): @classmethod def suitable(cls, url): - return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url) + return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url) + else super(YoutubeChannelIE, cls).suitable(url)) + + 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._TEMPLATE_URL % channel_id + 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 @@ -1776,9 +2002,13 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): channel_playlist_id = self._html_search_meta( 'channelId', channel_page, 'channel id', default=None) if not channel_playlist_id: - channel_playlist_id = self._search_regex( - r'data-(?:channel-external-|yt)id="([^"]+)"', - channel_page, 'channel id', default=None) + 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( @@ -1801,36 +2031,114 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): for video_id, video_title in self.extract_videos_from_page(channel_page)] return self.playlist_result(entries, channel_id) + 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/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' - _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^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': { - 'title': 'TheLinuxFoundation', + '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, }] @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_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls) - if any(ie.suitable(url) for ie in other_ies): + 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 _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)/(?P<id>[^/]+))/live' + IE_NAME = 'youtube:live' + + _TESTS = [{ + 'url': 'http://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': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live', + 'only_matching': True, + }] + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + 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}$', video_id): + return self.url_result(video_id, YoutubeIE.ie_key()) + return self.url_result(base_url) + class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor): IE_DESC = 'YouTube.com user/channel playlists' @@ -1885,7 +2193,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE): 'spf': 'navigate', } url_query.update(self._EXTRA_QUERY_ARGS) - result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query) + result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query) data = self._download_json( result_url, video_id='query "%s"' % query, note='Downloading page %s' % pagenum, @@ -1914,10 +2222,11 @@ class YoutubeSearchDateIE(YoutubeSearchIE): _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'} -class YoutubeSearchURLIE(InfoExtractor): +class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor): IE_DESC = 'YouTube.com search URLs' IE_NAME = 'youtube:search_url' _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)' + _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, @@ -1932,32 +2241,8 @@ class YoutubeSearchURLIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) query = compat_urllib_parse_unquote_plus(mobj.group('query')) - webpage = self._download_webpage(url, query) - result_code = self._search_regex( - r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML') - - part_codes = re.findall( - r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code) - entries = [] - for part_code in part_codes: - part_title = self._html_search_regex( - [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False) - part_url_snippet = self._html_search_regex( - r'(?s)href="([^"]+)"', part_code, 'item URL') - part_url = compat_urlparse.urljoin( - 'https://www.youtube.com/', part_url_snippet) - entries.append({ - '_type': 'url', - 'url': part_url, - 'title': part_title, - }) - - return { - '_type': 'playlist', - 'entries': entries, - 'title': query, - } + return self.playlist_result(self._process_page(webpage), playlist_title=query) class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):