X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/179629569ec7a2eda04957287c17fab7e55b1a7f..d2e1a98478a93d5d191b83a9aa545df270ebf323:/youtube_dl/extractor/youtube.py diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index afb655c..3690f80 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1,472 +1,901 @@ # coding: utf-8 +from __future__ import unicode_literals + + +import itertools import json -import netrc +import os.path import re -import socket -import itertools +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_urlparse, compat_str, - +) +from ..utils import ( clean_html, - get_element_by_id, ExtractorError, + float_or_none, + get_element_by_attribute, + get_element_by_id, + int_or_none, + OnDemandPagedList, + orderedSet, unescapeHTML, unified_strdate, - orderedSet, + uppercase_escape, ) -class YoutubeIE(InfoExtractor): - IE_DESC = u'YouTube.com' - _VALID_URL = r"""^ +class YoutubeBaseInfoExtractor(InfoExtractor): + """Provide base functions for Youtube extractors""" + _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' + _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 _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 + + galx = self._search_regex(r'(?s)]* id="gaia_secondfactorform"', login_results) is not None: + tfa_code = self._get_tfa_info() + + if tfa_code is None: + self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor ') + self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)') + return False + + # Unlike the first login form, secTok and timeStmp are both required for the TFA form + + match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) + if match is None: + self._downloader.report_warning('Failed to get secTok - did the page structure change?') + secTok = match.group(1) + match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U) + if match is None: + self._downloader.report_warning('Failed to get timeStmp - did the page structure change?') + timeStmp = match.group(1) + + tfa_form_strs = { + 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', + 'smsToken': '', + 'smsUserPin': tfa_code, + 'smsVerifyPin': 'Verify', + + 'PersistentCookie': 'yes', + 'checkConnection': '', + 'checkedDomains': 'youtube', + 'pstMsg': '1', + 'secTok': secTok, + 'timeStmp': timeStmp, + 'service': 'youtube', + 'hl': 'en_US', + } + tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items()) + tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii') + + tfa_req = compat_urllib_request.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="gaia_secondfactorform"', tfa_results) is not None: + self._downloader.report_warning('Two-factor code expired. 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 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|movie(?:_popup)?(?:\.php)?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) + (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx) v= ) - )? # optional -> youtube.com/xxxx is OK + )) + |youtu\.be/ # just youtu.be/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 + (?!.*?&list=) # 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': 'mp4', - '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}, + '6': {'ext': 'flv', 'width': 450, 'height': 270}, + '13': {'ext': '3gp'}, + '17': {'ext': '3gp', 'width': 176, 'height': 144}, + '18': {'ext': 'mp4', 'width': 640, 'height': 360}, + '22': {'ext': 'mp4', 'width': 1280, 'height': 720}, + '34': {'ext': 'flv', 'width': 640, 'height': 360}, + '35': {'ext': 'flv', 'width': 854, 'height': 480}, + '36': {'ext': '3gp', 'width': 320, 'height': 240}, + '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}, + '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}, + '43': {'ext': 'webm', 'width': 640, 'height': 360}, + '44': {'ext': 'webm', 'width': 854, 'height': 480}, + '45': {'ext': 'webm', 'width': 1280, 'height': 720}, + '46': {'ext': 'webm', 'width': 1920, 'height': 1080}, + + + # 3d videos + '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}, + '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}, + '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}, + '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}, + '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}, + '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}, + '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}, + + # Apple HTTP Live Streaming + '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, + '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}, + '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}, + '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}, + '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}, + '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}, + '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}, + + # DASH mp4 video + '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559) + '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, + '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, + '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}, + + # Dash mp4 audio + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}, + + # Dash webm + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'}, + '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'}, + + # Dash webm audio + '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}, + '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}, + + # Dash webm audio with opus inside + '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}, + '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}, + '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}, + + # RTMP (unnamed) + '_rtmp': {'protocol': 'rtmp'}, } - IE_NAME = u'youtube' + + IE_NAME = 'youtube' _TESTS = [ { - u"url": u"http://www.youtube.com/watch?v=BaW_jenozKc", - u"file": u"BaW_jenozKc.mp4", - u"info_dict": { - u"title": u"youtube-dl test video \"'/\\ä↭𝕐", - u"uploader": u"Philipp Hagemeister", - u"uploader_id": u"phihag", - u"upload_date": u"20121002", - u"description": u"test chars: \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ." + 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc', + 'info_dict': { + 'id': 'BaW_jenozKc', + 'ext': 'mp4', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', + 'uploader': 'Philipp Hagemeister', + 'uploader_id': 'phihag', + 'upload_date': '20121002', + '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'], + 'like_count': int, + 'dislike_count': int, } }, { - u"url": u"http://www.youtube.com/watch?v=1ltcDfZMA3U", - u"file": u"1ltcDfZMA3U.flv", - u"note": u"Test VEVO video (#897)", - u"info_dict": { - u"upload_date": u"20070518", - u"title": u"Maps - It Will Find You", - u"description": u"Music video by Maps performing It Will Find You.", - u"uploader": u"MuteUSA", - u"uploader_id": u"MuteUSA" + 'url': 'http://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]', + 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f', + 'uploader': 'Icona Pop', + 'uploader_id': 'IconaPop', } }, { - u"url": u"http://www.youtube.com/watch?v=UxxajLWwzqY", - u"file": u"UxxajLWwzqY.mp4", - u"note": u"Test generic use_cipher_signature video (#897)", - u"info_dict": { - u"upload_date": u"20120506", - u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]", - u"description": u"md5:b085c9804f5ab69f4adea963a2dceb3c", - u"uploader": u"IconaPop", - u"uploader_id": u"IconaPop" + '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)', + 'description': 'md5:64249768eec3bc4276236606ea996373', + 'uploader': 'justintimberlakeVEVO', + 'uploader_id': 'justintimberlakeVEVO', } }, { - u"url": u"https://www.youtube.com/watch?v=07FYdnEawAQ", - u"file": u"07FYdnEawAQ.mp4", - u"note": u"Test VEVO video with age protection (#956)", - u"info_dict": { - u"upload_date": u"20130703", - u"title": u"Justin Timberlake - Tunnel Vision (Explicit)", - u"description": u"md5:64249768eec3bc4276236606ea996373", - u"uploader": u"justintimberlakeVEVO", - u"uploader_id": u"justintimberlakeVEVO" + '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' } }, + { + 'url': 'http://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', + 'description': '', + 'uploader': '8KVIDEO', + 'title': 'UHDTV TEST 8K VIDEO.mp4' + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', + }, + }, + # 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', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', + }, + }, + # 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', + 'description': 'md5:2acfda1b285bdd478ccec22f9918199d', + 'uploader': 'TaylorSwiftVEVO', + 'uploader_id': 'TaylorSwiftVEVO', + 'upload_date': '20140818', + }, + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', + }, + }, + # 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', + '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': 'http://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', + 'upload_date': '20140605', + }, + }, + # Age-gate video with encrypted signature + { + 'url': 'http://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', + 'upload_date': '20110629', + }, + }, + # 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', + 'description': 'md5:12c56784b8032162bb936a5f76d55360', + 'uploader': 'deadmau5', + 'title': 'Deadmau5 - Some Chords (HD)', + }, + 'expected_warnings': [ + 'DASH manifest missing', + ] + }, + # Olympics (https://github.com/rg3/youtube-dl/issues/4431) + { + 'url': 'lqQg6PlCWgI', + 'info_dict': { + 'id': 'lqQg6PlCWgI', + 'ext': 'mp4', + 'upload_date': '20120731', + 'uploader_id': 'olympic', + 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', + 'uploader': 'Olympics', + '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', + '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', + 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人', + }, + } ] - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - if YoutubePlaylistIE.suitable(url) or YoutubeSubscriptionsIE.suitable(url): return False - return re.match(cls._VALID_URL, url, re.VERBOSE) is not None - - def report_lang(self): - """Report attempt to set language.""" - self.to_screen(u'Setting language') - - def report_video_webpage_download(self, video_id): - """Report attempt to download video webpage.""" - self.to_screen(u'%s: Downloading video webpage' % video_id) + def __init__(self, *args, **kwargs): + super(YoutubeIE, self).__init__(*args, **kwargs) + self._player_cache = {} def report_video_info_webpage_download(self, video_id): """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Downloading video info webpage' % video_id) - - def report_video_subtitles_download(self, video_id): - """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Checking available subtitles' % video_id) - - def report_video_subtitles_request(self, video_id, sub_lang, format): - """Report attempt to download video info webpage.""" - self.to_screen(u'%s: Downloading video subtitles for %s.%s' % (video_id, sub_lang, format)) - - def report_video_subtitles_available(self, video_id, sub_lang_list): - """Report available subtitles.""" - sub_lang = ",".join(list(sub_lang_list.keys())) - self.to_screen(u'%s: Available subtitles for video: %s' % (video_id, sub_lang)) + self.to_screen('%s: Downloading video info webpage' % video_id) def report_information_extraction(self, video_id): """Report attempt to extract video information.""" - self.to_screen(u'%s: Extracting video information' % video_id) + self.to_screen('%s: Extracting video information' % video_id) def report_unavailable_format(self, video_id, format): """Report extracted video URL.""" - self.to_screen(u'%s: Format %s not available' % (video_id, format)) + self.to_screen('%s: Format %s not available' % (video_id, format)) def report_rtmp_download(self): """Indicate the download will use the RTMP protocol.""" - self.to_screen(u'RTMP download detected') - - def _decrypt_signature(self, s): + self.to_screen('RTMP download detected') + + def _signature_cache_id(self, example_sig): + """ Return a string representation of a signature """ + return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) + + def _extract_signature_function(self, video_id, player_url, example_sig): + id_m = re.match( + r'.*?-(?P[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?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 len(s) == 88: - return s[48] + s[81:67:-1] + s[82] + s[66:62:-1] + s[85] + s[61:48:-1] + s[67] + s[47:12:-1] + s[3] + s[11:3:-1] + s[2] + s[12] - elif len(s) == 87: - return s[62] + s[82:62:-1] + s[83] + s[61:52:-1] + s[0] + s[51:2:-1] - elif len(s) == 86: - return s[2:63] + s[82] + s[64:82] + s[63] - elif len(s) == 85: - return s[76] + s[82:76:-1] + s[83] + s[75:60:-1] + s[0] + s[59:50:-1] + s[1] + s[49:2:-1] - elif len(s) == 84: - return s[83:36:-1] + s[2] + s[35:26:-1] + s[3] + s[25:3:-1] + s[26] - elif len(s) == 83: - return s[:81] - elif len(s) == 82: - return s[36] + s[79:67:-1] + s[81] + s[66:40:-1] + s[33] + s[39:36:-1] + s[40] + s[35] + s[0] + s[67] + s[32:0:-1] + s[34] - - else: - raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s))) + if player_url is None: + raise ExtractorError('Cannot decrypt signature without player_url') - def _get_available_subtitles(self, video_id): - self.report_video_subtitles_download(video_id) - request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) + 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' % 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 ['sbv', 'vtt', 'srt']: + 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_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) + self.to_screen('%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 + err_msg = 'Couldn\'t find automatic captions for %s' % video_id if mobj is None: - return [(err_msg, None, None)] + self._downloader.report_warning(err_msg) + return {} player_config = json.loads(mobj.group(1)) 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', + args = player_config['args'] + caption_url = args['ttsurl'] + timestamp = args['timestamp'] + # We get the available subtitles + list_params = compat_urllib_parse.urlencode({ + 'type': 'list', + 'tlangs': 1, + 'asrs': 1, }) - 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: - 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 - - (username, password) = self._get_login_info() - - # No authentication to be performed - if username is None: - return + 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 ['sbv', 'vtt', 'srt']: + 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 + # 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 {} - request = compat_urllib_request.Request(self._LOGIN_URL) - try: - login_page = compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err)) - return - - galx = None - dsh = None - match = re.search(re.compile(r']* id="gaia_loginform"', login_results) is not None: - self._downloader.report_warning(u'unable to log in: bad username or password') - return - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self._downloader.report_warning(u'unable to log in: %s' % compat_str(err)) - return - - # Confirm age - age_form = { - 'next_url': '/', - 'action_confirm': 'Confirm', - } - request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) - try: - self.report_age_confirmation() - compat_urllib_request.urlopen(request).read().decode('utf-8') - except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err)) - - 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 _parse_dash_manifest( + self, video_id, dash_manifest_url, player_url, age_gate): + 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 + dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) + dash_doc = self._download_xml( + dash_manifest_url, video_id, + note='Downloading DASH manifest', + errnote='Could not download DASH manifest') + + formats = [] + for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): + url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL') + if url_el is None: + continue + format_id = r.attrib['id'] + video_url = url_el.text + filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength')) + f = { + 'format_id': format_id, + 'url': video_url, + 'width': int_or_none(r.attrib.get('width')), + 'height': int_or_none(r.attrib.get('height')), + 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000), + 'asr': int_or_none(r.attrib.get('audioSamplingRate')), + 'filesize': filesize, + 'fps': int_or_none(r.attrib.get('frameRate')), + } + try: + existing_format = next( + fo for fo in formats + if fo['format_id'] == format_id) + except StopIteration: + full_info = self._formats.get(format_id, {}).copy() + full_info.update(f) + formats.append(full_info) + else: + existing_format.update(f) + return formats + def _real_extract(self, url): - if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url): - self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply youtube-dl BaW_jenozKc ).') + proto = ( + 'http' if self._downloader.params.get('prefer_insecure', False) + else 'https') # 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 # Get video info - self.report_video_info_webpage_download(video_id) + embed_webpage = None if re.search(r'player-age-gate-content">', video_webpage) is not None: - self.report_age_confirmation() age_gate = True # We simulate the access to the video from www.youtube.com/v/{video_id} # this can be viewed without login into Youtube - data = compat_urllib_parse.urlencode({'video_id': video_id, - 'el': 'embedded', - 'gl': 'US', - 'hl': 'en', - 'eurl': 'https://youtube.googleapis.com/v/' + video_id, - 'asv': 3, - 'sts':'1588', - }) - video_info_url = 'https://www.youtube.com/get_video_info?' + data - video_info_webpage = self._download_webpage(video_info_url, video_id, - note=False, - errnote='unable to download video info webpage') + 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) else: age_gate = False - for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']: - video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en' - % (video_id, el_type)) - video_info_webpage = self._download_webpage(video_info_url, video_id, - note=False, - errnote='unable to download video info webpage') - video_info = compat_parse_qs(video_info_webpage) - if 'token' in video_info: - break + try: + # Try looking directly into the video webpage + mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage) + if not mobj: + raise ValueError('Could not find ytplayer.config') # caught below + json_code = uppercase_escape(mobj.group(1)) + ytplayer_config = json.loads(json_code) + args = ytplayer_config['args'] + # Convert to the same format returned by compat_parse_qs + video_info = dict((k, [v]) for k, v in args.items()) + if 'url_encoded_fmt_stream_map' not in args: + raise ValueError('No stream_map present') # caught below + except ValueError: + # We fallback to the get_video_info pages (used by the embed page) + self.report_video_info_webpage_download(video_id) + for el_type in ['&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') + video_info = compat_parse_qs(video_info_webpage) + if 'token' in video_info: + break if 'token' not in video_info: if 'reason' in video_info: - raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], 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) + + if 'view_count' in video_info: + view_count = int(video_info['view_count'][0]) + else: + view_count = None # Check for "rental" videos if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info: - raise ExtractorError(u'"rental" videos not supported') + raise ExtractorError('"rental" videos not supported') # Start extracting information self.report_information_extraction(video_id) # uploader if 'author' not in video_info: - raise ExtractorError(u'Unable to extract uploader name') + raise ExtractorError('Unable to extract uploader name') video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0]) # uploader_id @@ -475,12 +904,14 @@ class YoutubeIE(InfoExtractor): if mobj is not None: video_uploader_id = mobj.group(1) else: - self._downloader.report_warning(u'unable to extract uploader nickname') + self._downloader.report_warning('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]) + if 'title' in video_info: + video_title = video_info['title'][0] + else: + self._downloader.report_warning('Unable to extract video title') + video_title = '_' # thumbnail image # We try first to get a high quality image: @@ -489,241 +920,426 @@ class YoutubeIE(InfoExtractor): if m_thumb is not None: video_thumbnail = m_thumb.group(1) elif 'thumbnail_url' not in video_info: - self._downloader.report_warning(u'unable to extract video thumbnail') - video_thumbnail = '' + 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]) # upload date upload_date = None - mobj = re.search(r'id="eow-date.*?>(.*?)', video_webpage, re.DOTALL) + mobj = re.search(r'(?s)id="eow-date.*?>(.*?)', video_webpage) + if mobj is None: + mobj = re.search( + r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)', + video_webpage) if mobj is not None: upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) upload_date = unified_strdate(upload_date) + m_cat_container = self._search_regex( + r'(?s)]*>\s*Category\s*\s*]*>(.*?)', + video_webpage, 'categories', default=None) + if m_cat_container: + category = self._html_search_regex( + r'(?s)(.*?)', m_cat_container, 'category', + default=None) + video_categories = None if category is None else [category] + else: + video_categories = None + # description video_description = get_element_by_id("eow-description", video_webpage) if video_description: + video_description = re.sub(r'''(?x) + + [^<]+ + + ''', r'\1', video_description) video_description = clean_html(video_description) else: fd_mobj = re.search(r']*>.*?([\d,]+)\s*' % re.escape(count_name), + video_webpage, count_name, default=None) + if count is not None: + return int(count.replace(',', '')) + return 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: - self._downloader.report_warning(sub_error) - - if self._downloader.params.get('writeautomaticsub', False): - video_subtitles = self._request_automatic_caption(video_id, video_webpage) - (sub_error, sub_lang, sub) = video_subtitles[0] - if sub_error: - self._downloader.report_warning(sub_error) - - if self._downloader.params.get('allsubtitles', False): - video_subtitles = self._extract_all_subtitles(video_id) - 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) - if not mobj: - raise ValueError('Could not find vevo ID') - info = json.loads(mobj.group(1)) - args = info['args'] - # Easy way to know if the 's' value is in url_encoded_fmt_stream_map - # this signatures are encrypted - m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map']) - if m_s is not None: - self.to_screen(u'%s: Encrypted signatures detected.' % video_id) - video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] - except ValueError: - pass + 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: + 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) url_map = {} - for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','): + 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: - if self._downloader.params.get('verbose'): - s = url_data['s'][0] - if age_gate: - player_version = self._search_regex(r'ad3-(.+?)\.swf', - video_info['ad3_module'][0], 'flash player', - fatal=False) - player = 'flash player %s' % player_version + 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 = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage, + player_version = self._search_regex( + r'html5player-([^/]+?)(?:/html5player)?\.js', + player_url, 'html5 player', fatal=False) - self.to_screen('encrypted signature length %d (%d.%d), itag %s, %s' % - (len(s), len(s.split('.')[0]), len(s.split('.')[1]), url_data['itag'][0], player)) - 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[-1], url_map[existing_formats[-1]])] # worst quality - elif req_format in ('-1', 'all'): - video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats - else: - # Specific formats. We pick the first in a slash-delimeted sequence. - # For example, if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'. - req_formats = req_format.split('/') - video_url_list = None - for rf in req_formats: - if rf in url_map: - video_url_list = [(rf, url_map[rf])] - break - if video_url_list is None: - raise ExtractorError(u'requested format not available') + 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' + url_map[format_id] = url + formats = _map_to_format_list(url_map) + 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) 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 + 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 = video_info.get('dashmpd') + if dash_mpd: + dash_manifest_url = dash_mpd[0] + try: + dash_formats = self._parse_dash_manifest( + video_id, dash_manifest_url, player_url, age_gate) + except (ExtractorError, KeyError) as e: + self.report_warning( + 'Skipping DASH manifest: %r' % e, video_id) + else: + # Hide the formats we found through non-DASH + dash_keys = set(df['format_id'] for df in dash_formats) + for f in formats: + if f['format_id'] in dash_keys: + f['format_id'] = 'nondash-%s' % f['format_id'] + f['preference'] = f.get('preference', 0) - 10000 + formats.extend(dash_formats) + + # Check for malformed aspect ratio + stretched_m = re.search( + r'[0-9]+):(?P[0-9]+)">', + video_webpage) + if stretched_m: + ratio = float(stretched_m.group('w')) / float(stretched_m.group('h')) + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio + + self._sort_formats(formats) + + return { + 'id': video_id, + 'uploader': video_uploader, + 'uploader_id': video_uploader_id, + 'upload_date': upload_date, + 'title': video_title, + 'thumbnail': video_thumbnail, + 'description': video_description, + 'categories': video_categories, + '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, + } -class YoutubePlaylistIE(InfoExtractor): - IE_DESC = u'YouTube.com playlists' - _VALID_URL = r"""(?: + +class YoutubePlaylistIE(YoutubeBaseInfoExtractor): + IE_DESC = 'YouTube.com playlists' + _VALID_URL = r"""(?x)(?: (?:https?://)? (?:\w+\.)? youtube\.com/ (?: - (?:course|view_play_list|my_playlists|artist|playlist|watch) + (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries) \? (?:.*?&)*? (?:p|a|list)= | p/ ) - ((?:PL|EC|UU)?[0-9A-Za-z-_]{10,}) + ( + (?: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|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' + _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' + _VIDEO_RE = r'href="\s*/watch\?v=(?P[0-9A-Za-z_-]{11})&[^"]*?index=(?P\d+)' + 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': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + 'playlist_count': 4, + 'info_dict': { + 'title': 'JODA15', + 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + } + }, { + 'note': 'Embedded SWF player', + 'url': 'http://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_mincout': 21, + }] - @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 + def _real_initialize(self): + self._login() + + def _ids_to_results(self, ids): + return [ + self.url_result(vid_id, 'Youtube', video_id=vid_id) + for vid_id in ids] + + 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') + 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) def _real_extract(self, url): # Extract playlist id - mobj = re.match(self._VALID_URL, url, re.VERBOSE) + mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - # Download playlist videos from API + raise ExtractorError('Invalid URL: %s' % url) playlist_id = mobj.group(1) or mobj.group(2) - page_num = 1 - videos = [] - while True: - url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, self._MAX_RESULTS * (page_num - 1) + 1) - page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num) - - 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']['$t'] - if 'entry' not in response['feed']: - # Number of videos is a multiple of self._MAX_RESULTS + # Check if it's a video-specific URL + query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + if 'v' in query_dict: + video_id = query_dict['v'][0] + if self._downloader.params.get('noplaylist'): + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + return self.url_result(video_id, 'Youtube', video_id=video_id) + else: + self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) + + if playlist_id.startswith('RD') or playlist_id.startswith('UL'): + # Mixes require a custom extraction process + return self._extract_mix(playlist_id) + + url = self._TEMPLATE_URL % playlist_id + page = self._download_webpage(url, playlist_id) + more_widget_html = content_html = page + + # Check if the playlist exists or is private + if re.search(r'
[^<]*?(The|This) playlist (does not exist|is private)[^<]*?
', page) is not None: + raise ExtractorError( + 'The playlist doesn\'t exist or is private, use --username or ' + '--netrc to access it.', + expected=True) + + # Extract the video ids from the playlist pages + ids = [] + + for page_num in itertools.count(1): + matches = re.finditer(self._VIDEO_RE, content_html) + # We remove the duplicates and the link with index 0 + # (it's not the first video of the playlist) + new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0') + ids.extend(new_ids) + + mobj = re.search(r'data-uix-load-more-href="/?(?P[^"]+)"', more_widget_html) + if not mobj: break - for entry in response['feed']['entry']: - index = entry['yt$position']['$t'] - if 'media$group' in entry and 'media$player' in entry['media$group']: - videos.append((index, entry['media$group']['media$player']['url'])) - - if len(response['feed']['entry']) < self._MAX_RESULTS: + 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 - page_num += 1 + more_widget_html = more['load_more_widget_html'] - videos = [v[1] for v in sorted(videos)] + playlist_title = self._html_search_regex( + r'(?s)

\s*(.*?)\s*

', + page, 'title') - url_results = [self.url_result(vurl, 'Youtube') for vurl in videos] - return [self.playlist_result(url_results, playlist_id, playlist_title)] + url_results = self._ids_to_results(ids) + return self.playlist_result(url_results, playlist_id, playlist_title) class YoutubeChannelIE(InfoExtractor): - IE_DESC = u'YouTube.com channels' - _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" - _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en' - _MORE_PAGES_INDICATOR = 'yt-uix-load-more' - _MORE_PAGES_URL = 'http://www.youtube.com/channel_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' - IE_NAME = u'youtube:channel' + IE_DESC = 'YouTube.com channels' + _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P[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': 'UCKfVa3S1e4PHvxWcwyMMg8w', + } + }] def extract_videos_from_page(self, page): ids_in_page = [] @@ -733,134 +1349,150 @@ class YoutubeChannelIE(InfoExtractor): return ids_in_page 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) + channel_id = self._match_id(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) - - page = json.loads(page) - - ids_in_page = self.extract_videos_from_page(page['content_html']) - video_ids.extend(ids_in_page) - - if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: + url = 'https://www.youtube.com/channel/%s/videos' % channel_id + channel_page = self._download_webpage(url, channel_id) + 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 + video_ids = self.extract_videos_from_page(channel_page) + entries = [ + self.url_result(video_id, 'Youtube', video_id=video_id) + for video_id in video_ids] + return self.playlist_result(entries, channel_id) + + def _entries(): + more_widget_html = content_html = channel_page + for pagenum in itertools.count(1): + + ids_in_page = self.extract_videos_from_page(content_html) + for video_id in ids_in_page: + yield self.url_result( + video_id, 'Youtube', video_id=video_id) + + mobj = re.search( + r'data-uix-load-more-href="/?(?P[^"]+)"', + more_widget_html) + if not mobj: break - self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) + more = self._download_json( + 'https://youtube.com/%s' % mobj.group('more'), channel_id, + 'Downloading page #%s' % (pagenum + 1), + transform_source=uppercase_escape) + content_html = more['content_html'] + more_widget_html = more['load_more_widget_html'] - urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids] - url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls] - return [self.playlist_result(url_entries, channel_id)] + return self.playlist_result(_entries(), channel_id) class YoutubeUserIE(InfoExtractor): - IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' - _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/user/)|ytuser:)([A-Za-z0-9_-]+)' - _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s' + 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[A-Za-z0-9_-]+)' + _TEMPLATE_URL = 'https://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' + _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' + IE_NAME = 'youtube:user' + + _TESTS = [{ + 'url': 'https://www.youtube.com/user/TheLinuxFoundation', + 'playlist_mincount': 320, + 'info_dict': { + 'title': 'TheLinuxFoundation', + } + }, { + 'url': 'ytuser:phihag', + '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) + @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): + return False + else: + return super(YoutubeUserIE, cls).suitable(url) - username = mobj.group(1) + def _real_extract(self, url): + username = self._match_id(url) # 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 - - while True: + def download_page(pagenum): start_index = pagenum * self._GDATA_PAGE_SIZE + 1 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)) - - # Extract video identifiers - ids_in_page = [] - - 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)) + page = self._download_webpage( + gdata_url, username, + 'Downloading video ids from %d to %d' % ( + start_index, start_index + self._GDATA_PAGE_SIZE)) - video_ids.extend(ids_in_page) + try: + response = json.loads(page) + except ValueError as err: + raise ExtractorError('Invalid JSON in API response: ' + compat_str(err)) + if 'entry' not in response['feed']: + return - # 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. + # Extract video identifiers + entries = response['feed']['entry'] + for entry in entries: + title = entry['title']['$t'] + video_id = entry['id']['$t'].split('/')[-1] + yield { + '_type': 'url', + 'url': video_id, + 'ie_key': 'Youtube', + 'id': video_id, + 'title': title, + } + url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE) - if len(ids_in_page) < self._GDATA_PAGE_SIZE: - break + return self.playlist_result(url_results, playlist_title=username) - pagenum += 1 - - urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids] - url_results = [self.url_result(rurl, 'Youtube') for rurl in urls] - return [self.playlist_result(url_results, playlist_title = username)] class YoutubeSearchIE(SearchInfoExtractor): - IE_DESC = u'YouTube.com searches' + IE_DESC = 'YouTube.com searches' _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' _MAX_RESULTS = 1000 - IE_NAME = u'youtube:search' + IE_NAME = 'youtube:search' _SEARCH_KEY = 'ytsearch' - 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 _get_n_results(self, query, n): """Get a specified number of results for a query""" video_ids = [] pagenum = 0 limit = n - - 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'] - - if not 'items' in api_response: - raise ExtractorError(u'[youtube] No video results') + PAGE_SIZE = 50 + + while (PAGE_SIZE * pagenum) < limit: + result_url = self._API_URL % ( + compat_urllib_parse.quote_plus(query.encode('utf-8')), + (PAGE_SIZE * pagenum) + 1) + data_json = self._download_webpage( + result_url, video_id='query "%s"' % query, + note='Downloading page %s' % (pagenum + 1), + errnote='Unable to download API page') + data = json.loads(data_json) + api_response = data['data'] + + if 'items' not in api_response: + raise ExtractorError( + '[youtube] No video results', expected=True) new_ids = list(video['id'] for video in api_response['items']) video_ids += new_ids @@ -870,57 +1502,276 @@ class YoutubeSearchIE(SearchInfoExtractor): 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] + videos = [self.url_result(video_id, 'Youtube', video_id=video_id) + for video_id in video_ids] return self.playlist_result(videos, query) +class YoutubeSearchDateIE(YoutubeSearchIE): + IE_NAME = YoutubeSearchIE.IE_NAME + ':date' + _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published' + _SEARCH_KEY = 'ytsearchdate' + IE_DESC = 'YouTube.com searches, newest videos first' + + +class YoutubeSearchURLIE(InfoExtractor): + IE_DESC = 'YouTube.com search URLs' + IE_NAME = 'youtube:search_url' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P[^&]+)(?:[&]|$)' + _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', + } + }] + + 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)
    ', webpage, 'result HTML') + + part_codes = re.findall( + r'(?s)

    (.*?)

    ', result_code) + entries = [] + for part_code in part_codes: + part_title = self._html_search_regex( + [r'(?s)title="([^"]+)"', r'>([^<]+)'], 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, + } + + class YoutubeShowIE(InfoExtractor): - IE_DESC = u'YouTube.com (multi-season) shows' - _VALID_URL = r'https?://www\.youtube\.com/show/(.*)' - IE_NAME = u'youtube:show' + IE_DESC = 'YouTube.com (multi-season) shows' + _VALID_URL = r'https?://www\.youtube\.com/show/(?P[^?#]*)' + IE_NAME = 'youtube:show' + _TESTS = [{ + 'url': 'http://www.youtube.com/show/airdisasters', + 'playlist_mincount': 3, + 'info_dict': { + 'id': 'airdisasters', + 'title': 'Air Disasters', + } + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - show_name = mobj.group(1) - webpage = self._download_webpage(url, show_name, u'Downloading show webpage') + playlist_id = mobj.group('id') + webpage = self._download_webpage( + url, playlist_id, 'Downloading show webpage') # There's one playlist for each season of the show m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage)) - self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons))) - return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons] + self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons))) + entries = [ + self.url_result( + 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist') + for season in m_seasons + ] + title = self._og_search_title(webpage, fatal=False) + + return { + '_type': 'playlist', + 'id': playlist_id, + 'title': title, + 'entries': entries, + } -class YoutubeSubscriptionsIE(YoutubeIE): - """It's a subclass of YoutubeIE because we need to login""" - IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)' - _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' - IE_NAME = u'youtube:subscriptions' - _FEED_TEMPLATE = 'http://www.youtube.com/feed_ajax?action_load_system_feed=1&feed_name=subscriptions&paging=%s' - _PAGING_STEP = 30 +class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): + """ + Base class for extractors that fetch info from + http://www.youtube.com/feed_ajax + Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties. + """ + _LOGIN_REQUIRED = True + # use action_load_personal_feed instead of action_load_system_feed + _PERSONAL_FEED = False - # Overwrite YoutubeIE properties we don't want - _TESTS = [] - @classmethod - def suitable(cls, url): - return re.match(cls._VALID_URL, url) is not None + @property + def _FEED_TEMPLATE(self): + action = 'action_load_system_feed' + if self._PERSONAL_FEED: + action = 'action_load_personal_feed' + return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME) + + @property + def IE_NAME(self): + return 'youtube:%s' % self._FEED_NAME def _real_initialize(self): - (username, password) = self._get_login_info() - if username is None: - raise ExtractorError(u'No login info available, needed for downloading the Youtube subscriptions.', expected=True) - super(YoutubeSubscriptionsIE, self)._real_initialize() + self._login() def _real_extract(self, url): feed_entries = [] - # The step argument is available only in 2.7 or higher - for i in itertools.count(0): - paging = i*self._PAGING_STEP - info = self._download_webpage(self._FEED_TEMPLATE % paging, 'feed', - u'Downloading page %s' % i) - info = json.loads(info) - feed_html = info['feed_html'] - m_ids = re.finditer(r'"/watch\?v=(.*?)"', feed_html) + paging = 0 + for i in itertools.count(1): + info = self._download_json( + self._FEED_TEMPLATE % paging, + '%s feed' % self._FEED_NAME, + 'Downloading page %s' % i, + transform_source=uppercase_escape) + feed_html = info.get('feed_html') or info.get('content_html') + load_more_widget_html = info.get('load_more_widget_html') or feed_html + m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html) ids = orderedSet(m.group(1) for m in m_ids) - feed_entries.extend(self.url_result(id, 'Youtube') for id in ids) - if info['paging'] is None: + feed_entries.extend( + self.url_result(video_id, 'Youtube', video_id=video_id) + for video_id in ids) + mobj = re.search( + r'data-uix-load-more-href="/?[^"]+paging=(?P\d+)', + load_more_widget_html) + if mobj is None: break - return self.playlist_result(feed_entries, playlist_title='Youtube Subscriptions') + paging = mobj.group('paging') + return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) + + +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 YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): + IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' + _FEED_NAME = 'watch_later' + _PLAYLIST_TITLE = 'Youtube Watch Later' + _PERSONAL_FEED = True + + +class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): + IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)' + _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory' + _FEED_NAME = 'history' + _PERSONAL_FEED = True + _PLAYLIST_TITLE = 'Youtube Watch History' + + +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 + + 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 YoutubeSubscriptionsIE(YoutubePlaylistIE): + IE_NAME = 'youtube:subscriptions' + IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' + _TESTS = [] + + def _real_extract(self, url): + title = 'Youtube Subscriptions' + page = self._download_webpage('https://www.youtube.com/feed/subscriptions', 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) + new_ids = orderedSet(matches) + ids.extend(new_ids) + + 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'), title, + 'Downloading page #%s' % page_num, + transform_source=uppercase_escape) + content_html = more['content_html'] + more_widget_html = more['load_more_widget_html'] + + return { + '_type': 'playlist', + 'title': title, + 'entries': self._ids_to_results(ids), + } + + +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=[^&]*| + )? + | + attribution_link\?a=[^&]+ + ) + $ + ''' + + _TESTS = [{ + 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041', + 'only_matching': True, + }, { + 'url': 'http://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, + }] + + 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 ' + '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' + ' or simply youtube-dl BaW_jenozKc .', + expected=True) + + +class YoutubeTruncatedIDIE(InfoExtractor): + IE_NAME = 'youtube:truncated_id' + IE_DESC = False # Do not list + _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P[0-9A-Za-z_-]{1,10})$' + + _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)