X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/1931fa5f23ac75ea766517aa1b5ebf267ad3f97d..7e09b23b42767b13b618b123e3832ff657bd3976:/youtube_dl/extractor/youtube.py diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 3c9b0b5..6071d72 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1,7 +1,8 @@ # coding: utf-8 -import errno -import io +from __future__ import unicode_literals + + import itertools import json import os.path @@ -21,22 +22,21 @@ from ..utils import ( compat_str, clean_html, - get_cachedir, get_element_by_id, get_element_by_attribute, ExtractorError, int_or_none, - PagedList, + OnDemandPagedList, unescapeHTML, unified_strdate, orderedSet, - write_json_file, uppercase_escape, ) class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' + _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor' _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NETRC_MACHINE = 'youtube' @@ -46,48 +46,57 @@ class YoutubeBaseInfoExtractor(InfoExtractor): def _set_language(self): return bool(self._download_webpage( self._LANG_URL, None, - note=u'Setting language', errnote='unable to set language', + note='Setting language', errnote='unable to set language', fatal=False)) 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(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) - return False + 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=u'Downloading login page', - errnote=u'unable to fetch login page', fatal=False) + 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(u'unable to log in: bad username or password') + self._downloader.report_warning('unable to log in: bad username or password') return False return True @@ -114,24 +185,25 @@ class YoutubeBaseInfoExtractor(InfoExtractor): self._download_webpage( req, None, - note=u'Confirming age', errnote=u'Unable to confirm age') - return True + note='Confirming age', errnote='Unable to confirm age', + fatal=False) def _real_initialize(self): if self._downloader is None: return - if not self._set_language(): - return + if self._get_login_info()[0] is not None: + if not self._set_language(): + return if not self._login(): return self._confirm_age() class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): - IE_DESC = u'YouTube.com' + IE_DESC = 'YouTube.com' _VALID_URL = r"""(?x)^ ( - (?:https?://|//)? # http(s):// or protocol-independent URL (optional) + (?: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/| @@ -140,7 +212,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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) (?:\?|\#!?) # the params delimiter ? or # or #! @@ -149,10 +221,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): ) )) |youtu\.be/ # just youtu.be/xxxx - |https?://(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= + |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= ) )? # all until now is optional -> you can pass the naked 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 $""" _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' @@ -214,6 +287,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '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}, @@ -223,130 +297,137 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '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'}, # Dash webm audio - '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 48, 'preference': -50}, + '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 mov + '298': {'ext': 'mov', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, + '299': {'ext': 'mov', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}, + '266': {'ext': 'mov', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}, + # 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\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 .", - u"categories": [u'Science & Technology'], + '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=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:fea86fda2d5a5784273df5c7cc994d9f", - u"uploader": u"Icona Pop", - u"uploader_id": u"IconaPop" + '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"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': '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"//www.YouTube.com/watch?v=yZIXLfi8CZQ", - u"file": u"yZIXLfi8CZQ.mp4", - u"note": u"Embed-only video (#1746)", - u"info_dict": { - u"upload_date": u"20120608", - u"title": u"Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012", - u"description": u"md5:09b78bd971f1e3e289601dfba15ca4f7", - u"uploader": u"SET India", - u"uploader_id": u"setindia" + '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' } }, { - u"url": u"http://www.youtube.com/watch?v=a9LDPn-MO4I", - u"file": u"a9LDPn-MO4I.m4a", - u"note": u"256k DASH audio (format 141) via DASH manifest", - u"info_dict": { - u"upload_date": "20121002", - u"uploader_id": "8KVIDEO", - u"description": "No description available.", - u"uploader": "8KVIDEO", - u"title": "UHDTV TEST 8K VIDEO.mp4" + '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' }, - u"params": { - u"youtube_include_dash_manifest": True, - u"format": "141", + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', }, }, # DASH manifest with encrypted signature { - u'url': u'https://www.youtube.com/watch?v=IB3lcPjvWLA', - u'info_dict': { - u'id': u'IB3lcPjvWLA', - u'ext': u'm4a', - u'title': u'Afrojack - The Spark ft. Spree Wilson', - u'description': u'md5:9717375db5a9a3992be4668bbf3bc0a8', - u'uploader': u'AfrojackVEVO', - u'uploader_id': u'AfrojackVEVO', - u'upload_date': u'20131011', + 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA', + 'info_dict': { + 'id': 'IB3lcPjvWLA', + 'ext': 'm4a', + 'title': 'Afrojack - The Spark ft. Spree Wilson', + 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8', + 'uploader': 'AfrojackVEVO', + 'uploader_id': 'AfrojackVEVO', + 'upload_date': '20131011', }, - u"params": { - u'youtube_include_dash_manifest': True, - u'format': '141', + 'params': { + 'youtube_include_dash_manifest': True, + 'format': '141', }, }, ] - - @classmethod - def suitable(cls, url): - """Receives a URL and returns True if suitable for this IE.""" - if YoutubePlaylistIE.suitable(url): return False - return re.match(cls._VALID_URL, url) is not None - 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) + 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') + self.to_screen('RTMP download detected') def _signature_cache_id(self, example_sig): """ Return a string representation of a signature """ - return u'.'.join(compat_str(len(part)) for part in example_sig.split('.')) + 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( @@ -361,61 +442,42 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): func_id = '%s_%s_%s' % ( player_type, player_id, self._signature_cache_id(example_sig)) assert os.path.basename(func_id) == func_id - cache_dir = get_cachedir(self._downloader.params) - cache_enabled = cache_dir is not None - if cache_enabled: - cache_fn = os.path.join(os.path.expanduser(cache_dir), - u'youtube-sigfuncs', - func_id + '.json') - try: - with io.open(cache_fn, 'r', encoding='utf-8') as cachef: - cache_spec = json.load(cachef) - return lambda s: u''.join(s[i] for i in cache_spec) - except IOError: - pass # No cache available + 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) if player_type == 'js': code = self._download_webpage( player_url, video_id, - note=u'Downloading %s player %s' % (player_type, player_id), - errnote=u'Download of %s failed' % player_url) + note='Downloading %s player %s' % (player_type, player_id), + 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=u'Downloading %s player %s' % (player_type, player_id), - errnote=u'Download of %s failed' % player_url) + note='Downloading %s player %s' % (player_type, player_id), + 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 - if cache_enabled: - try: - test_string = u''.join(map(compat_chr, range(len(example_sig)))) - cache_res = res(test_string) - cache_spec = [ord(c) for c in cache_res] - try: - os.makedirs(os.path.dirname(cache_fn)) - except OSError as ose: - if ose.errno != errno.EEXIST: - raise - write_json_file(cache_spec, cache_fn) - except Exception: - tb = traceback.format_exc() - self._downloader.report_warning( - u'Writing cache to %r failed: %s' % (cache_fn, tb)) + if cache_spec is None: + 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 = u'' if start == 0 else str(start) - ends = (u':%d' % (end+step)) if end + step >= 0 else u':' - steps = u'' if step == 1 else (u':%d' % step) - return u's[%s%s%s]' % (starts, ends, steps) + 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 start = '(Never used)' # Quelch pyflakes warnings - start will be @@ -432,26 +494,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): start = prev continue else: - yield u's[%d]' % prev + yield 's[%d]' % prev if step is None: - yield u's[%d]' % i + yield 's[%d]' % i else: yield _genslice(start, i, step) - test_string = u''.join(map(compat_chr, range(len(example_sig)))) + 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 = u' + '.join(gen_sig_code(cache_spec)) + expr_code = ' + '.join(gen_sig_code(cache_spec)) signature_id_tuple = '(%s)' % ( ', '.join(compat_str(len(p)) for p in example_sig.split('.'))) - code = (u'if tuple(len(p) for p in s.split(\'.\')) == %s:\n' - u' return %s\n') % (signature_id_tuple, expr_code) - self.to_screen(u'Extracted signature function:\n' + code) + 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'signature=([$a-zA-Z]+)', jscode, - u'Initial JS player signature function name') + 'Initial JS player signature function name') jsi = JSInterpreter(jscode) initial_function = jsi.extract_function(funcname) @@ -459,19 +521,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): def _parse_sig_swf(self, file_contents): swfi = SWFInterpreter(file_contents) - TARGET_CLASSNAME = u'SignatureDecipher' + TARGET_CLASSNAME = 'SignatureDecipher' searched_class = swfi.extract_class(TARGET_CLASSNAME) - initial_function = swfi.extract_function(searched_class, u'decipher') + initial_function = swfi.extract_function(searched_class, 'decipher') return lambda s: initial_function([s]) def _decrypt_signature(self, s, video_id, player_url, age_gate=False): """Turn the encrypted s field into a working signature""" if player_url is None: - raise ExtractorError(u'Cannot decrypt signature without player_url') + raise ExtractorError('Cannot decrypt signature without player_url') - if player_url.startswith(u'//'): - player_url = u'https:' + player_url + if player_url.startswith('//'): + player_url = 'https:' + player_url try: player_id = (player_url, self._signature_cache_id(s)) if player_id not in self._player_cache: @@ -486,7 +548,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): except Exception as e: tb = traceback.format_exc() raise ExtractorError( - u'Signature extraction failed: ' + tb, cause=e) + 'Signature extraction failed: ' + tb, cause=e) def _get_available_subtitles(self, video_id, webpage): try: @@ -494,23 +556,25 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '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(u'unable to download video subtitles: %s' % compat_str(err)) + self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err)) return {} lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list) sub_lang_list = {} for l in lang_list: lang = l[1] + if lang in sub_lang_list: + continue params = compat_urllib_parse.urlencode({ 'lang': lang, 'v': video_id, 'fmt': self._downloader.params.get('subtitlesformat', 'srt'), 'name': unescapeHTML(l[0]).encode('utf-8'), }) - url = u'https://www.youtube.com/api/timedtext?' + params + url = 'https://www.youtube.com/api/timedtext?' + params sub_lang_list[lang] = url if not sub_lang_list: - self._downloader.report_warning(u'video doesn\'t have subtitles') + self._downloader.report_warning('video doesn\'t have subtitles') return {} return sub_lang_list @@ -518,9 +582,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): """We need the webpage for getting the captions url, pass it as an argument to speed up the process.""" sub_format = self._downloader.params.get('subtitlesformat', 'srt') - 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' % video_id + err_msg = 'Couldn\'t find automatic captions for %s' % video_id if mobj is None: self._downloader.report_warning(err_msg) return {} @@ -539,7 +603,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): caption_list = self._download_xml(list_url, video_id) original_lang_node = caption_list.find('track') if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr' : - self._downloader.report_warning(u'Video doesn\'t have automatic captions') + self._downloader.report_warning('Video doesn\'t have automatic captions') return {} original_lang = original_lang_node.attrib['lang_code'] @@ -565,7 +629,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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 @@ -576,7 +640,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): urls = filter(lambda l: l and not l.startswith('#'), lines) return urls - manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest') + 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') @@ -585,12 +649,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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=u'Searching for annotations.', errnote=u'Unable to download video annotations.') + return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.') def _real_extract(self, url): proto = ( - u'http' if self._downloader.params.get('prefer_insecure', False) - else u'https') + '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) @@ -600,6 +664,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # Get video webpage url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id + pref_cookies = [ + c for c in self._downloader.cookiejar + if c.domain == '.youtube.com' and c.name == 'PREF'] + for pc in pref_cookies: + if 'hl=' in pc.value: + pc.value = re.sub(r'hl=[^&]+', 'hl=en', pc.value) + else: + if pc.value: + pc.value += '&' + pc.value += 'hl=en' video_webpage = self._download_webpage(url, video_id) # Attempt to extract SWF player URL @@ -641,11 +715,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if 'token' not in video_info: if 'reason' in video_info: raise ExtractorError( - u'YouTube said: %s' % video_info['reason'][0], + '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', + '"token" parameter not in video info for unknown reason', video_id=video_id) if 'view_count' in video_info: @@ -655,14 +729,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # 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 @@ -671,14 +745,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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' in video_info: video_title = video_info['title'][0] else: - self._downloader.report_warning(u'Unable to extract video title') - video_title = u'_' + self._downloader.report_warning('Unable to extract video title') + video_title = '_' # thumbnail image # We try first to get a high quality image: @@ -687,7 +761,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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') + 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]) @@ -703,7 +777,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split()) upload_date = unified_strdate(upload_date) - m_cat_container = get_element_by_id("eow-category", video_webpage) + m_cat_container = self._search_regex( + r'(?s)]*>\s*Category\s*\s*]*>(.*?)', + video_webpage, 'categories', fatal=False) if m_cat_container: category = self._html_search_regex( r'(?s)(.*?)', m_cat_container, 'category', @@ -730,17 +806,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if fd_mobj: video_description = unescapeHTML(fd_mobj.group(1)) else: - video_description = u'' + video_description = '' - def _extract_count(klass): + def _extract_count(count_name): count = self._search_regex( - r'class="%s">([\d,]+)' % re.escape(klass), - video_webpage, klass, default=None) + r'id="watch-%s"[^>]*>.*?([\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(u'likes-count') - dislike_count = _extract_count(u'dislikes-count') + like_count = _extract_count('like') + dislike_count = _extract_count('dislike') # subtitles video_subtitles = self.extract_subtitles(video_id, video_webpage) @@ -750,7 +826,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): return if 'length_seconds' not in video_info: - self._downloader.report_warning(u'unable to extract video duration') + self._downloader.report_warning('unable to extract video duration') video_duration = None else: video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])) @@ -771,13 +847,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): # Easy way to know if the 's' value is in url_encoded_fmt_stream_map # this signatures are encrypted if 'url_encoded_fmt_stream_map' not in args: - raise ValueError(u'No stream_map present') # caught below + raise ValueError('No stream_map present') # caught below re_signature = re.compile(r'[&,]s=') m_s = re_signature.search(args['url_encoded_fmt_stream_map']) if m_s is not None: - self.to_screen(u'%s: Encrypted signatures detected.' % video_id) + self.to_screen('%s: Encrypted signatures detected.' % video_id) video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']] - m_s = re_signature.search(args.get('adaptive_fmts', u'')) + m_s = re_signature.search(args.get('adaptive_fmts', '')) if m_s is not None: if 'adaptive_fmts' in video_info: video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts'] @@ -827,12 +903,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if not age_gate: jsplayer_url_json = self._search_regex( r'"assets":.+?"js":\s*("[^"]+")', - video_webpage, u'JS player URL') + video_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, u'age gate player URL') + video_webpage, 'age gate player URL') player_url = json.loads(player_url_json) if self._downloader.params.get('verbose'): @@ -843,17 +919,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if player_url.endswith('swf'): player_version = self._search_regex( r'-(.+?)(?:/watch_as3)?\.swf$', player_url, - u'flash player', fatal=False) + 'flash player', fatal=False) player_desc = 'flash player %s' % player_version else: player_version = self._search_regex( r'html5player-([^/]+?)(?:/html5player)?\.js', player_url, 'html5 player', fatal=False) - player_desc = u'html5 player %s' % player_version + player_desc = 'html5 player %s' % player_version parts_sizes = self._signature_cache_id(encrypted_sig) - self.to_screen(u'{%s} signature length %s, %s' % + self.to_screen('{%s} signature length %s, %s' % (format_id, parts_sizes, player_desc)) signature = self._decrypt_signature( @@ -868,10 +944,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): url_map = self._extract_from_m3u8(manifest_url, video_id) formats = _map_to_format_list(url_map) else: - raise ExtractorError(u'no conn, hlsvp or url_encoded_fmt_stream_map information found in video info') + 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', False)): + if self._downloader.params.get('youtube_include_dash_manifest', True): try: # The DASH manifest used needs to be the one from the original video_webpage. # The one found in get_video_info seems to be using different signatures. @@ -889,9 +965,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url) dash_doc = self._download_xml( dash_manifest_url, video_id, - note=u'Downloading DASH manifest', - errnote=u'Could not download DASH manifest') - for r in dash_doc.findall(u'.//{urn:mpeg:DASH:schema:MPD:2011}Representation'): + note='Downloading DASH manifest', + errnote='Could not download DASH manifest') + 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 @@ -917,7 +993,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): existing_format.update(f) except (ExtractorError, KeyError) as e: - self.report_warning(u'Skipping DASH manifest: %s' % e, video_id) + self.report_warning('Skipping DASH manifest: %s' % e, video_id) self._sort_formats(formats) @@ -942,13 +1018,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): } class YoutubePlaylistIE(YoutubeBaseInfoExtractor): - IE_DESC = u'YouTube.com playlists' + 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/ ) @@ -964,27 +1040,86 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s' _MORE_PAGES_INDICATOR = r'data-link-type="next"' _VIDEO_RE = r'href="\s*/watch\?v=(?P[0-9A-Za-z_-]{11})&[^"]*?index=(?P\d+)' - IE_NAME = u'youtube:playlist' + IE_NAME = 'youtube:playlist' + _TESTS = [{ + 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re', + 'info_dict': { + 'title': 'ytdl test PL', + }, + 'playlist_count': 3, + }, { + 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx', + 'info_dict': { + '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', + }, + 'playlist_count': 95, + }, { + 'note': 'issue #673', + 'url': 'PLBB231211A4F62143', + 'info_dict': { + 'title': '[OLD]Team Fortress 2 (Class-based LP)', + }, + 'playlist_mincount': 26, + }, { + 'note': 'Large playlist', + 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', + 'info_dict': { + 'title': 'Uploads from Cauchemar', + }, + 'playlist_mincount': 799, + }, { + 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl', + 'info_dict': { + 'title': 'YDL_safe_search', + }, + 'playlist_count': 2, + }, { + 'note': 'embedded', + 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', + 'playlist_count': 4, + 'info_dict': { + 'title': 'JODA15', + } + }, { + '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', + } + }] 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] + 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 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, u'Downloading Youtube mix') + 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_span = ( + search_title('playlist-title') or + search_title('title long-title') or + search_title('title')) title = clean_html(title_span) - video_re = r'''(?x)data-video-username=".*?".*? - href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id) - ids = orderedSet(re.findall(video_re, webpage, flags=re.DOTALL)) + 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) @@ -993,7 +1128,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): # Extract playlist id mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError('Invalid URL: %s' % url) playlist_id = mobj.group(1) or mobj.group(2) # Check if it's a video-specific URL @@ -1001,17 +1136,17 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): if 'v' in query_dict: video_id = query_dict['v'][0] if self._downloader.params.get('noplaylist'): - self.to_screen(u'Downloading just video %s because of --no-playlist' % video_id) + 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(u'Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) + self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) if playlist_id.startswith('RD'): # Mixes require a custom extraction process return self._extract_mix(playlist_id) if playlist_id.startswith('TL'): - raise ExtractorError(u'For downloading YouTube.com top lists, use ' - u'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) + raise ExtractorError('For downloading YouTube.com top lists, use ' + 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True) url = self._TEMPLATE_URL % playlist_id page = self._download_webpage(url, playlist_id) @@ -1020,7 +1155,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): # 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( - u'The playlist doesn\'t exist or is private, use --username or ' + 'The playlist doesn\'t exist or is private, use --username or ' '--netrc to access it.', expected=True) @@ -1047,26 +1182,36 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): playlist_title = self._html_search_regex( r'(?s)

\s*(.*?)\s*

', - page, u'title') + page, 'title') url_results = self._ids_to_results(ids) return self.playlist_result(url_results, playlist_id, playlist_title) class YoutubeTopListIE(YoutubePlaylistIE): - IE_NAME = u'youtube:toplist' - IE_DESC = (u'YouTube.com top lists, "yttoplist:{channel}:{list title}"' - u' (Example: "yttoplist:music:Top Tracks")') + IE_NAME = 'youtube:toplist' + IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"' + ' (Example: "yttoplist:music:Top Tracks")') _VALID_URL = r'yttoplist:(?P.*?):(?P.*?)$' + _TESTS = [{ + 'url': 'yttoplist:music:Trending', + 'playlist_mincount': 5, + 'skip': 'Only works for logged-in users', + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) channel = mobj.group('chann') title = mobj.group('title') query = compat_urllib_parse.urlencode({'title': title}) - playlist_re = 'href="([^"]+?%s.*?)"' % re.escape(query) - channel_page = self._download_webpage('https://www.youtube.com/%s' % channel, title) - link = self._html_search_regex(playlist_re, channel_page, u'list') + channel_page = self._download_webpage( + 'https://www.youtube.com/%s' % channel, title) + link = self._html_search_regex( + r'''(?x) + <a\s+href="([^"]+)".*?>\s* + <span\s+class="branded-page-module-title-text">\s* + <span[^>]*>.*?%s.*?</span>''' % re.escape(query), + channel_page, 'list') url = compat_urlparse.urljoin('https://www.youtube.com/', link) video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"' @@ -1074,9 +1219,10 @@ class YoutubeTopListIE(YoutubePlaylistIE): # sometimes the webpage doesn't contain the videos # retry until we get them for i in itertools.count(0): - msg = u'Downloading Youtube mix' + msg = 'Downloading Youtube mix' if i > 0: msg += ', retry #%d' % i + webpage = self._download_webpage(url, title, msg) ids = orderedSet(re.findall(video_re, webpage)) if ids: @@ -1086,11 +1232,16 @@ class YoutubeTopListIE(YoutubePlaylistIE): class YoutubeChannelIE(InfoExtractor): - IE_DESC = u'YouTube.com channels' + IE_DESC = 'YouTube.com channels' _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)" _MORE_PAGES_INDICATOR = 'yt-uix-load-more' _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s' - IE_NAME = u'youtube:channel' + IE_NAME = 'youtube:channel' + _TESTS = [{ + 'note': 'paginated channel', + 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'playlist_mincount': 91, + }] def extract_videos_from_page(self, page): ids_in_page = [] @@ -1103,7 +1254,7 @@ class YoutubeChannelIE(InfoExtractor): # Extract channel id mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError('Invalid URL: %s' % url) # Download channel page channel_id = mobj.group(1) @@ -1125,7 +1276,7 @@ class YoutubeChannelIE(InfoExtractor): for pagenum in itertools.count(1): url = self._MORE_PAGES_URL % (pagenum, channel_id) page = self._download_json( - url, channel_id, note=u'Downloading page #%s' % pagenum, + url, channel_id, note='Downloading page #%s' % pagenum, transform_source=uppercase_escape) ids_in_page = self.extract_videos_from_page(page['content_html']) @@ -1134,7 +1285,7 @@ class YoutubeChannelIE(InfoExtractor): if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']: break - self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) + self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids))) url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id) for video_id in video_ids] @@ -1142,12 +1293,23 @@ class YoutubeChannelIE(InfoExtractor): class YoutubeUserIE(InfoExtractor): - IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)' + 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/)([A-Za-z0-9_-]+)' _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' _GDATA_PAGE_SIZE = 50 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json' - IE_NAME = u'youtube:user' + 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, + }] @classmethod def suitable(cls, url): @@ -1161,7 +1323,7 @@ class YoutubeUserIE(InfoExtractor): # Extract username mobj = re.match(self._VALID_URL, url) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError('Invalid URL: %s' % url) username = mobj.group(1) @@ -1176,13 +1338,13 @@ class YoutubeUserIE(InfoExtractor): 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' % ( + 'Downloading video ids from %d to %d' % ( start_index, start_index + self._GDATA_PAGE_SIZE)) try: response = json.loads(page) except ValueError as err: - raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err)) + raise ExtractorError('Invalid JSON in API response: ' + compat_str(err)) if 'entry' not in response['feed']: return @@ -1198,16 +1360,16 @@ class YoutubeUserIE(InfoExtractor): 'id': video_id, 'title': title, } - url_results = PagedList(download_page, self._GDATA_PAGE_SIZE) + url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE) return self.playlist_result(url_results, playlist_title=username) class YoutubeSearchIE(SearchInfoExtractor): - IE_DESC = u'YouTube.com searches' - _API_URL = u'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc' + 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 _get_n_results(self, query, n): @@ -1223,15 +1385,15 @@ class YoutubeSearchIE(SearchInfoExtractor): compat_urllib_parse.quote_plus(query.encode('utf-8')), (PAGE_SIZE * pagenum) + 1) data_json = self._download_webpage( - result_url, video_id=u'query "%s"' % query, - note=u'Downloading page %s' % (pagenum + 1), - errnote=u'Unable to download API page') + 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( - u'[youtube] No video results', expected=True) + '[youtube] No video results', expected=True) new_ids = list(video['id'] for video in api_response['items']) video_ids += new_ids @@ -1250,13 +1412,20 @@ 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 = u'YouTube.com searches, newest videos first' + IE_DESC = 'YouTube.com searches, newest videos first' class YoutubeSearchURLIE(InfoExtractor): - IE_DESC = u'YouTube.com search URLs' - IE_NAME = u'youtube:search_url' + IE_DESC = 'YouTube.com search URLs' + IE_NAME = 'youtube:search_url' _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)' + _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) @@ -1264,7 +1433,7 @@ class YoutubeSearchURLIE(InfoExtractor): webpage = self._download_webpage(url, query) result_code = self._search_regex( - r'(?s)<ol class="item-section"(.*?)</ol>', webpage, u'result HTML') + r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML') part_codes = re.findall( r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code) @@ -1290,18 +1459,39 @@ class YoutubeSearchURLIE(InfoExtractor): 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<id>[^?#]*)' + 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 YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): @@ -1323,7 +1513,7 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): @property def IE_NAME(self): - return u'youtube:%s' % self._FEED_NAME + return 'youtube:%s' % self._FEED_NAME def _real_initialize(self): self._login() @@ -1333,9 +1523,10 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): paging = 0 for i in itertools.count(1): info = self._download_json(self._FEED_TEMPLATE % paging, - u'%s feed' % self._FEED_NAME, - u'Downloading page %s' % i) + '%s feed' % self._FEED_NAME, + 'Downloading page %s' % i) 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( @@ -1343,50 +1534,82 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): for video_id in ids) mobj = re.search( r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)', - feed_html) + load_more_widget_html) if mobj is None: break paging = mobj.group('paging') return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE) -class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): - IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' - _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' - _FEED_NAME = 'subscriptions' - _PLAYLIST_TITLE = u'Youtube Subscriptions' - class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): - IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' + IE_DESC = 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)' _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' _FEED_NAME = 'recommended' - _PLAYLIST_TITLE = u'Youtube Recommended videos' + _PLAYLIST_TITLE = 'Youtube Recommended videos' class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor): - IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' + IE_DESC = 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)' _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater' _FEED_NAME = 'watch_later' - _PLAYLIST_TITLE = u'Youtube Watch Later' + _PLAYLIST_TITLE = 'Youtube Watch Later' _PERSONAL_FEED = True class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): - IE_DESC = u'Youtube watch history, "ythistory" keyword (requires authentication)' - _VALID_URL = u'https?://www\.youtube\.com/feed/history|:ythistory' + IE_DESC = 'Youtube watch history, "ythistory" keyword (requires authentication)' + _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory' _FEED_NAME = 'history' _PERSONAL_FEED = True - _PLAYLIST_TITLE = u'Youtube Watch History' + _PLAYLIST_TITLE = 'Youtube Watch History' class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): - IE_NAME = u'youtube:favorites' - IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)' + IE_NAME = 'youtube:favorites' + IE_DESC = 'YouTube.com favourite videos, "ytfav" keyword (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, u'favourites playlist id') + 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>[^"]+)"', 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 @@ -1408,9 +1631,9 @@ class YoutubeTruncatedURLIE(InfoExtractor): def _real_extract(self, url): raise ExtractorError( - u'Did you forget to quote the URL? Remember that & is a meta ' - u'character in most shells, so you want to put the URL in quotes, ' - u'like youtube-dl ' - u'"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' - u' or simply youtube-dl BaW_jenozKc .', + '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)