X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fyoutube.py;h=e28db2b5a57c7208ac61ab90a026b0d26e050ba5;hb=9b583dca4cf3b623323de8fadf6dc851b7111fd2;hp=072e711c2e0105c3a69908cff70bf3758542587d;hpb=246168bd72a8f031adb243e9465ef52f62fb502c;p=youtube-dl diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 072e711c2..e28db2b5a 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1,7 +1,5 @@ # coding: utf-8 -import errno -import io import itertools import json import os.path @@ -21,7 +19,6 @@ from ..utils import ( compat_str, clean_html, - get_cachedir, get_element_by_id, get_element_by_attribute, ExtractorError, @@ -30,13 +27,13 @@ from ..utils import ( 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' @@ -50,12 +47,19 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 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 + return True login_page = self._download_webpage( self._LOGIN_URL, None, @@ -73,6 +77,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): u'Email': username, u'GALX': galx, u'Passwd': password, + u'PersistentCookie': u'yes', u'_utf8': u'霱', u'bgresponse': u'js_disabled', @@ -88,6 +93,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): u'uilel': u'3', u'hl': u'en_US', } + # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode # chokes on unicode login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) @@ -99,6 +105,68 @@ class YoutubeBaseInfoExtractor(InfoExtractor): note=u'Logging in', errnote=u'unable to log in', fatal=False) if login_results is False: return False + + if re.search(r'id="errormsg_0_Passwd"', login_results) is not None: + raise ExtractorError(u'Please use your account password and a two-factor code instead of an application-specific password.', expected=True) + + # Two-Factor + # TODO add SMS and phone call support - these require making a request and then prompting the user + + if re.search(r'(?i)]* id="gaia_secondfactorform"', login_results) is not None: + tfa_code = self._get_tfa_info() + + if tfa_code is None: + self._downloader.report_warning(u'Two-factor authentication required. Provide it with --twofactor ') + self._downloader.report_warning(u'(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(u'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(u'Failed to get timeStmp - did the page structure change?') + timeStmp = match.group(1) + + tfa_form_strs = { + u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1', + u'smsToken': u'', + u'smsUserPin': tfa_code, + u'smsVerifyPin': u'Verify', + + u'PersistentCookie': u'yes', + u'checkConnection': u'', + u'checkedDomains': u'youtube', + u'pstMsg': u'1', + u'secTok': secTok, + u'timeStmp': timeStmp, + u'service': u'youtube', + u'hl': u'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=u'Submitting TFA code', errnote=u'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(u'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(u'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(u'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') return False @@ -131,7 +199,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): IE_DESC = u'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/| @@ -149,7 +217,7 @@ 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 @@ -225,7 +293,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # 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}, # RTMP (unnamed) @@ -244,6 +312,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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'], + 'like_count': int, + 'dislike_count': int, } }, { @@ -344,29 +414,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): """Indicate the download will use the RTMP protocol.""" self.to_screen(u'RTMP download detected') - def _extract_signature_function(self, video_id, player_url, slen): + 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('.')) + + def _extract_signature_function(self, video_id, player_url, example_sig): id_m = re.match( - r'.*-(?P[a-zA-Z0-9_-]+)(?:/watch_as3)?\.(?P[a-z]+)$', + r'.*-(?P[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P[a-z]+)$', 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_%d' % (player_type, player_id, slen) + 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(u'youtube-sigfuncs', func_id) + if cache_spec is not None: + return lambda s: u''.join(s[i] for i in cache_spec) if player_type == 'js': code = self._download_webpage( @@ -384,25 +452,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: assert False, 'Invalid player type %r' % player_type - if cache_enabled: - try: - test_string = u''.join(map(compat_chr, range(slen))) - 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 = u''.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(u'youtube-sigfuncs', func_id, cache_spec) return res - def _print_sig_code(self, func, slen): + 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) @@ -431,11 +489,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: yield _genslice(start, i, step) - test_string = u''.join(map(compat_chr, range(slen))) + test_string = u''.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)) - code = u'if len(s) == %d:\n return %s\n' % (slen, expr_code) + 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) def _parse_sig_js(self, jscode): @@ -463,20 +524,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): if player_url.startswith(u'//'): player_url = u'https:' + player_url try: - player_id = (player_url, len(s)) + 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, len(s) + 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, len(s)) + self._print_sig_code(func, s) return func(s) except Exception as e: tb = traceback.format_exc() raise ExtractorError( - u'Automatic signature extraction failed: ' + tb, cause=e) + u'Signature extraction failed: ' + tb, cause=e) def _get_available_subtitles(self, video_id, webpage): try: @@ -491,6 +552,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): 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, @@ -609,7 +672,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): data = compat_urllib_parse.urlencode({ 'video_id': video_id, 'eurl': 'https://youtube.googleapis.com/v/' + video_id, - 'sts':'16268', + 'sts': self._search_regex( + r'"sts"\s*:\s*(\d+)', video_webpage, 'sts'), }) video_info_url = proto + '://www.youtube.com/get_video_info?' + data video_info_webpage = self._download_webpage(video_info_url, video_id, @@ -692,7 +756,9 @@ 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', @@ -721,15 +787,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): else: video_description = u'' - 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(u'like') + dislike_count = _extract_count(u'dislike') # subtitles video_subtitles = self.extract_subtitles(video_id, video_webpage) @@ -803,50 +869,54 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor): url_map = {} 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: - encrypted_sig = url_data['s'][0] - - if not age_gate: - jsplayer_url_json = self._search_regex( - r'"assets":.+?"js":\s*("[^"]+")', - video_webpage, u'JS player URL') - player_url = json.loads(jsplayer_url_json) + 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] + + if not age_gate: + jsplayer_url_json = self._search_regex( + r'"assets":.+?"js":\s*("[^"]+")', + video_webpage, u'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') + player_url = json.loads(player_url_json) + + if self._downloader.params.get('verbose'): if player_url is None: - player_url_json = self._search_regex( - r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', - video_webpage, u'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' + player_version = 'unknown' + player_desc = 'unknown' + else: + if player_url.endswith('swf'): + player_version = self._search_regex( + r'-(.+?)(?:/watch_as3)?\.swf$', player_url, + u'flash player', fatal=False) + player_desc = 'flash player %s' % player_version else: - if player_url.endswith('swf'): - player_version = self._search_regex( - r'-(.+?)(?:/watch_as3)?\.swf$', player_url, - u'flash player', fatal=False) - player_desc = 'flash player %s' % player_version - else: - player_version = self._search_regex( - r'html5player-(.+?)\.js', video_webpage, - 'html5 player', fatal=False) - player_desc = u'html5 player %s' % player_version - - parts_sizes = u'.'.join(compat_str(len(part)) for part in encrypted_sig.split('.')) - self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' % - (len(encrypted_sig), parts_sizes, url_data['itag'][0], 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[url_data['itag'][0]] = url + player_version = self._search_regex( + r'html5player-([^/]+?)(?:/html5player)?\.js', + player_url, + 'html5 player', fatal=False) + player_desc = u'html5 player %s' % player_version + + parts_sizes = self._signature_cache_id(encrypted_sig) + self.to_screen(u'{%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] @@ -955,21 +1025,26 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor): 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, u'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) @@ -1062,6 +1137,7 @@ class YoutubeTopListIE(YoutubePlaylistIE): msg = u'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: @@ -1321,6 +1397,7 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): u'%s feed' % self._FEED_NAME, u'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( @@ -1328,18 +1405,12 @@ class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor): for video_id in ids) mobj = re.search( r'data-uix-load-more-href="/?[^"]+paging=(?P\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)' _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?' @@ -1372,6 +1443,43 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): return self.url_result(playlist_id, 'YoutubePlaylist') +class YoutubeSubscriptionsIE(YoutubePlaylistIE): + IE_NAME = u'youtube:subscriptions' + IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' + _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?' + + def _real_extract(self, url): + title = u'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