X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/f46044c66663049e286c20ee015db99d47d9dd8a..7537b0e3b51e7491205fd2198e8b8b9ae334e50d:/youtube_dl/extractor/vk.py diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 7dea8c5..8b6dc0e 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -1,38 +1,96 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals +import collections import re -import json from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urllib_parse, - compat_urllib_request, -) +from ..compat import compat_urlparse from ..utils import ( + clean_html, ExtractorError, + get_element_by_class, + int_or_none, orderedSet, + remove_start, + str_or_none, + str_to_int, unescapeHTML, - unified_strdate, + unified_timestamp, + url_or_none, + urlencode_postdata, ) +from .dailymotion import DailymotionIE +from .pladform import PladformIE +from .vimeo import VimeoIE +from .youtube import YoutubeIE -class VKIE(InfoExtractor): - IE_NAME = 'vk.com' - _VALID_URL = r'https?://(?:m\.)?vk\.com/(?:video_ext\.php\?.*?\boid=(?P-?\d+).*?\bid=(?P\d+)|(?:.+?\?.*?z=)?video(?P[^s].*?)(?:\?|%2F|$))' +class VKBaseIE(InfoExtractor): _NETRC_MACHINE = 'vk' + def _login(self): + username, password = self._get_login_info() + if username is None: + return + + login_page, url_handle = self._download_webpage_handle( + 'https://vk.com', None, 'Downloading login page') + + login_form = self._hidden_inputs(login_page) + + login_form.update({ + 'email': username.encode('cp1251'), + 'pass': password.encode('cp1251'), + }) + + # vk serves two same remixlhk cookies in Set-Cookie header and expects + # first one to be actually set + self._apply_first_set_cookie_header(url_handle, 'remixlhk') + + login_page = self._download_webpage( + 'https://login.vk.com/?act=login', None, + note='Logging in', + data=urlencode_postdata(login_form)) + + if re.search(r'onLoginFailed', login_page): + raise ExtractorError( + 'Unable to login, incorrect username and/or password', expected=True) + + def _real_initialize(self): + self._login() + + +class VKIE(VKBaseIE): + IE_NAME = 'vk' + IE_DESC = 'VK' + _VALID_URL = r'''(?x) + https?:// + (?: + (?: + (?:(?:m|new)\.)?vk\.com/video_| + (?:www\.)?daxab.com/ + ) + ext\.php\?(?P.*?\boid=(?P-?\d+).*?\bid=(?P\d+).*)| + (?: + (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video| + (?:www\.)?daxab.com/embed/ + ) + (?P-?\d+_\d+)(?:.*\blist=(?P[\da-f]+))? + ) + ''' _TESTS = [ { 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', - 'md5': '0deae91935c54e00003c2a00646315f0', + 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { - 'id': '162222515', - 'ext': 'flv', + 'id': '-77521_162222515', + 'ext': 'mp4', 'title': 'ProtivoGunz - Хуёвая песня', - 'uploader': 're:Noize MC.*', + 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', + 'uploader_id': '-77521', 'duration': 195, + 'timestamp': 1329049880, 'upload_date': '20120212', }, }, @@ -40,12 +98,14 @@ class VKIE(InfoExtractor): 'url': 'http://vk.com/video205387401_165548505', 'md5': '6c0aeb2e90396ba97035b9cbde548700', 'info_dict': { - 'id': '165548505', + 'id': '205387401_165548505', 'ext': 'mp4', - 'uploader': 'Tom Cruise', 'title': 'No name', + 'uploader': 'Tom Cruise', + 'uploader_id': '205387401', 'duration': 9, - 'upload_date': '20130721' + 'timestamp': 1374364108, + 'upload_date': '20130720', } }, { @@ -53,13 +113,15 @@ class VKIE(InfoExtractor): 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1', 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a', 'info_dict': { - 'id': '162925554', + 'id': '32194266_162925554', 'ext': 'mp4', 'uploader': 'Vladimir Gavrin', 'title': 'Lin Dan', 'duration': 101, 'upload_date': '20120730', - } + 'view_count': int, + }, + 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -68,12 +130,13 @@ class VKIE(InfoExtractor): 'md5': 'a590bcaf3d543576c9bd162812387666', 'note': 'Only available for registered users', 'info_dict': { - 'id': '164049491', + 'id': '-8871596_164049491', 'ext': 'mp4', 'uploader': 'Триллеры', 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]', 'duration': 8352, - 'upload_date': '20121218' + 'upload_date': '20121218', + 'view_count': int, }, 'skip': 'Requires vk account credentials', }, @@ -81,7 +144,7 @@ class VKIE(InfoExtractor): 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d', 'md5': '4d7a5ef8cf114dfa09577e57b2993202', 'info_dict': { - 'id': '168067957', + 'id': '-43215063_168067957', 'ext': 'mp4', 'uploader': 'Киномания - лучшее из мира кино', 'title': ' ', @@ -95,141 +158,487 @@ class VKIE(InfoExtractor): 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6', 'note': 'ivi.ru embed', 'info_dict': { - 'id': '60690', + 'id': '-43215063_169084319', 'ext': 'mp4', 'title': 'Книга Илая', 'duration': 6771, 'upload_date': '20140626', + 'view_count': int, }, 'skip': 'Only works from Russia', }, + { + # video (removed?) only available with list id + 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4', + 'md5': '091287af5402239a1051c37ec7b92913', + 'info_dict': { + 'id': '30481095_171201961', + 'ext': 'mp4', + 'title': 'ТюменцевВВ_09.07.2015', + 'uploader': 'Anton Ivanov', + 'duration': 109, + 'upload_date': '20150709', + 'view_count': int, + }, + 'skip': 'Removed', + }, + { + # youtube embed + 'url': 'https://vk.com/video276849682_170681728', + 'info_dict': { + 'id': 'V3K4mi0SYkc', + 'ext': 'mp4', + 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", + 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', + 'duration': 178, + 'upload_date': '20130116', + 'uploader': "Children's Joy Foundation Inc.", + 'uploader_id': 'thecjf', + 'view_count': int, + }, + }, + { + # dailymotion embed + 'url': 'https://vk.com/video-37468416_456239855', + 'info_dict': { + 'id': 'k3lz2cmXyRuJQSjGHUv', + 'ext': 'mp4', + 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f', + # TODO: fix test by fixing dailymotion description extraction + 'description': 'md5:c651358f03c56f1150b555c26d90a0fd', + 'uploader': 'AniLibria.Tv', + 'upload_date': '20160914', + 'uploader_id': 'x1p5vl5', + 'timestamp': 1473877246, + }, + 'params': { + 'skip_download': True, + }, + }, + { + # video key is extra_data not url\d+ + 'url': 'http://vk.com/video-110305615_171782105', + 'md5': 'e13fcda136f99764872e739d13fac1d1', + 'info_dict': { + 'id': '-110305615_171782105', + 'ext': 'mp4', + 'title': 'S-Dance, репетиции к The way show', + 'uploader': 'THE WAY SHOW | 17 апреля', + 'uploader_id': '-110305615', + 'timestamp': 1454859345, + 'upload_date': '20160207', + }, + 'params': { + 'skip_download': True, + }, + }, + { + # finished live stream, postlive_mp4 + 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2', + 'info_dict': { + 'id': '-387766_456242764', + 'ext': 'mp4', + 'title': 'ИгроМир 2016 День 1 — Игромания Утром', + 'uploader': 'Игромания', + 'duration': 5239, + # TODO: use act=show to extract view_count + # 'view_count': int, + 'upload_date': '20160929', + 'uploader_id': '-387766', + 'timestamp': 1475137527, + }, + }, + { + # live stream, hls and rtmp links, most likely already finished live + # stream by the time you are reading this comment + 'url': 'https://vk.com/video-140332_456239111', + 'only_matching': True, + }, { # removed video, just testing that we match the pattern 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a', 'only_matching': True, }, - ] - - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - - login_form = { - 'act': 'login', - 'role': 'al_frame', - 'expire': '1', - 'email': username, - 'pass': password, - } - - request = compat_urllib_request.Request('https://login.vk.com/?act=login', - compat_urllib_parse.urlencode(login_form).encode('utf-8')) - login_page = self._download_webpage(request, None, note='Logging in as %s' % username) - - if re.search(r'onLoginFailed', login_page): - raise ExtractorError('Unable to login, incorrect username and/or password', expected=True) - - def _real_initialize(self): - self._login() + { + # age restricted video, requires vk account credentials + 'url': 'https://vk.com/video205387401_164765225', + 'only_matching': True, + }, + { + # pladform embed + 'url': 'https://vk.com/video-76116461_171554880', + 'only_matching': True, + }, + { + 'url': 'http://new.vk.com/video205387401_165548505', + 'only_matching': True, + }, + { + # This video is no longer available, because its author has been blocked. + 'url': 'https://vk.com/video-10639516_456240611', + 'only_matching': True, + }, + { + # The video is not available in your region. + 'url': 'https://vk.com/video-51812607_171445436', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') - if not video_id: + if video_id: + info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id + # Some videos (removed?) can only be downloaded with list id specified + list_id = mobj.group('list_id') + if list_id: + info_url += '&list=%s' % list_id + else: + info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query') video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id')) - info_url = 'http://vk.com/al_video.php?act=show&al=1&video=%s' % video_id info_page = self._download_webpage(info_url, video_id) + error_message = self._html_search_regex( + [r'(?s)]+class="video_layer_message"[^>]*>(.+?)', + r'(?s)]+id="video_ext_msg"[^>]*>(.+?)'], + info_page, 'error message', default=None) + if error_message: + raise ExtractorError(error_message, expected=True) + + if re.search(r'/login\.php\?.*\bact=security_check', info_page): + raise ExtractorError( + 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', + expected=True) + + ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.' + ERRORS = { r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': - 'Video %s has been removed from public access due to rightholder complaint.', + ERROR_COPYRIGHT, + + r'>The video .*? was removed from public access by request of the copyright holder.<': + ERROR_COPYRIGHT, r'Please log in or <': 'Video %s is only available for registered users, ' 'use --username and --password options to provide account credentials.', r'Unknown error': - 'Video %s does not exist.' + 'Video %s does not exist.', + + r'Видео временно недоступно': + 'Video %s is temporarily unavailable.', + + r'Access denied': + 'Access denied to video %s.', + + r'Видеозапись недоступна, так как её автор был заблокирован.': + 'Video %s is no longer available, because its author has been blocked.', + + r'This video is no longer available, because its author has been blocked.': + 'Video %s is no longer available, because its author has been blocked.', + + r'This video is no longer available, because it has been deleted.': + 'Video %s is no longer available, because it has been deleted.', + + r'The video .+? is not available in your region.': + 'Video %s is not available in your region.', } for error_re, error_msg in ERRORS.items(): if re.search(error_re, info_page): raise ExtractorError(error_msg % video_id, expected=True) - m_yt = re.search(r'src="(http://www.youtube.com/.*?)"', info_page) - if m_yt is not None: - self.to_screen('Youtube video detected') - return self.url_result(m_yt.group(1), 'Youtube') + youtube_url = YoutubeIE._extract_url(info_page) + if youtube_url: + return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) + + vimeo_url = VimeoIE._extract_url(url, info_page) + if vimeo_url is not None: + return self.url_result(vimeo_url) + + pladform_url = PladformIE._extract_url(info_page) + if pladform_url: + return self.url_result(pladform_url) m_rutube = re.search( - r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page) + r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page) if m_rutube is not None: - self.to_screen('rutube video detected') rutube_url = self._proto_relative_url( m_rutube.group(1).replace('\\', '')) return self.url_result(rutube_url) - m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.*?});', info_page) + dailymotion_urls = DailymotionIE._extract_urls(info_page) + if dailymotion_urls: + return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key()) + + m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page) if m_opts: - m_opts_url = re.search(r"url\s*:\s*'([^']+)", m_opts.group(1)) + m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1)) if m_opts_url: opts_url = m_opts_url.group(1) if opts_url.startswith('//'): opts_url = 'http:' + opts_url return self.url_result(opts_url) - data_json = self._search_regex(r'var vars = ({.*?});', info_page, 'vars') - data = json.loads(data_json) - - # Extract upload date - upload_date = None - mobj = re.search(r'id="mv_date_wrap".*?Added ([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page) - if mobj is not None: - mobj.group(1) + ' ' + mobj.group(2) - upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2)) - - formats = [{ - 'format_id': k, - 'url': v, - 'width': int(k[len('url'):]), - } for k, v in data.items() - if k.startswith('url')] + # vars does not look to be served anymore since 24.10.2016 + data = self._parse_json( + self._search_regex( + r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'), + video_id, fatal=False) + + # is served instead + if not data: + data = self._parse_json( + self._search_regex( + [r'\s*({.+?})\s*', r'\s*({.+})'], + info_page, 'json', default='{}'), + video_id) + if data: + data = data['player']['params'][0] + + if not data: + data = self._parse_json( + self._search_regex( + r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page, + 'player params', default='{}'), + video_id) + if data: + data = data['params'][0] + + #