X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvk.py;h=cd22df25a2623e63ccbdabe791e2c497ac09c656;hb=89f257d6e57131a266efae629334fe5f4bcf96e9;hp=01960b827254a3190fb463919f9f8ce2efead676;hpb=a2973eb59733c5f86a249c627d654b789020bc7d;p=youtube-dl diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 01960b827..cd22df25a 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -1,41 +1,100 @@ # encoding: utf-8 from __future__ import unicode_literals +import collections import re import json +import sys from .common import InfoExtractor from ..compat import ( compat_str, - compat_urllib_parse, - compat_urllib_request, + compat_urlparse, ) from ..utils import ( + clean_html, ExtractorError, + get_element_by_class, + int_or_none, orderedSet, + remove_start, str_to_int, unescapeHTML, unified_strdate, + urlencode_postdata, ) from .vimeo import VimeoIE +from .pladform import PladformIE -class VKIE(InfoExtractor): +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'), + }) + + # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header + # and expects the first one to be set rather than second (see + # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201). + # As of RFC6265 the newer one cookie should be set into cookie store + # what actually happens. + # We will workaround this VK issue by resetting the remixlhk cookie to + # the first one manually. + for header, cookies in url_handle.headers.items(): + if header.lower() != 'set-cookie': + continue + if sys.version_info[0] >= 3: + cookies = cookies.encode('iso-8859-1') + cookies = cookies.decode('utf-8') + remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies) + if remixlhk: + value, domain = remixlhk.groups() + self._set_cookie(domain, 'remixlhk', value) + break + + login_page = self._download_webpage( + 'https://login.vk.com/?act=login', None, + note='Logging in as %s' % username, + data=urlencode_postdata(login_form)) + + if re.search(r'onLoginFailed', login_page): + raise ExtractorError( + 'Unable to login, incorrect username and/or password', expected=True) + + def _real_initialize(self): + self._login() + + +class VKIE(VKBaseIE): IE_NAME = 'vk' IE_DESC = 'VK' _VALID_URL = r'''(?x) https?:// (?: - (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P-?\d+).*?\bid=(?P\d+)| (?: - (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video| - (?:www\.)?biqle\.ru/watch/ + (?:(?:m|new)\.)?vk\.com/video_| + (?:www\.)?daxab.com/ + ) + ext\.php\?(?P.*?\boid=(?P-?\d+).*?\bid=(?P\d+).*)| + (?: + (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video| + (?:www\.)?daxab.com/embed/ ) - (?P[^s].*?)(?:\?(?:.*\blist=(?P[\da-f]+))?|%2F|$) + (?P-?\d+_\d+)(?:.*\blist=(?P[\da-f]+))? ) ''' - _NETRC_MACHINE = 'vk' - _TESTS = [ { 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', @@ -75,7 +134,8 @@ class VKIE(InfoExtractor): 'duration': 101, 'upload_date': '20120730', 'view_count': int, - } + }, + 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -140,16 +200,29 @@ class VKIE(InfoExtractor): 'url': 'https://vk.com/video276849682_170681728', 'info_dict': { 'id': 'V3K4mi0SYkc', - 'ext': 'mp4', + 'ext': 'webm', 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", - 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', - 'duration': 179, + 'description': 'md5:d9903938abdc74c738af77f527ca0596', + 'duration': 178, 'upload_date': '20130116', 'uploader': "Children's Joy Foundation", 'uploader_id': 'thecjf', 'view_count': int, }, }, + { + # video key is extra_data not url\d+ + 'url': 'http://vk.com/video-110305615_171782105', + 'md5': 'e13fcda136f99764872e739d13fac1d1', + 'info_dict': { + 'id': '171782105', + 'ext': 'mp4', + 'title': 'S-Dance, репетиции к The way show', + 'uploader': 'THE WAY SHOW | 17 апреля', + 'upload_date': '20160207', + 'view_count': int, + }, + }, { # removed video, just testing that we match the pattern 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a', @@ -161,58 +234,35 @@ class VKIE(InfoExtractor): 'only_matching': True, }, { - # vk wrapper - 'url': 'http://www.biqle.ru/watch/847655_160197695', + # pladform embed + 'url': 'https://vk.com/video-76116461_171554880', + 'only_matching': True, + }, + { + 'url': 'http://new.vk.com/video205387401_165548505', 'only_matching': True, } ] - def _login(self): - (username, password) = self._get_login_info() - if username is None: - return - - login_page = self._download_webpage( - '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'), - }) - - 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() - 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&al=1&module=video&video=%s' % 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 = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % 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 - info_page = self._download_webpage(info_url, video_id) error_message = self._html_search_regex( - r'(?s)]+class="video_layer_message"[^>]*>(.+?)', + [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) @@ -254,10 +304,13 @@ class VKIE(InfoExtractor): 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) @@ -284,17 +337,22 @@ class VKIE(InfoExtractor): view_count = None views = self._html_search_regex( r'"mv_views_count_number"[^>]*>(.+?\bviews?)<', - info_page, 'view count', fatal=False) + info_page, 'view count', default=None) if views: view_count = str_to_int(self._search_regex( r'([\d,.]+)', views, 'view count', fatal=False)) - formats = [{ - 'format_id': k, - 'url': v, - 'width': int(k[len('url'):]), - } for k, v in data.items() - if k.startswith('url')] + formats = [] + for k, v in data.items(): + if not k.startswith('url') and not k.startswith('cache') and k != 'extra_data' or not v: + continue + height = int_or_none(self._search_regex( + r'^(?:url|cache)(\d+)', k, 'height', default=None)) + formats.append({ + 'format_id': k, + 'url': v, + 'height': height, + }) self._sort_formats(formats) return { @@ -309,10 +367,10 @@ class VKIE(InfoExtractor): } -class VKUserVideosIE(InfoExtractor): +class VKUserVideosIE(VKBaseIE): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://vk\.com/videos(?P-?[0-9]+)$' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ 'url': 'http://vk.com/videos205387401', @@ -324,6 +382,15 @@ class VKUserVideosIE(InfoExtractor): }, { 'url': 'http://vk.com/videos-77521', 'only_matching': True, + }, { + 'url': 'http://vk.com/videos-97664626?section=all', + 'only_matching': True, + }, { + 'url': 'http://m.vk.com/videos205387401', + 'only_matching': True, + }, { + 'url': 'http://new.vk.com/videos205387401', + 'only_matching': True, }] def _real_extract(self, url): @@ -341,3 +408,131 @@ class VKUserVideosIE(InfoExtractor): webpage, 'title', default=page_id)) return self.playlist_result(entries, page_id, title) + + +class VKWallPostIE(VKBaseIE): + IE_NAME = 'vk:wallpost' + _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P-?\d+_\d+)))' + _TESTS = [{ + # public page URL, audio playlist + 'url': 'https://vk.com/bs.official?w=wall-23538238_35', + 'info_dict': { + 'id': '23538238_35', + 'title': 'Black Shadow - Wall post 23538238_35', + 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c', + }, + 'playlist': [{ + 'md5': '5ba93864ec5b85f7ce19a9af4af080f6', + 'info_dict': { + 'id': '135220665_111806521', + 'ext': 'mp3', + 'title': 'Black Shadow - Слепое Верование', + 'duration': 370, + 'uploader': 'Black Shadow', + 'artist': 'Black Shadow', + 'track': 'Слепое Верование', + }, + }, { + 'md5': '4cc7e804579122b17ea95af7834c9233', + 'info_dict': { + 'id': '135220665_111802303', + 'ext': 'mp3', + 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!', + 'duration': 423, + 'uploader': 'Black Shadow', + 'artist': 'Black Shadow', + 'track': 'Война - Негасимое Бездны Пламя!', + }, + 'params': { + 'skip_download': True, + }, + }], + 'params': { + 'usenetrc': True, + }, + 'skip': 'Requires vk account credentials', + }, { + # single YouTube embed, no leading - + 'url': 'https://vk.com/wall85155021_6319', + 'info_dict': { + 'id': '85155021_6319', + 'title': 'Sergey Gorbunov - Wall post 85155021_6319', + }, + 'playlist_count': 1, + 'params': { + 'usenetrc': True, + }, + 'skip': 'Requires vk account credentials', + }, { + # wall page URL + 'url': 'https://vk.com/wall-23538238_35', + 'only_matching': True, + }, { + # mobile wall page URL + 'url': 'https://m.vk.com/wall-23538238_35', + 'only_matching': True, + }] + + def _real_extract(self, url): + post_id = self._match_id(url) + + wall_url = 'https://vk.com/wall%s' % post_id + + post_id = remove_start(post_id, '-') + + webpage = self._download_webpage(wall_url, post_id) + + error = self._html_search_regex( + r'>Error\s*]+class=["\']body["\'][^>]*>([^<]+)', + webpage, 'error', default=None) + if error: + raise ExtractorError('VK said: %s' % error, expected=True) + + description = clean_html(get_element_by_class('wall_post_text', webpage)) + uploader = clean_html(get_element_by_class('author', webpage)) + thumbnail = self._og_search_thumbnail(webpage) + + entries = [] + + audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage) + if audio_ids: + al_audio = self._download_webpage( + 'https://vk.com/al_audio.php', post_id, + note='Downloading audio info', fatal=False, + data=urlencode_postdata({ + 'act': 'reload_audio', + 'al': '1', + 'ids': ','.join(audio_ids) + })) + if al_audio: + Audio = collections.namedtuple( + 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration']) + audios = self._parse_json( + self._search_regex( + r'(.+?)', al_audio, 'audios', default='[]'), + post_id, fatal=False, transform_source=unescapeHTML) + if isinstance(audios, list): + for audio in audios: + a = Audio._make(audio[:6]) + entries.append({ + 'id': '%s_%s' % (a.user_id, a.id), + 'url': a.url, + 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id, + 'thumbnail': thumbnail, + 'duration': a.duration, + 'uploader': uploader, + 'artist': a.artist, + 'track': a.track, + }) + + for video in re.finditer( + r']+href=(["\'])(?P/video(?:-?[\d_]+).*?)\1', webpage): + entries.append(self.url_result( + compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key())) + + title = 'Wall post %s' % post_id + + return self.playlist_result( + orderedSet(entries), post_id, + '%s - %s' % (uploader, title) if uploader else title, + description)