From 3c989818e7dc7706da069312bbdd040165a97517 Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Fri, 25 Oct 2019 19:35:07 +0100 Subject: [PATCH] [vk] improve extraction - add support for Odnoklassniki embeds - update tests - extract more video from user lists(closes #4470) - fix wall post audio extraction(closes #18332) - improve error detection(closes #22568) --- youtube_dl/extractor/vk.py | 329 +++++++++++++++++++------------------ 1 file changed, 173 insertions(+), 156 deletions(-) diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 8b6dc0e24..c289fcad3 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -12,7 +12,6 @@ get_element_by_class, int_or_none, orderedSet, - remove_start, str_or_none, str_to_int, unescapeHTML, @@ -21,6 +20,7 @@ urlencode_postdata, ) from .dailymotion import DailymotionIE +from .odnoklassniki import OdnoklassnikiIE from .pladform import PladformIE from .vimeo import VimeoIE from .youtube import YoutubeIE @@ -60,6 +60,18 @@ def _login(self): def _real_initialize(self): self._login() + def _download_payload(self, path, video_id, data, fatal=True): + data['al'] = 1 + code, payload = self._download_json( + 'https://vk.com/%s.php' % path, video_id, + data=urlencode_postdata(data), fatal=fatal, + headers={'X-Requested-With': 'XMLHttpRequest'})['payload'] + if code == '3': + self.raise_login_required() + elif code == '8': + raise ExtractorError(clean_html(payload[0][1:-1]), expected=True) + return payload + class VKIE(VKBaseIE): IE_NAME = 'vk' @@ -96,7 +108,6 @@ class VKIE(VKBaseIE): }, { 'url': 'http://vk.com/video205387401_165548505', - 'md5': '6c0aeb2e90396ba97035b9cbde548700', 'info_dict': { 'id': '205387401_165548505', 'ext': 'mp4', @@ -110,18 +121,18 @@ class VKIE(VKBaseIE): }, { 'note': 'Embedded video', - 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1', - 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a', + 'url': 'https://vk.com/video_ext.php?oid=-77521&id=162222515&hash=87b046504ccd8bfa', + 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { - 'id': '32194266_162925554', + 'id': '-77521_162222515', 'ext': 'mp4', - 'uploader': 'Vladimir Gavrin', - 'title': 'Lin Dan', - 'duration': 101, - 'upload_date': '20120730', - 'view_count': int, + 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', + 'title': 'ProtivoGunz - Хуёвая песня', + 'duration': 195, + 'upload_date': '20120212', + 'timestamp': 1329049880, + 'uploader_id': '-77521', }, - 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -138,18 +149,19 @@ class VKIE(VKBaseIE): 'upload_date': '20121218', 'view_count': int, }, - 'skip': 'Requires vk account credentials', + 'skip': 'Removed', }, { 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d', - 'md5': '4d7a5ef8cf114dfa09577e57b2993202', 'info_dict': { 'id': '-43215063_168067957', 'ext': 'mp4', - 'uploader': 'Киномания - лучшее из мира кино', + 'uploader': 'Bro Mazter', 'title': ' ', 'duration': 7291, 'upload_date': '20140328', + 'uploader_id': '223413403', + 'timestamp': 1396018030, }, 'skip': 'Requires vk account credentials', }, @@ -165,7 +177,7 @@ class VKIE(VKBaseIE): 'upload_date': '20140626', 'view_count': int, }, - 'skip': 'Only works from Russia', + 'skip': 'Removed', }, { # video (removed?) only available with list id @@ -247,6 +259,9 @@ class VKIE(VKBaseIE): 'uploader_id': '-387766', 'timestamp': 1475137527, }, + 'params': { + 'skip_download': True, + }, }, { # live stream, hls and rtmp links, most likely already finished live @@ -288,80 +303,94 @@ def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') + mv_data = {} if video_id: - info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id + data = { + 'act': 'show_inline', + '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 + data['list'] = list_id + + payload = self._download_payload('al_video', video_id, data) + info_page = payload[1] + opts = payload[-1] + mv_data = opts.get('mvData') or {} + player = opts.get('player') or {} else: - info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query') video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id')) - info_page = self._download_webpage(info_url, video_id) + info_page = self._download_webpage( + 'http://vk.com/video_ext.php?' + mobj.group('embed_query'), 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) + 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) + 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.' + 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.<': - ERROR_COPYRIGHT, + ERRORS = { + r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': + ERROR_COPYRIGHT, - r'>The video .*? was removed from public access by request of the copyright holder.<': - 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'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.', + r'Unknown error': + 'Video %s does not exist.', - r'Видео временно недоступно': - 'Video %s is temporarily unavailable.', + r'Видео временно недоступно': + 'Video %s is temporarily unavailable.', - r'Access denied': - 'Access denied to video %s.', + 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'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 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'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.', - } + 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) + for error_re, error_msg in ERRORS.items(): + if re.search(error_re, info_page): + raise ExtractorError(error_msg % video_id, expected=True) + + player = self._parse_json(self._search_regex( + r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', + info_page, 'player params'), video_id) youtube_url = YoutubeIE._extract_url(info_page) if youtube_url: - return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) + return self.url_result(youtube_url, YoutubeIE.ie_key()) vimeo_url = VimeoIE._extract_url(url, info_page) if vimeo_url is not None: - return self.url_result(vimeo_url) + return self.url_result(vimeo_url, VimeoIE.ie_key()) pladform_url = PladformIE._extract_url(info_page) if pladform_url: - return self.url_result(pladform_url) + return self.url_result(pladform_url, PladformIE.ie_key()) m_rutube = re.search( r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page) @@ -374,6 +403,10 @@ def _real_extract(self, url): if dailymotion_urls: return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key()) + odnoklassniki_url = OdnoklassnikiIE._extract_url(info_page) + if odnoklassniki_url: + return self.url_result(odnoklassniki_url, OdnoklassnikiIE.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*'((?!/\b)[^']+)", m_opts.group(1)) @@ -383,38 +416,7 @@ def _real_extract(self, url): opts_url = 'http:' + opts_url return self.url_result(opts_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] - - #