diff options
| author | Sergey M․ <[email protected]> | 2016-02-09 21:14:02 +0600 |
|---|---|---|
| committer | Sergey M․ <[email protected]> | 2016-02-09 21:14:02 +0600 |
| commit | 16f1430ba6fc1dc3fe0baf587f4461f37b62466a (patch) | |
| tree | 3422cb67ee7886906d7a3d76db98a2d1ad53dbe3 | |
| parent | 085ad71157dd6588cece2df0bfba6815f234564f (diff) | |
| download | youtube-dl-16f1430ba6fc1dc3fe0baf587f4461f37b62466a.tar.gz youtube-dl-16f1430ba6fc1dc3fe0baf587f4461f37b62466a.zip | |
[mailru] Prefer metaUrl API (Closes #8474)
| -rw-r--r-- | youtube_dl/extractor/mailru.py | 36 |
1 files changed, 34 insertions, 2 deletions
diff --git a/youtube_dl/extractor/mailru.py b/youtube_dl/extractor/mailru.py index ab1300185..09424620b 100644 --- a/youtube_dl/extractor/mailru.py +++ b/youtube_dl/extractor/mailru.py @@ -42,6 +42,22 @@ class MailRuIE(InfoExtractor): }, 'skip': 'Not accessible from Travis CI server', }, + { + # only available via metaUrl API + 'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html', + 'md5': '3b26d2491c6949d031a32b96bd97c096', + 'info_dict': { + 'id': '56664382_502', + 'ext': 'mp4', + 'title': ':8336', + 'timestamp': 1449094163, + 'upload_date': '20151202', + 'uploader': '[email protected]', + 'uploader_id': '[email protected]', + 'duration': 6001, + }, + 'skip': 'Not accessible from Travis CI server', + } ] def _real_extract(self, url): @@ -51,8 +67,24 @@ class MailRuIE(InfoExtractor): if not video_id: video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix') - video_data = self._download_json( - 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id, video_id, 'Downloading video JSON') + webpage = self._download_webpage(url, video_id) + + video_data = None + + page_config = self._parse_json(self._search_regex( + r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>', + webpage, 'page config', default='{}'), video_id, fatal=False) + if page_config: + meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') + if meta_url: + video_data = self._download_json( + meta_url, video_id, 'Downloading video meta JSON', fatal=False) + + # Fallback old approach + if not video_data: + video_data = self._download_json( + 'http://api.video.mail.ru/videos/%s.json?new=1' % video_id, + video_id, 'Downloading video JSON') author = video_data['author'] uploader = author['name'] |