From cb354c8f6218ecd722b218a08935f7bd7eecabd3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Thu, 10 Oct 2013 21:01:45 +0200 Subject: [PATCH] [yahoo] Download the info from another page The 'meta' field is not always in the video webpage --- youtube_dl/extractor/yahoo.py | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index 5bdd5d591b..464b498f58 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -50,6 +50,21 @@ def _real_extract(self, url): webpage, u'items', flags=re.MULTILINE) items = json.loads(items_json) info = items['mediaItems']['query']['results']['mediaObj'][0] + # The 'meta' field is not always in the video webpage, we request it + # from another page + long_id = info['id'] + query = ('SELECT * FROM yahoo.media.video.streams WHERE id="%s"' + ' AND plrs="86Gj0vCaSzV_Iuf6hNylf2"' % long_id) + data = compat_urllib_parse.urlencode({ + 'q': query, + 'env': 'prod', + 'format': 'json', + }) + query_result_json = self._download_webpage( + 'http://video.query.yahoo.com/v1/public/yql?' + data, + video_id, u'Downloading video info') + query_result = json.loads(query_result_json) + info = query_result['query']['results']['mediaObj'][0] meta = info['meta'] formats = []