From eb9a15be6091d09f46d4306516b58cf2fe91e647 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 30 Sep 2017 22:47:03 +0700 Subject: [PATCH] [yahoo] Add support for custom brigthcove embeds (closes #14210) --- youtube_dl/extractor/yahoo.py | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index 38f82bf44..e92337969 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -28,7 +28,7 @@ class YahooIE(InfoExtractor): IE_DESC = 'Yahoo screen and movies' - _VALID_URL = r'(?P(?Phttps?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P.+)?-(?P[0-9]+)(?:-[a-z]+)?(?:\.html)?)' + _VALID_URL = r'(?Phttps?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P.+)?-)?(?P[0-9]+)(?:-[a-z]+)?(?:\.html)?' _TESTS = [ { 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', @@ -227,13 +227,28 @@ class YahooIE(InfoExtractor): 'skip_download': True, }, }, + { + # custom brightcove + 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/', + 'info_dict': { + 'id': '5575377707001', + 'ext': 'mp4', + 'title': "Clown entertainers say 'It' is hurting their business", + 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.', + 'timestamp': 1505341164, + 'upload_date': '20170913', + 'uploader_id': '2376984109001', + }, + 'params': { + 'skip_download': True, + }, + } ] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') or self._match_id(url) page_id = mobj.group('id') - url = mobj.group('url') + display_id = mobj.group('display_id') or page_id host = mobj.group('host') webpage, urlh = self._download_webpage_handle(url, display_id) if 'err=404' in urlh.geturl(): @@ -262,6 +277,14 @@ def _real_extract(self, url): if bc_url: return self.url_result(bc_url, BrightcoveNewIE.ie_key()) + brightcove_id = self._search_regex( + r']+data-video-id=["\'](\d+)', webpage, 'brightcove id', + default=None) + if brightcove_id: + return self.url_result( + 'http://players.brightcove.net/2376984109001/default_default/index.html?videoId=%s' % brightcove_id, + BrightcoveNewIE.ie_key()) + # Query result is often embedded in webpage as JSON. Sometimes explicit requests # to video API results in a failure with geo restriction reason therefore using # embedded query result when present sounds reasonable.