From ac2547f5ffc30a352207336194e7bbb0435d01a7 Mon Sep 17 00:00:00 2001 From: Alex Van't Hof Date: Thu, 31 Oct 2013 01:57:22 -0400 Subject: [PATCH] [teamcoco] Fix video url extraction for some videos Video url extraction failed for some videos, e.g. http://teamcoco.com/video/old-time-baseball The url extracted was also occasionally suboptimal quality, e.g. http://teamcoco.com/video/louis-ck-interview-george-w-bush --- youtube_dl/extractor/teamcoco.py | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py index c910110ca..76246c7cc 100644 --- a/youtube_dl/extractor/teamcoco.py +++ b/youtube_dl/extractor/teamcoco.py @@ -3,6 +3,7 @@ from .common import InfoExtractor from ..utils import ( ExtractorError, + RegexNotFoundError, ) @@ -11,7 +12,7 @@ class TeamcocoIE(InfoExtractor): _TEST = { u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush', u'file': u'19705.mp4', - u'md5': u'27b6f7527da5acf534b15f21b032656e', + u'md5': u'cde9ba0fa3506f5f017ce11ead928f9a', u'info_dict': { u"description": u"Louis C.K. got starstruck by George W. Bush, so what? Part one.", u"title": u"Louis C.K. Interview Pt. 1 11/3/11" @@ -33,8 +34,21 @@ def _real_extract(self, url): data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id data = self._download_webpage(data_url, video_id, 'Downloading data webpage') - video_url = self._html_search_regex(r']*type="high".*?>(.*?)', - data, u'video URL') + + qualities = [ '1080p', '720p', '1000k', '480p', '500k' ] + best_quality_idx = len(qualities)+1 # First regex match may not be optimal + for idx, quality in enumerate(qualities): + regex = r']*type="(?:high|standard)".*?>(.*%s.*)' % quality + try: + url = self._html_search_regex(regex, data, u'video URL') + if idx < best_quality_idx: + video_url = url + best_quality_idx = idx + except RegexNotFoundError: + # Just catch fatal exc. Don't want the fatal=False warning + continue + if not video_url: + raise RegexNotFoundError(u'Unable to extract video URL') return [{ 'id': video_id,