Archives/yt-dlp
2
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-15 18:12:58 +00:00
Code Issues Projects Releases Wiki Activity

[teamcoco] Fix video url extraction for some videos

Browse Source 
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
This commit is contained in:
Alex Van't Hof 2013-10-31 01:57:22 -04:00
parent 9f1109a564
commit ac2547f5ff
1 changed files with 17 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
youtube_dl/extractor/teamcoco.py
View File

@ -3,6 +3,7 @@
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
RegexNotFoundError,
) )
@ -11,7 +12,7 @@ class TeamcocoIE(InfoExtractor):
_TEST = { _TEST = {
u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush', u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
u'file': u'19705.mp4', u'file': u'19705.mp4',
u'md5': u'27b6f7527da5acf534b15f21b032656e', u'md5': u'cde9ba0fa3506f5f017ce11ead928f9a',
u'info_dict': { u'info_dict': {
u"description": u"Louis C.K. got starstruck by George W. Bush, so what? Part one.", 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" 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_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
data = self._download_webpage(data_url, video_id, 'Downloading data webpage') data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
video_url = self._html_search_regex(r'<file [^>]*type="high".*?>(.*?)</file>',
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'<file [^>]*type="(?:high|standard)".*?>(.*%s.*)</file>' % 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 [{ return [{
'id': video_id, 'id': video_id,