1
0
mirror of https://github.com/l1ving/youtube-dl synced 2025-01-27 14:05:17 +08:00

Improved regexes and returns info dict now.

This commit is contained in:
Sander van den Oever 2016-02-27 22:28:39 +01:00
parent 0b6645fdb5
commit 29331a4d18

View File

@ -7,7 +7,7 @@ from .common import InfoExtractor
class CloserToTruthIE(InfoExtractor):
_VALID_URL = r'http?://(?:www\.)?closertotruth\.com/series/\S+#video-(?P<id>\w+)'
_VALID_URL = r'https?://(?:www\.)?closertotruth\.com/series/[^#]+#video-(?P<id>\w+)'
_TESTS = [{
'url': 'http://closertotruth.com/series/solutions-the-mind-body-problem#video-3688',
'md5': '2aa5b8971633d86fe32152827846a5b4',
@ -27,12 +27,17 @@ class CloserToTruthIE(InfoExtractor):
video_title = self._search_regex(r'<title>(.+) \|.+</title>', webpage, 'video title')
entry_id = self._search_regex(r'<a href="\S+" id="video-' + video_id + '" data-kaltura="(\w+)">.+<span.+<\/a>', webpage, "video entry_id")
entry_id = self._search_regex(r'<a[^>]+id="video-%s"[^>]+data-kaltura="([^"]+)' % video_id, webpage, "video entry_id")
interviewee_name = re.sub(r'(<[^>]+>)', '', self._search_regex(r'<a href="\S+" id="video-' + video_id + '" data-kaltura="\w+">(.+)<span.+<\/a>', webpage, "video interviewee_name"))
video_title = video_title + ' - ' + interviewee_name
# extract the partner id for kaltura.com
p_id = self._search_regex(r'<script src="http://cdnapi\.kaltura\.com/p/(?P<p>\w+)/sp/\w+/\S+/partner_id/\w+"></script>', webpage, "kaltura partner_id")
p_id = self._search_regex(r'<script[^>]+src=["\'].+?partner_id/(\d+)', webpage, "kaltura partner_id")
return self.url_result('kaltura:%s:%s' % (p_id, entry_id), 'Kaltura', entry_id, video_title)
return {
'_type': 'url_transparent',
'id': entry_id,
'url': 'kaltura:%s:%s' % (p_id, entry_id),
'ie_key': 'Kaltura',
'title': video_title
}