[tlc] fix extraction and update extractor to use BrightcoveNewIE

This commit is contained in:
remitamine 2016-03-14 21:53:00 +01:00
parent 203f3d779a
commit bf475e1990

View file

@ -4,12 +4,12 @@
from .common import InfoExtractor
from .brightcove import BrightcoveLegacyIE
from ..compat import compat_urlparse
from ..compat import compat_parse_qs
class TlcDeIE(InfoExtractor):
IE_NAME = 'tlc.de'
_VALID_URL = r'http://www\.tlc\.de/sendungen/[^/]+/videos/(?P<title>[^/?]+)'
_VALID_URL = r'http://www\.tlc\.de/(?:[^/]+/)*videos/(?P<title>[^/?#]+)?(?:.*#(?P<id>\d+))?'
_TEST = {
'url': 'http://www.tlc.de/sendungen/breaking-amish/videos/#3235167922001',
@ -17,32 +17,23 @@ class TlcDeIE(InfoExtractor):
'id': '3235167922001',
'ext': 'mp4',
'title': 'Breaking Amish: Die Welt da draußen',
'uploader': 'Discovery Networks - Germany',
'description': (
'Vier Amische und eine Mennonitin wagen in New York'
' den Sprung in ein komplett anderes Leben. Begleitet sie auf'
' ihrem spannenden Weg.'),
'timestamp': 1396598084,
'upload_date': '20140404',
'uploader_id': '1659832546',
},
}
BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1659832546/default_default/index.html?videoId=%s'
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
title = mobj.group('title')
webpage = self._download_webpage(url, title)
iframe_url = self._search_regex(
'<iframe src="(http://www\.tlc\.de/wp-content/.+?)"', webpage,
'iframe url')
# Otherwise we don't get the correct 'BrightcoveExperience' element,
# example: http://www.tlc.de/sendungen/cake-boss/videos/cake-boss-cannoli-drama/
iframe_url = iframe_url.replace('.htm?', '.php?')
url_fragment = compat_urlparse.urlparse(url).fragment
if url_fragment:
# Since the fragment is not send to the server, we always get the same iframe
iframe_url = re.sub(r'playlist=(\d+)', 'playlist=%s' % url_fragment, iframe_url)
iframe = self._download_webpage(iframe_url, title)
return {
'_type': 'url',
'url': BrightcoveLegacyIE._extract_brightcove_url(iframe),
'ie': BrightcoveLegacyIE.ie_key(),
}
brightcove_id = mobj.group('id')
if not brightcove_id:
title = mobj.group('title')
webpage = self._download_webpage(url, title)
brightcove_legacy_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
brightcove_id = compat_parse_qs(brightcove_legacy_url)['@videoPlayer'][0]
return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)