[mlb] Fallback to extracting video id from webpage for all URLs that does not contain it explicitly (Closes #5630)

This commit is contained in:
Sergey M․ 2015-05-08 20:07:53 +06:00
parent 43837189c1
commit d1feb30811

View File

@ -10,7 +10,21 @@
class MLBIE(InfoExtractor): class MLBIE(InfoExtractor):
_VALID_URL = r'https?://m(?:lb)?\.(?:[\da-z_-]+\.)?mlb\.com/(?:(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v|(?:shared/video/embed/(?:embed|m-internal-embed)\.html|[^/]+/video/play\.jsp)\?.*?\bcontent_id=)(?P<id>n?\d+)' _VALID_URL = r'''(?x)
https?://
m(?:lb)?\.(?:[\da-z_-]+\.)?mlb\.com/
(?:
(?:
(?:.*?/)?video/(?:topic/[\da-z_-]+/)?v|
(?:
shared/video/embed/(?:embed|m-internal-embed)\.html|
[^/]+/video/play\.jsp
)\?.*?\bcontent_id=
)
(?P<id>n?\d+)|
(?P<path>.+?)
)
'''
_TESTS = [ _TESTS = [
{ {
'url': 'http://m.mlb.com/sea/video/topic/51231442/v34698933/nymsea-ackley-robs-a-home-run-with-an-amazing-catch/?c_id=sea', 'url': 'http://m.mlb.com/sea/video/topic/51231442/v34698933/nymsea-ackley-robs-a-home-run-with-an-amazing-catch/?c_id=sea',
@ -95,6 +109,12 @@ def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id = mobj.group('id')
if not video_id:
video_path = mobj.group('path')
webpage = self._download_webpage(url, video_path)
video_id = self._search_regex(
r'data-videoid="(\d+)"', webpage, 'video id')
detail = self._download_xml( detail = self._download_xml(
'http://m.mlb.com/gen/multimedia/detail/%s/%s/%s/%s.xml' 'http://m.mlb.com/gen/multimedia/detail/%s/%s/%s/%s.xml'
% (video_id[-3], video_id[-2], video_id[-1], video_id), video_id) % (video_id[-3], video_id[-2], video_id[-1], video_id), video_id)