[cinemassacre] Keep both extraction approaches and make more robust (Closes #4109)
This commit is contained in:
parent
81b22aee8b
commit
dab647a7b6
1 changed files with 47 additions and 31 deletions
|
@ -42,11 +42,12 @@ class CinemassacreIE(InfoExtractor):
|
||||||
|
|
||||||
webpage = self._download_webpage(url, display_id)
|
webpage = self._download_webpage(url, display_id)
|
||||||
video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
|
video_date = mobj.group('date_Y') + mobj.group('date_m') + mobj.group('date_d')
|
||||||
mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?:Cinemassacre-)?(?P<video_id>.+?))"', webpage)
|
mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?P<full_video_id>(?:Cinemassacre-)?(?P<video_id>.+?)))"', webpage)
|
||||||
if not mobj:
|
if not mobj:
|
||||||
raise ExtractorError('Can\'t extract embed url and video id')
|
raise ExtractorError('Can\'t extract embed url and video id')
|
||||||
playerdata_url = mobj.group('embed_url')
|
playerdata_url = mobj.group('embed_url')
|
||||||
video_id = mobj.group('video_id')
|
video_id = mobj.group('video_id')
|
||||||
|
full_video_id = mobj.group('full_video_id')
|
||||||
|
|
||||||
video_title = self._html_search_regex(
|
video_title = self._html_search_regex(
|
||||||
r'<title>(?P<title>.+?)\|', webpage, 'title')
|
r'<title>(?P<title>.+?)\|', webpage, 'title')
|
||||||
|
@ -60,37 +61,52 @@ class CinemassacreIE(InfoExtractor):
|
||||||
vidurl = self._search_regex(
|
vidurl = self._search_regex(
|
||||||
r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
|
r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/')
|
||||||
|
|
||||||
videolist_url = self._search_regex(
|
videolist_url = None
|
||||||
r"file\s*:\s*'(http.+?/jwplayer\.smil)'", playerdata, 'jwplayer.smil')
|
|
||||||
videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
|
|
||||||
|
|
||||||
formats = []
|
mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata)
|
||||||
baseurl = vidurl[:vidurl.rfind('/')+1]
|
if mobj:
|
||||||
for video in videolist.findall('.//video'):
|
videoserver = mobj.group('videoserver')
|
||||||
src = video.get('src')
|
mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata)
|
||||||
if not src:
|
vidid = mobj.group('vidid') if mobj else full_video_id
|
||||||
continue
|
videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid)
|
||||||
file_ = src.partition(':')[-1]
|
else:
|
||||||
width = int_or_none(video.get('width'))
|
mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata)
|
||||||
height = int_or_none(video.get('height'))
|
if mobj:
|
||||||
bitrate = int_or_none(video.get('system-bitrate'))
|
videolist_url = mobj.group('smil')
|
||||||
format = {
|
|
||||||
'url': baseurl + file_,
|
if videolist_url:
|
||||||
'format_id': src.rpartition('.')[0].rpartition('_')[-1],
|
videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML')
|
||||||
}
|
formats = []
|
||||||
if width or height:
|
baseurl = vidurl[:vidurl.rfind('/')+1]
|
||||||
format.update({
|
for video in videolist.findall('.//video'):
|
||||||
'tbr': bitrate // 1000 if bitrate else None,
|
src = video.get('src')
|
||||||
'width': width,
|
if not src:
|
||||||
'height': height,
|
continue
|
||||||
})
|
file_ = src.partition(':')[-1]
|
||||||
else:
|
width = int_or_none(video.get('width'))
|
||||||
format.update({
|
height = int_or_none(video.get('height'))
|
||||||
'abr': bitrate // 1000 if bitrate else None,
|
bitrate = int_or_none(video.get('system-bitrate'))
|
||||||
'vcodec': 'none',
|
format = {
|
||||||
})
|
'url': baseurl + file_,
|
||||||
formats.append(format)
|
'format_id': src.rpartition('.')[0].rpartition('_')[-1],
|
||||||
self._sort_formats(formats)
|
}
|
||||||
|
if width or height:
|
||||||
|
format.update({
|
||||||
|
'tbr': bitrate // 1000 if bitrate else None,
|
||||||
|
'width': width,
|
||||||
|
'height': height,
|
||||||
|
})
|
||||||
|
else:
|
||||||
|
format.update({
|
||||||
|
'abr': bitrate // 1000 if bitrate else None,
|
||||||
|
'vcodec': 'none',
|
||||||
|
})
|
||||||
|
formats.append(format)
|
||||||
|
self._sort_formats(formats)
|
||||||
|
else:
|
||||||
|
formats = [{
|
||||||
|
'url': vidurl,
|
||||||
|
}]
|
||||||
|
|
||||||
return {
|
return {
|
||||||
'id': video_id,
|
'id': video_id,
|
||||||
|
|
Loading…
Reference in a new issue