mirror of
https://github.com/ytdl-org/youtube-dl
synced 2024-11-05 01:48:01 +09:00
[arte] Mark dead code as such
This commit is contained in:
parent
d5822b96b0
commit
9b3a760bbb
@ -48,31 +48,32 @@ class ArteTvIE(InfoExtractor):
|
||||
|
||||
return info
|
||||
|
||||
def extractLiveStream(self, url):
|
||||
video_lang = url.split('/')[-4]
|
||||
info = self.grep_webpage(
|
||||
url,
|
||||
r'src="(.*?/videothek_js.*?\.js)',
|
||||
0,
|
||||
[
|
||||
(1, 'url', u'Invalid URL: %s' % url)
|
||||
]
|
||||
)
|
||||
http_host = url.split('/')[2]
|
||||
next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
|
||||
info = self.grep_webpage(
|
||||
next_url,
|
||||
r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
|
||||
'(http://.*?\.swf).*?' +
|
||||
'(rtmp://.*?)\'',
|
||||
re.DOTALL,
|
||||
[
|
||||
(1, 'path', u'could not extract video path: %s' % url),
|
||||
(2, 'player', u'could not extract video player: %s' % url),
|
||||
(3, 'url', u'could not extract video url: %s' % url)
|
||||
]
|
||||
)
|
||||
video_url = u'%s/%s' % (info.get('url'), info.get('path'))
|
||||
# TODO implement Live Stream
|
||||
# def extractLiveStream(self, url):
|
||||
# video_lang = url.split('/')[-4]
|
||||
# info = self.grep_webpage(
|
||||
# url,
|
||||
# r'src="(.*?/videothek_js.*?\.js)',
|
||||
# 0,
|
||||
# [
|
||||
# (1, 'url', u'Invalid URL: %s' % url)
|
||||
# ]
|
||||
# )
|
||||
# http_host = url.split('/')[2]
|
||||
# next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
|
||||
# info = self.grep_webpage(
|
||||
# next_url,
|
||||
# r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
|
||||
# '(http://.*?\.swf).*?' +
|
||||
# '(rtmp://.*?)\'',
|
||||
# re.DOTALL,
|
||||
# [
|
||||
# (1, 'path', u'could not extract video path: %s' % url),
|
||||
# (2, 'player', u'could not extract video player: %s' % url),
|
||||
# (3, 'url', u'could not extract video url: %s' % url)
|
||||
# ]
|
||||
# )
|
||||
# video_url = u'%s/%s' % (info.get('url'), info.get('path'))
|
||||
|
||||
def extractPlus7Stream(self, url):
|
||||
video_lang = url.split('/')[-3]
|
||||
@ -126,8 +127,9 @@ class ArteTvIE(InfoExtractor):
|
||||
self.report_extraction(video_id)
|
||||
|
||||
if re.search(self._LIVE_URL, video_id) is not None:
|
||||
self.extractLiveStream(url)
|
||||
return
|
||||
raise ExtractorError(u'Arte live streams are not yet supported, sorry')
|
||||
# self.extractLiveStream(url)
|
||||
# return
|
||||
else:
|
||||
info = self.extractPlus7Stream(url)
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user