Browse Source

[cbsnews:livevideo] fix extraction and extract m3u8 formats

totalwebcasting
Remita Amine 8 years ago
parent
commit
7fd57de6fb
1 changed files with 15 additions and 16 deletions
  1. +15
    -16
      youtube_dl/extractor/cbsnews.py

+ 15
- 16
youtube_dl/extractor/cbsnews.py View File

@ -9,6 +9,7 @@ from ..utils import (
class CBSNewsIE(CBSIE): class CBSNewsIE(CBSIE):
IE_NAME = 'cbsnews'
IE_DESC = 'CBS News' IE_DESC = 'CBS News'
_VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)' _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
@ -68,15 +69,16 @@ class CBSNewsIE(CBSIE):
class CBSNewsLiveVideoIE(InfoExtractor): class CBSNewsLiveVideoIE(InfoExtractor):
IE_NAME = 'cbsnews:livevideo'
IE_DESC = 'CBS News Live Videos' IE_DESC = 'CBS News Live Videos'
_VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[\da-z_-]+)'
_VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[^/?#]+)'
# Live videos get deleted soon. See http://www.cbsnews.com/live/ for the latest examples # Live videos get deleted soon. See http://www.cbsnews.com/live/ for the latest examples
_TEST = { _TEST = {
'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/', 'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
'info_dict': { 'info_dict': {
'id': 'clinton-sanders-prepare-to-face-off-in-nh', 'id': 'clinton-sanders-prepare-to-face-off-in-nh',
'ext': 'flv',
'ext': 'mp4',
'title': 'Clinton, Sanders Prepare To Face Off In NH', 'title': 'Clinton, Sanders Prepare To Face Off In NH',
'duration': 334, 'duration': 334,
}, },
@ -84,25 +86,22 @@ class CBSNewsLiveVideoIE(InfoExtractor):
} }
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
display_id = self._match_id(url)
video_info = self._parse_json(self._html_search_regex(
r'data-story-obj=\'({.+?})\'', webpage, 'video JSON info'), video_id)['story']
video_info = self._download_json(
'http://feeds.cbsn.cbsnews.com/rundown/story', display_id, query={
'device': 'desktop',
'dvr_slug': display_id,
})
hdcore_sign = 'hdcore=3.3.1'
f4m_formats = self._extract_f4m_formats(video_info['url'] + '&' + hdcore_sign, video_id)
if f4m_formats:
for entry in f4m_formats:
# URLs without the extra param induce an 404 error
entry.update({'extra_param_to_segment_url': hdcore_sign})
self._sort_formats(f4m_formats)
formats = self._extract_akamai_formats(video_info['url'], display_id)
self._sort_formats(formats)
return { return {
'id': video_id,
'id': display_id,
'display_id': display_id,
'title': video_info['headline'], 'title': video_info['headline'],
'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'), 'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
'duration': parse_duration(video_info.get('segmentDur')), 'duration': parse_duration(video_info.get('segmentDur')),
'formats': f4m_formats,
'formats': formats,
} }

Loading…
Cancel
Save