administration/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-10 03:10:38 +00:00
Code Issues Projects Releases Wiki Activity

[localnews8] Fix extractor (Closes #9539)

Browse Source
This commit is contained in:
Sergey M․ 2016-05-20 22:31:08 +06:00
parent 6756602be6
commit 1846e9ade0
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
1 changed files with 28 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
youtube_dl/extractor/localnews8.py
View File

@ -1,29 +1,47 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class LocalNews8IE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?localnews8\.com/.+?/(?P<id>[0-9]+)'
_VALID_URL = r'https?://(?:www\.)?localnews8\.com/(?:[^/]+/)*(?P<display_id>[^/]+)/(?P<id>[0-9]+)'
_TEST = {
'url': 'http://www.localnews8.com/news/rexburg-business-turns-carbon-fiber-scraps-into-wedding-rings/35183304',
'md5': '477bdb188f177788c65db27ecb56649b',
'md5': 'be4d48aea61aa2bde7be2ee47691ad20',
'info_dict': {
'id': '35183304',
'display_id': 'rexburg-business-turns-carbon-fiber-scraps-into-wedding-rings',
'ext': 'mp4',
'title': 'Rexburg business turns carbon fiber scraps into wedding ring',
'description': 'The process was first invented by Lamborghini and less than a dozen companies around the world use it.',
'duration': '153',
'timestamp': '1441844822',
'duration': 153,
'timestamp': 1441844822,
'upload_date': '20150910',
'uploader_id': 'api',
}}
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
display_id = mobj.group('display_id')
partner_id = self._search_regex(r'partnerId\s*:\s*"(\d+)"', webpage, video_id)
kaltura_id = self._search_regex(r'var\s+videoIdString\s*=\s*"kaltura:(.+)";', webpage, video_id)
webpage = self._download_webpage(url, display_id)
return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura')
partner_id = self._search_regex(
r'partnerId\s*[:=]\s*(["\'])(?P<id>\d+)\1',
webpage, 'partner id', group='id')
kaltura_id = self._search_regex(
r'videoIdString\s*[:=]\s*(["\'])kaltura:(?P<id>[0-9a-z_]+)\1',
webpage, 'videl id', group='id')
return {
'_type': 'url_transparent',
'url': 'kaltura:%s:%s' % (partner_id, kaltura_id),
'ie_key': 'Kaltura',
'id': video_id,
'display_id': display_id,
}