github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-24 17:31:14 +00:00
Code Issues Projects Releases Packages Wiki Activity

[bloomberg] Fix extraction (fixes #2154)

Browse source 
Stop using the OoyalaIE, extract the f4m url instead.
This commit is contained in:
Jaime Marquínez Ferrándiz 2014-03-29 11:55:12 +01:00
parent 056b56688a
commit 7e70ac36b3
1 changed files with 22 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

36
youtube_dl/extractor/bloomberg.py
View file

@ -1,22 +1,21 @@
from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from .ooyala import OoyalaIE
class BloombergIE(InfoExtractor): class BloombergIE(InfoExtractor):
_VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html' _VALID_URL = r'https?://www\.bloomberg\.com/video/(?P<name>.+?)\.html'
_TEST = { _TEST = {
u'url': u'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html', 'url': 'http://www.bloomberg.com/video/shah-s-presentation-on-foreign-exchange-strategies-qurhIVlJSB6hzkVi229d8g.html',
u'file': u'12bzhqZTqQHmmlA8I-i0NpzJgcG5NNYX.mp4', 'md5': '7bf08858ff7c203c870e8a6190e221e5',
u'info_dict': { 'info_dict': {
u'title': u'Shah\'s Presentation on Foreign-Exchange Strategies', 'id': 'qurhIVlJSB6hzkVi229d8g',
u'description': u'md5:abc86e5236f9f0e4866c59ad36736686', 'ext': 'flv',
}, 'title': 'Shah\'s Presentation on Foreign-Exchange Strategies',
u'params': { 'description': 'md5:0681e0d30dcdfc6abf34594961d8ea88',
# Requires ffmpeg (m3u8 manifest)
u'skip_download': True,
}, },
} }
@ -24,7 +23,16 @@ def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
name = mobj.group('name') name = mobj.group('name')
webpage = self._download_webpage(url, name) webpage = self._download_webpage(url, name)
embed_code = self._search_regex( f4m_url = self._search_regex(
r'<source src="https?://[^/]+/[^/]+/[^/]+/([^/]+)', webpage, r'<source src="(https?://[^"]+\.f4m.*?)"', webpage,
'embed code') 'f4m url')
return OoyalaIE._build_url_result(embed_code) title = re.sub(': Video$', '', self._og_search_title(webpage))
return {
'id': name.split('-')[-1],
'title': title,
'url': f4m_url,
'ext': 'flv',
'description': self._og_search_description(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
}