github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-21 20:46:36 -05:00
Code Issues Projects Releases Packages Wiki Activity

[infoq] Fix extraction for HTTP URLs (closes #7739)

Browse source
This commit is contained in:
Yen Chi Hsuan 2015-12-13 17:29:27 +08:00
parent f6abca506e
commit 22d07ba4e4
1 changed files with 11 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
youtube_dl/extractor/infoq.py
View file

@ -3,10 +3,7 @@
import base64 import base64
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import compat_urllib_parse_unquote
compat_urllib_parse_unquote,
compat_urlparse,
)
class InfoQIE(InfoExtractor): class InfoQIE(InfoExtractor):
@ -45,9 +42,11 @@ def _real_extract(self, url):
video_filename = playpath.split('/')[-1] video_filename = playpath.split('/')[-1]
video_id, extension = video_filename.split('.') video_id, extension = video_filename.split('.')
http_base = self._search_regex( http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
r'EXPRESSINSTALL_SWF\s*=\s*[^"]*"((?:https?:)?//[^/"]+/)', webpage,
'HTTP base URL') policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
formats = [{ formats = [{
'format_id': 'rtmp', 'format_id': 'rtmp',
@ -56,7 +55,11 @@ def _real_extract(self, url):
'play_path': playpath, 'play_path': playpath,
}, { }, {
'format_id': 'http', 'format_id': 'http',
'url': compat_urlparse.urljoin(url, http_base) + real_id, 'url': http_video_url,
'http_headers': {
'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
policy, signature, key_pair_id),
},
}] }]
self._sort_formats(formats) self._sort_formats(formats)