[infoq] fix http format downloading

This commit is contained in:
Remita Amine 2017-10-12 17:39:51 +00:00
parent 2637fadc38
commit e650659b94

View File

@ -8,7 +8,10 @@
compat_urllib_parse_unquote, compat_urllib_parse_unquote,
compat_urlparse, compat_urlparse,
) )
from ..utils import determine_ext from ..utils import (
determine_ext,
update_url_query,
)
from .bokecc import BokeCCBaseIE from .bokecc import BokeCCBaseIE
@ -68,21 +71,22 @@ def _extract_rtmp_video(self, webpage):
'play_path': playpath, 'play_path': playpath,
}] }]
def _extract_cookies(self, webpage): def _extract_cf_auth(self, webpage):
policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy') policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature') 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') key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
return 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % ( return {
policy, signature, key_pair_id) 'Policy': policy,
'Signature': signature,
'Key-Pair-Id': key_pair_id,
}
def _extract_http_video(self, webpage): def _extract_http_video(self, webpage):
http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL') http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage))
return [{ return [{
'format_id': 'http_video', 'format_id': 'http_video',
'url': http_video_url, 'url': http_video_url,
'http_headers': {
'Cookie': self._extract_cookies(webpage)
},
}] }]
def _extract_http_audio(self, webpage, video_id): def _extract_http_audio(self, webpage, video_id):
@ -91,22 +95,20 @@ def _extract_http_audio(self, webpage, video_id):
if not http_audio_url: if not http_audio_url:
return [] return []
cookies_header = {'Cookie': self._extract_cookies(webpage)}
# base URL is found in the Location header in the response returned by # base URL is found in the Location header in the response returned by
# GET https://www.infoq.com/mp3download.action?filename=... when logged in. # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url) http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
# audio file seem to be missing some times even if there is a download link # audio file seem to be missing some times even if there is a download link
# so probe URL to make sure # so probe URL to make sure
if not self._is_valid_url(http_audio_url, video_id, headers=cookies_header): if not self._is_valid_url(http_audio_url, video_id):
return [] return []
return [{ return [{
'format_id': 'http_audio', 'format_id': 'http_audio',
'url': http_audio_url, 'url': http_audio_url,
'vcodec': 'none', 'vcodec': 'none',
'http_headers': cookies_header,
}] }]
def _real_extract(self, url): def _real_extract(self, url):