github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-12-27 00:51:15 +00:00
Code Issues Projects Releases Packages Wiki Activity

[shared] Modernize and make more robust

Browse source
This commit is contained in:
Sergey M․ 2016-07-27 23:39:02 +07:00
parent 289a16b4f3
commit 05c8268c81
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
1 changed files with 10 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

18
youtube_dl/extractor/shared.py
View file

@ -6,7 +6,6 @@
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
int_or_none, int_or_none,
sanitized_Request,
urlencode_postdata, urlencode_postdata,
) )
@ -46,21 +45,24 @@ def _real_extract(self, url):
download_form = self._hidden_inputs(webpage) download_form = self._hidden_inputs(webpage)
request = sanitized_Request(
urlh.geturl(), urlencode_postdata(download_form))
request.add_header('Content-Type', 'application/x-www-form-urlencoded')
video_page = self._download_webpage( video_page = self._download_webpage(
request, video_id, 'Downloading video page') urlh.geturl(), video_id, 'Downloading video page',
data=urlencode_postdata(download_form),
headers={
'Content-Type': 'application/x-www-form-urlencoded',
'Referer': urlh.geturl(),
})
video_url = self._html_search_regex( video_url = self._html_search_regex(
r'data-url="([^"]+)"', video_page, 'video URL') r'data-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
video_page, 'video URL', group='url')
title = base64.b64decode(self._html_search_meta( title = base64.b64decode(self._html_search_meta(
'full:title', webpage, 'title').encode('utf-8')).decode('utf-8') 'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
filesize = int_or_none(self._html_search_meta( filesize = int_or_none(self._html_search_meta(
'full:size', webpage, 'file size', fatal=False)) 'full:size', webpage, 'file size', fatal=False))
thumbnail = self._html_search_regex( thumbnail = self._html_search_regex(
r'data-poster="([^"]+)"', video_page, 'thumbnail', default=None) r'data-poster=(["\'])(?P<url>(?:(?!\1).)+)\1',
video_page, 'thumbnail', default=None, group='url')
return { return {
'id': video_id, 'id': video_id,