mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-11-27 02:42:30 +00:00
cefecac12c
to avoid using same folder when using pip install for example
72 lines
2.7 KiB
Python
72 lines
2.7 KiB
Python
# coding: utf-8
|
|
from __future__ import unicode_literals
|
|
|
|
import string
|
|
import random
|
|
import time
|
|
|
|
from .common import InfoExtractor
|
|
|
|
|
|
class DoodStreamIE(InfoExtractor):
|
|
_VALID_URL = r'https?://(?:www\.)?dood\.(?:to|watch)/[ed]/(?P<id>[a-z0-9]+)'
|
|
_TESTS = [{
|
|
'url': 'http://dood.to/e/5s1wmbdacezb',
|
|
'md5': '4568b83b31e13242b3f1ff96c55f0595',
|
|
'info_dict': {
|
|
'id': '5s1wmbdacezb',
|
|
'ext': 'mp4',
|
|
'title': 'Kat Wonders - Monthly May 2020',
|
|
'description': 'Kat Wonders - Monthly May 2020 | DoodStream.com',
|
|
'thumbnail': 'https://img.doodcdn.com/snaps/flyus84qgl2fsk4g.jpg',
|
|
}
|
|
}, {
|
|
'url': 'https://dood.to/d/jzrxn12t2s7n',
|
|
'md5': '3207e199426eca7c2aa23c2872e6728a',
|
|
'info_dict': {
|
|
'id': 'jzrxn12t2s7n',
|
|
'ext': 'mp4',
|
|
'title': 'Stacy Cruz Cute ALLWAYSWELL',
|
|
'description': 'Stacy Cruz Cute ALLWAYSWELL | DoodStream.com',
|
|
'thumbnail': 'https://img.doodcdn.com/snaps/8edqd5nppkac3x8u.jpg',
|
|
}
|
|
}]
|
|
|
|
def _real_extract(self, url):
|
|
video_id = self._match_id(url)
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
if '/d/' in url:
|
|
url = "https://dood.to" + self._html_search_regex(
|
|
r'<iframe src="(/e/[a-z0-9]+)"', webpage, 'embed')
|
|
video_id = self._match_id(url)
|
|
webpage = self._download_webpage(url, video_id)
|
|
|
|
title = self._html_search_meta(['og:title', 'twitter:title'],
|
|
webpage, default=None)
|
|
thumb = self._html_search_meta(['og:image', 'twitter:image'],
|
|
webpage, default=None)
|
|
token = self._html_search_regex(r'[?&]token=([a-z0-9]+)[&\']', webpage, 'token')
|
|
description = self._html_search_meta(
|
|
['og:description', 'description', 'twitter:description'],
|
|
webpage, default=None)
|
|
auth_url = 'https://dood.to' + self._html_search_regex(
|
|
r'(/pass_md5.*?)\'', webpage, 'pass_md5')
|
|
headers = {
|
|
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:53.0) Gecko/20100101 Firefox/66.0',
|
|
'referer': url
|
|
}
|
|
|
|
webpage = self._download_webpage(auth_url, video_id, headers=headers)
|
|
final_url = webpage + ''.join([random.choice(string.ascii_letters + string.digits) for _ in range(10)]) + "?token=" + token + "&expiry=" + str(int(time.time() * 1000))
|
|
|
|
return {
|
|
'id': video_id,
|
|
'title': title,
|
|
'url': final_url,
|
|
'http_headers': headers,
|
|
'ext': 'mp4',
|
|
'description': description,
|
|
'thumbnail': thumb,
|
|
}
|