youtube-dl/youtube_dl/extractor/openload.py

109 lines
3.5 KiB
Python
Raw Normal View History

# coding: utf-8
from __future__ import unicode_literals, division
from .common import InfoExtractor
from ..compat import (
compat_chr,
compat_ord,
)
from ..utils import (
2016-04-24 20:03:29 +02:00
determine_ext,
ExtractorError,
)
class OpenloadIE(InfoExtractor):
2016-09-08 13:29:05 +02:00
_VALID_URL = r'https?://openload\.(?:co|io)/(?:f|embed)/(?P<id>[a-zA-Z0-9-_]+)'
_TESTS = [{
'url': 'https://openload.co/f/kUEfGclsU9o',
'md5': 'bf1c059b004ebc7a256f89408e65c36e',
'info_dict': {
'id': 'kUEfGclsU9o',
'ext': 'mp4',
'title': 'skyrim_no-audio_1080.mp4',
'thumbnail': 're:^https?://.*\.jpg$',
},
}, {
'url': 'https://openload.co/embed/rjC09fkPLYs',
'info_dict': {
'id': 'rjC09fkPLYs',
'ext': 'mp4',
'title': 'movie.mp4',
'thumbnail': 're:^https?://.*\.jpg$',
'subtitles': {
'en': [{
'ext': 'vtt',
}],
},
},
'params': {
'skip_download': True, # test subtitles only
},
}, {
'url': 'https://openload.co/embed/kUEfGclsU9o/skyrim_no-audio_1080.mp4',
'only_matching': True,
}, {
'url': 'https://openload.io/f/ZAn6oz-VZGE/',
'only_matching': True,
}, {
'url': 'https://openload.co/f/_-ztPaZtMhM/',
'only_matching': True,
2016-04-24 20:22:55 +02:00
}, {
# unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
# for title and ext
'url': 'https://openload.co/embed/Sxz5sADo82g/',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id)
if 'File not found' in webpage or 'deleted by the owner' in webpage:
raise ExtractorError('File not found', expected=True)
# The following decryption algorithm is written by @yokrysty and
# declared to be freely used in youtube-dl
# See https://github.com/rg3/youtube-dl/issues/10408
enc_data = self._html_search_regex(
r'<span[^>]*>([^<]+)</span>\s*<span[^>]*>[^<]+</span>\s*<span[^>]+id="streamurl"',
webpage, 'encrypted data')
magic = compat_ord(enc_data[-1])
video_url_chars = []
for idx, c in enumerate(enc_data):
j = compat_ord(c)
if j == magic:
j -= 1
elif j == magic - 1:
j += 1
if j >= 33 and j <= 126:
j = ((j + 14) % 94) + 33
if idx == len(enc_data) - 1:
j += 2
video_url_chars += compat_chr(j)
video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars)
title = self._og_search_title(webpage, default=None) or self._search_regex(
r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
'title', default=None) or self._html_search_meta(
'description', webpage, 'title', fatal=True)
entries = self._parse_html5_media_entries(url, webpage, video_id)
subtitles = entries[0]['subtitles'] if entries else None
info_dict = {
'id': video_id,
'title': title,
2016-04-24 20:26:06 +02:00
'thumbnail': self._og_search_thumbnail(webpage, default=None),
'url': video_url,
# Seems all videos have extensions in their titles
'ext': determine_ext(title),
'subtitles': subtitles,
}
return info_dict