X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fopenload.py;h=fc7ff43a62ba4a1a294d368d0b8fbaeb9abdb2ed;hb=39e7277ed16c1647d636c766d57870121f5f2d68;hp=7f19b1ba5c3c355977c694334694b51f71a9840c;hpb=315cb86a95d6e5c67fef27a88d96477786072b93;p=youtube-dl diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 7f19b1ba5..fc7ff43a6 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -1,25 +1,18 @@ # coding: utf-8 -from __future__ import unicode_literals, division +from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_chr, - compat_ord, -) +from ..compat import compat_chr from ..utils import ( determine_ext, ExtractorError, ) -from ..jsinterp import ( - JSInterpreter, - _NAME_RE -) class OpenloadIE(InfoExtractor): - _VALID_URL = r'https?://openload\.(?:co|io)/(?:f|embed)/(?P[a-zA-Z0-9-_]+)' + _VALID_URL = r'https?://(?:openload\.(?:co|io)|oload\.tv)/(?:f|embed)/(?P[a-zA-Z0-9-_]+)' _TESTS = [{ 'url': 'https://openload.co/f/kUEfGclsU9o', @@ -28,7 +21,7 @@ class OpenloadIE(InfoExtractor): 'id': 'kUEfGclsU9o', 'ext': 'mp4', 'title': 'skyrim_no-audio_1080.mp4', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', }, }, { 'url': 'https://openload.co/embed/rjC09fkPLYs', @@ -36,7 +29,7 @@ class OpenloadIE(InfoExtractor): 'id': 'rjC09fkPLYs', 'ext': 'mp4', 'title': 'movie.mp4', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'subtitles': { 'en': [{ 'ext': 'vtt', @@ -60,45 +53,16 @@ class OpenloadIE(InfoExtractor): # for title and ext 'url': 'https://openload.co/embed/Sxz5sADo82g/', 'only_matching': True, + }, { + 'url': 'https://oload.tv/embed/KnG-kKZdcfY/', + 'only_matching': True, }] - def openload_decode(self, txt): - symbol_dict = { - '(゚Д゚) [゚Θ゚]': '_', - '(゚Д゚) [゚ω゚ノ]': 'a', - '(゚Д゚) [゚Θ゚ノ]': 'b', - '(゚Д゚) [\'c\']': 'c', - '(゚Д゚) [゚ー゚ノ]': 'd', - '(゚Д゚) [゚Д゚ノ]': 'e', - '(゚Д゚) [1]': 'f', - '(゚Д゚) [\'o\']': 'o', - '(o゚ー゚o)': 'u', - '(゚Д゚) [\'c\']': 'c', - '((゚ー゚) + (o^_^o))': '7', - '((o^_^o) +(o^_^o) +(c^_^o))': '6', - '((゚ー゚) + (゚Θ゚))': '5', - '(-~3)': '4', - '(-~-~1)': '3', - '(-~1)': '2', - '(-~0)': '1', - '((c^_^o)-(c^_^o))': '0', - } - delim = '(゚Д゚)[゚ε゚]+' - end_token = '(゚Д゚)[゚o゚]' - symbols = '|'.join(map(re.escape, symbol_dict.keys())) - txt = re.sub('(%s)\+\s?' % symbols, lambda m: symbol_dict[m.group(1)], txt) - ret = '' - for aacode in re.findall(r'{0}\+\s?{1}(.*?){0}'.format(re.escape(end_token), re.escape(delim)), txt): - for aachar in aacode.split(delim): - if aachar.isdigit(): - ret += compat_chr(int(aachar, 8)) - else: - m = re.match(r'^u([\da-f]{4})$', aachar) - if m: - ret += compat_chr(int(m.group(1), 16)) - else: - self.report_warning("Cannot decode: %s" % aachar) - return ret + @staticmethod + def _extract_urls(webpage): + return re.findall( + r']+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)', + webpage) def _real_extract(self, url): video_id = self._match_id(url) @@ -107,36 +71,26 @@ class OpenloadIE(InfoExtractor): 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']*>([^<]+)\s*]*>[^<]+\s*]+id="streamurl"', - webpage, 'encrypted data') + ol_id = self._search_regex( + ']+id="[^"]+"[^>]*>([0-9A-Za-z]+)', + webpage, 'openload ID') - enc_code = self._html_search_regex(r']+>(゚ω゚[^<]+)', - webpage, 'encrypted code') + first_char = int(ol_id[0]) + urlcode = [] + num = 1 - js_code = self.openload_decode(enc_code) - jsi = JSInterpreter(js_code) + while num < len(ol_id): + i = ord(ol_id[num]) + key = 0 + if i <= 90: + key = i - 65 + elif i >= 97: + key = 25 + i - 97 + urlcode.append((key, compat_chr(int(ol_id[num + 2:num + 5]) // int(ol_id[num + 1]) - first_char))) + num += 5 - m_offset_fun = self._search_regex(r'slice\(0\s*-\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript offset function') - m_diff_fun = self._search_regex(r'charCodeAt\(0\)\s*\+\s*(%s)\(\)' % _NAME_RE, js_code, 'javascript diff function') - - offset = jsi.call_function(m_offset_fun) - diff = jsi.call_function(m_diff_fun) - - video_url_chars = [] - - for idx, c in enumerate(enc_data): - j = compat_ord(c) - if j >= 33 and j <= 126: - j = ((j + 14) % 94) + 33 - if idx == len(enc_data) - offset: - j += diff - video_url_chars += compat_chr(j) - - video_url = 'https://openload.co/stream/%s?mime=true' % ''.join(video_url_chars) + video_url = 'https://openload.co/stream/' + ''.join( + [value for _, value in sorted(urlcode, key=lambda x: x[0])]) title = self._og_search_title(webpage, default=None) or self._search_regex( r']+class=["\']title["\'][^>]*>([^<]+)', webpage, @@ -152,8 +106,7 @@ class OpenloadIE(InfoExtractor): 'thumbnail': self._og_search_thumbnail(webpage, default=None), 'url': video_url, # Seems all videos have extensions in their titles - 'ext': determine_ext(title), + 'ext': determine_ext(title, 'mp4'), 'subtitles': subtitles, } - return info_dict