From 6c20a0bb99e626db6870747b6329ad9c9064c123 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Fri, 9 Dec 2016 02:15:16 +0800 Subject: [PATCH] [openload] Fix extraction (closes #10408) --- ChangeLog | 1 + youtube_dl/extractor/openload.py | 90 +++++--------------------------- 2 files changed, 14 insertions(+), 77 deletions(-) diff --git a/ChangeLog b/ChangeLog index bf5f26943..9d7de1f95 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,6 +1,7 @@ version Extractors +* [openload] Fix extraction (#10408) + [thisoldhouse] Recognize /tv-episode/ URLs (#11271) version 2016.12.01 diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 7f19b1ba5..84aa12585 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -1,21 +1,12 @@ # coding: utf-8 -from __future__ import unicode_literals, division - -import re +from __future__ import unicode_literals 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): @@ -62,44 +53,6 @@ class OpenloadIE(InfoExtractor): '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 - def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id) @@ -107,36 +60,20 @@ 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="[a-zA-Z0-9]+x"[^>]*>([0-9]+)', + webpage, 'openload ID') - enc_code = self._html_search_regex(r']+>(゚ω゚[^<]+)', - webpage, 'encrypted code') + first_two_chars = int(float(ol_id[0:][:2])) + urlcode = '' + num = 2 - js_code = self.openload_decode(enc_code) - jsi = JSInterpreter(js_code) + while num < len(ol_id): + urlcode += compat_chr(int(float(ol_id[num:][:3])) - + first_two_chars * int(float(ol_id[num + 3:][:2]))) + 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/' + urlcode title = self._og_search_title(webpage, default=None) or self._search_regex( r']+class=["\']title["\'][^>]*>([^<]+)', webpage, @@ -155,5 +92,4 @@ class OpenloadIE(InfoExtractor): 'ext': determine_ext(title), 'subtitles': subtitles, } - return info_dict