X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fopenload.py;h=d8036b54acd8defc75992e912af90a28c99f8d87;hb=1c9c8de29e83bbfc38027dabcd9642f1c41a64b0;hp=71021d573f1fbe7af75b278a9ed519cb73f02517;hpb=2bfeee69b976fe049761dd3012e30b637ee05a58;p=youtube-dl diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 71021d573..d8036b54a 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -5,88 +5,124 @@ import re from .common import InfoExtractor from ..compat import compat_chr -from ..utils import encode_base_n +from ..utils import ( + determine_ext, + ExtractorError, +) class OpenloadIE(InfoExtractor): - _VALID_URL = r'https://openload.co/f/(?P[a-zA-Z0-9]+)' + _VALID_URL = r'https?://(?:openload\.(?:co|io)|oload\.tv)/(?:f|embed)/(?P[a-zA-Z0-9-_]+)' - _TEST = { + _TESTS = [{ 'url': 'https://openload.co/f/kUEfGclsU9o', 'md5': 'bf1c059b004ebc7a256f89408e65c36e', 'info_dict': { 'id': 'kUEfGclsU9o', 'ext': 'mp4', 'title': 'skyrim_no-audio_1080.mp4', + 'thumbnail': r're:^https?://.*\.jpg$', }, - } + }, { + 'url': 'https://openload.co/embed/rjC09fkPLYs', + 'info_dict': { + 'id': 'rjC09fkPLYs', + 'ext': 'mp4', + 'title': 'movie.mp4', + 'thumbnail': r're:^https?://.*\.jpg$', + '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, + }, { + # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout + # for title and ext + 'url': 'https://openload.co/embed/Sxz5sADo82g/', + 'only_matching': True, + }, { + 'url': 'https://oload.tv/embed/KnG-kKZdcfY/', + 'only_matching': True, + }] @staticmethod - def openload_level2_debase(m): - radix, num = int(m.group(1)) + 27, int(m.group(2)) - return '"' + encode_base_n(num, radix) + '"' + def _extract_urls(webpage): + return re.findall( + r']+src=["\']((?:https?://)?(?:openload\.(?:co|io)|oload\.tv)/embed/[a-zA-Z0-9-_]+)', + webpage) - @classmethod - def openload_level2(cls, txt): - # The function name is ǃ \u01c3 - # Using escaped unicode literals does not work in Python 3.2 - return re.sub(r'ǃ\((\d+),(\d+)\)', cls.openload_level2_debase, txt, re.UNICODE).replace('"+"', '') + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage('https://openload.co/embed/%s/' % video_id, video_id) - # Openload uses a variant of aadecode - # openload_decode and related functions are originally written by - # vitas@matfyz.cz and released with public domain - # See https://github.com/rg3/youtube-dl/issues/8489 - @classmethod - def openload_decode(cls, txt): - symbol_table = [ - ('_', '(ﾟДﾟ) [ﾟΘﾟ]'), - ('a', '(ﾟДﾟ) [ﾟωﾟﾉ]'), - ('b', '(ﾟДﾟ) [ﾟΘﾟﾉ]'), - ('c', '(ﾟДﾟ) [\'c\']'), - ('d', '(ﾟДﾟ) [ﾟｰﾟﾉ]'), - ('e', '(ﾟДﾟ) [ﾟДﾟﾉ]'), - ('f', '(ﾟДﾟ) [1]'), + if 'File not found' in webpage or 'deleted by the owner' in webpage: + raise ExtractorError('File not found', expected=True) - ('o', '(ﾟДﾟ) [\'o\']'), - ('u', '(oﾟｰﾟo)'), - ('c', '(ﾟДﾟ) [\'c\']'), + ol_id = self._search_regex( + ']+id="[^"]+"[^>]*>([0-9A-Za-z]+)', + webpage, 'openload ID') - ('7', '((ﾟｰﾟ) + (o^_^o))'), - ('6', '((o^_^o) +(o^_^o) +(c^_^o))'), - ('5', '((ﾟｰﾟ) + (ﾟΘﾟ))'), - ('4', '(-~3)'), - ('3', '(-~-~1)'), - ('2', '(-~1)'), - ('1', '(-~0)'), - ('0', '((c^_^o)-(c^_^o))'), - ] - delim = '(ﾟДﾟ)[ﾟεﾟ]+' - ret = '' - for aachar in txt.split(delim): - for val, pat in symbol_table: - aachar = aachar.replace(pat, val) - aachar = aachar.replace('+ ', '') - m = re.match(r'^\d+', aachar) - if m: - ret += compat_chr(int(m.group(0), 8)) - else: - m = re.match(r'^u([\da-f]+)', aachar) - if m: - ret += compat_chr(int(m.group(1), 16)) - return cls.openload_level2(ret) + decoded = '' + a = ol_id[0:24] + b = [] + for i in range(0, len(a), 8): + b.append(int(a[i:i + 8] or '0', 16)) + ol_id = ol_id[24:] + j = 0 + k = 0 + while j < len(ol_id): + c = 128 + d = 0 + e = 0 + f = 0 + _more = True + while _more: + if j + 1 >= len(ol_id): + c = 143 + f = int(ol_id[j:j + 2] or '0', 16) + j += 2 + d += (f & 127) << e + e += 7 + _more = f >= c + g = d ^ b[k % 3] + for i in range(4): + char_dec = (g >> 8 * i) & (c + 127) + char = compat_chr(char_dec) + if char != '#': + decoded += char + k += 1 - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - code = self._search_regex( - r']+>\s*]+>([^<]+)', - webpage, 'JS code') + video_url = 'https://openload.co/stream/%s?mime=true' + video_url = video_url % decoded + + title = self._og_search_title(webpage, default=None) or self._search_regex( + r']+class=["\']title["\'][^>]*>([^<]+)', webpage, + 'title', default=None) or self._html_search_meta( + 'description', webpage, 'title', fatal=True) - video_url = self._search_regex( - r'return\s+"(https?://[^"]+)"', self.openload_decode(code), 'video URL') + entries = self._parse_html5_media_entries(url, webpage, video_id) + subtitles = entries[0]['subtitles'] if entries else None - return { + info_dict = { 'id': video_id, - 'title': self._og_search_title(webpage), + 'title': title, + 'thumbnail': self._og_search_thumbnail(webpage, default=None), 'url': video_url, + # Seems all videos have extensions in their titles + 'ext': determine_ext(title, 'mp4'), + 'subtitles': subtitles, } + return info_dict