X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fopenload.py;h=292476ef86cf8680fc59e58488f84cae44da7947;hb=a88d461dff67205fcec684426afbcbeb4b0e7cf5;hp=1b57462b5072c36c2d6bd52c93146a78e1eb4ec2;hpb=e2713d32f49f1bfa830cc755a96691c39da88290;p=youtube-dl.git diff --git a/youtube_dl/extractor/openload.py b/youtube_dl/extractor/openload.py index 1b57462b5..292476ef8 100644 --- a/youtube_dl/extractor/openload.py +++ b/youtube_dl/extractor/openload.py @@ -4,17 +4,16 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_chr from ..utils import ( determine_ext, - encode_base_n, ExtractorError, - mimetype2ext, + get_element_by_id, + PhantomJSwrapper, ) 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', @@ -23,7 +22,23 @@ 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', + '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', @@ -31,97 +46,63 @@ class OpenloadIE(InfoExtractor): }, { '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, }] + _USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36' + @staticmethod - def openload_level2_debase(m): - radix, num = int(m.group(1)) + 27, int(m.group(2)) - return '"' + encode_base_n(num, radix) + '"' - - @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('"+"', '') - - # 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]'), - - ('o', '(ﾟДﾟ) [\'o\']'), - ('u', '(oﾟｰﾟo)'), - ('c', '(ﾟДﾟ) [\'c\']'), - - ('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) + 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) - webpage = self._download_webpage(url, video_id) + url = 'https://openload.co/embed/%s/' % video_id + headers = { + 'User-Agent': self._USER_AGENT, + } + + webpage = self._download_webpage(url, video_id, headers=headers) - if 'File not found' in webpage: - raise ExtractorError('File not found', expected=True) + if 'File not found' in webpage or 'deleted by the owner' in webpage: + raise ExtractorError('File not found', expected=True, video_id=video_id) - code = self._search_regex( - r'\s*\s*]+>[^>]+\s*]+>([^<]+)', - webpage, 'JS code') + phantom = PhantomJSwrapper(self, required_version='2.0') + webpage, _ = phantom.get(url, html=webpage, video_id=video_id, headers=headers) - decoded = self.openload_decode(code) + decoded_id = get_element_by_id('streamurl', webpage) - video_url = self._search_regex( - r'return\s+"(https?://[^"]+)"', decoded, 'video URL') + video_url = 'https://openload.co/stream/%s?mime=true' % decoded_id 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) - ext = mimetype2ext(self._search_regex( - r'window\.vt\s*=\s*(["\'])(?P.+?)\1', decoded, - 'mimetype', default=None, group='mimetype')) or determine_ext( - video_url, 'mp4') + entries = self._parse_html5_media_entries(url, webpage, video_id) + entry = entries[0] if entries else {} + subtitles = entry.get('subtitles') - return { + info_dict = { 'id': video_id, 'title': title, - 'ext': ext, - 'thumbnail': self._og_search_thumbnail(webpage, default=None), + 'thumbnail': entry.get('thumbnail') or 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, + 'http_headers': headers, } + return info_dict