From: Sergey M․ Date: Sun, 25 Feb 2018 11:25:40 +0000 (+0700) Subject: [streamango] Fix formats extraction, improve and simplify (closes #14256) X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=c106237d5652e659071114f9d16661df335ae93c;p=youtube-dl [streamango] Fix formats extraction, improve and simplify (closes #14256) --- diff --git a/youtube_dl/extractor/streamango.py b/youtube_dl/extractor/streamango.py index 58c03f11a..08d84495a 100644 --- a/youtube_dl/extractor/streamango.py +++ b/youtube_dl/extractor/streamango.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_chr from ..utils import ( determine_ext, int_or_none, @@ -38,26 +39,26 @@ class StreamangoIE(InfoExtractor): }] def _real_extract(self, url): - def decrypt_src(str_, val): - k = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA' - str_ = re.sub(r'[^A-Za-z0-9+/=]', '', str_) - src = '' + def decrypt_src(encoded, val): + ALPHABET = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA' + encoded = re.sub(r'[^A-Za-z0-9+/=]', '', encoded) + decoded = '' sm = [None] * 4 i = 0 - str_len = len(str_) + str_len = len(encoded) while i < str_len: for j in range(4): - sm[j % 4] = k.index(str_[i]) + sm[j % 4] = ALPHABET.index(encoded[i]) i += 1 - charCode = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val - src += chr(charCode) - if (sm[2] != 0x40): - charCode = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2) - src += chr(charCode) - if (sm[3] != 0x40): - charCode = ((sm[2] & 0x3) << 0x6) | sm[3] - src += chr(charCode) - return src + char_code = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val + decoded += compat_chr(char_code) + if sm[2] != 0x40: + char_code = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2) + decoded += compat_chr(char_code) + if sm[3] != 0x40: + char_code = ((sm[2] & 0x3) << 0x6) | sm[3] + decoded += compat_chr(char_code) + return decoded video_id = self._match_id(url) @@ -66,22 +67,27 @@ class StreamangoIE(InfoExtractor): title = self._og_search_title(webpage, default=video_id) formats = [] - for format_ in re.findall(r'\(\s*({[^}]*\bsrc\s*:\s*[^}]*})', webpage): - mobj = re.search(r'(src\s*:\s*[^(]\(([^)]*)\)[\s,]*)', format_) + for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage): + mobj = re.search(r'(src\s*:\s*[^(]+\(([^)]*)\)[\s,]*)', format_) if mobj is None: continue + format_ = format_.replace(mobj.group(0), '') video = self._parse_json( - format_, video_id, transform_source=js_to_json, fatal=False) - if not video: - continue + format_, video_id, transform_source=js_to_json, + fatal=False) or {} - mobj = re.search(r'[\'"](?P[^\'"]+)[\'"]\s*,\s*(?P\d+)', mobj.group(1)) + mobj = re.search( + r'([\'"])(?P(?:(?!\1).)+)\1\s*,\s*(?P\d+)', + mobj.group(1)) if mobj is None: continue src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val'))) + if not src: + continue + ext = determine_ext(src, default_ext=None) if video.get('type') == 'application/dash+xml' or ext == 'mpd': formats.extend(self._extract_mpd_formats(