[streamango] Fix formats extraction, improve and simplify (closes #14256)

master
Sergey M․ 6 years ago
parent 6e72ea4775
commit c106237d56
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D
  1. 48
      youtube_dl/extractor/streamango.py

@ -4,6 +4,7 @@ from __future__ import unicode_literals
import re import re
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import compat_chr
from ..utils import ( from ..utils import (
determine_ext, determine_ext,
int_or_none, int_or_none,
@ -38,26 +39,26 @@ class StreamangoIE(InfoExtractor):
}] }]
def _real_extract(self, url): def _real_extract(self, url):
def decrypt_src(str_, val): def decrypt_src(encoded, val):
k = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA' ALPHABET = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA'
str_ = re.sub(r'[^A-Za-z0-9+/=]', '', str_) encoded = re.sub(r'[^A-Za-z0-9+/=]', '', encoded)
src = '' decoded = ''
sm = [None] * 4 sm = [None] * 4
i = 0 i = 0
str_len = len(str_) str_len = len(encoded)
while i < str_len: while i < str_len:
for j in range(4): for j in range(4):
sm[j % 4] = k.index(str_[i]) sm[j % 4] = ALPHABET.index(encoded[i])
i += 1 i += 1
charCode = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val char_code = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val
src += chr(charCode) decoded += compat_chr(char_code)
if (sm[2] != 0x40): if sm[2] != 0x40:
charCode = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2) char_code = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2)
src += chr(charCode) decoded += compat_chr(char_code)
if (sm[3] != 0x40): if sm[3] != 0x40:
charCode = ((sm[2] & 0x3) << 0x6) | sm[3] char_code = ((sm[2] & 0x3) << 0x6) | sm[3]
src += chr(charCode) decoded += compat_chr(char_code)
return src return decoded
video_id = self._match_id(url) video_id = self._match_id(url)
@ -66,22 +67,27 @@ class StreamangoIE(InfoExtractor):
title = self._og_search_title(webpage, default=video_id) title = self._og_search_title(webpage, default=video_id)
formats = [] formats = []
for format_ in re.findall(r'\(\s*({[^}]*\bsrc\s*:\s*[^}]*})', webpage): for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
mobj = re.search(r'(src\s*:\s*[^(]\(([^)]*)\)[\s,]*)', format_) mobj = re.search(r'(src\s*:\s*[^(]+\(([^)]*)\)[\s,]*)', format_)
if mobj is None: if mobj is None:
continue continue
format_ = format_.replace(mobj.group(0), '') format_ = format_.replace(mobj.group(0), '')
video = self._parse_json( video = self._parse_json(
format_, video_id, transform_source=js_to_json, fatal=False) format_, video_id, transform_source=js_to_json,
if not video: fatal=False) or {}
continue
mobj = re.search(r'[\'"](?P<src>[^\'"]+)[\'"]\s*,\s*(?P<val>\d+)', mobj.group(1)) mobj = re.search(
r'([\'"])(?P<src>(?:(?!\1).)+)\1\s*,\s*(?P<val>\d+)',
mobj.group(1))
if mobj is None: if mobj is None:
continue continue
src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val'))) src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val')))
if not src:
continue
ext = determine_ext(src, default_ext=None) ext = determine_ext(src, default_ext=None)
if video.get('type') == 'application/dash+xml' or ext == 'mpd': if video.get('type') == 'application/dash+xml' or ext == 'mpd':
formats.extend(self._extract_mpd_formats( formats.extend(self._extract_mpd_formats(

Loading…
Cancel
Save