_ Git - youtube-dl/blob - youtube_dl/extractor/streamango.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_chr
   8 from ..utils import (
   9     determine_ext,
  10     int_or_none,
  11     js_to_json,
  12 )
  13
  14
  15 class StreamangoIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?streamango\.com/(?:f|embed)/(?P<id>[^/?#&]+)'
  17     _TESTS = [{
  18         'url': 'https://streamango.com/f/clapasobsptpkdfe/20170315_150006_mp4',
  19         'md5': 'e992787515a182f55e38fc97588d802a',
  20         'info_dict': {
  21             'id': 'clapasobsptpkdfe',
  22             'ext': 'mp4',
  23             'title': '20170315_150006.mp4',
  24         }
  25     }, {
  26         # no og:title
  27         'url': 'https://streamango.com/embed/foqebrpftarclpob/asdf_asd_2_mp4',
  28         'info_dict': {
  29             'id': 'foqebrpftarclpob',
  30             'ext': 'mp4',
  31             'title': 'foqebrpftarclpob',
  32         },
  33         'params': {
  34             'skip_download': True,
  35         },
  36     }, {
  37         'url': 'https://streamango.com/embed/clapasobsptpkdfe/20170315_150006_mp4',
  38         'only_matching': True,
  39     }]
  40
  41     def _real_extract(self, url):
  42         def decrypt_src(encoded, val):
  43             ALPHABET = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA'
  44             encoded = re.sub(r'[^A-Za-z0-9+/=]', '', encoded)
  45             decoded = ''
  46             sm = [None] * 4
  47             i = 0
  48             str_len = len(encoded)
  49             while i < str_len:
  50                 for j in range(4):
  51                     sm[j % 4] = ALPHABET.index(encoded[i])
  52                     i += 1
  53                 char_code = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val
  54                 decoded += compat_chr(char_code)
  55                 if sm[2] != 0x40:
  56                     char_code = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2)
  57                     decoded += compat_chr(char_code)
  58                 if sm[3] != 0x40:
  59                     char_code = ((sm[2] & 0x3) << 0x6) | sm[3]
  60                     decoded += compat_chr(char_code)
  61             return decoded
  62
  63         video_id = self._match_id(url)
  64
  65         webpage = self._download_webpage(url, video_id)
  66
  67         title = self._og_search_title(webpage, default=video_id)
  68
  69         formats = []
  70         for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
  71             mobj = re.search(r'(src\s*:\s*[^(]+\(([^)]*)\)[\s,]*)', format_)
  72             if mobj is None:
  73                 continue
  74
  75             format_ = format_.replace(mobj.group(0), '')
  76
  77             video = self._parse_json(
  78                 format_, video_id, transform_source=js_to_json,
  79                 fatal=False) or {}
  80
  81             mobj = re.search(
  82                 r'([\'"])(?P<src>(?:(?!\1).)+)\1\s*,\s*(?P<val>\d+)',
  83                 mobj.group(1))
  84             if mobj is None:
  85                 continue
  86
  87             src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val')))
  88             if not src:
  89                 continue
  90
  91             ext = determine_ext(src, default_ext=None)
  92             if video.get('type') == 'application/dash+xml' or ext == 'mpd':
  93                 formats.extend(self._extract_mpd_formats(
  94                     src, video_id, mpd_id='dash', fatal=False))
  95             else:
  96                 formats.append({
  97                     'url': src,
  98                     'ext': ext or 'mp4',
  99                     'width': int_or_none(video.get('width')),
 100                     'height': int_or_none(video.get('height')),
 101                     'tbr': int_or_none(video.get('bitrate')),
 102                 })
 103         self._sort_formats(formats)
 104
 105         return {
 106             'id': video_id,
 107             'url': url,
 108             'title': title,
 109             'formats': formats,
 110         }