Raphaël G. Git Repositories - youtubedl/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     ExtractorError,
  11     int_or_none,
  12     js_to_json,
  13 )
  14
  15
  16 class StreamangoIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?streamango\.com/(?:f|embed)/(?P<id>[^/?#&]+)'
  18     _TESTS = [{
  19         'url': 'https://streamango.com/f/clapasobsptpkdfe/20170315_150006_mp4',
  20         'md5': 'e992787515a182f55e38fc97588d802a',
  21         'info_dict': {
  22             'id': 'clapasobsptpkdfe',
  23             'ext': 'mp4',
  24             'title': '20170315_150006.mp4',
  25         }
  26     }, {
  27         # no og:title
  28         'url': 'https://streamango.com/embed/foqebrpftarclpob/asdf_asd_2_mp4',
  29         'info_dict': {
  30             'id': 'foqebrpftarclpob',
  31             'ext': 'mp4',
  32             'title': 'foqebrpftarclpob',
  33         },
  34         'params': {
  35             'skip_download': True,
  36         },
  37         'skip': 'gone',
  38     }, {
  39         'url': 'https://streamango.com/embed/clapasobsptpkdfe/20170315_150006_mp4',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         def decrypt_src(encoded, val):
  45             ALPHABET = '=/+9876543210zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA'
  46             encoded = re.sub(r'[^A-Za-z0-9+/=]', '', encoded)
  47             decoded = ''
  48             sm = [None] * 4
  49             i = 0
  50             str_len = len(encoded)
  51             while i < str_len:
  52                 for j in range(4):
  53                     sm[j % 4] = ALPHABET.index(encoded[i])
  54                     i += 1
  55                 char_code = ((sm[0] << 0x2) | (sm[1] >> 0x4)) ^ val
  56                 decoded += compat_chr(char_code)
  57                 if sm[2] != 0x40:
  58                     char_code = ((sm[1] & 0xf) << 0x4) | (sm[2] >> 0x2)
  59                     decoded += compat_chr(char_code)
  60                 if sm[3] != 0x40:
  61                     char_code = ((sm[2] & 0x3) << 0x6) | sm[3]
  62                     decoded += compat_chr(char_code)
  63             return decoded
  64
  65         video_id = self._match_id(url)
  66
  67         webpage = self._download_webpage(url, video_id)
  68
  69         title = self._og_search_title(webpage, default=video_id)
  70
  71         formats = []
  72         for format_ in re.findall(r'({[^}]*\bsrc\s*:\s*[^}]*})', webpage):
  73             mobj = re.search(r'(src\s*:\s*[^(]+\(([^)]*)\)[\s,]*)', format_)
  74             if mobj is None:
  75                 continue
  76
  77             format_ = format_.replace(mobj.group(0), '')
  78
  79             video = self._parse_json(
  80                 format_, video_id, transform_source=js_to_json,
  81                 fatal=False) or {}
  82
  83             mobj = re.search(
  84                 r'([\'"])(?P<src>(?:(?!\1).)+)\1\s*,\s*(?P<val>\d+)',
  85                 mobj.group(1))
  86             if mobj is None:
  87                 continue
  88
  89             src = decrypt_src(mobj.group('src'), int_or_none(mobj.group('val')))
  90             if not src:
  91                 continue
  92
  93             ext = determine_ext(src, default_ext=None)
  94             if video.get('type') == 'application/dash+xml' or ext == 'mpd':
  95                 formats.extend(self._extract_mpd_formats(
  96                     src, video_id, mpd_id='dash', fatal=False))
  97             else:
  98                 formats.append({
  99                     'url': src,
 100                     'ext': ext or 'mp4',
 101                     'width': int_or_none(video.get('width')),
 102                     'height': int_or_none(video.get('height')),
 103                     'tbr': int_or_none(video.get('bitrate')),
 104                 })
 105
 106         if not formats:
 107             error = self._search_regex(
 108                 r'<p[^>]+\bclass=["\']lead[^>]+>(.+?)</p>', webpage,
 109                 'error', default=None)
 110             if not error and '>Sorry' in webpage:
 111                 error = 'Video %s is not available' % video_id
 112             if error:
 113                 raise ExtractorError(error, expected=True)
 114
 115         self._sort_formats(formats)
 116
 117         return {
 118             'id': video_id,
 119             'url': url,
 120             'title': title,
 121             'formats': formats,
 122         }