Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vvvvid.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     str_or_none,
  11 )
  12
  13
  14 class VVVVIDIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?vvvvid\.it/#!(?:show|anime|film|series)/(?P<show_id>\d+)/[^/]+/(?P<season_id>\d+)/(?P<id>[0-9]+)'
  16     _TESTS = [{
  17         # video_type == 'video/vvvvid'
  18         'url': 'https://www.vvvvid.it/#!show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048/ping-pong',
  19         'md5': 'b8d3cecc2e981adc3835adf07f6df91b',
  20         'info_dict': {
  21             'id': '489048',
  22             'ext': 'mp4',
  23             'title': 'Ping Pong',
  24         },
  25     }, {
  26         # video_type == 'video/rcs'
  27         'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
  28         'md5': '33e0edfba720ad73a8782157fdebc648',
  29         'info_dict': {
  30             'id': '482493',
  31             'ext': 'mp4',
  32             'title': 'Episodio 01',
  33         },
  34     }]
  35     _conn_id = None
  36
  37     def _real_initialize(self):
  38         self._conn_id = self._download_json(
  39             'https://www.vvvvid.it/user/login',
  40             None, headers=self.geo_verification_headers())['data']['conn_id']
  41
  42     def _real_extract(self, url):
  43         show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()
  44         response = self._download_json(
  45             'https://www.vvvvid.it/vvvvid/ondemand/%s/season/%s' % (show_id, season_id),
  46             video_id, headers=self.geo_verification_headers(), query={
  47                 'conn_id': self._conn_id,
  48             })
  49         if response['result'] == 'error':
  50             raise ExtractorError('%s said: %s' % (
  51                 self.IE_NAME, response['message']), expected=True)
  52
  53         vid = int(video_id)
  54         video_data = list(filter(
  55             lambda episode: episode.get('video_id') == vid, response['data']))[0]
  56         formats = []
  57
  58         # vvvvid embed_info decryption algorithm is reverse engineered from function $ds(h) at vvvvid.js
  59         def ds(h):
  60             g = "MNOPIJKL89+/4567UVWXQRSTEFGHABCDcdefYZabstuvopqr0123wxyzklmnghij"
  61
  62             def f(m):
  63                 l = []
  64                 o = 0
  65                 b = False
  66                 m_len = len(m)
  67                 while ((not b) and o < m_len):
  68                     n = m[o] << 2
  69                     o += 1
  70                     k = -1
  71                     j = -1
  72                     if o < m_len:
  73                         n += m[o] >> 4
  74                         o += 1
  75                         if o < m_len:
  76                             k = (m[o - 1] << 4) & 255
  77                             k += m[o] >> 2
  78                             o += 1
  79                             if o < m_len:
  80                                 j = (m[o - 1] << 6) & 255
  81                                 j += m[o]
  82                                 o += 1
  83                             else:
  84                                 b = True
  85                         else:
  86                             b = True
  87                     else:
  88                         b = True
  89                     l.append(n)
  90                     if k != -1:
  91                         l.append(k)
  92                     if j != -1:
  93                         l.append(j)
  94                 return l
  95
  96             c = []
  97             for e in h:
  98                 c.append(g.index(e))
  99
 100             c_len = len(c)
 101             for e in range(c_len * 2 - 1, -1, -1):
 102                 a = c[e % c_len] ^ c[(e + 1) % c_len]
 103                 c[e % c_len] = a
 104
 105             c = f(c)
 106             d = ''
 107             for e in c:
 108                 d += chr(e)
 109
 110             return d
 111
 112         for quality in ('_sd', ''):
 113             embed_code = video_data.get('embed_info' + quality)
 114             if not embed_code:
 115                 continue
 116             embed_code = ds(embed_code)
 117             video_type = video_data.get('video_type')
 118             if video_type in ('video/rcs', 'video/kenc'):
 119                 formats.extend(self._extract_akamai_formats(
 120                     embed_code, video_id))
 121             else:
 122                 formats.extend(self._extract_wowza_formats(
 123                     'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
 124         self._sort_formats(formats)
 125
 126         return {
 127             'id': video_id,
 128             'title': video_data['title'],
 129             'formats': formats,
 130             'thumbnail': video_data.get('thumbnail'),
 131             'duration': int_or_none(video_data.get('length')),
 132             'series': video_data.get('show_title'),
 133             'season_id': season_id,
 134             'season_number': video_data.get('season_number'),
 135             'episode_id': str_or_none(video_data.get('id')),
 136             'epidode_number': int_or_none(video_data.get('number')),
 137             'episode_title': video_data['title'],
 138             'view_count': int_or_none(video_data.get('views')),
 139             'like_count': int_or_none(video_data.get('video_likes')),
 140         }