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         'params': {
  26             'skip_download': True,
  27         },
  28     }, {
  29         # video_type == 'video/rcs'
  30         'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
  31         'md5': '33e0edfba720ad73a8782157fdebc648',
  32         'info_dict': {
  33             'id': '482493',
  34             'ext': 'mp4',
  35             'title': 'Episodio 01',
  36         },
  37         'params': {
  38             'skip_download': True,
  39         },
  40     }]
  41     _conn_id = None
  42
  43     def _real_initialize(self):
  44         self._conn_id = self._download_json(
  45             'https://www.vvvvid.it/user/login',
  46             None, headers=self.geo_verification_headers())['data']['conn_id']
  47
  48     def _real_extract(self, url):
  49         show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()
  50         response = self._download_json(
  51             'https://www.vvvvid.it/vvvvid/ondemand/%s/season/%s' % (show_id, season_id),
  52             video_id, headers=self.geo_verification_headers(), query={
  53                 'conn_id': self._conn_id,
  54             })
  55         if response['result'] == 'error':
  56             raise ExtractorError('%s said: %s' % (
  57                 self.IE_NAME, response['message']), expected=True)
  58
  59         vid = int(video_id)
  60         video_data = list(filter(
  61             lambda episode: episode.get('video_id') == vid, response['data']))[0]
  62         formats = []
  63
  64         # vvvvid embed_info decryption algorithm is reverse engineered from function $ds(h) at vvvvid.js
  65         def ds(h):
  66             g = "MNOPIJKL89+/4567UVWXQRSTEFGHABCDcdefYZabstuvopqr0123wxyzklmnghij"
  67
  68             def f(m):
  69                 l = []
  70                 o = 0
  71                 b = False
  72                 m_len = len(m)
  73                 while ((not b) and o < m_len):
  74                     n = m[o] << 2
  75                     o += 1
  76                     k = -1
  77                     j = -1
  78                     if o < m_len:
  79                         n += m[o] >> 4
  80                         o += 1
  81                         if o < m_len:
  82                             k = (m[o - 1] << 4) & 255
  83                             k += m[o] >> 2
  84                             o += 1
  85                             if o < m_len:
  86                                 j = (m[o - 1] << 6) & 255
  87                                 j += m[o]
  88                                 o += 1
  89                             else:
  90                                 b = True
  91                         else:
  92                             b = True
  93                     else:
  94                         b = True
  95                     l.append(n)
  96                     if k != -1:
  97                         l.append(k)
  98                     if j != -1:
  99                         l.append(j)
 100                 return l
 101
 102             c = []
 103             for e in h:
 104                 c.append(g.index(e))
 105
 106             c_len = len(c)
 107             for e in range(c_len * 2 - 1, -1, -1):
 108                 a = c[e % c_len] ^ c[(e + 1) % c_len]
 109                 c[e % c_len] = a
 110
 111             c = f(c)
 112             d = ''
 113             for e in c:
 114                 d += chr(e)
 115
 116             return d
 117
 118         for quality in ('_sd', ''):
 119             embed_code = video_data.get('embed_info' + quality)
 120             if not embed_code:
 121                 continue
 122             embed_code = ds(embed_code)
 123             video_type = video_data.get('video_type')
 124             if video_type in ('video/rcs', 'video/kenc'):
 125                 embed_code = re.sub(r'https?://([^/]+)/z/', r'https://\1/i/', embed_code).replace('/manifest.f4m', '/master.m3u8')
 126                 if video_type == 'video/kenc':
 127                     kenc = self._download_json(
 128                         'https://www.vvvvid.it/kenc', video_id, query={
 129                             'action': 'kt',
 130                             'conn_id': self._conn_id,
 131                             'url': embed_code,
 132                         }, fatal=False) or {}
 133                     kenc_message = kenc.get('message')
 134                     if kenc_message:
 135                         embed_code += '?' + ds(kenc_message)
 136                 formats.extend(self._extract_m3u8_formats(
 137                     embed_code, video_id, 'mp4',
 138                     m3u8_id='hls', fatal=False))
 139             else:
 140                 formats.extend(self._extract_wowza_formats(
 141                     'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
 142         self._sort_formats(formats)
 143
 144         return {
 145             'id': video_id,
 146             'title': video_data['title'],
 147             'formats': formats,
 148             'thumbnail': video_data.get('thumbnail'),
 149             'duration': int_or_none(video_data.get('length')),
 150             'series': video_data.get('show_title'),
 151             'season_id': season_id,
 152             'season_number': video_data.get('season_number'),
 153             'episode_id': str_or_none(video_data.get('id')),
 154             'episode_number': int_or_none(video_data.get('number')),
 155             'episode_title': video_data['title'],
 156             'view_count': int_or_none(video_data.get('views')),
 157             'like_count': int_or_none(video_data.get('video_likes')),
 158         }