youtube_dlc/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_BASE = r'https?://(?:www\.)?vvvvid\.it/(?:#!)?(?:show|anime|film|series)/'
  16     _VALID_URL = r'%s(?P<show_id>\d+)/[^/]+/(?P<season_id>\d+)/(?P<id>[0-9]+)' % _VALID_URL_BASE
  17     _TESTS = [{
  18         # video_type == 'video/vvvvid'
  19         'url': 'https://www.vvvvid.it/#!show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048/ping-pong',
  20         'md5': 'b8d3cecc2e981adc3835adf07f6df91b',
  21         'info_dict': {
  22             'id': '489048',
  23             'ext': 'mp4',
  24             'title': 'Ping Pong',
  25             'duration': 239,
  26             'series': '"Perché dovrei guardarlo?" di Dario Moccia',
  27             'season_id': '437',
  28             'episode': 'Ping Pong',
  29             'episode_number': 1,
  30             'episode_id': '3334',
  31             'view_count': int,
  32             'like_count': int,
  33             'repost_count': int,
  34         },
  35         'params': {
  36             'skip_download': True,
  37         },
  38     }, {
  39         # video_type == 'video/rcs'
  40         'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
  41         'md5': '33e0edfba720ad73a8782157fdebc648',
  42         'info_dict': {
  43             'id': '482493',
  44             'ext': 'mp4',
  45             'title': 'Episodio 01',
  46         },
  47         'params': {
  48             'skip_download': True,
  49         },
  50     }, {
  51         'url': 'https://www.vvvvid.it/show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048',
  52         'only_matching': True
  53     }]
  54     _conn_id = None
  55
  56     def _real_initialize(self):
  57         self._conn_id = self._download_json(
  58             'https://www.vvvvid.it/user/login',
  59             None, headers=self.geo_verification_headers())['data']['conn_id']
  60
  61     def _download_info(self, show_id, path, video_id, fatal=True):
  62         response = self._download_json(
  63             'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path),
  64             video_id, headers=self.geo_verification_headers(), query={
  65                 'conn_id': self._conn_id,
  66             }, fatal=fatal)
  67         if not (response or fatal):
  68             return
  69         if response.get('result') == 'error':
  70             raise ExtractorError('%s said: %s' % (
  71                 self.IE_NAME, response['message']), expected=True)
  72         return response['data']
  73
  74     def _extract_common_video_info(self, video_data):
  75         return {
  76             'thumbnail': video_data.get('thumbnail'),
  77             'episode_id': str_or_none(video_data.get('id')),
  78         }
  79
  80     def _real_extract(self, url):
  81         show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()
  82
  83         response = self._download_info(
  84             show_id, 'season/%s' % season_id, video_id)
  85
  86         vid = int(video_id)
  87         video_data = list(filter(
  88             lambda episode: episode.get('video_id') == vid, response))[0]
  89         title = video_data['title']
  90         formats = []
  91
  92         # vvvvid embed_info decryption algorithm is reverse engineered from function $ds(h) at vvvvid.js
  93         def ds(h):
  94             g = "MNOPIJKL89+/4567UVWXQRSTEFGHABCDcdefYZabstuvopqr0123wxyzklmnghij"
  95
  96             def f(m):
  97                 l = []
  98                 o = 0
  99                 b = False
 100                 m_len = len(m)
 101                 while ((not b) and o < m_len):
 102                     n = m[o] << 2
 103                     o += 1
 104                     k = -1
 105                     j = -1
 106                     if o < m_len:
 107                         n += m[o] >> 4
 108                         o += 1
 109                         if o < m_len:
 110                             k = (m[o - 1] << 4) & 255
 111                             k += m[o] >> 2
 112                             o += 1
 113                             if o < m_len:
 114                                 j = (m[o - 1] << 6) & 255
 115                                 j += m[o]
 116                                 o += 1
 117                             else:
 118                                 b = True
 119                         else:
 120                             b = True
 121                     else:
 122                         b = True
 123                     l.append(n)
 124                     if k != -1:
 125                         l.append(k)
 126                     if j != -1:
 127                         l.append(j)
 128                 return l
 129
 130             c = []
 131             for e in h:
 132                 c.append(g.index(e))
 133
 134             c_len = len(c)
 135             for e in range(c_len * 2 - 1, -1, -1):
 136                 a = c[e % c_len] ^ c[(e + 1) % c_len]
 137                 c[e % c_len] = a
 138
 139             c = f(c)
 140             d = ''
 141             for e in c:
 142                 d += chr(e)
 143
 144             return d
 145
 146         info = {}
 147
 148         def metadata_from_url(r_url):
 149             if not info and r_url:
 150                 mobj = re.search(r'_(?:S(\d+))?Ep(\d+)', r_url)
 151                 if mobj:
 152                     info['episode_number'] = int(mobj.group(2))
 153                     season_number = mobj.group(1)
 154                     if season_number:
 155                         info['season_number'] = int(season_number)
 156
 157         for quality in ('_sd', ''):
 158             embed_code = video_data.get('embed_info' + quality)
 159             if not embed_code:
 160                 continue
 161             embed_code = ds(embed_code)
 162             video_type = video_data.get('video_type')
 163             if video_type in ('video/rcs', 'video/kenc'):
 164                 if video_type == 'video/kenc':
 165                     kenc = self._download_json(
 166                         'https://www.vvvvid.it/kenc', video_id, query={
 167                             'action': 'kt',
 168                             'conn_id': self._conn_id,
 169                             'url': embed_code,
 170                         }, fatal=False) or {}
 171                     kenc_message = kenc.get('message')
 172                     if kenc_message:
 173                         embed_code += '?' + ds(kenc_message)
 174                 formats.extend(self._extract_akamai_formats(embed_code, video_id))
 175             else:
 176                 formats.extend(self._extract_wowza_formats(
 177                     'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
 178             metadata_from_url(embed_code)
 179
 180         self._sort_formats(formats)
 181
 182         metadata_from_url(video_data.get('thumbnail'))
 183         info.update(self._extract_common_video_info(video_data))
 184         info.update({
 185             'id': video_id,
 186             'title': title,
 187             'formats': formats,
 188             'duration': int_or_none(video_data.get('length')),
 189             'series': video_data.get('show_title'),
 190             'season_id': season_id,
 191             'episode': title,
 192             'view_count': int_or_none(video_data.get('views')),
 193             'like_count': int_or_none(video_data.get('video_likes')),
 194             'repost_count': int_or_none(video_data.get('video_shares')),
 195         })
 196         return info
 197
 198
 199 class VVVVIDShowIE(VVVVIDIE):
 200     _VALID_URL = r'(?P<base_url>%s(?P<id>\d+)(?:/(?P<show_title>[^/?&#]+))?)/?(?:[?#&]|$)' % VVVVIDIE._VALID_URL_BASE
 201     _TESTS = [{
 202         'url': 'https://www.vvvvid.it/show/156/psyco-pass',
 203         'info_dict': {
 204             'id': '156',
 205             'title': 'Psycho-Pass',
 206             'description': 'md5:94d572c0bd85894b193b8aebc9a3a806',
 207         },
 208         'playlist_count': 46,
 209     }, {
 210         'url': 'https://www.vvvvid.it/show/156',
 211         'only_matching': True,
 212     }]
 213
 214     def _real_extract(self, url):
 215         base_url, show_id, show_title = re.match(self._VALID_URL, url).groups()
 216
 217         seasons = self._download_info(
 218             show_id, 'seasons/', show_title)
 219
 220         show_info = self._download_info(
 221             show_id, 'info/', show_title, fatal=False)
 222
 223         entries = []
 224         for season in (seasons or []):
 225             episodes = season.get('episodes') or []
 226             for episode in episodes:
 227                 if episode.get('playable') is False:
 228                     continue
 229                 season_id = str_or_none(episode.get('season_id'))
 230                 video_id = str_or_none(episode.get('video_id'))
 231                 if not (season_id and video_id):
 232                     continue
 233                 info = self._extract_common_video_info(episode)
 234                 info.update({
 235                     '_type': 'url',
 236                     'ie_key': VVVVIDIE.ie_key(),
 237                     'url': '/'.join([base_url, season_id, video_id]),
 238                     'title': episode.get('title'),
 239                     'description': episode.get('description'),
 240                     'season_id': season_id,
 241                 })
 242                 entries.append(info)
 243
 244         return self.playlist_result(
 245             entries, show_id, show_info.get('title'), show_info.get('description'))