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

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import json
   5 import random
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..aes import (
  10     aes_cbc_decrypt,
  11     aes_cbc_encrypt,
  12 )
  13 from ..compat import compat_b64decode
  14 from ..utils import (
  15     bytes_to_intlist,
  16     bytes_to_long,
  17     extract_attributes,
  18     ExtractorError,
  19     intlist_to_bytes,
  20     js_to_json,
  21     int_or_none,
  22     long_to_bytes,
  23     pkcs1pad,
  24 )
  25
  26
  27 class DaisukiMottoIE(InfoExtractor):
  28     _VALID_URL = r'https?://motto\.daisuki\.net/framewatch/embed/[^/]+/(?P<id>[0-9a-zA-Z]{3})'
  29
  30     _TEST = {
  31         'url': 'http://motto.daisuki.net/framewatch/embed/embedDRAGONBALLSUPERUniverseSurvivalsaga/V2e/760/428',
  32         'info_dict': {
  33             'id': 'V2e',
  34             'ext': 'mp4',
  35             'title': '#117 SHOWDOWN OF LOVE! ANDROIDS VS UNIVERSE 2!!',
  36             'subtitles': {
  37                 'mul': [{
  38                     'ext': 'ttml',
  39                 }],
  40             },
  41         },
  42         'params': {
  43             'skip_download': True,  # AES-encrypted HLS stream
  44         },
  45     }
  46
  47     # The public key in PEM format can be found in clientlibs_anime_watch.min.js
  48     _RSA_KEY = (0xc5524c25e8e14b366b3754940beeb6f96cb7e2feef0b932c7659a0c5c3bf173d602464c2df73d693b513ae06ff1be8f367529ab30bf969c5640522181f2a0c51ea546ae120d3d8d908595e4eff765b389cde080a1ef7f1bbfb07411cc568db73b7f521cedf270cbfbe0ddbc29b1ac9d0f2d8f4359098caffee6d07915020077d, 65537)
  49
  50     def _real_extract(self, url):
  51         video_id = self._match_id(url)
  52
  53         webpage = self._download_webpage(url, video_id)
  54
  55         flashvars = self._parse_json(self._search_regex(
  56             r'(?s)var\s+flashvars\s*=\s*({.+?});', webpage, 'flashvars'),
  57             video_id, transform_source=js_to_json)
  58
  59         iv = [0] * 16
  60
  61         data = {}
  62         for key in ('device_cd', 'mv_id', 'ss1_prm', 'ss2_prm', 'ss3_prm', 'ss_id'):
  63             data[key] = flashvars.get(key, '')
  64
  65         encrypted_rtn = None
  66
  67         # Some AES keys are rejected. Try it with different AES keys
  68         for idx in range(5):
  69             aes_key = [random.randint(0, 254) for _ in range(32)]
  70             padded_aeskey = intlist_to_bytes(pkcs1pad(aes_key, 128))
  71
  72             n, e = self._RSA_KEY
  73             encrypted_aeskey = long_to_bytes(pow(bytes_to_long(padded_aeskey), e, n))
  74             init_data = self._download_json(
  75                 'http://motto.daisuki.net/fastAPI/bgn/init/',
  76                 video_id, query={
  77                     's': flashvars.get('s', ''),
  78                     'c': flashvars.get('ss3_prm', ''),
  79                     'e': url,
  80                     'd': base64.b64encode(intlist_to_bytes(aes_cbc_encrypt(
  81                         bytes_to_intlist(json.dumps(data)),
  82                         aes_key, iv))).decode('ascii'),
  83                     'a': base64.b64encode(encrypted_aeskey).decode('ascii'),
  84                 }, note='Downloading JSON metadata' + (' (try #%d)' % (idx + 1) if idx > 0 else ''))
  85
  86             if 'rtn' in init_data:
  87                 encrypted_rtn = init_data['rtn']
  88                 break
  89
  90             self._sleep(5, video_id)
  91
  92         if encrypted_rtn is None:
  93             raise ExtractorError('Failed to fetch init data')
  94
  95         rtn = self._parse_json(
  96             intlist_to_bytes(aes_cbc_decrypt(bytes_to_intlist(
  97                 compat_b64decode(encrypted_rtn)),
  98                 aes_key, iv)).decode('utf-8').rstrip('\0'),
  99             video_id)
 100
 101         title = rtn['title_str']
 102
 103         formats = self._extract_m3u8_formats(
 104             rtn['play_url'], video_id, ext='mp4', entry_protocol='m3u8_native')
 105
 106         subtitles = {}
 107         caption_url = rtn.get('caption_url')
 108         if caption_url:
 109             # mul: multiple languages
 110             subtitles['mul'] = [{
 111                 'url': caption_url,
 112                 'ext': 'ttml',
 113             }]
 114
 115         return {
 116             'id': video_id,
 117             'title': title,
 118             'formats': formats,
 119             'subtitles': subtitles,
 120         }
 121
 122
 123 class DaisukiMottoPlaylistIE(InfoExtractor):
 124     _VALID_URL = r'https?://motto\.daisuki\.net/(?P<id>information)/'
 125
 126     _TEST = {
 127         'url': 'http://motto.daisuki.net/information/',
 128         'info_dict': {
 129             'title': 'DRAGON BALL SUPER',
 130         },
 131         'playlist_mincount': 117,
 132     }
 133
 134     def _real_extract(self, url):
 135         playlist_id = self._match_id(url)
 136
 137         webpage = self._download_webpage(url, playlist_id)
 138
 139         entries = []
 140         for li in re.findall(r'(<li[^>]+?data-product_id="[a-zA-Z0-9]{3}"[^>]+>)', webpage):
 141             attr = extract_attributes(li)
 142             ad_id = attr.get('data-ad_id')
 143             product_id = attr.get('data-product_id')
 144             if ad_id and product_id:
 145                 episode_id = attr.get('data-chapter')
 146                 entries.append({
 147                     '_type': 'url_transparent',
 148                     'url': 'http://motto.daisuki.net/framewatch/embed/%s/%s/760/428' % (ad_id, product_id),
 149                     'episode_id': episode_id,
 150                     'episode_number': int_or_none(episode_id),
 151                     'ie_key': 'DaisukiMotto',
 152                 })
 153
 154         return self.playlist_result(entries, playlist_title='DRAGON BALL SUPER')