yt_dlp/extractor/iwara.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_urllib_parse_urlparse
   5 from ..utils import (
   6     int_or_none,
   7     mimetype2ext,
   8     remove_end,
   9     url_or_none,
  10     unified_strdate,
  11     strip_or_none,
  12 )
  13
  14
  15 class IwaraIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.|ecchi\.)?iwara\.tv/videos/(?P<id>[a-zA-Z0-9]+)'
  17     _TESTS = [{
  18         'url': 'http://iwara.tv/videos/amVwUl1EHpAD9RD',
  19         # md5 is unstable
  20         'info_dict': {
  21             'id': 'amVwUl1EHpAD9RD',
  22             'ext': 'mp4',
  23             'title': '【MMD R-18】ガールフレンド carry_me_off',
  24             'age_limit': 18,
  25             'thumbnail': 'https://i.iwara.tv/sites/default/files/videos/thumbnails/7951/thumbnail-7951_0001.png',
  26             'uploader': 'Reimu丨Action',
  27             'upload_date': '20150828',
  28             'description': 'md5:1d4905ce48c66c9299c617f08e106e0f',
  29         },
  30     }, {
  31         'url': 'http://ecchi.iwara.tv/videos/Vb4yf2yZspkzkBO',
  32         'md5': '7e5f1f359cd51a027ba4a7b7710a50f0',
  33         'info_dict': {
  34             'id': '0B1LvuHnL-sRFNXB1WHNqbGw4SXc',
  35             'ext': 'mp4',
  36             'title': '[3D Hentai] Kyonyu × Genkai × Emaki Shinobi Girls.mp4',
  37             'age_limit': 18,
  38         },
  39         'add_ie': ['GoogleDrive'],
  40     }, {
  41         'url': 'http://www.iwara.tv/videos/nawkaumd6ilezzgq',
  42         # md5 is unstable
  43         'info_dict': {
  44             'id': '6liAP9s2Ojc',
  45             'ext': 'mp4',
  46             'age_limit': 18,
  47             'title': '[MMD] Do It Again Ver.2 [1080p 60FPS] (Motion,Camera,Wav+DL)',
  48             'description': 'md5:590c12c0df1443d833fbebe05da8c47a',
  49             'upload_date': '20160910',
  50             'uploader': 'aMMDsork',
  51             'uploader_id': 'UCVOFyOSCyFkXTYYHITtqB7A',
  52         },
  53         'add_ie': ['Youtube'],
  54     }]
  55
  56     def _real_extract(self, url):
  57         video_id = self._match_id(url)
  58
  59         webpage, urlh = self._download_webpage_handle(url, video_id)
  60
  61         hostname = compat_urllib_parse_urlparse(urlh.geturl()).hostname
  62         # ecchi is 'sexy' in Japanese
  63         age_limit = 18 if hostname.split('.')[0] == 'ecchi' else 0
  64
  65         video_data = self._download_json('http://www.iwara.tv/api/video/%s' % video_id, video_id)
  66
  67         if not video_data:
  68             iframe_url = self._html_search_regex(
  69                 r'<iframe[^>]+src=([\'"])(?P<url>[^\'"]+)\1',
  70                 webpage, 'iframe URL', group='url')
  71             return {
  72                 '_type': 'url_transparent',
  73                 'url': iframe_url,
  74                 'age_limit': age_limit,
  75             }
  76
  77         title = remove_end(self._html_extract_title(webpage), ' | Iwara')
  78
  79         thumbnail = self._html_search_regex(
  80             r'poster=[\'"]([^\'"]+)', webpage, 'thumbnail', default=None)
  81
  82         uploader = self._html_search_regex(
  83             r'class="username">([^<]+)', webpage, 'uploader', fatal=False)
  84
  85         upload_date = unified_strdate(self._html_search_regex(
  86             r'作成日:([^\s]+)', webpage, 'upload_date', fatal=False))
  87
  88         description = strip_or_none(self._search_regex(
  89             r'<p>(.+?(?=</div))', webpage, 'description', fatal=False,
  90             flags=re.DOTALL))
  91
  92         formats = []
  93         for a_format in video_data:
  94             format_uri = url_or_none(a_format.get('uri'))
  95             if not format_uri:
  96                 continue
  97             format_id = a_format.get('resolution')
  98             height = int_or_none(self._search_regex(
  99                 r'(\d+)p', format_id, 'height', default=None))
 100             formats.append({
 101                 'url': self._proto_relative_url(format_uri, 'https:'),
 102                 'format_id': format_id,
 103                 'ext': mimetype2ext(a_format.get('mime')) or 'mp4',
 104                 'height': height,
 105                 'width': int_or_none(height / 9.0 * 16.0 if height else None),
 106                 'quality': 1 if format_id == 'Source' else 0,
 107             })
 108
 109         self._sort_formats(formats)
 110
 111         return {
 112             'id': video_id,
 113             'title': title,
 114             'age_limit': age_limit,
 115             'formats': formats,
 116             'thumbnail': self._proto_relative_url(thumbnail, 'https:'),
 117             'uploader': uploader,
 118             'upload_date': upload_date,
 119             'description': description,
 120         }