yt_dlp/extractor/miomio.py

   1 import random
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_urlparse
   5 from ..utils import (
   6     xpath_text,
   7     int_or_none,
   8     ExtractorError,
   9     sanitized_Request,
  10 )
  11
  12
  13 class MioMioIE(InfoExtractor):
  14     IE_NAME = 'miomio.tv'
  15     _VALID_URL = r'https?://(?:www\.)?miomio\.tv/watch/cc(?P<id>[0-9]+)'
  16     _TESTS = [{
  17         # "type=video" in flashvars
  18         'url': 'http://www.miomio.tv/watch/cc88912/',
  19         'info_dict': {
  20             'id': '88912',
  21             'ext': 'flv',
  22             'title': '【SKY】字幕 铠武昭和VS平成 假面骑士大战FEAT战队 魔星字幕组 字幕',
  23             'duration': 5923,
  24         },
  25         'skip': 'Unable to load videos',
  26     }, {
  27         'url': 'http://www.miomio.tv/watch/cc184024/',
  28         'info_dict': {
  29             'id': '43729',
  30             'title': '《动漫同人插画绘制》',
  31         },
  32         'playlist_mincount': 86,
  33         'skip': 'Unable to load videos',
  34     }, {
  35         'url': 'http://www.miomio.tv/watch/cc173113/',
  36         'info_dict': {
  37             'id': '173113',
  38             'title': 'The New Macbook 2015 上手试玩与简评'
  39         },
  40         'playlist_mincount': 2,
  41         'skip': 'Unable to load videos',
  42     }, {
  43         # new 'h5' player
  44         'url': 'http://www.miomio.tv/watch/cc273997/',
  45         'md5': '0b27a4b4495055d826813f8c3a6b2070',
  46         'info_dict': {
  47             'id': '273997',
  48             'ext': 'mp4',
  49             'title': 'マツコの知らない世界【劇的進化SP！ビニール傘＆冷凍食品2016】 1_2 - 16 05 31',
  50         },
  51         'skip': 'Unable to load videos',
  52     }]
  53
  54     def _extract_mioplayer(self, webpage, video_id, title, http_headers):
  55         xml_config = self._search_regex(
  56             r'flashvars="type=(?:sina|video)&amp;(.+?)&amp;',
  57             webpage, 'xml config')
  58
  59         # skipping the following page causes lags and eventually connection drop-outs
  60         self._request_webpage(
  61             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  62             video_id)
  63
  64         vid_config_request = sanitized_Request(
  65             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  66             headers=http_headers)
  67
  68         # the following xml contains the actual configuration information on the video file(s)
  69         vid_config = self._download_xml(vid_config_request, video_id)
  70
  71         if not int_or_none(xpath_text(vid_config, 'timelength')):
  72             raise ExtractorError('Unable to load videos!', expected=True)
  73
  74         entries = []
  75         for f in vid_config.findall('./durl'):
  76             segment_url = xpath_text(f, 'url', 'video url')
  77             if not segment_url:
  78                 continue
  79             order = xpath_text(f, 'order', 'order')
  80             segment_id = video_id
  81             segment_title = title
  82             if order:
  83                 segment_id += '-%s' % order
  84                 segment_title += ' part %s' % order
  85             entries.append({
  86                 'id': segment_id,
  87                 'url': segment_url,
  88                 'title': segment_title,
  89                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  90                 'http_headers': http_headers,
  91             })
  92
  93         return entries
  94
  95     def _download_chinese_webpage(self, *args, **kwargs):
  96         # Requests with English locales return garbage
  97         headers = {
  98             'Accept-Language': 'zh-TW,en-US;q=0.7,en;q=0.3',
  99         }
 100         kwargs.setdefault('headers', {}).update(headers)
 101         return self._download_webpage(*args, **kwargs)
 102
 103     def _real_extract(self, url):
 104         video_id = self._match_id(url)
 105         webpage = self._download_chinese_webpage(
 106             url, video_id)
 107
 108         title = self._html_search_meta(
 109             'description', webpage, 'title', fatal=True)
 110
 111         mioplayer_path = self._search_regex(
 112             r'src="(/mioplayer(?:_h5)?/[^"]+)"', webpage, 'ref_path')
 113
 114         if '_h5' in mioplayer_path:
 115             player_url = compat_urlparse.urljoin(url, mioplayer_path)
 116             player_webpage = self._download_chinese_webpage(
 117                 player_url, video_id,
 118                 note='Downloading player webpage', headers={'Referer': url})
 119             entries = self._parse_html5_media_entries(player_url, player_webpage, video_id)
 120             http_headers = {'Referer': player_url}
 121         else:
 122             http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
 123             entries = self._extract_mioplayer(webpage, video_id, title, http_headers)
 124
 125         if len(entries) == 1:
 126             segment = entries[0]
 127             segment['id'] = video_id
 128             segment['title'] = title
 129             segment['http_headers'] = http_headers
 130             return segment
 131
 132         return {
 133             '_type': 'multi_video',
 134             'id': video_id,
 135             'entries': entries,
 136             'title': title,
 137             'http_headers': http_headers,
 138         }