yt_dlp/extractor/miomio.py

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