youtube_dl/extractor/miomio.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urlparse
   8 from ..utils import (
   9     xpath_text,
  10     int_or_none,
  11     ExtractorError,
  12     sanitized_Request,
  13 )
  14
  15
  16 class MioMioIE(InfoExtractor):
  17     IE_NAME = 'miomio.tv'
  18     _VALID_URL = r'https?://(?:www\.)?miomio\.tv/watch/cc(?P<id>[0-9]+)'
  19     _TESTS = [{
  20         # "type=video" in flashvars
  21         'url': 'http://www.miomio.tv/watch/cc88912/',
  22         'info_dict': {
  23             'id': '88912',
  24             'ext': 'flv',
  25             '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',
  26             'duration': 5923,
  27         },
  28         'skip': 'Unable to load videos',
  29     }, {
  30         'url': 'http://www.miomio.tv/watch/cc184024/',
  31         'info_dict': {
  32             'id': '43729',
  33             'title': '《动漫同人插画绘制》',
  34         },
  35         'playlist_mincount': 86,
  36         'skip': 'Unable to load videos',
  37     }, {
  38         'url': 'http://www.miomio.tv/watch/cc173113/',
  39         'info_dict': {
  40             'id': '173113',
  41             'title': 'The New Macbook 2015 上手试玩与简评'
  42         },
  43         'playlist_mincount': 2,
  44         'skip': 'Unable to load videos',
  45     }, {
  46         # new 'h5' player
  47         'url': 'http://www.miomio.tv/watch/cc273997/',
  48         'md5': '0b27a4b4495055d826813f8c3a6b2070',
  49         'info_dict': {
  50             'id': '273997',
  51             'ext': 'mp4',
  52             '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',
  53         },
  54         'skip': 'Unable to load videos',
  55     }]
  56
  57     def _extract_mioplayer(self, webpage, video_id, title, http_headers):
  58         xml_config = self._search_regex(
  59             r'flashvars="type=(?:sina|video)&amp;(.+?)&amp;',
  60             webpage, 'xml config')
  61
  62         # skipping the following page causes lags and eventually connection drop-outs
  63         self._request_webpage(
  64             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  65             video_id)
  66
  67         vid_config_request = sanitized_Request(
  68             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  69             headers=http_headers)
  70
  71         # the following xml contains the actual configuration information on the video file(s)
  72         vid_config = self._download_xml(vid_config_request, video_id)
  73
  74         if not int_or_none(xpath_text(vid_config, 'timelength')):
  75             raise ExtractorError('Unable to load videos!', expected=True)
  76
  77         entries = []
  78         for f in vid_config.findall('./durl'):
  79             segment_url = xpath_text(f, 'url', 'video url')
  80             if not segment_url:
  81                 continue
  82             order = xpath_text(f, 'order', 'order')
  83             segment_id = video_id
  84             segment_title = title
  85             if order:
  86                 segment_id += '-%s' % order
  87                 segment_title += ' part %s' % order
  88             entries.append({
  89                 'id': segment_id,
  90                 'url': segment_url,
  91                 'title': segment_title,
  92                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  93                 'http_headers': http_headers,
  94             })
  95
  96         return entries
  97
  98     def _download_chinese_webpage(self, *args, **kwargs):
  99         # Requests with English locales return garbage
 100         headers = {
 101             'Accept-Language': 'zh-TW,en-US;q=0.7,en;q=0.3',
 102         }
 103         kwargs.setdefault('headers', {}).update(headers)
 104         return self._download_webpage(*args, **kwargs)
 105
 106     def _real_extract(self, url):
 107         video_id = self._match_id(url)
 108         webpage = self._download_chinese_webpage(
 109             url, video_id)
 110
 111         title = self._html_search_meta(
 112             'description', webpage, 'title', fatal=True)
 113
 114         mioplayer_path = self._search_regex(
 115             r'src="(/mioplayer(?:_h5)?/[^"]+)"', webpage, 'ref_path')
 116
 117         if '_h5' in mioplayer_path:
 118             player_url = compat_urlparse.urljoin(url, mioplayer_path)
 119             player_webpage = self._download_chinese_webpage(
 120                 player_url, video_id,
 121                 note='Downloading player webpage', headers={'Referer': url})
 122             entries = self._parse_html5_media_entries(player_url, player_webpage, video_id)
 123             http_headers = {'Referer': player_url}
 124         else:
 125             http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
 126             entries = self._extract_mioplayer(webpage, video_id, title, http_headers)
 127
 128         if len(entries) == 1:
 129             segment = entries[0]
 130             segment['id'] = video_id
 131             segment['title'] = title
 132             segment['http_headers'] = http_headers
 133             return segment
 134
 135         return {
 136             '_type': 'multi_video',
 137             'id': video_id,
 138             'entries': entries,
 139             'title': title,
 140             'http_headers': http_headers,
 141         }