_ Git - youtube-dl/blob - 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         'md5': '317a5f7f6b544ce8419b784ca8edae65',
  23         'info_dict': {
  24             'id': '88912',
  25             'ext': 'flv',
  26             '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',
  27             'duration': 5923,
  28         },
  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': 'This video takes time too long for retrieving the URL',
  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     }, {
  45         # new 'h5' player
  46         'url': 'http://www.miomio.tv/watch/cc273295/',
  47         'md5': '',
  48         'info_dict': {
  49             'id': '273295',
  50             'ext': 'mp4',
  51             'title': 'アウト×デラックス 20160526',
  52         },
  53         'params': {
  54             # intermittent HTTP 500
  55             'skip_download': True,
  56         },
  57     }]
  58
  59     def _extract_mioplayer(self, webpage, video_id, title, http_headers):
  60         xml_config = self._search_regex(
  61             r'flashvars="type=(?:sina|video)&amp;(.+?)&amp;',
  62             webpage, 'xml config')
  63
  64         # skipping the following page causes lags and eventually connection drop-outs
  65         self._request_webpage(
  66             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  67             video_id)
  68
  69         vid_config_request = sanitized_Request(
  70             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  71             headers=http_headers)
  72
  73         # the following xml contains the actual configuration information on the video file(s)
  74         vid_config = self._download_xml(vid_config_request, video_id)
  75
  76         if not int_or_none(xpath_text(vid_config, 'timelength')):
  77             raise ExtractorError('Unable to load videos!', expected=True)
  78
  79         entries = []
  80         for f in vid_config.findall('./durl'):
  81             segment_url = xpath_text(f, 'url', 'video url')
  82             if not segment_url:
  83                 continue
  84             order = xpath_text(f, 'order', 'order')
  85             segment_id = video_id
  86             segment_title = title
  87             if order:
  88                 segment_id += '-%s' % order
  89                 segment_title += ' part %s' % order
  90             entries.append({
  91                 'id': segment_id,
  92                 'url': segment_url,
  93                 'title': segment_title,
  94                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  95                 'http_headers': http_headers,
  96             })
  97
  98         return entries
  99
 100     def _real_extract(self, url):
 101         video_id = self._match_id(url)
 102         webpage = self._download_webpage(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_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)
 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         }