Raphaël G. Git Repositories - youtubedl/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         '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         'params': {
  29             # The server provides broken file
  30             'skip_download': True,
  31         }
  32     }, {
  33         'url': 'http://www.miomio.tv/watch/cc184024/',
  34         'info_dict': {
  35             'id': '43729',
  36             'title': '《动漫同人插画绘制》',
  37         },
  38         'playlist_mincount': 86,
  39         'skip': 'Unable to load videos',
  40     }, {
  41         'url': 'http://www.miomio.tv/watch/cc173113/',
  42         'info_dict': {
  43             'id': '173113',
  44             'title': 'The New Macbook 2015 上手试玩与简评'
  45         },
  46         'playlist_mincount': 2,
  47         'skip': 'Unable to load videos',
  48     }, {
  49         # new 'h5' player
  50         'url': 'http://www.miomio.tv/watch/cc273295/',
  51         'md5': '',
  52         'info_dict': {
  53             'id': '273295',
  54             'ext': 'mp4',
  55             'title': 'アウト×デラックス 20160526',
  56         },
  57         'params': {
  58             # intermittent HTTP 500
  59             'skip_download': True,
  60         },
  61     }]
  62
  63     def _extract_mioplayer(self, webpage, video_id, title, http_headers):
  64         xml_config = self._search_regex(
  65             r'flashvars="type=(?:sina|video)&amp;(.+?)&amp;',
  66             webpage, 'xml config')
  67
  68         # skipping the following page causes lags and eventually connection drop-outs
  69         self._request_webpage(
  70             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  71             video_id)
  72
  73         vid_config_request = sanitized_Request(
  74             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  75             headers=http_headers)
  76
  77         # the following xml contains the actual configuration information on the video file(s)
  78         vid_config = self._download_xml(vid_config_request, video_id)
  79
  80         if not int_or_none(xpath_text(vid_config, 'timelength')):
  81             raise ExtractorError('Unable to load videos!', expected=True)
  82
  83         entries = []
  84         for f in vid_config.findall('./durl'):
  85             segment_url = xpath_text(f, 'url', 'video url')
  86             if not segment_url:
  87                 continue
  88             order = xpath_text(f, 'order', 'order')
  89             segment_id = video_id
  90             segment_title = title
  91             if order:
  92                 segment_id += '-%s' % order
  93                 segment_title += ' part %s' % order
  94             entries.append({
  95                 'id': segment_id,
  96                 'url': segment_url,
  97                 'title': segment_title,
  98                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  99                 'http_headers': http_headers,
 100             })
 101
 102         return entries
 103
 104     def _real_extract(self, url):
 105         video_id = self._match_id(url)
 106         webpage = self._download_webpage(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_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)
 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         }