_ 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 ..utils import (
   8     xpath_text,
   9     int_or_none,
  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         'url': 'http://www.miomio.tv/watch/cc179734/',
  18         'md5': '48de02137d0739c15b440a224ad364b9',
  19         'info_dict': {
  20             'id': '179734',
  21             'ext': 'flv',
  22             'title': '手绘动漫鬼泣但丁全程画法',
  23             'duration': 354,
  24         },
  25     }, {
  26         'url': 'http://www.miomio.tv/watch/cc184024/',
  27         'info_dict': {
  28             'id': '43729',
  29             'title': '《动漫同人插画绘制》',
  30         },
  31         'playlist_mincount': 86,
  32     }]
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36         webpage = self._download_webpage(url, video_id)
  37
  38         title = self._html_search_meta(
  39             'description', webpage, 'title', fatal=True)
  40
  41         mioplayer_path = self._search_regex(
  42             r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path')
  43
  44         xml_config = self._search_regex(
  45             r'flashvars="type=sina&amp;(.+?)&amp;',
  46             webpage, 'xml config')
  47
  48         # skipping the following page causes lags and eventually connection drop-outs
  49         self._request_webpage(
  50             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/xml.php?id=%s&r=%s' % (id, random.randint(100, 999)),
  51             video_id)
  52
  53         # the following xml contains the actual configuration information on the video file(s)
  54         vid_config = self._download_xml(
  55             'http://www.miomio.tv/mioplayer/mioplayerconfigfiles/sina.php?{0}'.format(xml_config),
  56             video_id)
  57
  58         http_headers = {
  59             'Referer': 'http://www.miomio.tv%s' % mioplayer_path,
  60         }
  61
  62         entries = []
  63         for f in vid_config.findall('./durl'):
  64             segment_url = xpath_text(f, 'url', 'video url')
  65             if not segment_url:
  66                 continue
  67             order = xpath_text(f, 'order', 'order')
  68             segment_id = video_id
  69             segment_title = title
  70             if order:
  71                 segment_id += '-%s' % order
  72                 segment_title += ' part %s' % order
  73             entries.append({
  74                 'id': segment_id,
  75                 'url': segment_url,
  76                 'title': segment_title,
  77                 'duration': int_or_none(xpath_text(f, 'length', 'duration'), 1000),
  78                 'http_headers': http_headers,
  79             })
  80
  81         if len(entries) == 1:
  82             segment = entries[0]
  83             segment['id'] = video_id
  84             segment['title'] = title
  85             return segment
  86
  87         return {
  88             '_type': 'multi_video',
  89             'id': video_id,
  90             'entries': entries,
  91             'title': title,
  92             'http_headers': http_headers,
  93         }