_ Git - youtube-dl/blob - youtube_dl/extractor/sportbox.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urlparse
   8 from ..utils import (
   9     js_to_json,
  10     unified_strdate,
  11 )
  12
  13
  14 class SportBoxEmbedIE(InfoExtractor):
  15     _VALID_URL = r'https?://news\.sportbox\.ru/vdl/player(?:/[^/]+/|\?.*?\bn?id=)(?P<id>\d+)'
  16     _TESTS = [{
  17         'url': 'http://news.sportbox.ru/vdl/player/ci/211355',
  18         'info_dict': {
  19             'id': '211355',
  20             'ext': 'mp4',
  21             'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
  22             'thumbnail': r're:^https?://.*\.jpg$',
  23         },
  24         'params': {
  25             # m3u8 download
  26             'skip_download': True,
  27         },
  28     }, {
  29         'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
  30         'only_matching': True,
  31     }]
  32
  33     @staticmethod
  34     def _extract_urls(webpage):
  35         return re.findall(
  36             r'<iframe[^>]+src="(https?://news\.sportbox\.ru/vdl/player[^"]+)"',
  37             webpage)
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41
  42         webpage = self._download_webpage(url, video_id)
  43
  44         formats = []
  45
  46         def cleanup_js(code):
  47             # desktop_advert_config contains complex Javascripts and we don't need it
  48             return js_to_json(re.sub(r'desktop_advert_config.*', '', code))
  49
  50         jwplayer_data = self._parse_json(self._search_regex(
  51             r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
  52             transform_source=cleanup_js)
  53
  54         hls_url = jwplayer_data.get('hls_url')
  55         if hls_url:
  56             formats.extend(self._extract_m3u8_formats(
  57                 hls_url, video_id, ext='mp4', m3u8_id='hls'))
  58
  59         rtsp_url = jwplayer_data.get('rtsp_url')
  60         if rtsp_url:
  61             formats.append({
  62                 'url': rtsp_url,
  63                 'format_id': 'rtsp',
  64             })
  65
  66         self._sort_formats(formats)
  67
  68         title = jwplayer_data['node_title']
  69         thumbnail = jwplayer_data.get('image_url')
  70
  71         return {
  72             'id': video_id,
  73             'title': title,
  74             'thumbnail': thumbnail,
  75             'formats': formats,
  76         }