_ 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 ..utils import (
   8     determine_ext,
   9     int_or_none,
  10     js_to_json,
  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             'duration': 292,
  24             'view_count': int,
  25         },
  26         'params': {
  27             # m3u8 download
  28             'skip_download': True,
  29         },
  30     }, {
  31         'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
  32         'only_matching': True,
  33     }, {
  34         'url': 'https://news.sportbox.ru/vdl/player/media/193095',
  35         'only_matching': True,
  36     }]
  37
  38     @staticmethod
  39     def _extract_urls(webpage):
  40         return re.findall(
  41             r'<iframe[^>]+src="(https?://news\.sportbox\.ru/vdl/player[^"]+)"',
  42             webpage)
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46
  47         webpage = self._download_webpage(url, video_id)
  48
  49         wjplayer_data = self._parse_json(
  50             self._search_regex(
  51                 r'(?s)var\s+playerOptions\s*=\s*({.+?});', webpage, 'wjplayer settings'),
  52             video_id, transform_source=js_to_json)
  53
  54         wjplayer_data['sources'] = self._parse_json(
  55             self._search_regex(
  56                 r'(?s)playerOptions\.sources\s*=\s*(\[.+?\]);', webpage, 'wjplayer sources'),
  57             video_id, transform_source=js_to_json)
  58
  59         title = self._html_search_meta(
  60             ['og:title', 'twitter:title'], webpage) or self._html_search_regex(
  61             r'<title>(.+?)</title>', webpage, 'title', fatal=False) or video_id
  62
  63         formats = []
  64         for source in wjplayer_data['sources']:
  65             src = source.get('src')
  66             if not src:
  67                 continue
  68             if determine_ext(src) == 'm3u8':
  69                 formats.extend(self._extract_m3u8_formats(
  70                     src, video_id, 'mp4', entry_protocol='m3u8_native',
  71                     m3u8_id='hls', fatal=False))
  72             else:
  73                 formats.append({
  74                     'url': src,
  75                 })
  76         self._sort_formats(formats)
  77
  78         view_count = int_or_none(self._search_regex(
  79             r'Просмотров\s*:\s*(\d+)', webpage, 'view count', default=None))
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'thumbnail': wjplayer_data.get('poster'),
  85             'duration': int_or_none(wjplayer_data.get('duration')),
  86             'view_count': view_count,
  87             'formats': formats,
  88         }