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 SportBoxIE(InfoExtractor):
  15     _VALID_URL = r'https?://news\.sportbox\.ru/(?:[^/]+/)+spbvideo_NI\d+_(?P<display_id>.+)'
  16     _TESTS = [{
  17         'url': 'http://news.sportbox.ru/Vidy_sporta/Avtosport/Rossijskij/spbvideo_NI483529_Gonka-2-zaezd-Obyedinenniy-2000-klassi-Turing-i-S',
  18         'md5': 'ff56a598c2cf411a9a38a69709e97079',
  19         'info_dict': {
  20             'id': '80822',
  21             'ext': 'mp4',
  22             'title': 'Гонка 2  заезд ««Объединенный 2000»: классы Туринг и Супер-продакшн',
  23             'description': 'md5:3d72dc4a006ab6805d82f037fdc637ad',
  24             'thumbnail': 're:^https?://.*\.jpg$',
  25             'upload_date': '20140928',
  26         },
  27         'params': {
  28             # m3u8 download
  29             'skip_download': True,
  30         },
  31     }, {
  32         'url': 'http://news.sportbox.ru/Vidy_sporta/billiard/spbvideo_NI486287_CHempionat-mira-po-dinamichnoy-piramide-4',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://news.sportbox.ru/video/no_ads/spbvideo_NI536574_V_Novorossijske_proshel_detskij_turnir_Pole_slavy_bojevoj?ci=211355',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         display_id = mobj.group('display_id')
  42
  43         webpage = self._download_webpage(url, display_id)
  44
  45         player = self._search_regex(
  46             r'src="/?(vdl/player/[^"]+)"', webpage, 'player')
  47
  48         title = self._html_search_regex(
  49             [r'"nodetitle"\s*:\s*"([^"]+)"', r'class="node-header_{1,2}title">([^<]+)'],
  50             webpage, 'title')
  51         description = self._og_search_description(webpage) or self._html_search_meta(
  52             'description', webpage, 'description')
  53         thumbnail = self._og_search_thumbnail(webpage)
  54         upload_date = unified_strdate(self._html_search_meta(
  55             'dateCreated', webpage, 'upload date'))
  56
  57         return {
  58             '_type': 'url_transparent',
  59             'url': compat_urlparse.urljoin(url, '/%s' % player),
  60             'display_id': display_id,
  61             'title': title,
  62             'description': description,
  63             'thumbnail': thumbnail,
  64             'upload_date': upload_date,
  65         }
  66
  67
  68 class SportBoxEmbedIE(InfoExtractor):
  69     _VALID_URL = r'https?://news\.sportbox\.ru/vdl/player(?:/[^/]+/|\?.*?\bn?id=)(?P<id>\d+)'
  70     _TESTS = [{
  71         'url': 'http://news.sportbox.ru/vdl/player/ci/211355',
  72         'info_dict': {
  73             'id': '211355',
  74             'ext': 'mp4',
  75             'title': 'В Новороссийске прошел детский турнир «Поле славы боевой»',
  76             'thumbnail': 're:^https?://.*\.jpg$',
  77         },
  78         'params': {
  79             # m3u8 download
  80             'skip_download': True,
  81         },
  82     }, {
  83         'url': 'http://news.sportbox.ru/vdl/player?nid=370908&only_player=1&autostart=false&playeri=2&height=340&width=580',
  84         'only_matching': True,
  85     }]
  86
  87     @staticmethod
  88     def _extract_urls(webpage):
  89         return re.findall(
  90             r'<iframe[^>]+src="(https?://news\.sportbox\.ru/vdl/player[^"]+)"',
  91             webpage)
  92
  93     def _real_extract(self, url):
  94         video_id = self._match_id(url)
  95
  96         webpage = self._download_webpage(url, video_id)
  97
  98         formats = []
  99
 100         def cleanup_js(code):
 101             # desktop_advert_config contains complex Javascripts and we don't need it
 102             return js_to_json(re.sub(r'desktop_advert_config.*', '', code))
 103
 104         jwplayer_data = self._parse_json(self._search_regex(
 105             r'(?s)player\.setup\(({.+?})\);', webpage, 'jwplayer settings'), video_id,
 106             transform_source=cleanup_js)
 107
 108         hls_url = jwplayer_data.get('hls_url')
 109         if hls_url:
 110             formats.extend(self._extract_m3u8_formats(
 111                 hls_url, video_id, ext='mp4', m3u8_id='hls'))
 112
 113         rtsp_url = jwplayer_data.get('rtsp_url')
 114         if rtsp_url:
 115             formats.append({
 116                 'url': rtsp_url,
 117                 'format_id': 'rtsp',
 118             })
 119
 120         self._sort_formats(formats)
 121
 122         title = jwplayer_data['node_title']
 123         thumbnail = jwplayer_data.get('image_url')
 124
 125         return {
 126             'id': video_id,
 127             'title': title,
 128             'thumbnail': thumbnail,
 129             'formats': formats,
 130         }