yt_dlp/extractor/firsttv.py

   1 from .common import InfoExtractor
   2 from ..compat import (
   3     compat_str,
   4     compat_urlparse,
   5 )
   6 from ..utils import (
   7     int_or_none,
   8     qualities,
   9     unified_strdate,
  10     url_or_none,
  11 )
  12
  13
  14 class FirstTVIE(InfoExtractor):
  15     IE_NAME = '1tv'
  16     IE_DESC = 'Первый канал'
  17     _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>[^/?#]+)'
  18
  19     _TESTS = [{
  20         # single format
  21         'url': 'http://www.1tv.ru/shows/naedine-so-vsemi/vypuski/gost-lyudmila-senchina-naedine-so-vsemi-vypusk-ot-12-02-2015',
  22         'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
  23         'info_dict': {
  24             'id': '40049',
  25             'ext': 'mp4',
  26             'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
  27             'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$',
  28             'upload_date': '20150212',
  29             'duration': 2694,
  30         },
  31     }, {
  32         # multiple formats
  33         'url': 'http://www.1tv.ru/shows/dobroe-utro/pro-zdorove/vesennyaya-allergiya-dobroe-utro-fragment-vypuska-ot-07042016',
  34         'info_dict': {
  35             'id': '364746',
  36             'ext': 'mp4',
  37             'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
  38             'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$',
  39             'upload_date': '20160407',
  40             'duration': 179,
  41             'formats': 'mincount:3',
  42         },
  43         'params': {
  44             'skip_download': True,
  45         },
  46     }, {
  47         'url': 'http://www.1tv.ru/news/issue/2016-12-01/14:00',
  48         'info_dict': {
  49             'id': '14:00',
  50             'title': 'Выпуск новостей в 14:00   1 декабря 2016 года. Новости. Первый канал',
  51             'description': 'md5:2e921b948f8c1ff93901da78ebdb1dfd',
  52         },
  53         'playlist_count': 13,
  54     }, {
  55         'url': 'http://www.1tv.ru/shows/tochvtoch-supersezon/vystupleniya/evgeniy-dyatlov-vladimir-vysockiy-koni-priveredlivye-toch-v-toch-supersezon-fragment-vypuska-ot-06-11-2016',
  56         'only_matching': True,
  57     }]
  58
  59     def _real_extract(self, url):
  60         display_id = self._match_id(url)
  61
  62         webpage = self._download_webpage(url, display_id)
  63         playlist_url = compat_urlparse.urljoin(url, self._search_regex(
  64             r'data-playlist-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
  65             webpage, 'playlist url', group='url'))
  66
  67         parsed_url = compat_urlparse.urlparse(playlist_url)
  68         qs = compat_urlparse.parse_qs(parsed_url.query)
  69         item_ids = qs.get('videos_ids[]') or qs.get('news_ids[]')
  70
  71         items = self._download_json(playlist_url, display_id)
  72
  73         if item_ids:
  74             items = [
  75                 item for item in items
  76                 if item.get('uid') and compat_str(item['uid']) in item_ids]
  77         else:
  78             items = [items[0]]
  79
  80         entries = []
  81         QUALITIES = ('ld', 'sd', 'hd', )
  82
  83         for item in items:
  84             title = item['title']
  85             quality = qualities(QUALITIES)
  86             formats = []
  87             path = None
  88             for f in item.get('mbr', []):
  89                 src = url_or_none(f.get('src'))
  90                 if not src:
  91                     continue
  92                 tbr = int_or_none(self._search_regex(
  93                     r'_(\d{3,})\.mp4', src, 'tbr', default=None))
  94                 if not path:
  95                     path = self._search_regex(
  96                         r'//[^/]+/(.+?)_\d+\.mp4', src,
  97                         'm3u8 path', default=None)
  98                 formats.append({
  99                     'url': src,
 100                     'format_id': f.get('name'),
 101                     'tbr': tbr,
 102                     'source_preference': quality(f.get('name')),
 103                     # quality metadata of http formats may be incorrect
 104                     'preference': -10,
 105                 })
 106             # m3u8 URL format is reverse engineered from [1] (search for
 107             # master.m3u8). dashEdges (that is currently balancer-vod.1tv.ru)
 108             # is taken from [2].
 109             # 1. http://static.1tv.ru/player/eump1tv-current/eump-1tv.all.min.js?rnd=9097422834:formatted
 110             # 2. http://static.1tv.ru/player/eump1tv-config/config-main.js?rnd=9097422834
 111             if not path and len(formats) == 1:
 112                 path = self._search_regex(
 113                     r'//[^/]+/(.+?$)', formats[0]['url'],
 114                     'm3u8 path', default=None)
 115             if path:
 116                 if len(formats) == 1:
 117                     m3u8_path = ','
 118                 else:
 119                     tbrs = [compat_str(t) for t in sorted(f['tbr'] for f in formats)]
 120                     m3u8_path = '_,%s,%s' % (','.join(tbrs), '.mp4')
 121                 formats.extend(self._extract_m3u8_formats(
 122                     'http://balancer-vod.1tv.ru/%s%s.urlset/master.m3u8'
 123                     % (path, m3u8_path),
 124                     display_id, 'mp4',
 125                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 126
 127             thumbnail = item.get('poster') or self._og_search_thumbnail(webpage)
 128             duration = int_or_none(item.get('duration') or self._html_search_meta(
 129                 'video:duration', webpage, 'video duration', fatal=False))
 130             upload_date = unified_strdate(self._html_search_meta(
 131                 'ya:ovs:upload_date', webpage, 'upload date', default=None))
 132
 133             entries.append({
 134                 'id': compat_str(item.get('id') or item['uid']),
 135                 'thumbnail': thumbnail,
 136                 'title': title,
 137                 'upload_date': upload_date,
 138                 'duration': int_or_none(duration),
 139                 'formats': formats
 140             })
 141
 142         title = self._html_search_regex(
 143             (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
 144              r"'title'\s*:\s*'([^']+)'"),
 145             webpage, 'title', default=None) or self._og_search_title(
 146             webpage, default=None)
 147         description = self._html_search_regex(
 148             r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
 149             webpage, 'description', default=None) or self._html_search_meta(
 150             'description', webpage, 'description', default=None)
 151
 152         return self.playlist_result(entries, display_id, title, description)