yt_dlp/extractor/firsttv.py

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