Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/firsttv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_str,
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     int_or_none,
  11     qualities,
  12     unified_strdate,
  13     url_or_none,
  14 )
  15
  16
  17 class FirstTVIE(InfoExtractor):
  18     IE_NAME = '1tv'
  19     IE_DESC = 'Первый канал'
  20     _VALID_URL = r'https?://(?:www\.)?1tv\.ru/(?:[^/]+/)+(?P<id>[^/?#]+)'
  21
  22     _TESTS = [{
  23         # single format
  24         'url': 'http://www.1tv.ru/shows/naedine-so-vsemi/vypuski/gost-lyudmila-senchina-naedine-so-vsemi-vypusk-ot-12-02-2015',
  25         'md5': 'a1b6b60d530ebcf8daacf4565762bbaf',
  26         'info_dict': {
  27             'id': '40049',
  28             'ext': 'mp4',
  29             'title': 'Гость Людмила Сенчина. Наедине со всеми. Выпуск от 12.02.2015',
  30             'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$',
  31             'upload_date': '20150212',
  32             'duration': 2694,
  33         },
  34     }, {
  35         # multiple formats
  36         'url': 'http://www.1tv.ru/shows/dobroe-utro/pro-zdorove/vesennyaya-allergiya-dobroe-utro-fragment-vypuska-ot-07042016',
  37         'info_dict': {
  38             'id': '364746',
  39             'ext': 'mp4',
  40             'title': 'Весенняя аллергия. Доброе утро. Фрагмент выпуска от 07.04.2016',
  41             'thumbnail': r're:^https?://.*\.(?:jpg|JPG)$',
  42             'upload_date': '20160407',
  43             'duration': 179,
  44             'formats': 'mincount:3',
  45         },
  46         'params': {
  47             'skip_download': True,
  48         },
  49     }, {
  50         'url': 'http://www.1tv.ru/news/issue/2016-12-01/14:00',
  51         'info_dict': {
  52             'id': '14:00',
  53             'title': 'Выпуск новостей в 14:00   1 декабря 2016 года. Новости. Первый канал',
  54             'description': 'md5:2e921b948f8c1ff93901da78ebdb1dfd',
  55         },
  56         'playlist_count': 13,
  57     }, {
  58         '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',
  59         'only_matching': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         display_id = self._match_id(url)
  64
  65         webpage = self._download_webpage(url, display_id)
  66         playlist_url = compat_urlparse.urljoin(url, self._search_regex(
  67             r'data-playlist-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
  68             webpage, 'playlist url', group='url'))
  69
  70         parsed_url = compat_urlparse.urlparse(playlist_url)
  71         qs = compat_urlparse.parse_qs(parsed_url.query)
  72         item_ids = qs.get('videos_ids[]') or qs.get('news_ids[]')
  73
  74         items = self._download_json(playlist_url, display_id)
  75
  76         if item_ids:
  77             items = [
  78                 item for item in items
  79                 if item.get('uid') and compat_str(item['uid']) in item_ids]
  80         else:
  81             items = [items[0]]
  82
  83         entries = []
  84         QUALITIES = ('ld', 'sd', 'hd', )
  85
  86         for item in items:
  87             title = item['title']
  88             quality = qualities(QUALITIES)
  89             formats = []
  90             path = None
  91             for f in item.get('mbr', []):
  92                 src = url_or_none(f.get('src'))
  93                 if not src:
  94                     continue
  95                 tbr = int_or_none(self._search_regex(
  96                     r'_(\d{3,})\.mp4', src, 'tbr', default=None))
  97                 if not path:
  98                     path = self._search_regex(
  99                         r'//[^/]+/(.+?)_\d+\.mp4', src,
 100                         'm3u8 path', default=None)
 101                 formats.append({
 102                     'url': src,
 103                     'format_id': f.get('name'),
 104                     'tbr': tbr,
 105                     'source_preference': quality(f.get('name')),
 106                     # quality metadata of http formats may be incorrect
 107                     'preference': -1,
 108                 })
 109             # m3u8 URL format is reverse engineered from [1] (search for
 110             # master.m3u8). dashEdges (that is currently balancer-vod.1tv.ru)
 111             # is taken from [2].
 112             # 1. http://static.1tv.ru/player/eump1tv-current/eump-1tv.all.min.js?rnd=9097422834:formatted
 113             # 2. http://static.1tv.ru/player/eump1tv-config/config-main.js?rnd=9097422834
 114             if not path and len(formats) == 1:
 115                 path = self._search_regex(
 116                     r'//[^/]+/(.+?$)', formats[0]['url'],
 117                     'm3u8 path', default=None)
 118             if path:
 119                 if len(formats) == 1:
 120                     m3u8_path = ','
 121                 else:
 122                     tbrs = [compat_str(t) for t in sorted(f['tbr'] for f in formats)]
 123                     m3u8_path = '_,%s,%s' % (','.join(tbrs), '.mp4')
 124                 formats.extend(self._extract_m3u8_formats(
 125                     'http://balancer-vod.1tv.ru/%s%s.urlset/master.m3u8'
 126                     % (path, m3u8_path),
 127                     display_id, 'mp4',
 128                     entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
 129             self._sort_formats(formats)
 130
 131             thumbnail = item.get('poster') or self._og_search_thumbnail(webpage)
 132             duration = int_or_none(item.get('duration') or self._html_search_meta(
 133                 'video:duration', webpage, 'video duration', fatal=False))
 134             upload_date = unified_strdate(self._html_search_meta(
 135                 'ya:ovs:upload_date', webpage, 'upload date', default=None))
 136
 137             entries.append({
 138                 'id': compat_str(item.get('id') or item['uid']),
 139                 'thumbnail': thumbnail,
 140                 'title': title,
 141                 'upload_date': upload_date,
 142                 'duration': int_or_none(duration),
 143                 'formats': formats
 144             })
 145
 146         title = self._html_search_regex(
 147             (r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>',
 148              r"'title'\s*:\s*'([^']+)'"),
 149             webpage, 'title', default=None) or self._og_search_title(
 150             webpage, default=None)
 151         description = self._html_search_regex(
 152             r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>',
 153             webpage, 'description', default=None) or self._html_search_meta(
 154             'description', webpage, 'description', default=None)
 155
 156         return self.playlist_result(entries, display_id, title, description)