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