Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/tv2.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     int_or_none,
  10     float_or_none,
  11     parse_iso8601,
  12     remove_end,
  13 )
  14
  15
  16 class TV2IE(InfoExtractor):
  17     _VALID_URL = 'http://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
  18     _TEST = {
  19         'url': 'http://www.tv2.no/v/916509/',
  20         'md5': '9cb9e3410b18b515d71892f27856e9b1',
  21         'info_dict': {
  22             'id': '916509',
  23             'ext': 'flv',
  24             'title': 'Se Gryttens hyllest av Steven Gerrard',
  25             'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.',
  26             'timestamp': 1431715610,
  27             'upload_date': '20150515',
  28             'duration': 156.967,
  29             'view_count': int,
  30             'categories': list,
  31         }
  32     }
  33
  34     def _real_extract(self, url):
  35         video_id = self._match_id(url)
  36
  37         formats = []
  38         format_urls = []
  39         for protocol in ('HDS', 'HLS'):
  40             data = self._download_json(
  41                 'http://sumo.tv2.no/api/web/asset/%s/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % (video_id, protocol),
  42                 video_id, 'Downloading play JSON')['playback']
  43             for item in data['items']['item']:
  44                 video_url = item.get('url')
  45                 if not video_url or video_url in format_urls:
  46                     continue
  47                 format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
  48                 if not self._is_valid_url(video_url, video_id, format_id):
  49                     continue
  50                 format_urls.append(video_url)
  51                 ext = determine_ext(video_url)
  52                 if ext == 'f4m':
  53                     formats.extend(self._extract_f4m_formats(
  54                         video_url, video_id, f4m_id=format_id))
  55                 elif ext == 'm3u8':
  56                     formats.extend(self._extract_m3u8_formats(
  57                         video_url, video_id, 'mp4', m3u8_id=format_id))
  58                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
  59                     pass
  60                 else:
  61                     formats.append({
  62                         'url': video_url,
  63                         'format_id': format_id,
  64                         'tbr': int_or_none(item.get('bitrate')),
  65                         'filesize': int_or_none(item.get('fileSize')),
  66                     })
  67         self._sort_formats(formats)
  68
  69         asset = self._download_json(
  70             'http://sumo.tv2.no/api/web/asset/%s.json' % video_id,
  71             video_id, 'Downloading metadata JSON')['asset']
  72
  73         title = asset['title']
  74         description = asset.get('description')
  75         timestamp = parse_iso8601(asset.get('createTime'))
  76         duration = float_or_none(asset.get('accurateDuration') or asset.get('duration'))
  77         view_count = int_or_none(asset.get('views'))
  78         categories = asset.get('keywords', '').split(',')
  79
  80         thumbnails = [{
  81             'id': thumbnail.get('@type'),
  82             'url': thumbnail.get('url'),
  83         } for _, thumbnail in asset.get('imageVersions', {}).items()]
  84
  85         return {
  86             'id': video_id,
  87             'url': video_url,
  88             'title': title,
  89             'description': description,
  90             'thumbnails': thumbnails,
  91             'timestamp': timestamp,
  92             'duration': duration,
  93             'view_count': view_count,
  94             'categories': categories,
  95             'formats': formats,
  96         }
  97
  98
  99 class TV2ArticleIE(InfoExtractor):
 100     _VALID_URL = 'http://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
 101     _TESTS = [{
 102         'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
 103         'info_dict': {
 104             'id': '6930542',
 105             'title': 'Russen hetses etter pingvintyveri – innrømmer å ha åpnet luken på buret',
 106             'description': 'md5:339573779d3eea3542ffe12006190954',
 107         },
 108         'playlist_count': 2,
 109     }, {
 110         'url': 'http://www.tv2.no/a/6930542',
 111         'only_matching': True,
 112     }]
 113
 114     def _real_extract(self, url):
 115         playlist_id = self._match_id(url)
 116
 117         webpage = self._download_webpage(url, playlist_id)
 118
 119         entries = [
 120             self.url_result('http://www.tv2.no/v/%s' % video_id, 'TV2')
 121             for video_id in re.findall(r'data-assetid="(\d+)"', webpage)]
 122
 123         title = remove_end(self._og_search_title(webpage), ' - TV2.no')
 124         description = remove_end(self._og_search_description(webpage), ' - TV2.no')
 125
 126         return self.playlist_result(entries, playlist_id, title, description)