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 = r'https?://(?:www\.)?tv2\.no/v/(?P<id>\d+)'
  18     _TEST = {
  19         'url': 'http://www.tv2.no/v/916509/',
  20         'info_dict': {
  21             'id': '916509',
  22             'ext': 'mp4',
  23             'title': 'Se Frode Gryttens hyllest av Steven Gerrard',
  24             'description': 'TV 2 Sportens huspoet tar avskjed med Liverpools kaptein Steven Gerrard.',
  25             'timestamp': 1431715610,
  26             'upload_date': '20150515',
  27             'duration': 156.967,
  28             'view_count': int,
  29             'categories': list,
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         formats = []
  41         format_urls = []
  42         for protocol in ('HDS', 'HLS'):
  43             data = self._download_json(
  44                 'http://sumo.tv2.no/api/web/asset/%s/play.json?protocol=%s&videoFormat=SMIL+ISMUSP' % (video_id, protocol),
  45                 video_id, 'Downloading play JSON')['playback']
  46             for item in data['items']['item']:
  47                 video_url = item.get('url')
  48                 if not video_url or video_url in format_urls:
  49                     continue
  50                 format_id = '%s-%s' % (protocol.lower(), item.get('mediaFormat'))
  51                 if not self._is_valid_url(video_url, video_id, format_id):
  52                     continue
  53                 format_urls.append(video_url)
  54                 ext = determine_ext(video_url)
  55                 if ext == 'f4m':
  56                     formats.extend(self._extract_f4m_formats(
  57                         video_url, video_id, f4m_id=format_id))
  58                 elif ext == 'm3u8':
  59                     formats.extend(self._extract_m3u8_formats(
  60                         video_url, video_id, 'mp4', m3u8_id=format_id))
  61                 elif ext == 'ism' or video_url.endswith('.ism/Manifest'):
  62                     pass
  63                 else:
  64                     formats.append({
  65                         'url': video_url,
  66                         'format_id': format_id,
  67                         'tbr': int_or_none(item.get('bitrate')),
  68                         'filesize': int_or_none(item.get('fileSize')),
  69                     })
  70         self._sort_formats(formats)
  71
  72         asset = self._download_json(
  73             'http://sumo.tv2.no/api/web/asset/%s.json' % video_id,
  74             video_id, 'Downloading metadata JSON')['asset']
  75
  76         title = asset['title']
  77         description = asset.get('description')
  78         timestamp = parse_iso8601(asset.get('createTime'))
  79         duration = float_or_none(asset.get('accurateDuration') or asset.get('duration'))
  80         view_count = int_or_none(asset.get('views'))
  81         categories = asset.get('keywords', '').split(',')
  82
  83         thumbnails = [{
  84             'id': thumbnail.get('@type'),
  85             'url': thumbnail.get('url'),
  86         } for _, thumbnail in asset.get('imageVersions', {}).items()]
  87
  88         return {
  89             'id': video_id,
  90             'url': video_url,
  91             'title': title,
  92             'description': description,
  93             'thumbnails': thumbnails,
  94             'timestamp': timestamp,
  95             'duration': duration,
  96             'view_count': view_count,
  97             'categories': categories,
  98             'formats': formats,
  99         }
 100
 101
 102 class TV2ArticleIE(InfoExtractor):
 103     _VALID_URL = r'https?://(?:www\.)?tv2\.no/(?:a|\d{4}/\d{2}/\d{2}(/[^/]+)+)/(?P<id>\d+)'
 104     _TESTS = [{
 105         'url': 'http://www.tv2.no/2015/05/16/nyheter/alesund/krim/pingvin/6930542',
 106         'info_dict': {
 107             'id': '6930542',
 108             'title': 'Russen hetses etter pingvintyveri – innrømmer å ha åpnet luken på buret',
 109             'description': 'md5:339573779d3eea3542ffe12006190954',
 110         },
 111         'playlist_count': 2,
 112     }, {
 113         'url': 'http://www.tv2.no/a/6930542',
 114         'only_matching': True,
 115     }]
 116
 117     def _real_extract(self, url):
 118         playlist_id = self._match_id(url)
 119
 120         webpage = self._download_webpage(url, playlist_id)
 121
 122         entries = [
 123             self.url_result('http://www.tv2.no/v/%s' % video_id, 'TV2')
 124             for video_id in re.findall(r'data-assetid="(\d+)"', webpage)]
 125
 126         title = remove_end(self._og_search_title(webpage), ' - TV2.no')
 127         description = remove_end(self._og_search_description(webpage), ' - TV2.no')
 128
 129         return self.playlist_result(entries, playlist_id, title, description)