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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urllib_parse_urlparse
   6 from ..utils import (
   7     determine_ext,
   8     float_or_none,
   9 )
  10
  11
  12 class SpiegeltvIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?spiegel\.tv/(?:#/)?filme/(?P<id>[\-a-z0-9]+)'
  14     _TESTS = [{
  15         'url': 'http://www.spiegel.tv/filme/flug-mh370/',
  16         'info_dict': {
  17             'id': 'flug-mh370',
  18             'ext': 'm4v',
  19             'title': 'Flug MH370',
  20             'description': 'Das Rätsel um die Boeing 777 der Malaysia-Airlines',
  21             'thumbnail': 're:http://.*\.jpg$',
  22         },
  23         'params': {
  24             # m3u8 download
  25             'skip_download': True,
  26         }
  27     }, {
  28         'url': 'http://www.spiegel.tv/#/filme/alleskino-die-wahrheit-ueber-maenner/',
  29         'only_matching': True,
  30     }]
  31
  32     def _real_extract(self, url):
  33         if '/#/' in url:
  34             url = url.replace('/#/', '/')
  35         video_id = self._match_id(url)
  36         webpage = self._download_webpage(url, video_id)
  37         title = self._html_search_regex(r'<h1.*?>(.*?)</h1>', webpage, 'title')
  38
  39         apihost = 'http://spiegeltv-ivms2-restapi.s3.amazonaws.com'
  40         version_json = self._download_json(
  41             '%s/version.json' % apihost, video_id,
  42             note='Downloading version information')
  43         version_name = version_json['version_name']
  44
  45         slug_json = self._download_json(
  46             '%s/%s/restapi/slugs/%s.json' % (apihost, version_name, video_id),
  47             video_id,
  48             note='Downloading object information')
  49         oid = slug_json['object_id']
  50
  51         media_json = self._download_json(
  52             '%s/%s/restapi/media/%s.json' % (apihost, version_name, oid),
  53             video_id, note='Downloading media information')
  54         uuid = media_json['uuid']
  55         is_wide = media_json['is_wide']
  56
  57         server_json = self._download_json(
  58             'http://spiegeltv-prod-static.s3.amazonaws.com/projectConfigs/projectConfig.json',
  59             video_id, note='Downloading server information')
  60
  61         format = '16x9' if is_wide else '4x3'
  62
  63         formats = []
  64         for streamingserver in server_json['streamingserver']:
  65             endpoint = streamingserver.get('endpoint')
  66             if not endpoint:
  67                 continue
  68             play_path = 'mp4:%s_spiegeltv_0500_%s.m4v' % (uuid, format)
  69             if endpoint.startswith('rtmp'):
  70                 formats.append({
  71                     'url': endpoint,
  72                     'format_id': 'rtmp',
  73                     'app': compat_urllib_parse_urlparse(endpoint).path[1:],
  74                     'play_path': play_path,
  75                     'player_path': 'http://prod-static.spiegel.tv/frontend-076.swf',
  76                     'ext': 'flv',
  77                     'rtmp_live': True,
  78                 })
  79             elif determine_ext(endpoint) == 'm3u8':
  80                 formats.append({
  81                     'url': endpoint.replace('[video]', play_path),
  82                     'ext': 'm4v',
  83                     'format_id': 'hls',  # Prefer hls since it allows to workaround georestriction
  84                     'protocol': 'm3u8',
  85                     'preference': 1,
  86                     'http_headers': {
  87                         'Accept-Encoding': 'deflate',  # gzip causes trouble on the server side
  88                     },
  89                 })
  90             else:
  91                 formats.append({
  92                     'url': endpoint,
  93                 })
  94         self._check_formats(formats, video_id)
  95
  96         thumbnails = []
  97         for image in media_json['images']:
  98             thumbnails.append({
  99                 'url': image['url'],
 100                 'width': image['width'],
 101                 'height': image['height'],
 102             })
 103
 104         description = media_json['subtitle']
 105         duration = float_or_none(media_json.get('duration_in_ms'), scale=1000)
 106
 107         return {
 108             'id': video_id,
 109             'title': title,
 110             'description': description,
 111             'duration': duration,
 112             'thumbnails': thumbnails,
 113             'formats': formats,
 114         }