yt_dlp/extractor/dctp.py

   1 from .common import InfoExtractor
   2 from ..compat import compat_str
   3 from ..utils import (
   4     float_or_none,
   5     int_or_none,
   6     unified_timestamp,
   7     url_or_none,
   8 )
   9
  10
  11 class DctpTvIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
  13     _TESTS = [{
  14         # 4x3
  15         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
  16         'md5': '3ffbd1556c3fe210724d7088fad723e3',
  17         'info_dict': {
  18             'id': '95eaa4f33dad413aa17b4ee613cccc6c',
  19             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
  20             'ext': 'm4v',
  21             'title': 'Videoinstallation für eine Kaufhausfassade',
  22             'description': 'Kurzfilm',
  23             'thumbnail': r're:^https?://.*\.jpg$',
  24             'duration': 71.24,
  25             'timestamp': 1302172322,
  26             'upload_date': '20110407',
  27         },
  28     }, {
  29         # 16x9
  30         'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
  31         'only_matching': True,
  32     }]
  33
  34     _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'
  35
  36     def _real_extract(self, url):
  37         display_id = self._match_id(url)
  38
  39         version = self._download_json(
  40             '%s/version.json' % self._BASE_URL, display_id,
  41             'Downloading version JSON')
  42
  43         restapi_base = '%s/%s/restapi' % (
  44             self._BASE_URL, version['version_name'])
  45
  46         info = self._download_json(
  47             '%s/slugs/%s.json' % (restapi_base, display_id), display_id,
  48             'Downloading video info JSON')
  49
  50         media = self._download_json(
  51             '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])),
  52             display_id, 'Downloading media JSON')
  53
  54         uuid = media['uuid']
  55         title = media['title']
  56         is_wide = media.get('is_wide')
  57         formats = []
  58
  59         def add_formats(suffix):
  60             templ = 'https://%%s/%s_dctp_%s.m4v' % (uuid, suffix)
  61             formats.extend([{
  62                 'format_id': 'hls-' + suffix,
  63                 'url': templ % 'cdn-segments.dctp.tv' + '/playlist.m3u8',
  64                 'protocol': 'm3u8_native',
  65             }, {
  66                 'format_id': 's3-' + suffix,
  67                 'url': templ % 'completed-media.s3.amazonaws.com',
  68             }, {
  69                 'format_id': 'http-' + suffix,
  70                 'url': templ % 'cdn-media.dctp.tv',
  71             }])
  72
  73         add_formats('0500_' + ('16x9' if is_wide else '4x3'))
  74         if is_wide:
  75             add_formats('720p')
  76
  77         thumbnails = []
  78         images = media.get('images')
  79         if isinstance(images, list):
  80             for image in images:
  81                 if not isinstance(image, dict):
  82                     continue
  83                 image_url = url_or_none(image.get('url'))
  84                 if not image_url:
  85                     continue
  86                 thumbnails.append({
  87                     'url': image_url,
  88                     'width': int_or_none(image.get('width')),
  89                     'height': int_or_none(image.get('height')),
  90                 })
  91
  92         return {
  93             'id': uuid,
  94             'display_id': display_id,
  95             'title': title,
  96             'alt_title': media.get('subtitle'),
  97             'description': media.get('description') or media.get('teaser'),
  98             'timestamp': unified_timestamp(media.get('created')),
  99             'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
 100             'thumbnails': thumbnails,
 101             'formats': formats,
 102         }