_ Git - youtube-dl/blob - youtube_dl/extractor/dctp.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     float_or_none,
   8     int_or_none,
   9     unified_timestamp,
  10 )
  11
  12
  13 class DctpTvIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
  15     _TESTS = [{
  16         # 4x3
  17         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
  18         'info_dict': {
  19             'id': '95eaa4f33dad413aa17b4ee613cccc6c',
  20             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
  21             'ext': 'flv',
  22             'title': 'Videoinstallation für eine Kaufhausfassade',
  23             'description': 'Kurzfilm',
  24             'thumbnail': r're:^https?://.*\.jpg$',
  25             'duration': 71.24,
  26             'timestamp': 1302172322,
  27             'upload_date': '20110407',
  28         },
  29         'params': {
  30             # rtmp download
  31             'skip_download': True,
  32         },
  33     }, {
  34         # 16x9
  35         'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
  36         'only_matching': True,
  37     }]
  38
  39     _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'
  40
  41     def _real_extract(self, url):
  42         display_id = self._match_id(url)
  43
  44         version = self._download_json(
  45             '%s/version.json' % self._BASE_URL, display_id,
  46             'Downloading version JSON')
  47
  48         restapi_base = '%s/%s/restapi' % (
  49             self._BASE_URL, version['version_name'])
  50
  51         info = self._download_json(
  52             '%s/slugs/%s.json' % (restapi_base, display_id), display_id,
  53             'Downloading video info JSON')
  54
  55         media = self._download_json(
  56             '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])),
  57             display_id, 'Downloading media JSON')
  58
  59         uuid = media['uuid']
  60         title = media['title']
  61         ratio = '16x9' if media.get('is_wide') else '4x3'
  62         play_path = 'mp4:%s_dctp_0500_%s.m4v' % (uuid, ratio)
  63
  64         servers = self._download_json(
  65             'http://www.dctp.tv/streaming_servers/', display_id,
  66             note='Downloading server list JSON', fatal=False)
  67
  68         if servers:
  69             endpoint = next(
  70                 server['endpoint']
  71                 for server in servers
  72                 if isinstance(server.get('endpoint'), compat_str) and
  73                 'cloudfront' in server['endpoint'])
  74         else:
  75             endpoint = 'rtmpe://s2pqqn4u96e4j8.cloudfront.net/cfx/st/'
  76
  77         app = self._search_regex(
  78             r'^rtmpe?://[^/]+/(?P<app>.*)$', endpoint, 'app')
  79
  80         formats = [{
  81             'url': endpoint,
  82             'app': app,
  83             'play_path': play_path,
  84             'page_url': url,
  85             'player_url': 'http://svm-prod-dctptv-static.s3.amazonaws.com/dctptv-relaunch2012-110.swf',
  86             'ext': 'flv',
  87         }]
  88
  89         thumbnails = []
  90         images = media.get('images')
  91         if isinstance(images, list):
  92             for image in images:
  93                 if not isinstance(image, dict):
  94                     continue
  95                 image_url = image.get('url')
  96                 if not image_url or not isinstance(image_url, compat_str):
  97                     continue
  98                 thumbnails.append({
  99                     'url': image_url,
 100                     'width': int_or_none(image.get('width')),
 101                     'height': int_or_none(image.get('height')),
 102                 })
 103
 104         return {
 105             'id': uuid,
 106             'display_id': display_id,
 107             'title': title,
 108             'alt_title': media.get('subtitle'),
 109             'description': media.get('description') or media.get('teaser'),
 110             'timestamp': unified_timestamp(media.get('created')),
 111             'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
 112             'thumbnails': thumbnails,
 113             'formats': formats,
 114         }