]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/dctp.py
[cbs] extract smpte and vtt subtitles
[youtube-dl.git] / 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     url_or_none,
11 )
12
13
14 class DctpTvIE(InfoExtractor):
15     _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
16     _TESTS = [{
17         # 4x3
18         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
19         'info_dict': {
20             'id': '95eaa4f33dad413aa17b4ee613cccc6c',
21             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
22             'ext': 'flv',
23             'title': 'Videoinstallation für eine Kaufhausfassade',
24             'description': 'Kurzfilm',
25             'thumbnail': r're:^https?://.*\.jpg$',
26             'duration': 71.24,
27             'timestamp': 1302172322,
28             'upload_date': '20110407',
29         },
30         'params': {
31             # rtmp download
32             'skip_download': True,
33         },
34     }, {
35         # 16x9
36         'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
37         'only_matching': True,
38     }]
39
40     _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'
41
42     def _real_extract(self, url):
43         display_id = self._match_id(url)
44
45         version = self._download_json(
46             '%s/version.json' % self._BASE_URL, display_id,
47             'Downloading version JSON')
48
49         restapi_base = '%s/%s/restapi' % (
50             self._BASE_URL, version['version_name'])
51
52         info = self._download_json(
53             '%s/slugs/%s.json' % (restapi_base, display_id), display_id,
54             'Downloading video info JSON')
55
56         media = self._download_json(
57             '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])),
58             display_id, 'Downloading media JSON')
59
60         uuid = media['uuid']
61         title = media['title']
62         ratio = '16x9' if media.get('is_wide') else '4x3'
63         play_path = 'mp4:%s_dctp_0500_%s.m4v' % (uuid, ratio)
64
65         servers = self._download_json(
66             'http://www.dctp.tv/streaming_servers/', display_id,
67             note='Downloading server list JSON', fatal=False)
68
69         if servers:
70             endpoint = next(
71                 server['endpoint']
72                 for server in servers
73                 if url_or_none(server.get('endpoint')) and
74                 'cloudfront' in server['endpoint'])
75         else:
76             endpoint = 'rtmpe://s2pqqn4u96e4j8.cloudfront.net/cfx/st/'
77
78         app = self._search_regex(
79             r'^rtmpe?://[^/]+/(?P<app>.*)$', endpoint, 'app')
80
81         formats = [{
82             'url': endpoint,
83             'app': app,
84             'play_path': play_path,
85             'page_url': url,
86             'player_url': 'http://svm-prod-dctptv-static.s3.amazonaws.com/dctptv-relaunch2012-110.swf',
87             'ext': 'flv',
88         }]
89
90         thumbnails = []
91         images = media.get('images')
92         if isinstance(images, list):
93             for image in images:
94                 if not isinstance(image, dict):
95                     continue
96                 image_url = url_or_none(image.get('url'))
97                 if not image_url:
98                     continue
99                 thumbnails.append({
100                     'url': image_url,
101                     'width': int_or_none(image.get('width')),
102                     'height': int_or_none(image.get('height')),
103                 })
104
105         return {
106             'id': uuid,
107             'display_id': display_id,
108             'title': title,
109             'alt_title': media.get('subtitle'),
110             'description': media.get('description') or media.get('teaser'),
111             'timestamp': unified_timestamp(media.get('created')),
112             'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
113             'thumbnails': thumbnails,
114             'formats': formats,
115         }