]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/cbsnews.py
[cbs] extract smpte and vtt subtitles
[youtube-dl.git] / youtube_dl / extractor / cbsnews.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from .cbs import CBSIE
6 from ..utils import (
7     parse_duration,
8 )
9
10
11 class CBSNewsIE(CBSIE):
12     IE_NAME = 'cbsnews'
13     IE_DESC = 'CBS News'
14     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
15
16     _TESTS = [
17         {
18             # 60 minutes
19             'url': 'http://www.cbsnews.com/news/artificial-intelligence-positioned-to-be-a-game-changer/',
20             'info_dict': {
21                 'id': '_B6Ga3VJrI4iQNKsir_cdFo9Re_YJHE_',
22                 'ext': 'mp4',
23                 'title': 'Artificial Intelligence',
24                 'description': 'md5:8818145f9974431e0fb58a1b8d69613c',
25                 'thumbnail': r're:^https?://.*\.jpg$',
26                 'duration': 1606,
27                 'uploader': 'CBSI-NEW',
28                 'timestamp': 1498431900,
29                 'upload_date': '20170625',
30             },
31             'params': {
32                 # m3u8 download
33                 'skip_download': True,
34             },
35         },
36         {
37             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
38             'info_dict': {
39                 'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
40                 'ext': 'mp4',
41                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
42                 'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
43                 'upload_date': '20140404',
44                 'timestamp': 1396650660,
45                 'uploader': 'CBSI-NEW',
46                 'thumbnail': r're:^https?://.*\.jpg$',
47                 'duration': 205,
48                 'subtitles': {
49                     'en': [{
50                         'ext': 'ttml',
51                     }],
52                 },
53             },
54             'params': {
55                 # m3u8 download
56                 'skip_download': True,
57             },
58         },
59         {
60             # 48 hours
61             'url': 'http://www.cbsnews.com/news/maria-ridulph-murder-will-the-nations-oldest-cold-case-to-go-to-trial-ever-get-solved/',
62             'info_dict': {
63                 'id': 'QpM5BJjBVEAUFi7ydR9LusS69DPLqPJ1',
64                 'ext': 'mp4',
65                 'title': 'Cold as Ice',
66                 'description': 'Can a childhood memory of a friend\'s murder solve a 1957 cold case? "48 Hours" correspondent Erin Moriarty has the latest.',
67                 'upload_date': '20170604',
68                 'timestamp': 1496538000,
69                 'uploader': 'CBSI-NEW',
70             },
71             'params': {
72                 'skip_download': True,
73             },
74         },
75     ]
76
77     def _real_extract(self, url):
78         video_id = self._match_id(url)
79
80         webpage = self._download_webpage(url, video_id)
81
82         video_info = self._parse_json(self._html_search_regex(
83             r'(?:<ul class="media-list items" id="media-related-items"[^>]*><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
84             webpage, 'video JSON info', default='{}'), video_id, fatal=False)
85
86         if video_info:
87             item = video_info['item'] if 'item' in video_info else video_info
88         else:
89             state = self._parse_json(self._search_regex(
90                 r'data-cbsvideoui-options=(["\'])(?P<json>{.+?})\1', webpage,
91                 'playlist JSON info', group='json'), video_id)['state']
92             item = state['playlist'][state['pid']]
93
94         return self._extract_video_info(item['mpxRefId'], 'cbsnews')
95
96
97 class CBSNewsLiveVideoIE(InfoExtractor):
98     IE_NAME = 'cbsnews:livevideo'
99     IE_DESC = 'CBS News Live Videos'
100     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/live/video/(?P<id>[^/?#]+)'
101
102     # Live videos get deleted soon. See http://www.cbsnews.com/live/ for the latest examples
103     _TEST = {
104         'url': 'http://www.cbsnews.com/live/video/clinton-sanders-prepare-to-face-off-in-nh/',
105         'info_dict': {
106             'id': 'clinton-sanders-prepare-to-face-off-in-nh',
107             'ext': 'mp4',
108             'title': 'Clinton, Sanders Prepare To Face Off In NH',
109             'duration': 334,
110         },
111         'skip': 'Video gone',
112     }
113
114     def _real_extract(self, url):
115         display_id = self._match_id(url)
116
117         video_info = self._download_json(
118             'http://feeds.cbsn.cbsnews.com/rundown/story', display_id, query={
119                 'device': 'desktop',
120                 'dvr_slug': display_id,
121             })
122
123         formats = self._extract_akamai_formats(video_info['url'], display_id)
124         self._sort_formats(formats)
125
126         return {
127             'id': display_id,
128             'display_id': display_id,
129             'title': video_info['headline'],
130             'thumbnail': video_info.get('thumbnail_url_hd') or video_info.get('thumbnail_url_sd'),
131             'duration': parse_duration(video_info.get('segmentDur')),
132             'formats': formats,
133         }