]>
Commit | Line | Data |
---|---|---|
1 | from __future__ import unicode_literals | |
2 | ||
3 | import json | |
4 | import re | |
5 | ||
6 | from .common import InfoExtractor | |
7 | from ..compat import ( | |
8 | compat_urlparse, | |
9 | ) | |
10 | from ..utils import ExtractorError | |
11 | ||
12 | ||
13 | class UstreamIE(InfoExtractor): | |
14 | _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)' | |
15 | IE_NAME = 'ustream' | |
16 | _TEST = { | |
17 | 'url': 'http://www.ustream.tv/recorded/20274954', | |
18 | 'md5': '088f151799e8f572f84eb62f17d73e5c', | |
19 | 'info_dict': { | |
20 | 'id': '20274954', | |
21 | 'ext': 'flv', | |
22 | 'uploader': 'Young Americans for Liberty', | |
23 | 'title': 'Young Americans for Liberty February 7, 2012 2:28 AM', | |
24 | }, | |
25 | } | |
26 | ||
27 | def _real_extract(self, url): | |
28 | m = re.match(self._VALID_URL, url) | |
29 | video_id = m.group('videoID') | |
30 | ||
31 | # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990) | |
32 | if m.group('type') == 'embed/recorded': | |
33 | video_id = m.group('videoID') | |
34 | desktop_url = 'http://www.ustream.tv/recorded/' + video_id | |
35 | return self.url_result(desktop_url, 'Ustream') | |
36 | if m.group('type') == 'embed': | |
37 | video_id = m.group('videoID') | |
38 | webpage = self._download_webpage(url, video_id) | |
39 | desktop_video_id = self._html_search_regex( | |
40 | r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id') | |
41 | desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id | |
42 | return self.url_result(desktop_url, 'Ustream') | |
43 | ||
44 | params = self._download_json( | |
45 | 'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({ | |
46 | 'brandId': 1, | |
47 | 'videoId': int(video_id), | |
48 | 'autoplay': False, | |
49 | }), video_id) | |
50 | ||
51 | if 'error' in params: | |
52 | raise ExtractorError(params['error']['message'], expected=True) | |
53 | ||
54 | video_url = params['flv'] | |
55 | ||
56 | webpage = self._download_webpage(url, video_id) | |
57 | ||
58 | self.report_extraction(video_id) | |
59 | ||
60 | video_title = self._html_search_regex(r'data-title="(?P<title>.+)"', | |
61 | webpage, 'title', default=None) | |
62 | ||
63 | if not video_title: | |
64 | try: | |
65 | video_title = params['moduleConfig']['meta']['title'] | |
66 | except KeyError: | |
67 | pass | |
68 | ||
69 | if not video_title: | |
70 | video_title = 'Ustream video ' + video_id | |
71 | ||
72 | uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>', | |
73 | webpage, 'uploader', fatal=False, flags=re.DOTALL) | |
74 | ||
75 | thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"', | |
76 | webpage, 'thumbnail', fatal=False) | |
77 | ||
78 | return { | |
79 | 'id': video_id, | |
80 | 'url': video_url, | |
81 | 'ext': 'flv', | |
82 | 'title': video_title, | |
83 | 'uploader': uploader, | |
84 | 'thumbnail': thumbnail, | |
85 | } | |
86 | ||
87 | ||
88 | class UstreamChannelIE(InfoExtractor): | |
89 | _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)' | |
90 | IE_NAME = 'ustream:channel' | |
91 | _TEST = { | |
92 | 'url': 'http://www.ustream.tv/channel/channeljapan', | |
93 | 'info_dict': { | |
94 | 'id': '10874166', | |
95 | }, | |
96 | 'playlist_mincount': 17, | |
97 | } | |
98 | ||
99 | def _real_extract(self, url): | |
100 | m = re.match(self._VALID_URL, url) | |
101 | display_id = m.group('slug') | |
102 | webpage = self._download_webpage(url, display_id) | |
103 | channel_id = self._html_search_meta('ustream:channel_id', webpage) | |
104 | ||
105 | BASE = 'http://www.ustream.tv' | |
106 | next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id | |
107 | video_ids = [] | |
108 | while next_url: | |
109 | reply = self._download_json( | |
110 | compat_urlparse.urljoin(BASE, next_url), display_id, | |
111 | note='Downloading video information (next: %d)' % (len(video_ids) + 1)) | |
112 | video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data'])) | |
113 | next_url = reply['nextUrl'] | |
114 | ||
115 | entries = [ | |
116 | self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream') | |
117 | for vid in video_ids] | |
118 | return { | |
119 | '_type': 'playlist', | |
120 | 'id': channel_id, | |
121 | 'display_id': display_id, | |
122 | 'entries': entries, | |
123 | } |