]> jfr.im git - yt-dlp.git/blame - youtube_dl/extractor/ustream.py
Fixed the ustream extractor to use the current ustream API
[yt-dlp.git] / youtube_dl / extractor / ustream.py
CommitLineData
0eb799ba
JMF
1from __future__ import unicode_literals
2
f8610ba1 3import json
78af8eb1
PH
4import re
5
6from .common import InfoExtractor
1cc79574 7from ..compat import (
74ac9bdd 8 compat_urlparse,
74ac9bdd 9)
762155cc 10from ..utils import ExtractorError
78af8eb1
PH
11
12
13class UstreamIE(InfoExtractor):
b702eceb 14 _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)'
0eb799ba 15 IE_NAME = 'ustream'
cd9fdccd 16 _TESTS = [{
0eb799ba 17 'url': 'http://www.ustream.tv/recorded/20274954',
0eb799ba
JMF
18 'md5': '088f151799e8f572f84eb62f17d73e5c',
19 'info_dict': {
9e875391
S
20 'id': '20274954',
21 'ext': 'flv',
22 'uploader': 'Young Americans for Liberty',
23 'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
0eb799ba 24 },
cd9fdccd
YCH
25 }, {
26 # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
27 # Title and uploader available only from params JSON
28 'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
29 'md5': '5a2abf40babeac9812ed20ae12d34e10',
30 'info_dict': {
31 'id': '59307601',
32 'ext': 'flv',
33 'title': '-CG11- Canada Games Figure Skating',
34 'uploader': 'sportscanadatv',
35 }
36 }]
78af8eb1
PH
37
38 def _real_extract(self, url):
39 m = re.match(self._VALID_URL, url)
b702eceb 40 video_id = m.group('videoID')
41
9e875391
S
42 # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
43 if m.group('type') == 'embed/recorded':
b702eceb 44 video_id = m.group('videoID')
b702eceb 45 desktop_url = 'http://www.ustream.tv/recorded/' + video_id
46 return self.url_result(desktop_url, 'Ustream')
5c386252 47 if m.group('type') == 'embed':
48 video_id = m.group('videoID')
49 webpage = self._download_webpage(url, video_id)
9e875391
S
50 desktop_video_id = self._html_search_regex(
51 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
5c386252 52 desktop_url = 'http://www.ustream.tv/recorded/' + desktop_video_id
53 return self.url_result(desktop_url, 'Ustream')
54
7fd4ed99 55 params = self._download_json('https://api.ustream.tv/videos/' + video_id + '.json', video_id)
f8610ba1 56
762155cc
YCH
57 if 'error' in params:
58 raise ExtractorError(params['error']['message'], expected=True)
59
7fd4ed99 60 video_url = params['video']['media_urls']['flv']
f8610ba1 61
78af8eb1
PH
62 webpage = self._download_webpage(url, video_id)
63
64 self.report_extraction(video_id)
65
66 video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
2a813727
YCH
67 webpage, 'title', default=None)
68
69 if not video_title:
70 try:
71 video_title = params['moduleConfig']['meta']['title']
72 except KeyError:
73 pass
74
75 if not video_title:
76 video_title = 'Ustream video ' + video_id
78af8eb1
PH
77
78 uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
cd9fdccd
YCH
79 webpage, 'uploader', fatal=False, flags=re.DOTALL, default=None)
80
81 if not uploader:
82 try:
83 uploader = params['moduleConfig']['meta']['userName']
84 except KeyError:
85 uploader = None
78af8eb1
PH
86
87 thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
9e1a5b84 88 webpage, 'thumbnail', fatal=False)
0eb799ba
JMF
89
90 return {
91 'id': video_id,
92 'url': video_url,
93 'ext': 'flv',
94 'title': video_title,
95 'uploader': uploader,
96 'thumbnail': thumbnail,
97 }
98
bfd5c93a 99
bfd5c93a 100class UstreamChannelIE(InfoExtractor):
101 _VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
0eb799ba 102 IE_NAME = 'ustream:channel'
22a6f150
PH
103 _TEST = {
104 'url': 'http://www.ustream.tv/channel/channeljapan',
105 'info_dict': {
106 'id': '10874166',
107 },
446a03bd 108 'playlist_mincount': 17,
22a6f150 109 }
bfd5c93a 110
111 def _real_extract(self, url):
112 m = re.match(self._VALID_URL, url)
22a6f150
PH
113 display_id = m.group('slug')
114 webpage = self._download_webpage(url, display_id)
44789f24 115 channel_id = self._html_search_meta('ustream:channel_id', webpage)
bfd5c93a 116
bfd5c93a 117 BASE = 'http://www.ustream.tv'
118 next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
a921f407 119 video_ids = []
bfd5c93a 120 while next_url:
22a6f150
PH
121 reply = self._download_json(
122 compat_urlparse.urljoin(BASE, next_url), display_id,
123 note='Downloading video information (next: %d)' % (len(video_ids) + 1))
a921f407 124 video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
bfd5c93a 125 next_url = reply['nextUrl']
bfd5c93a 126
22a6f150
PH
127 entries = [
128 self.url_result('http://www.ustream.tv/recorded/' + vid, 'Ustream')
129 for vid in video_ids]
130 return {
131 '_type': 'playlist',
132 'id': channel_id,
133 'display_id': display_id,
134 'entries': entries,
135 }