]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/threeqsdn.py
Update changelog.
[youtubedl] / youtube_dl / extractor / threeqsdn.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7 determine_ext,
8 js_to_json,
9 mimetype2ext,
10 )
11
12
13 class ThreeQSDNIE(InfoExtractor):
14 IE_NAME = '3qsdn'
15 IE_DESC = '3Q SDN'
16 _VALID_URL = r'https?://playout\.3qsdn\.com/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
17 _TESTS = [{
18 # ondemand from http://www.philharmonie.tv/veranstaltung/26/
19 'url': 'http://playout.3qsdn.com/0280d6b9-1215-11e6-b427-0cc47a188158?protocol=http',
20 'md5': 'ab040e37bcfa2e0c079f92cb1dd7f6cd',
21 'info_dict': {
22 'id': '0280d6b9-1215-11e6-b427-0cc47a188158',
23 'ext': 'mp4',
24 'title': '0280d6b9-1215-11e6-b427-0cc47a188158',
25 'is_live': False,
26 },
27 'expected_warnings': ['Failed to download MPD manifest'],
28 }, {
29 # live video stream
30 'url': 'https://playout.3qsdn.com/d755d94b-4ab9-11e3-9162-0025907ad44f?js=true',
31 'info_dict': {
32 'id': 'd755d94b-4ab9-11e3-9162-0025907ad44f',
33 'ext': 'mp4',
34 'title': 'd755d94b-4ab9-11e3-9162-0025907ad44f',
35 'is_live': False,
36 },
37 }, {
38 # live audio stream
39 'url': 'http://playout.3qsdn.com/9edf36e0-6bf2-11e2-a16a-9acf09e2db48',
40 'only_matching': True,
41 }, {
42 # live audio stream with some 404 URLs
43 'url': 'http://playout.3qsdn.com/ac5c3186-777a-11e2-9c30-9acf09e2db48',
44 'only_matching': True,
45 }, {
46 # geo restricted with 'This content is not available in your country'
47 'url': 'http://playout.3qsdn.com/d63a3ffe-75e8-11e2-9c30-9acf09e2db48',
48 'only_matching': True,
49 }, {
50 # geo restricted with 'playout.3qsdn.com/forbidden'
51 'url': 'http://playout.3qsdn.com/8e330f26-6ae2-11e2-a16a-9acf09e2db48',
52 'only_matching': True,
53 }, {
54 # live video with rtmp link
55 'url': 'https://playout.3qsdn.com/6092bb9e-8f72-11e4-a173-002590c750be',
56 'only_matching': True,
57 }]
58
59 @staticmethod
60 def _extract_url(webpage):
61 mobj = re.search(
62 r'<iframe[^>]+\b(?:data-)?src=(["\'])(?P<url>%s.*?)\1' % ThreeQSDNIE._VALID_URL, webpage)
63 if mobj:
64 return mobj.group('url')
65
66 def _real_extract(self, url):
67 video_id = self._match_id(url)
68
69 js = self._download_webpage(
70 'http://playout.3qsdn.com/%s' % video_id, video_id,
71 query={'js': 'true'})
72
73 if any(p in js for p in (
74 '>This content is not available in your country',
75 'playout.3qsdn.com/forbidden')):
76 self.raise_geo_restricted()
77
78 stream_content = self._search_regex(
79 r'streamContent\s*:\s*(["\'])(?P<content>.+?)\1', js,
80 'stream content', default='demand', group='content')
81
82 live = stream_content == 'live'
83
84 stream_type = self._search_regex(
85 r'streamType\s*:\s*(["\'])(?P<type>audio|video)\1', js,
86 'stream type', default='video', group='type')
87
88 formats = []
89 urls = set()
90
91 def extract_formats(item_url, item={}):
92 if not item_url or item_url in urls:
93 return
94 urls.add(item_url)
95 type_ = item.get('type')
96 ext = determine_ext(item_url, default_ext=None)
97 if type_ == 'application/dash+xml' or ext == 'mpd':
98 formats.extend(self._extract_mpd_formats(
99 item_url, video_id, mpd_id='mpd', fatal=False))
100 elif type_ in ('application/vnd.apple.mpegURL', 'application/x-mpegurl') or ext == 'm3u8':
101 formats.extend(self._extract_m3u8_formats(
102 item_url, video_id, 'mp4',
103 entry_protocol='m3u8' if live else 'm3u8_native',
104 m3u8_id='hls', fatal=False))
105 elif ext == 'f4m':
106 formats.extend(self._extract_f4m_formats(
107 item_url, video_id, f4m_id='hds', fatal=False))
108 else:
109 if not self._is_valid_url(item_url, video_id):
110 return
111 formats.append({
112 'url': item_url,
113 'format_id': item.get('quality'),
114 'ext': 'mp4' if item_url.startswith('rtsp') else mimetype2ext(type_) or ext,
115 'vcodec': 'none' if stream_type == 'audio' else None,
116 })
117
118 for item_js in re.findall(r'({.*?\b(?:src|source)\s*:\s*["\'].+?})', js):
119 f = self._parse_json(
120 item_js, video_id, transform_source=js_to_json, fatal=False)
121 if not f:
122 continue
123 extract_formats(f.get('src'), f)
124
125 # More relaxed version to collect additional URLs and acting
126 # as a future-proof fallback
127 for _, src in re.findall(r'\b(?:src|source)\s*:\s*(["\'])((?:https?|rtsp)://.+?)\1', js):
128 extract_formats(src)
129
130 self._sort_formats(formats)
131
132 title = self._live_title(video_id) if live else video_id
133
134 return {
135 'id': video_id,
136 'title': title,
137 'is_live': live,
138 'formats': formats,
139 }