]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/threeqsdn.py
Fix extraction from youtube.
[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', 'Failed to parse JSON'],
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': 're:^d755d94b-4ab9-11e3-9162-0025907ad44f [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
35 'is_live': True,
36 },
37 'params': {
38 'skip_download': True, # m3u8 downloads
39 },
40 'expected_warnings': ['Failed to download MPD manifest'],
41 }, {
42 # live audio stream
43 'url': 'http://playout.3qsdn.com/9edf36e0-6bf2-11e2-a16a-9acf09e2db48',
44 'only_matching': True,
45 }, {
46 # live audio stream with some 404 URLs
47 'url': 'http://playout.3qsdn.com/ac5c3186-777a-11e2-9c30-9acf09e2db48',
48 'only_matching': True,
49 }, {
50 # geo restricted with 'This content is not available in your country'
51 'url': 'http://playout.3qsdn.com/d63a3ffe-75e8-11e2-9c30-9acf09e2db48',
52 'only_matching': True,
53 }, {
54 # geo restricted with 'playout.3qsdn.com/forbidden'
55 'url': 'http://playout.3qsdn.com/8e330f26-6ae2-11e2-a16a-9acf09e2db48',
56 'only_matching': True,
57 }, {
58 # live video with rtmp link
59 'url': 'https://playout.3qsdn.com/6092bb9e-8f72-11e4-a173-002590c750be',
60 'only_matching': True,
61 }]
62
63 @staticmethod
64 def _extract_url(webpage):
65 mobj = re.search(
66 r'<iframe[^>]+\b(?:data-)?src=(["\'])(?P<url>%s.*?)\1' % ThreeQSDNIE._VALID_URL, webpage)
67 if mobj:
68 return mobj.group('url')
69
70 def _real_extract(self, url):
71 video_id = self._match_id(url)
72
73 js = self._download_webpage(
74 'http://playout.3qsdn.com/%s' % video_id, video_id,
75 query={'js': 'true'})
76
77 if any(p in js for p in (
78 '>This content is not available in your country',
79 'playout.3qsdn.com/forbidden')):
80 self.raise_geo_restricted()
81
82 stream_content = self._search_regex(
83 r'streamContent\s*:\s*(["\'])(?P<content>.+?)\1', js,
84 'stream content', default='demand', group='content')
85
86 live = stream_content == 'live'
87
88 stream_type = self._search_regex(
89 r'streamType\s*:\s*(["\'])(?P<type>audio|video)\1', js,
90 'stream type', default='video', group='type')
91
92 formats = []
93 urls = set()
94
95 def extract_formats(item_url, item={}):
96 if not item_url or item_url in urls:
97 return
98 urls.add(item_url)
99 ext = mimetype2ext(item.get('type')) or determine_ext(item_url, default_ext=None)
100 if ext == 'mpd':
101 formats.extend(self._extract_mpd_formats(
102 item_url, video_id, mpd_id='mpd', fatal=False))
103 elif ext == 'm3u8':
104 formats.extend(self._extract_m3u8_formats(
105 item_url, video_id, 'mp4',
106 entry_protocol='m3u8' if live else 'm3u8_native',
107 m3u8_id='hls', fatal=False))
108 elif ext == 'f4m':
109 formats.extend(self._extract_f4m_formats(
110 item_url, video_id, f4m_id='hds', fatal=False))
111 else:
112 if not self._is_valid_url(item_url, video_id):
113 return
114 formats.append({
115 'url': item_url,
116 'format_id': item.get('quality'),
117 'ext': 'mp4' if item_url.startswith('rtsp') else ext,
118 'vcodec': 'none' if stream_type == 'audio' else None,
119 })
120
121 for item_js in re.findall(r'({[^{]*?\b(?:src|source)\s*:\s*["\'].+?})', js):
122 f = self._parse_json(
123 item_js, video_id, transform_source=js_to_json, fatal=False)
124 if not f:
125 continue
126 extract_formats(f.get('src'), f)
127
128 # More relaxed version to collect additional URLs and acting
129 # as a future-proof fallback
130 for _, src in re.findall(r'\b(?:src|source)\s*:\s*(["\'])((?:https?|rtsp)://.+?)\1', js):
131 extract_formats(src)
132
133 self._sort_formats(formats)
134
135 title = self._live_title(video_id) if live else video_id
136
137 return {
138 'id': video_id,
139 'title': title,
140 'is_live': live,
141 'formats': formats,
142 }