Merge pull request #7320 from remitamine/adobetv
[youtube-dl] / youtube_dl / extractor / ooyala.py
1 from __future__ import unicode_literals
2 import re
3 import base64
4
5 from .common import InfoExtractor
6 from ..utils import (
7     int_or_none,
8     float_or_none,
9     ExtractorError,
10     unsmuggle_url,
11 )
12 from ..compat import compat_urllib_parse
13
14
15 class OoyalaBaseIE(InfoExtractor):
16
17     def _extract(self, content_tree_url, video_id, domain='example.org'):
18         content_tree = self._download_json(content_tree_url, video_id)['content_tree']
19         metadata = content_tree[list(content_tree)[0]]
20         embed_code = metadata['embed_code']
21         pcode = metadata.get('asset_pcode') or embed_code
22         video_info = {
23             'id': embed_code,
24             'title': metadata['title'],
25             'description': metadata.get('description'),
26             'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
27             'duration': int_or_none(metadata.get('duration')),
28         }
29
30         formats = []
31         for supported_format in ('mp4', 'm3u8', 'hds', 'rtmp'):
32             auth_data = self._download_json(
33                 'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?' % (pcode, embed_code) + compat_urllib_parse.urlencode({'domain': domain, 'supportedFormats': supported_format}),
34                 video_id, 'Downloading %s JSON' % supported_format)
35
36             cur_auth_data = auth_data['authorization_data'][embed_code]
37
38             if cur_auth_data['authorized']:
39                 for stream in cur_auth_data['streams']:
40                     url = base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8')
41                     delivery_type = stream['delivery_type']
42                     if delivery_type == 'remote_asset':
43                         video_info['url'] = url
44                         return video_info
45                     if delivery_type == 'hls':
46                         formats.extend(self._extract_m3u8_formats(url, embed_code, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
47                     elif delivery_type == 'hds':
48                         formats.extend(self._extract_f4m_formats(url, embed_code, -1, 'hds', fatal=False))
49                     else:
50                         formats.append({
51                             'url': url,
52                             'ext': stream.get('delivery_type'),
53                             'vcodec': stream.get('video_codec'),
54                             'format_id': '%s-%s-%sp' % (stream.get('profile'), delivery_type, stream.get('height')),
55                             'width': int_or_none(stream.get('width')),
56                             'height': int_or_none(stream.get('height')),
57                             'abr': int_or_none(stream.get('audio_bitrate')),
58                             'vbr': int_or_none(stream.get('video_bitrate')),
59                             'fps': float_or_none(stream.get('framerate')),
60                         })
61             else:
62                 raise ExtractorError('%s said: %s' % (self.IE_NAME, cur_auth_data['message']), expected=True)
63         self._sort_formats(formats)
64
65         video_info['formats'] = formats
66         return video_info
67
68
69 class OoyalaIE(OoyalaBaseIE):
70     _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
71
72     _TESTS = [
73         {
74             # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
75             'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
76             'info_dict': {
77                 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
78                 'ext': 'mp4',
79                 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
80                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
81                 'duration': 853386,
82             },
83         }, {
84             # Only available for ipad
85             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
86             'info_dict': {
87                 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
88                 'ext': 'mp4',
89                 'title': 'Simulation Overview - Levels of Simulation',
90                 'duration': 194948,
91             },
92         },
93         {
94             # Information available only through SAS api
95             # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
96             'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
97             'md5': 'a84001441b35ea492bc03736e59e7935',
98             'info_dict': {
99                 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
100                 'ext': 'mp4',
101                 'title': 'Divide Tool Path.mp4',
102                 'duration': 204405,
103             }
104         }
105     ]
106
107     @staticmethod
108     def _url_for_embed_code(embed_code):
109         return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
110
111     @classmethod
112     def _build_url_result(cls, embed_code):
113         return cls.url_result(cls._url_for_embed_code(embed_code),
114                               ie=cls.ie_key())
115
116     def _real_extract(self, url):
117         url, smuggled_data = unsmuggle_url(url, {})
118         embed_code = self._match_id(url)
119         domain = smuggled_data.get('domain')
120         content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/embed_code/%s/%s' % (embed_code, embed_code)
121         return self._extract(content_tree_url, embed_code, domain)
122
123
124 class OoyalaExternalIE(OoyalaBaseIE):
125     _VALID_URL = r'''(?x)
126                     (?:
127                         ooyalaexternal:|
128                         https?://.+?\.ooyala\.com/.*?\bexternalId=
129                     )
130                     (?P<partner_id>[^:]+)
131                     :
132                     (?P<id>.+)
133                     (?:
134                         :|
135                         .*?&pcode=
136                     )
137                     (?P<pcode>.+?)
138                     (?:&|$)
139                     '''
140
141     _TEST = {
142         'url': 'https://player.ooyala.com/player.js?externalId=espn:10365079&pcode=1kNG061cgaoolOncv54OAO1ceO-I&adSetCode=91cDU6NuXTGKz3OdjOxFdAgJVtQcKJnI&callback=handleEvents&hasModuleParams=1&height=968&playerBrandingId=7af3bd04449c444c964f347f11873075&targetReplaceId=videoPlayer&width=1656&wmode=opaque&allowScriptAccess=always',
143         'info_dict': {
144             'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
145             'ext': 'mp4',
146             'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
147             'duration': 1302000,
148         },
149         'params': {
150             # m3u8 download
151             'skip_download': True,
152         },
153     }
154
155     def _real_extract(self, url):
156         partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
157         content_tree_url = 'http://player.ooyala.com/player_api/v1/content_tree/external_id/%s/%s:%s' % (pcode, partner_id, video_id)
158         return self._extract(content_tree_url, video_id)