]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ooyala.py
Update README.md
[youtubedl] / youtube_dl / extractor / ooyala.py
1 from __future__ import unicode_literals
2
3 import base64
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_b64decode,
9 compat_str,
10 )
11 from ..utils import (
12 determine_ext,
13 ExtractorError,
14 float_or_none,
15 int_or_none,
16 try_get,
17 unsmuggle_url,
18 )
19
20
21 class OoyalaBaseIE(InfoExtractor):
22 _PLAYER_BASE = 'http://player.ooyala.com/'
23 _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
24 _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s'
25
26 def _extract(self, content_tree_url, video_id, domain=None, supportedformats=None, embed_token=None):
27 content_tree = self._download_json(content_tree_url, video_id)['content_tree']
28 metadata = content_tree[list(content_tree)[0]]
29 embed_code = metadata['embed_code']
30 pcode = metadata.get('asset_pcode') or embed_code
31 title = metadata['title']
32
33 auth_data = self._download_json(
34 self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code),
35 video_id, headers=self.geo_verification_headers(), query={
36 'domain': domain or 'player.ooyala.com',
37 'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
38 'embedToken': embed_token,
39 })['authorization_data'][embed_code]
40
41 urls = []
42 formats = []
43 streams = auth_data.get('streams') or [{
44 'delivery_type': 'hls',
45 'url': {
46 'data': base64.b64encode(('http://player.ooyala.com/hls/player/all/%s.m3u8' % embed_code).encode()).decode(),
47 }
48 }]
49 for stream in streams:
50 url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
51 if not url_data:
52 continue
53 s_url = compat_b64decode(url_data).decode('utf-8')
54 if not s_url or s_url in urls:
55 continue
56 urls.append(s_url)
57 ext = determine_ext(s_url, None)
58 delivery_type = stream.get('delivery_type')
59 if delivery_type == 'hls' or ext == 'm3u8':
60 formats.extend(self._extract_m3u8_formats(
61 re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
62 m3u8_id='hls', fatal=False))
63 elif delivery_type == 'hds' or ext == 'f4m':
64 formats.extend(self._extract_f4m_formats(
65 s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
66 elif delivery_type == 'dash' or ext == 'mpd':
67 formats.extend(self._extract_mpd_formats(
68 s_url, embed_code, mpd_id='dash', fatal=False))
69 elif delivery_type == 'smooth':
70 self._extract_ism_formats(
71 s_url, embed_code, ism_id='mss', fatal=False)
72 elif ext == 'smil':
73 formats.extend(self._extract_smil_formats(
74 s_url, embed_code, fatal=False))
75 else:
76 formats.append({
77 'url': s_url,
78 'ext': ext or delivery_type,
79 'vcodec': stream.get('video_codec'),
80 'format_id': delivery_type,
81 'width': int_or_none(stream.get('width')),
82 'height': int_or_none(stream.get('height')),
83 'abr': int_or_none(stream.get('audio_bitrate')),
84 'vbr': int_or_none(stream.get('video_bitrate')),
85 'fps': float_or_none(stream.get('framerate')),
86 })
87 if not formats and not auth_data.get('authorized'):
88 raise ExtractorError('%s said: %s' % (
89 self.IE_NAME, auth_data['message']), expected=True)
90 self._sort_formats(formats)
91
92 subtitles = {}
93 for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
94 sub_url = sub.get('url')
95 if not sub_url:
96 continue
97 subtitles[lang] = [{
98 'url': sub_url,
99 }]
100
101 return {
102 'id': embed_code,
103 'title': title,
104 'description': metadata.get('description'),
105 'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
106 'duration': float_or_none(metadata.get('duration'), 1000),
107 'subtitles': subtitles,
108 'formats': formats,
109 }
110
111
112 class OoyalaIE(OoyalaBaseIE):
113 _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
114
115 _TESTS = [
116 {
117 # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
118 'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
119 'info_dict': {
120 'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
121 'ext': 'mp4',
122 'title': 'Explaining Data Recovery from Hard Drives and SSDs',
123 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
124 'duration': 853.386,
125 },
126 # The video in the original webpage now uses PlayWire
127 'skip': 'Ooyala said: movie expired',
128 }, {
129 # Only available for ipad
130 'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
131 'info_dict': {
132 'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
133 'ext': 'mp4',
134 'title': 'Simulation Overview - Levels of Simulation',
135 'duration': 194.948,
136 },
137 },
138 {
139 # Information available only through SAS api
140 # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
141 'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
142 'md5': 'a84001441b35ea492bc03736e59e7935',
143 'info_dict': {
144 'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
145 'ext': 'mp4',
146 'title': 'Divide Tool Path.mp4',
147 'duration': 204.405,
148 }
149 },
150 {
151 # empty stream['url']['data']
152 'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
153 'only_matching': True,
154 }
155 ]
156
157 @staticmethod
158 def _url_for_embed_code(embed_code):
159 return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
160
161 @classmethod
162 def _build_url_result(cls, embed_code):
163 return cls.url_result(cls._url_for_embed_code(embed_code),
164 ie=cls.ie_key())
165
166 def _real_extract(self, url):
167 url, smuggled_data = unsmuggle_url(url, {})
168 embed_code = self._match_id(url)
169 domain = smuggled_data.get('domain')
170 supportedformats = smuggled_data.get('supportedformats')
171 embed_token = smuggled_data.get('embed_token')
172 content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
173 return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
174
175
176 class OoyalaExternalIE(OoyalaBaseIE):
177 _VALID_URL = r'''(?x)
178 (?:
179 ooyalaexternal:|
180 https?://.+?\.ooyala\.com/.*?\bexternalId=
181 )
182 (?P<partner_id>[^:]+)
183 :
184 (?P<id>.+)
185 (?:
186 :|
187 .*?&pcode=
188 )
189 (?P<pcode>.+?)
190 (?:&|$)
191 '''
192
193 _TEST = {
194 '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',
195 'info_dict': {
196 'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
197 'ext': 'mp4',
198 'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
199 'duration': 1302.0,
200 },
201 'params': {
202 # m3u8 download
203 'skip_download': True,
204 },
205 }
206
207 def _real_extract(self, url):
208 partner_id, video_id, pcode = re.match(self._VALID_URL, url).groups()
209 content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
210 return self._extract(content_tree_url, video_id)