]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/limelight.py
debian/rules: Adapt to new upstream build system.
[youtubedl] / youtube_dl / extractor / limelight.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import (
8 determine_ext,
9 float_or_none,
10 int_or_none,
11 )
12
13
14 class LimelightBaseIE(InfoExtractor):
15 _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
16 _API_URL = 'http://api.video.limelight.com/rest/organizations/%s/%s/%s/%s.json'
17
18 def _call_playlist_service(self, item_id, method, fatal=True):
19 return self._download_json(
20 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
21 item_id, 'Downloading PlaylistService %s JSON' % method, fatal=fatal)
22
23 def _call_api(self, organization_id, item_id, method):
24 return self._download_json(
25 self._API_URL % (organization_id, self._API_PATH, item_id, method),
26 item_id, 'Downloading API %s JSON' % method)
27
28 def _extract(self, item_id, pc_method, mobile_method, meta_method):
29 pc = self._call_playlist_service(item_id, pc_method)
30 metadata = self._call_api(pc['orgId'], item_id, meta_method)
31 mobile = self._call_playlist_service(item_id, mobile_method, fatal=False)
32 return pc, mobile, metadata
33
34 def _extract_info(self, streams, mobile_urls, properties):
35 video_id = properties['media_id']
36 formats = []
37
38 for stream in streams:
39 stream_url = stream.get('url')
40 if not stream_url:
41 continue
42 if '.f4m' in stream_url:
43 formats.extend(self._extract_f4m_formats(
44 stream_url, video_id, fatal=False))
45 else:
46 fmt = {
47 'url': stream_url,
48 'abr': float_or_none(stream.get('audioBitRate')),
49 'vbr': float_or_none(stream.get('videoBitRate')),
50 'fps': float_or_none(stream.get('videoFrameRate')),
51 'width': int_or_none(stream.get('videoWidthInPixels')),
52 'height': int_or_none(stream.get('videoHeightInPixels')),
53 'ext': determine_ext(stream_url)
54 }
55 rtmp = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
56 if rtmp:
57 format_id = 'rtmp'
58 if stream.get('videoBitRate'):
59 format_id += '-%d' % int_or_none(stream['videoBitRate'])
60 fmt.update({
61 'url': rtmp.group('url'),
62 'play_path': rtmp.group('playpath'),
63 'app': rtmp.group('app'),
64 'ext': 'flv',
65 'format_id': format_id,
66 })
67 formats.append(fmt)
68
69 for mobile_url in mobile_urls:
70 media_url = mobile_url.get('mobileUrl')
71 if not media_url:
72 continue
73 format_id = mobile_url.get('targetMediaPlatform')
74 if determine_ext(media_url) == 'm3u8':
75 formats.extend(self._extract_m3u8_formats(
76 media_url, video_id, 'mp4', 'm3u8_native',
77 m3u8_id=format_id, fatal=False))
78 else:
79 formats.append({
80 'url': media_url,
81 'format_id': format_id,
82 'preference': -1,
83 })
84
85 self._sort_formats(formats)
86
87 title = properties['title']
88 description = properties.get('description')
89 timestamp = int_or_none(properties.get('publish_date') or properties.get('create_date'))
90 duration = float_or_none(properties.get('duration_in_milliseconds'), 1000)
91 filesize = int_or_none(properties.get('total_storage_in_bytes'))
92 categories = [properties.get('category')]
93 tags = properties.get('tags', [])
94 thumbnails = [{
95 'url': thumbnail['url'],
96 'width': int_or_none(thumbnail.get('width')),
97 'height': int_or_none(thumbnail.get('height')),
98 } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
99
100 subtitles = {}
101 for caption in properties.get('captions', []):
102 lang = caption.get('language_code')
103 subtitles_url = caption.get('url')
104 if lang and subtitles_url:
105 subtitles.setdefault(lang, []).append({
106 'url': subtitles_url,
107 })
108 closed_captions_url = properties.get('closed_captions_url')
109 if closed_captions_url:
110 subtitles.setdefault('en', []).append({
111 'url': closed_captions_url,
112 'ext': 'ttml',
113 })
114
115 return {
116 'id': video_id,
117 'title': title,
118 'description': description,
119 'formats': formats,
120 'timestamp': timestamp,
121 'duration': duration,
122 'filesize': filesize,
123 'categories': categories,
124 'tags': tags,
125 'thumbnails': thumbnails,
126 'subtitles': subtitles,
127 }
128
129
130 class LimelightMediaIE(LimelightBaseIE):
131 IE_NAME = 'limelight'
132 _VALID_URL = r'''(?x)
133 (?:
134 limelight:media:|
135 https?://
136 (?:
137 link\.videoplatform\.limelight\.com/media/|
138 assets\.delvenetworks\.com/player/loader\.swf
139 )
140 \?.*?\bmediaId=
141 )
142 (?P<id>[a-z0-9]{32})
143 '''
144 _TESTS = [{
145 'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
146 'info_dict': {
147 'id': '3ffd040b522b4485b6d84effc750cd86',
148 'ext': 'flv',
149 'title': 'HaP and the HB Prince Trailer',
150 'description': 'md5:8005b944181778e313d95c1237ddb640',
151 'thumbnail': 're:^https?://.*\.jpeg$',
152 'duration': 144.23,
153 'timestamp': 1244136834,
154 'upload_date': '20090604',
155 },
156 'params': {
157 # rtmp download
158 'skip_download': True,
159 },
160 }, {
161 # video with subtitles
162 'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
163 'info_dict': {
164 'id': 'a3e00274d4564ec4a9b29b9466432335',
165 'ext': 'flv',
166 'title': '3Play Media Overview Video',
167 'description': '',
168 'thumbnail': 're:^https?://.*\.jpeg$',
169 'duration': 78.101,
170 'timestamp': 1338929955,
171 'upload_date': '20120605',
172 'subtitles': 'mincount:9',
173 },
174 'params': {
175 # rtmp download
176 'skip_download': True,
177 },
178 }, {
179 'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
180 'only_matching': True,
181 }]
182 _PLAYLIST_SERVICE_PATH = 'media'
183 _API_PATH = 'media'
184
185 def _real_extract(self, url):
186 video_id = self._match_id(url)
187
188 pc, mobile, metadata = self._extract(
189 video_id, 'getPlaylistByMediaId', 'getMobilePlaylistByMediaId', 'properties')
190
191 return self._extract_info(
192 pc['playlistItems'][0].get('streams', []),
193 mobile['mediaList'][0].get('mobileUrls', []) if mobile else [],
194 metadata)
195
196
197 class LimelightChannelIE(LimelightBaseIE):
198 IE_NAME = 'limelight:channel'
199 _VALID_URL = r'''(?x)
200 (?:
201 limelight:channel:|
202 https?://
203 (?:
204 link\.videoplatform\.limelight\.com/media/|
205 assets\.delvenetworks\.com/player/loader\.swf
206 )
207 \?.*?\bchannelId=
208 )
209 (?P<id>[a-z0-9]{32})
210 '''
211 _TESTS = [{
212 'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
213 'info_dict': {
214 'id': 'ab6a524c379342f9b23642917020c082',
215 'title': 'Javascript Sample Code',
216 },
217 'playlist_mincount': 3,
218 }, {
219 'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
220 'only_matching': True,
221 }]
222 _PLAYLIST_SERVICE_PATH = 'channel'
223 _API_PATH = 'channels'
224
225 def _real_extract(self, url):
226 channel_id = self._match_id(url)
227
228 pc, mobile, medias = self._extract(
229 channel_id, 'getPlaylistByChannelId',
230 'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1', 'media')
231
232 entries = [
233 self._extract_info(
234 pc['playlistItems'][i].get('streams', []),
235 mobile['mediaList'][i].get('mobileUrls', []) if mobile else [],
236 medias['media_list'][i])
237 for i in range(len(medias['media_list']))]
238
239 return self.playlist_result(entries, channel_id, pc['title'])
240
241
242 class LimelightChannelListIE(LimelightBaseIE):
243 IE_NAME = 'limelight:channel_list'
244 _VALID_URL = r'''(?x)
245 (?:
246 limelight:channel_list:|
247 https?://
248 (?:
249 link\.videoplatform\.limelight\.com/media/|
250 assets\.delvenetworks\.com/player/loader\.swf
251 )
252 \?.*?\bchannelListId=
253 )
254 (?P<id>[a-z0-9]{32})
255 '''
256 _TESTS = [{
257 'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
258 'info_dict': {
259 'id': '301b117890c4465c8179ede21fd92e2b',
260 'title': 'Website - Hero Player',
261 },
262 'playlist_mincount': 2,
263 }, {
264 'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
265 'only_matching': True,
266 }]
267 _PLAYLIST_SERVICE_PATH = 'channel_list'
268
269 def _real_extract(self, url):
270 channel_list_id = self._match_id(url)
271
272 channel_list = self._call_playlist_service(channel_list_id, 'getMobileChannelListById')
273
274 entries = [
275 self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
276 for channel in channel_list['channelList']]
277
278 return self.playlist_result(entries, channel_list_id, channel_list['title'])