]> jfr.im git - yt-dlp.git/blame - yt_dlp/extractor/limelight.py
[youtube] Force `hl=en` for comments (#594)
[yt-dlp.git] / yt_dlp / extractor / limelight.py
CommitLineData
ef5acfe3 1# coding: utf-8
2from __future__ import unicode_literals
3
4import re
5
6from .common import InfoExtractor
3444844b 7from ..compat import compat_HTTPError
ef5acfe3 8from ..utils import (
ef5acfe3 9 determine_ext,
d7fc5631
S
10 float_or_none,
11 int_or_none,
e5d39886 12 smuggle_url,
2b4e1ace 13 try_get,
454e5cdb 14 unsmuggle_url,
3444844b 15 ExtractorError,
ef5acfe3 16)
17
18
d7fc5631
S
19class LimelightBaseIE(InfoExtractor):
20 _PLAYLIST_SERVICE_URL = 'http://production-ps.lvp.llnw.net/r/PlaylistService/%s/%s/%s'
ef5acfe3 21
e5d39886
S
22 @classmethod
23 def _extract_urls(cls, webpage, source_url):
24 lm = {
25 'Media': 'media',
26 'Channel': 'channel',
27 'ChannelList': 'channel_list',
28 }
4ef91524
S
29
30 def smuggle(url):
31 return smuggle_url(url, {'source_url': source_url})
32
e5d39886
S
33 entries = []
34 for kind, video_id in re.findall(
35 r'LimelightPlayer\.doLoad(Media|Channel|ChannelList)\(["\'](?P<id>[a-z0-9]{32})',
36 webpage):
e5d39886 37 entries.append(cls.url_result(
4ef91524 38 smuggle('limelight:%s:%s' % (lm[kind], video_id)),
e5d39886
S
39 'Limelight%s' % kind, video_id))
40 for mobj in re.finditer(
41 # As per [1] class attribute should be exactly equal to
42 # LimelightEmbeddedPlayerFlash but numerous examples seen
43 # that don't exactly match it (e.g. [2]).
44 # 1. http://support.3playmedia.com/hc/en-us/articles/227732408-Limelight-Embedding-the-Captions-Plugin-with-the-Limelight-Player-on-Your-Webpage
45 # 2. http://www.sedona.com/FacilitatorTraining2017
46 r'''(?sx)
47 <object[^>]+class=(["\'])(?:(?!\1).)*\bLimelightEmbeddedPlayerFlash\b(?:(?!\1).)*\1[^>]*>.*?
48 <param[^>]+
49 name=(["\'])flashVars\2[^>]+
91bc57e4 50 value=(["\'])(?:(?!\3).)*(?P<kind>media|channel(?:List)?)Id=(?P<id>[a-z0-9]{32})
e5d39886 51 ''', webpage):
91bc57e4 52 kind, video_id = mobj.group('kind'), mobj.group('id')
e5d39886 53 entries.append(cls.url_result(
4ef91524 54 smuggle('limelight:%s:%s' % (kind, video_id)),
91bc57e4 55 'Limelight%s' % kind.capitalize(), video_id))
4ef91524
S
56 # http://support.3playmedia.com/hc/en-us/articles/115009517327-Limelight-Embedding-the-Audio-Description-Plugin-with-the-Limelight-Player-on-Your-Web-Page)
57 for video_id in re.findall(
58 r'(?s)LimelightPlayerUtil\.embed\s*\(\s*{.*?\bmediaId["\']\s*:\s*["\'](?P<id>[a-z0-9]{32})',
59 webpage):
60 entries.append(cls.url_result(
61 smuggle('limelight:media:%s' % video_id),
62 LimelightMediaIE.ie_key(), video_id))
e5d39886
S
63 return entries
64
454e5cdb
RA
65 def _call_playlist_service(self, item_id, method, fatal=True, referer=None):
66 headers = {}
67 if referer:
68 headers['Referer'] = referer
3444844b
RA
69 try:
70 return self._download_json(
71 self._PLAYLIST_SERVICE_URL % (self._PLAYLIST_SERVICE_PATH, item_id, method),
2e20cb36
RA
72 item_id, 'Downloading PlaylistService %s JSON' % method,
73 fatal=fatal, headers=headers)
3444844b
RA
74 except ExtractorError as e:
75 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
76 error = self._parse_json(e.cause.read().decode(), item_id)['detail']['contentAccessPermission']
77 if error == 'CountryDisabled':
78 self.raise_geo_restricted()
79 raise ExtractorError(error, expected=True)
80 raise
ef5acfe3 81
2e20cb36 82 def _extract(self, item_id, pc_method, mobile_method, referer=None):
454e5cdb 83 pc = self._call_playlist_service(item_id, pc_method, referer=referer)
2e20cb36
RA
84 mobile = self._call_playlist_service(
85 item_id, mobile_method, fatal=False, referer=referer)
86 return pc, mobile
87
88 def _extract_info(self, pc, mobile, i, referer):
89 get_item = lambda x, y: try_get(x, lambda x: x[y][i], dict) or {}
90 pc_item = get_item(pc, 'playlistItems')
91 mobile_item = get_item(mobile, 'mediaList')
92 video_id = pc_item.get('mediaId') or mobile_item['mediaId']
93 title = pc_item.get('title') or mobile_item['title']
d7fc5631 94
ef5acfe3 95 formats = []
f8fd510e 96 urls = []
2e20cb36 97 for stream in pc_item.get('streams', []):
d7fc5631 98 stream_url = stream.get('url')
06869367 99 if not stream_url or stream_url in urls:
100 continue
a06916d9 101 if not self.get_param('allow_unplayable_formats') and stream.get('drmProtected'):
d7fc5631 102 continue
f8fd510e 103 urls.append(stream_url)
e382b953
RA
104 ext = determine_ext(stream_url)
105 if ext == 'f4m':
8f1fddc8 106 formats.extend(self._extract_f4m_formats(
e382b953 107 stream_url, video_id, f4m_id='hds', fatal=False))
ef5acfe3 108 else:
109 fmt = {
d7fc5631
S
110 'url': stream_url,
111 'abr': float_or_none(stream.get('audioBitRate')),
d7fc5631 112 'fps': float_or_none(stream.get('videoFrameRate')),
e382b953 113 'ext': ext,
ef5acfe3 114 }
a6f3a162
RA
115 width = int_or_none(stream.get('videoWidthInPixels'))
116 height = int_or_none(stream.get('videoHeightInPixels'))
117 vbr = float_or_none(stream.get('videoBitRate'))
118 if width or height or vbr:
119 fmt.update({
120 'width': width,
121 'height': height,
122 'vbr': vbr,
123 })
124 else:
125 fmt['vcodec'] = 'none'
126 rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
ef5acfe3 127 if rtmp:
d7fc5631
S
128 format_id = 'rtmp'
129 if stream.get('videoBitRate'):
130 format_id += '-%d' % int_or_none(stream['videoBitRate'])
906420ca
S
131 http_format_id = format_id.replace('rtmp', 'http')
132
133 CDN_HOSTS = (
134 ('delvenetworks.com', 'cpl.delvenetworks.com'),
135 ('video.llnw.net', 's2.content.video.llnw.net'),
136 )
137 for cdn_host, http_host in CDN_HOSTS:
138 if cdn_host not in rtmp.group('host').lower():
139 continue
140 http_url = 'http://%s/%s' % (http_host, rtmp.group('playpath')[4:])
141 urls.append(http_url)
142 if self._is_valid_url(http_url, video_id, http_format_id):
143 http_fmt = fmt.copy()
144 http_fmt.update({
145 'url': http_url,
146 'format_id': http_format_id,
147 })
148 formats.append(http_fmt)
149 break
150
ef5acfe3 151 fmt.update({
152 'url': rtmp.group('url'),
153 'play_path': rtmp.group('playpath'),
154 'app': rtmp.group('app'),
d7fc5631
S
155 'ext': 'flv',
156 'format_id': format_id,
ef5acfe3 157 })
158 formats.append(fmt)
159
2e20cb36 160 for mobile_url in mobile_item.get('mobileUrls', []):
d7fc5631 161 media_url = mobile_url.get('mobileUrl')
d7fc5631 162 format_id = mobile_url.get('targetMediaPlatform')
b982cbdd 163 if not media_url or media_url in urls:
164 continue
165 if (format_id in ('Widevine', 'SmoothStreaming')
a06916d9 166 and not self.get_param('allow_unplayable_formats', False)):
e382b953 167 continue
f8fd510e 168 urls.append(media_url)
e382b953
RA
169 ext = determine_ext(media_url)
170 if ext == 'm3u8':
d7fc5631 171 formats.extend(self._extract_m3u8_formats(
8f1fddc8 172 media_url, video_id, 'mp4', 'm3u8_native',
173 m3u8_id=format_id, fatal=False))
e382b953
RA
174 elif ext == 'f4m':
175 formats.extend(self._extract_f4m_formats(
176 stream_url, video_id, f4m_id=format_id, fatal=False))
d7fc5631
S
177 else:
178 formats.append({
179 'url': media_url,
180 'format_id': format_id,
f983b875 181 'quality': -10,
e382b953 182 'ext': ext,
d7fc5631
S
183 })
184
ef5acfe3 185 self._sort_formats(formats)
186
d7fc5631 187 subtitles = {}
2e20cb36
RA
188 for flag in mobile_item.get('flags'):
189 if flag == 'ClosedCaptions':
190 closed_captions = self._call_playlist_service(
191 video_id, 'getClosedCaptionsDetailsByMediaId',
192 False, referer) or []
193 for cc in closed_captions:
194 cc_url = cc.get('webvttFileUrl')
195 if not cc_url:
196 continue
197 lang = cc.get('languageCode') or self._search_regex(r'/[a-z]{2}\.vtt', cc_url, 'lang', default='en')
198 subtitles.setdefault(lang, []).append({
199 'url': cc_url,
200 })
201 break
202
203 get_meta = lambda x: pc_item.get(x) or mobile_item.get(x)
ef5acfe3 204
205 return {
206 'id': video_id,
207 'title': title,
2e20cb36 208 'description': get_meta('description'),
ef5acfe3 209 'formats': formats,
2e20cb36
RA
210 'duration': float_or_none(get_meta('durationInMilliseconds'), 1000),
211 'thumbnail': get_meta('previewImageUrl') or get_meta('thumbnailImageUrl'),
ef5acfe3 212 'subtitles': subtitles,
213 }
214
215
d7fc5631 216class LimelightMediaIE(LimelightBaseIE):
ef5acfe3 217 IE_NAME = 'limelight'
79027c0e
S
218 _VALID_URL = r'''(?x)
219 (?:
220 limelight:media:|
221 https?://
222 (?:
223 link\.videoplatform\.limelight\.com/media/|
224 assets\.delvenetworks\.com/player/loader\.swf
225 )
226 \?.*?\bmediaId=
227 )
228 (?P<id>[a-z0-9]{32})
229 '''
9c544e25 230 _TESTS = [{
ef5acfe3 231 'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
ef5acfe3 232 'info_dict': {
233 'id': '3ffd040b522b4485b6d84effc750cd86',
e382b953 234 'ext': 'mp4',
ef5acfe3 235 'title': 'HaP and the HB Prince Trailer',
9c544e25 236 'description': 'md5:8005b944181778e313d95c1237ddb640',
ec85ded8 237 'thumbnail': r're:^https?://.*\.jpeg$',
d7fc5631 238 'duration': 144.23,
d7fc5631
S
239 },
240 'params': {
e382b953 241 # m3u8 download
d7fc5631
S
242 'skip_download': True,
243 },
9c544e25
S
244 }, {
245 # video with subtitles
246 'url': 'limelight:media:a3e00274d4564ec4a9b29b9466432335',
42b7a5af 247 'md5': '2fa3bad9ac321e23860ca23bc2c69e3d',
9c544e25
S
248 'info_dict': {
249 'id': 'a3e00274d4564ec4a9b29b9466432335',
42b7a5af 250 'ext': 'mp4',
9c544e25 251 'title': '3Play Media Overview Video',
ec85ded8 252 'thumbnail': r're:^https?://.*\.jpeg$',
9c544e25 253 'duration': 78.101,
2e20cb36
RA
254 # TODO: extract all languages that were accessible via API
255 # 'subtitles': 'mincount:9',
256 'subtitles': 'mincount:1',
9c544e25 257 },
79027c0e
S
258 }, {
259 'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
260 'only_matching': True,
9c544e25 261 }]
d7fc5631 262 _PLAYLIST_SERVICE_PATH = 'media'
ef5acfe3 263
264 def _real_extract(self, url):
454e5cdb 265 url, smuggled_data = unsmuggle_url(url, {})
ef5acfe3 266 video_id = self._match_id(url)
2e20cb36 267 source_url = smuggled_data.get('source_url')
5f95927a
S
268 self._initialize_geo_bypass({
269 'countries': smuggled_data.get('geo_countries'),
270 })
ef5acfe3 271
2e20cb36 272 pc, mobile = self._extract(
454e5cdb 273 video_id, 'getPlaylistByMediaId',
2e20cb36 274 'getMobilePlaylistByMediaId', source_url)
ef5acfe3 275
2e20cb36 276 return self._extract_info(pc, mobile, 0, source_url)
ef5acfe3 277
278
d7fc5631 279class LimelightChannelIE(LimelightBaseIE):
ef5acfe3 280 IE_NAME = 'limelight:channel'
79027c0e
S
281 _VALID_URL = r'''(?x)
282 (?:
283 limelight:channel:|
284 https?://
285 (?:
286 link\.videoplatform\.limelight\.com/media/|
287 assets\.delvenetworks\.com/player/loader\.swf
288 )
289 \?.*?\bchannelId=
290 )
291 (?P<id>[a-z0-9]{32})
292 '''
293 _TESTS = [{
ef5acfe3 294 'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
295 'info_dict': {
296 'id': 'ab6a524c379342f9b23642917020c082',
297 'title': 'Javascript Sample Code',
2e20cb36 298 'description': 'Javascript Sample Code - http://www.delvenetworks.com/sample-code/playerCode-demo.html',
ef5acfe3 299 },
300 'playlist_mincount': 3,
79027c0e
S
301 }, {
302 'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
303 'only_matching': True,
304 }]
d7fc5631 305 _PLAYLIST_SERVICE_PATH = 'channel'
ef5acfe3 306
307 def _real_extract(self, url):
454e5cdb 308 url, smuggled_data = unsmuggle_url(url, {})
ef5acfe3 309 channel_id = self._match_id(url)
2e20cb36 310 source_url = smuggled_data.get('source_url')
ef5acfe3 311
2e20cb36 312 pc, mobile = self._extract(
d7fc5631 313 channel_id, 'getPlaylistByChannelId',
454e5cdb 314 'getMobilePlaylistWithNItemsByChannelId?begin=0&count=-1',
2e20cb36 315 source_url)
ef5acfe3 316
d7fc5631 317 entries = [
2e20cb36
RA
318 self._extract_info(pc, mobile, i, source_url)
319 for i in range(len(pc['playlistItems']))]
ef5acfe3 320
2e20cb36
RA
321 return self.playlist_result(
322 entries, channel_id, pc.get('title'), mobile.get('description'))
ef5acfe3 323
324
d7fc5631 325class LimelightChannelListIE(LimelightBaseIE):
ef5acfe3 326 IE_NAME = 'limelight:channel_list'
79027c0e
S
327 _VALID_URL = r'''(?x)
328 (?:
329 limelight:channel_list:|
330 https?://
331 (?:
332 link\.videoplatform\.limelight\.com/media/|
333 assets\.delvenetworks\.com/player/loader\.swf
334 )
335 \?.*?\bchannelListId=
336 )
337 (?P<id>[a-z0-9]{32})
338 '''
339 _TESTS = [{
ef5acfe3 340 'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
341 'info_dict': {
342 'id': '301b117890c4465c8179ede21fd92e2b',
343 'title': 'Website - Hero Player',
344 },
345 'playlist_mincount': 2,
79027c0e
S
346 }, {
347 'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
348 'only_matching': True,
349 }]
d7fc5631 350 _PLAYLIST_SERVICE_PATH = 'channel_list'
ef5acfe3 351
352 def _real_extract(self, url):
353 channel_list_id = self._match_id(url)
354
2e20cb36
RA
355 channel_list = self._call_playlist_service(
356 channel_list_id, 'getMobileChannelListById')
ef5acfe3 357
d7fc5631
S
358 entries = [
359 self.url_result('limelight:channel:%s' % channel['id'], 'LimelightChannel')
360 for channel in channel_list['channelList']]
ef5acfe3 361
2e20cb36
RA
362 return self.playlist_result(
363 entries, channel_list_id, channel_list['title'])