]>
Commit | Line | Data |
---|---|---|
160c2773 | 1 | import base64 |
160c2773 RA |
2 | import re |
3 | import struct | |
add96eb9 | 4 | import urllib.parse |
f9934b96 | 5 | import xml.etree.ElementTree |
fbaaad49 | 6 | |
5996d21a | 7 | from .adobepass import AdobePassIE |
c8b232cc | 8 | from .common import InfoExtractor |
add96eb9 | 9 | from ..compat import compat_etree_fromstring |
3d2623a8 | 10 | from ..networking.exceptions import HTTPError |
1cc79574 | 11 | from ..utils import ( |
e897bd82 SS |
12 | ExtractorError, |
13 | UnsupportedError, | |
c8b232cc | 14 | clean_html, |
487c5b33 | 15 | dict_get, |
40158f55 | 16 | extract_attributes, |
1cc79574 PH |
17 | find_xpath_attr, |
18 | fix_xml_ampersands, | |
a6624898 | 19 | float_or_none, |
ed126900 | 20 | int_or_none, |
c8b232cc S |
21 | js_to_json, |
22 | mimetype2ext, | |
ed126900 | 23 | parse_iso8601, |
4dfbf869 | 24 | parse_qs, |
f455a934 | 25 | smuggle_url, |
b4eb0bc7 | 26 | str_or_none, |
29f7c58a | 27 | try_get, |
fd91257c S |
28 | unescapeHTML, |
29 | unsmuggle_url, | |
48254c3f | 30 | update_url_query, |
b4eb0bc7 | 31 | url_or_none, |
cfe50f04 | 32 | ) |
fbaaad49 | 33 | |
dd5bcdc4 | 34 | |
4fcaa4f4 | 35 | class BrightcoveLegacyIE(InfoExtractor): |
e721d857 | 36 | IE_NAME = 'brightcove:legacy' |
ec05fee4 | 37 | _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)' |
592882aa JMF |
38 | |
39 | _TESTS = [ | |
40 | { | |
4de1994b | 41 | # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ |
400e5810 | 42 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', |
400e5810 PH |
43 | 'md5': '5423e113865d26e40624dce2e4b45d95', |
44 | 'note': 'Test Brightcove downloads and detection in GenericIE', | |
45 | 'info_dict': { | |
96bef88f JMF |
46 | 'id': '2371591881001', |
47 | 'ext': 'mp4', | |
400e5810 PH |
48 | 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', |
49 | 'uploader': '8TV', | |
50 | 'description': 'md5:a950cc4285c43e44d763d036710cd9cd', | |
41f5492f | 51 | 'timestamp': 1368213670, |
52 | 'upload_date': '20130510', | |
4d4cd35f | 53 | 'uploader_id': '1589608506001', |
f455a934 RA |
54 | }, |
55 | 'skip': 'The player has been deactivated by the content owner', | |
592882aa JMF |
56 | }, |
57 | { | |
4de1994b | 58 | # From http://medianetwork.oracle.com/video/player/1785452137001 |
400e5810 | 59 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001', |
400e5810 | 60 | 'info_dict': { |
96bef88f JMF |
61 | 'id': '1785452137001', |
62 | 'ext': 'flv', | |
400e5810 PH |
63 | 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges', |
64 | 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.', | |
65 | 'uploader': 'Oracle', | |
41f5492f | 66 | 'timestamp': 1344975024, |
67 | 'upload_date': '20120814', | |
4d4cd35f | 68 | 'uploader_id': '1460825906', |
592882aa | 69 | }, |
f455a934 | 70 | 'skip': 'video not playable', |
592882aa | 71 | }, |
fc4a0c2a JMF |
72 | { |
73 | # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/ | |
400e5810 PH |
74 | 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001', |
75 | 'info_dict': { | |
76 | 'id': '2750934548001', | |
77 | 'ext': 'mp4', | |
78 | 'title': 'This Bracelet Acts as a Personal Thermostat', | |
79 | 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0', | |
f455a934 | 80 | # 'uploader': 'Mashable', |
41f5492f | 81 | 'timestamp': 1382041798, |
82 | 'upload_date': '20131017', | |
4d4cd35f | 83 | 'uploader_id': '1130468786001', |
fc4a0c2a JMF |
84 | }, |
85 | }, | |
77526143 JMF |
86 | { |
87 | # test that the default referer works | |
88 | # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ | |
400e5810 PH |
89 | 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', |
90 | 'info_dict': { | |
91 | 'id': '2878862109001', | |
92 | 'ext': 'mp4', | |
93 | 'title': 'Lost in Motion II', | |
94 | 'description': 'md5:363109c02998fee92ec02211bd8000df', | |
95 | 'uploader': 'National Ballet of Canada', | |
77526143 | 96 | }, |
c65aa4e9 | 97 | 'skip': 'Video gone', |
5524b242 JMF |
98 | }, |
99 | { | |
100 | # test flv videos served by akamaihd.net | |
101 | # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william | |
41f5492f | 102 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?%40videoPlayer=ref%3Aevent-stream-356&linkBaseURL=http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fvideos%2F1331655630249%2Freplay-uci-fort-william-2014-dh&playerKey=AQ%7E%7E%2CAAAApYJ7UqE%7E%2Cxqr_zXk0I-zzNndy8NlHogrCb5QdyZRf&playerID=1398061561001#__youtubedl_smuggle=%7B%22Referer%22%3A+%22http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fstories%2F1331655643987%2Freplay-uci-dh-world-cup-2014-from-fort-william%22%7D', |
5524b242 JMF |
103 | # The md5 checksum changes on each download |
104 | 'info_dict': { | |
41f5492f | 105 | 'id': '3750436379001', |
5524b242 JMF |
106 | 'ext': 'flv', |
107 | 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', | |
41f5492f | 108 | 'uploader': 'RBTV Old (do not use)', |
5524b242 | 109 | 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', |
41f5492f | 110 | 'timestamp': 1409122195, |
111 | 'upload_date': '20140827', | |
4d4cd35f | 112 | 'uploader_id': '710858724001', |
5524b242 | 113 | }, |
40158f55 | 114 | 'skip': 'Video gone', |
5524b242 | 115 | }, |
bd4e40df | 116 | { |
c65aa4e9 | 117 | # playlist with 'videoList' |
bd4e40df JMF |
118 | # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players |
119 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL', | |
120 | 'info_dict': { | |
121 | 'title': 'Sealife', | |
d47c26e1 | 122 | 'id': '3550319591001', |
bd4e40df JMF |
123 | }, |
124 | 'playlist_mincount': 7, | |
f455a934 | 125 | 'skip': 'Unsupported URL', |
bd4e40df | 126 | }, |
c65aa4e9 | 127 | { |
067aa17e | 128 | # playlist with 'playlistTab' (https://github.com/ytdl-org/youtube-dl/issues/9965) |
c65aa4e9 YCH |
129 | 'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg', |
130 | 'info_dict': { | |
131 | 'id': '1522758701001', | |
132 | 'title': 'Lesson 08', | |
133 | }, | |
134 | 'playlist_mincount': 10, | |
f455a934 | 135 | 'skip': 'Unsupported URL', |
c65aa4e9 | 136 | }, |
06d0ad9a YCH |
137 | { |
138 | # playerID inferred from bcpid | |
139 | # from http://www.un.org/chinese/News/story.asp?NewsID=27724 | |
140 | 'url': 'https://link.brightcove.com/services/player/bcpid1722935254001/?bctid=5360463607001&autoStart=false&secureConnections=true&width=650&height=350', | |
141 | 'only_matching': True, # Tested in GenericIE | |
add96eb9 | 142 | }, |
592882aa | 143 | ] |
cfe50f04 | 144 | |
f5a9e9df | 145 | _WEBPAGE_TESTS = [{ |
146 | # embedded brightcove video | |
147 | # it also tests brightcove videos that need to set the 'Referer' | |
148 | # in the http requests | |
149 | 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', | |
150 | 'info_dict': { | |
151 | 'id': '2765128793001', | |
152 | 'ext': 'mp4', | |
153 | 'title': 'Le cours de bourse : l’analyse technique', | |
154 | 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9', | |
155 | 'uploader': 'BFM BUSINESS', | |
156 | }, | |
157 | 'params': { | |
158 | 'skip_download': True, | |
159 | }, | |
160 | 'skip': '404 Not Found', | |
161 | }, { | |
162 | # embedded with itemprop embedURL and video id spelled as `idVideo` | |
163 | 'url': 'http://bfmbusiness.bfmtv.com/mediaplayer/chroniques/olivier-delamarche/', | |
164 | 'info_dict': { | |
165 | 'id': '5255628253001', | |
166 | 'ext': 'mp4', | |
167 | 'title': 'md5:37c519b1128915607601e75a87995fc0', | |
168 | 'description': 'md5:37f7f888b434bb8f8cc8dbd4f7a4cf26', | |
169 | 'uploader': 'BFM BUSINESS', | |
170 | 'uploader_id': '876450612001', | |
171 | 'timestamp': 1482255315, | |
172 | 'upload_date': '20161220', | |
173 | }, | |
174 | 'params': { | |
175 | 'skip_download': True, | |
176 | }, | |
177 | 'skip': 'Redirects, page gone', | |
178 | }, { | |
179 | # https://github.com/ytdl-org/youtube-dl/issues/2253 | |
180 | 'url': 'http://bcove.me/i6nfkrc3', | |
181 | 'md5': '0ba9446db037002366bab3b3eb30c88c', | |
182 | 'info_dict': { | |
183 | 'id': '3101154703001', | |
184 | 'ext': 'mp4', | |
185 | 'title': 'Still no power', | |
186 | 'uploader': 'thestar.com', | |
187 | 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.', | |
188 | }, | |
189 | 'skip': 'video gone', | |
190 | }, { | |
191 | # https://github.com/ytdl-org/youtube-dl/issues/3541 | |
192 | 'url': 'http://www.kijk.nl/sbs6/leermijvrouwenkennen/videos/jqMiXKAYan2S/aflevering-1', | |
193 | 'info_dict': { | |
194 | 'id': '3866516442001', | |
195 | 'ext': 'mp4', | |
196 | 'title': 'Leer mij vrouwen kennen: Aflevering 1', | |
197 | 'description': 'Leer mij vrouwen kennen: Aflevering 1', | |
198 | 'uploader': 'SBS Broadcasting', | |
199 | }, | |
200 | 'skip': 'Restricted to Netherlands, 404 Not Found', | |
201 | 'params': { | |
202 | 'skip_download': True, # m3u8 download | |
203 | }, | |
204 | }, { | |
205 | # Brightcove video in <iframe> | |
206 | 'url': 'http://www.un.org/chinese/News/story.asp?NewsID=27724', | |
207 | 'md5': '36d74ef5e37c8b4a2ce92880d208b968', | |
208 | 'info_dict': { | |
209 | 'id': '5360463607001', | |
210 | 'ext': 'mp4', | |
211 | 'title': '叙利亚失明儿童在废墟上演唱《心跳》 呼吁获得正常童年生活', | |
212 | 'description': '联合国儿童基金会中东和北非区域大使、作曲家扎德·迪拉尼(Zade Dirani)在3月15日叙利亚冲突爆发7周年纪念日之际发布了为叙利亚谱写的歌曲《心跳》(HEARTBEAT),为受到六年冲突影响的叙利亚儿童发出强烈呐喊,呼吁世界做出共同努力,使叙利亚儿童重新获得享有正常童年生活的权利。', | |
213 | 'uploader': 'United Nations', | |
214 | 'uploader_id': '1362235914001', | |
215 | 'timestamp': 1489593889, | |
216 | 'upload_date': '20170315', | |
217 | }, | |
218 | 'skip': '404 Not Found', | |
219 | }, { | |
220 | # Brightcove with UUID in videoPlayer | |
221 | 'url': 'http://www8.hp.com/cn/zh/home.html', | |
222 | 'info_dict': { | |
223 | 'id': '5255815316001', | |
224 | 'ext': 'mp4', | |
225 | 'title': 'Sprocket Video - China', | |
226 | 'description': 'Sprocket Video - China', | |
227 | 'uploader': 'HP-Video Gallery', | |
228 | 'timestamp': 1482263210, | |
229 | 'upload_date': '20161220', | |
230 | 'uploader_id': '1107601872001', | |
231 | }, | |
232 | 'params': { | |
233 | 'skip_download': True, # m3u8 download | |
234 | }, | |
235 | 'skip': 'video rotates...weekly?', | |
236 | }, { | |
237 | # Multiple brightcove videos | |
238 | # https://github.com/ytdl-org/youtube-dl/issues/2283 | |
239 | 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html', | |
240 | 'info_dict': { | |
241 | 'id': 'always-never', | |
242 | 'title': 'Always / Never - The New Yorker', | |
243 | }, | |
244 | 'playlist_count': 3, | |
245 | 'params': { | |
246 | 'extract_flat': False, | |
247 | 'skip_download': True, | |
248 | }, | |
249 | 'skip': 'Redirects, page gone', | |
250 | }, { | |
251 | # BrightcoveInPageEmbed embed | |
252 | 'url': 'http://www.geekandsundry.com/tabletop-bonus-wils-final-thoughts-on-dread/', | |
253 | 'info_dict': { | |
254 | 'id': '4238694884001', | |
255 | 'ext': 'flv', | |
256 | 'title': 'Tabletop: Dread, Last Thoughts', | |
257 | 'description': 'Tabletop: Dread, Last Thoughts', | |
258 | 'duration': 51690, | |
259 | }, | |
260 | 'skip': 'Redirects, page gone', | |
261 | }, { | |
262 | # Brightcove embed, with no valid 'renditions' but valid 'IOSRenditions' | |
263 | # This video can't be played in browsers if Flash disabled and UA set to iPhone, which is actually a false alarm | |
264 | 'url': 'https://dl.dropboxusercontent.com/u/29092637/interview.html', | |
265 | 'info_dict': { | |
266 | 'id': '4785848093001', | |
267 | 'ext': 'mp4', | |
268 | 'title': 'The Cardinal Pell Interview', | |
269 | 'description': 'Sky News Contributor Andrew Bolt interviews George Pell in Rome, following the Cardinal\'s evidence before the Royal Commission into Child Abuse. ', | |
270 | 'uploader': 'GlobeCast Australia - GlobeStream', | |
271 | 'uploader_id': '2733773828001', | |
272 | 'upload_date': '20160304', | |
273 | 'timestamp': 1457083087, | |
274 | }, | |
275 | 'params': { | |
276 | # m3u8 downloads | |
277 | 'skip_download': True, | |
278 | }, | |
279 | 'skip': '404 Not Found', | |
280 | }, { | |
281 | # Brightcove embed with whitespace around attribute names | |
282 | 'url': 'http://www.stack.com/video/3167554373001/learn-to-hit-open-three-pointers-with-damian-lillard-s-baseline-drift-drill', | |
283 | 'info_dict': { | |
284 | 'id': '3167554373001', | |
285 | 'ext': 'mp4', | |
286 | 'title': "Learn to Hit Open Three-Pointers With Damian Lillard's Baseline Drift Drill", | |
287 | 'description': 'md5:57bacb0e0f29349de4972bfda3191713', | |
288 | 'uploader_id': '1079349493', | |
289 | 'upload_date': '20140207', | |
290 | 'timestamp': 1391810548, | |
291 | }, | |
292 | 'params': { | |
293 | 'skip_download': True, | |
294 | }, | |
295 | 'skip': '410 Gone', | |
296 | }] | |
297 | ||
cfe50f04 | 298 | @classmethod |
a0566bbf | 299 | def _build_brightcove_url(cls, object_str): |
cfe50f04 JMF |
300 | """ |
301 | Build a Brightcove url from a xml string containing | |
302 | <object class="BrightcoveExperience">{params}</object> | |
303 | """ | |
46e28a84 | 304 | |
067aa17e | 305 | # Fix up some stupid HTML, see https://github.com/ytdl-org/youtube-dl/issues/1553 |
c1147c05 | 306 | object_str = re.sub(r'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>', |
46e28a84 | 307 | lambda m: m.group(1) + '/>', object_str) |
067aa17e | 308 | # Fix up some stupid XML, see https://github.com/ytdl-org/youtube-dl/issues/1608 |
400e5810 | 309 | object_str = object_str.replace('<--', '<!--') |
d6f714f3 JMF |
310 | # remove namespace to simplify extraction |
311 | object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str) | |
7b0817e8 | 312 | object_str = fix_xml_ampersands(object_str) |
46e28a84 | 313 | |
94c12557 | 314 | try: |
add96eb9 | 315 | object_doc = compat_etree_fromstring(object_str.encode()) |
f9934b96 | 316 | except xml.etree.ElementTree.ParseError: |
94c12557 | 317 | return |
7b0817e8 PH |
318 | |
319 | fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars') | |
47917f24 JMF |
320 | if fv_el is not None: |
321 | flashvars = dict( | |
322 | (k, v[0]) | |
add96eb9 | 323 | for k, v in urllib.parse.parse_qs(fv_el.attrib['value']).items()) |
47917f24 JMF |
324 | else: |
325 | flashvars = {} | |
7b0817e8 | 326 | |
5964b598 | 327 | data_url = object_doc.attrib.get('data', '') |
4dfbf869 | 328 | data_url_params = parse_qs(data_url) |
5964b598 | 329 | |
36de0a0e | 330 | def find_param(name): |
7b0817e8 PH |
331 | if name in flashvars: |
332 | return flashvars[name] | |
d214fdb8 JMF |
333 | node = find_xpath_attr(object_doc, './param', 'name', name) |
334 | if node is not None: | |
335 | return node.attrib['value'] | |
5964b598 | 336 | return data_url_params.get(name) |
7b0817e8 PH |
337 | |
338 | params = {} | |
339 | ||
add96eb9 | 340 | player_id = find_param('playerID') or find_param('playerId') |
341 | if player_id is None: | |
7b0817e8 | 342 | raise ExtractorError('Cannot find player ID') |
add96eb9 | 343 | params['playerID'] = player_id |
7b0817e8 | 344 | |
add96eb9 | 345 | player_key = find_param('playerKey') |
cfe50f04 | 346 | # Not all pages define this value |
add96eb9 | 347 | if player_key is not None: |
348 | params['playerKey'] = player_key | |
48254c3f | 349 | # These fields hold the id of the video |
add96eb9 | 350 | video_player = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList') |
351 | if video_player is not None: | |
352 | if isinstance(video_player, list): | |
353 | video_player = video_player[0] | |
354 | video_player = video_player.strip() | |
2e76bdc8 S |
355 | # UUID is also possible for videoPlayer (e.g. |
356 | # http://www.popcornflix.com/hoodies-vs-hooligans/7f2d2b87-bbf2-4623-acfb-ea942b4f01dd | |
357 | # or http://www8.hp.com/cn/zh/home.html) | |
358 | if not (re.match( | |
359 | r'^(?:\d+|[\da-fA-F]{8}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{12})$', | |
add96eb9 | 360 | video_player) or video_player.startswith('ref:')): |
049a0f4d | 361 | return None |
add96eb9 | 362 | params['@videoPlayer'] = video_player |
363 | link_base = find_param('linkBaseURL') | |
364 | if link_base is not None: | |
365 | params['linkBaseURL'] = link_base | |
af9cdee9 S |
366 | return cls._make_brightcove_url(params) |
367 | ||
368 | @classmethod | |
a0566bbf | 369 | def _build_brightcove_url_from_js(cls, object_js): |
af9cdee9 S |
370 | # The layout of JS is as follows: |
371 | # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) { | |
372 | # // build Brightcove <object /> XML | |
373 | # } | |
374 | m = re.search( | |
4f66c16f | 375 | r'''(?x)customBC\.createVideo\( |
af9cdee9 S |
376 | .*? # skipping width and height |
377 | ["\'](?P<playerID>\d+)["\']\s*,\s* # playerID | |
378 | ["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters | |
379 | # in length, however it's appended to itself | |
380 | # in places, so truncate | |
381 | ["\'](?P<videoID>\d+)["\'] # @videoPlayer | |
382 | ''', object_js) | |
383 | if m: | |
384 | return cls._make_brightcove_url(m.groupdict()) | |
385 | ||
386 | @classmethod | |
387 | def _make_brightcove_url(cls, params): | |
f455a934 RA |
388 | return update_url_query( |
389 | 'http://c.brightcove.com/services/viewer/htmlFederated', params) | |
fbaaad49 | 390 | |
eeb165e6 JMF |
391 | @classmethod |
392 | def _extract_brightcove_url(cls, webpage): | |
99877772 | 393 | """Try to extract the brightcove url from the webpage, returns None |
eeb165e6 JMF |
394 | if it can't be found |
395 | """ | |
99877772 PH |
396 | urls = cls._extract_brightcove_urls(webpage) |
397 | return urls[0] if urls else None | |
398 | ||
399 | @classmethod | |
400 | def _extract_brightcove_urls(cls, webpage): | |
401 | """Return a list of all Brightcove URLs from the webpage """ | |
117bec93 | 402 | |
e1ab5000 | 403 | url_m = re.search( |
53a664ed S |
404 | r'''(?x) |
405 | <meta\s+ | |
406 | (?:property|itemprop)=([\'"])(?:og:video|embedURL)\1[^>]+ | |
407 | content=([\'"])(?P<url>https?://(?:secure|c)\.brightcove.com/(?:(?!\2).)+)\2 | |
408 | ''', webpage) | |
117bec93 | 409 | if url_m: |
53a664ed | 410 | url = unescapeHTML(url_m.group('url')) |
381640e3 JMF |
411 | # Some sites don't add it, we can't download with this url, for example: |
412 | # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/ | |
53a664ed | 413 | if 'playerKey' in url or 'videoId' in url or 'idVideo' in url: |
381640e3 | 414 | return [url] |
117bec93 | 415 | |
99877772 | 416 | matches = re.findall( |
7b0817e8 PH |
417 | r'''(?sx)<object |
418 | (?: | |
99877772 | 419 | [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] | |
7b0817e8 | 420 | [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/ |
476e1095 | 421 | ).+?>\s*</object>''', |
7b0817e8 | 422 | webpage) |
b4e1576a | 423 | if matches: |
a0566bbf | 424 | return list(filter(None, [cls._build_brightcove_url(m) for m in matches])) |
b4e1576a | 425 | |
06d0ad9a YCH |
426 | matches = re.findall(r'(customBC\.createVideo\(.+?\);)', webpage) |
427 | if matches: | |
428 | return list(filter(None, [ | |
a0566bbf | 429 | cls._build_brightcove_url_from_js(custom_bc) |
06d0ad9a | 430 | for custom_bc in matches])) |
481ef51e | 431 | return [src for _, src in re.findall( |
06d0ad9a | 432 | r'<iframe[^>]+src=([\'"])((?:https?:)?//link\.brightcove\.com/services/player/(?!\1).+)\1', webpage)] |
eeb165e6 | 433 | |
f5a9e9df | 434 | def _extract_from_webpage(self, url, webpage): |
435 | bc_urls = self._extract_brightcove_urls(webpage) | |
436 | for bc_url in bc_urls: | |
437 | yield self.url_result(smuggle_url(bc_url, {'Referer': url}), BrightcoveLegacyIE) | |
438 | ||
fbaaad49 | 439 | def _real_extract(self, url): |
79f82953 PH |
440 | url, smuggled_data = unsmuggle_url(url, {}) |
441 | ||
51040b72 | 442 | # Change the 'videoId' and others field to '@videoPlayer' |
53a664ed | 443 | url = re.sub(r'(?<=[?&])(videoI(d|D)|idVideo|bctid)', '%40videoPlayer', url) |
51040b72 JMF |
444 | # Change bckey (used by bcove.me urls) to playerKey |
445 | url = re.sub(r'(?<=[?&])bckey', 'playerKey', url) | |
5ad28e7f | 446 | mobj = self._match_valid_url(url) |
6543f0dc | 447 | query_str = mobj.group('query') |
add96eb9 | 448 | query = urllib.parse.parse_qs(query_str) |
fbaaad49 | 449 | |
add96eb9 | 450 | video_player = query.get('@videoPlayer') |
451 | if video_player: | |
79f82953 | 452 | # We set the original url as the default 'Referer' header |
f455a934 | 453 | referer = query.get('linkBaseURL', [None])[0] or smuggled_data.get('Referer', url) |
add96eb9 | 454 | video_id = video_player[0] |
06d0ad9a YCH |
455 | if 'playerID' not in query: |
456 | mobj = re.search(r'/bcpid(\d+)', url) | |
457 | if mobj is not None: | |
458 | query['playerID'] = [mobj.group(1)] | |
160c2773 RA |
459 | publisher_id = query.get('publisherId') |
460 | if publisher_id and publisher_id[0].isdigit(): | |
461 | publisher_id = publisher_id[0] | |
462 | if not publisher_id: | |
160c2773 RA |
463 | player_key = query.get('playerKey') |
464 | if player_key and ',' in player_key[0]: | |
465 | player_key = player_key[0] | |
466 | else: | |
467 | player_id = query.get('playerID') | |
468 | if player_id and player_id[0].isdigit(): | |
f455a934 RA |
469 | headers = {} |
470 | if referer: | |
471 | headers['Referer'] = referer | |
160c2773 RA |
472 | player_page = self._download_webpage( |
473 | 'http://link.brightcove.com/services/player/bcpid' + player_id[0], | |
474 | video_id, headers=headers, fatal=False) | |
475 | if player_page: | |
476 | player_key = self._search_regex( | |
477 | r'<param\s+name="playerKey"\s+value="([\w~,-]+)"', | |
478 | player_page, 'player key', fatal=False) | |
479 | if player_key: | |
480 | enc_pub_id = player_key.split(',')[1].replace('~', '=') | |
481 | publisher_id = struct.unpack('>Q', base64.urlsafe_b64decode(enc_pub_id))[0] | |
f455a934 | 482 | if publisher_id: |
add96eb9 | 483 | brightcove_new_url = f'http://players.brightcove.net/{publisher_id}/default_default/index.html?videoId={video_id}' |
f455a934 RA |
484 | if referer: |
485 | brightcove_new_url = smuggle_url(brightcove_new_url, {'referrer': referer}) | |
486 | return self.url_result(brightcove_new_url, BrightcoveNewIE.ie_key(), video_id) | |
487 | # TODO: figure out if it's possible to extract playlistId from playerKey | |
488 | # elif 'playerKey' in query: | |
489 | # player_key = query['playerKey'] | |
490 | # return self._get_playlist_info(player_key[0]) | |
491 | raise UnsupportedError(url) | |
ed126900 | 492 | |
493 | ||
f5a9e9df | 494 | class BrightcoveNewBaseIE(AdobePassIE): |
495 | def _parse_brightcove_metadata(self, json_data, video_id, headers={}): | |
496 | title = json_data['name'].strip() | |
497 | ||
498 | formats, subtitles = [], {} | |
499 | sources = json_data.get('sources') or [] | |
500 | for source in sources: | |
501 | container = source.get('container') | |
502 | ext = mimetype2ext(source.get('type')) | |
503 | src = source.get('src') | |
504 | if ext == 'm3u8' or container == 'M2TS': | |
505 | if not src: | |
506 | continue | |
507 | fmts, subs = self._extract_m3u8_formats_and_subtitles( | |
508 | src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False) | |
509 | subtitles = self._merge_subtitles(subtitles, subs) | |
510 | elif ext == 'mpd': | |
511 | if not src: | |
512 | continue | |
513 | fmts, subs = self._extract_mpd_formats_and_subtitles(src, video_id, 'dash', fatal=False) | |
514 | subtitles = self._merge_subtitles(subtitles, subs) | |
515 | else: | |
516 | streaming_src = source.get('streaming_src') | |
517 | stream_name, app_name = source.get('stream_name'), source.get('app_name') | |
518 | if not src and not streaming_src and (not stream_name or not app_name): | |
519 | continue | |
520 | tbr = float_or_none(source.get('avg_bitrate'), 1000) | |
521 | height = int_or_none(source.get('height')) | |
522 | width = int_or_none(source.get('width')) | |
523 | f = { | |
524 | 'tbr': tbr, | |
525 | 'filesize': int_or_none(source.get('size')), | |
526 | 'container': container, | |
527 | 'ext': ext or container.lower(), | |
528 | } | |
529 | if width == 0 and height == 0: | |
530 | f.update({ | |
531 | 'vcodec': 'none', | |
532 | }) | |
533 | else: | |
534 | f.update({ | |
535 | 'width': width, | |
536 | 'height': height, | |
537 | 'vcodec': source.get('codec'), | |
538 | }) | |
539 | ||
540 | def build_format_id(kind): | |
541 | format_id = kind | |
542 | if tbr: | |
add96eb9 | 543 | format_id += f'-{int(tbr)}k' |
f5a9e9df | 544 | if height: |
add96eb9 | 545 | format_id += f'-{height}p' |
f5a9e9df | 546 | return format_id |
547 | ||
548 | if src or streaming_src: | |
549 | f.update({ | |
550 | 'url': src or streaming_src, | |
551 | 'format_id': build_format_id('http' if src else 'http-streaming'), | |
552 | 'source_preference': 0 if src else -1, | |
553 | }) | |
554 | else: | |
555 | f.update({ | |
556 | 'url': app_name, | |
557 | 'play_path': stream_name, | |
558 | 'format_id': build_format_id('rtmp'), | |
559 | }) | |
560 | fmts = [f] | |
561 | ||
562 | # https://support.brightcove.com/playback-api-video-fields-reference#key_systems_object | |
563 | if container == 'WVM' or source.get('key_systems') or ext == 'ism': | |
564 | for f in fmts: | |
565 | f['has_drm'] = True | |
566 | formats.extend(fmts) | |
567 | ||
568 | if not formats: | |
569 | errors = json_data.get('errors') | |
570 | if errors: | |
571 | error = errors[0] | |
572 | self.raise_no_formats( | |
573 | error.get('message') or error.get('error_subcode') or error['error_code'], expected=True) | |
574 | ||
44369c9a | 575 | headers.pop('Authorization', None) # or else http formats will give error 400 |
f5a9e9df | 576 | for f in formats: |
577 | f.setdefault('http_headers', {}).update(headers) | |
578 | ||
579 | for text_track in json_data.get('text_tracks', []): | |
580 | if text_track.get('kind') != 'captions': | |
581 | continue | |
582 | text_track_url = url_or_none(text_track.get('src')) | |
583 | if not text_track_url: | |
584 | continue | |
585 | lang = (str_or_none(text_track.get('srclang')) | |
586 | or str_or_none(text_track.get('label')) or 'en').lower() | |
587 | subtitles.setdefault(lang, []).append({ | |
588 | 'url': text_track_url, | |
589 | }) | |
590 | ||
591 | is_live = False | |
592 | duration = float_or_none(json_data.get('duration'), 1000) | |
593 | if duration is not None and duration <= 0: | |
594 | is_live = True | |
595 | ||
596 | common_res = [(160, 90), (320, 180), (480, 720), (640, 360), (768, 432), (1024, 576), (1280, 720), (1366, 768), (1920, 1080)] | |
597 | thumb_base_url = dict_get(json_data, ('poster', 'thumbnail')) | |
598 | thumbnails = [{ | |
599 | 'url': re.sub(r'\d+x\d+', f'{w}x{h}', thumb_base_url), | |
600 | 'width': w, | |
601 | 'height': h, | |
602 | } for w, h in common_res] if thumb_base_url else None | |
603 | ||
604 | return { | |
605 | 'id': video_id, | |
606 | 'title': title, | |
607 | 'description': clean_html(json_data.get('description')), | |
608 | 'thumbnails': thumbnails, | |
609 | 'duration': duration, | |
610 | 'timestamp': parse_iso8601(json_data.get('published_at')), | |
611 | 'uploader_id': json_data.get('account_id'), | |
612 | 'formats': formats, | |
613 | 'subtitles': subtitles, | |
614 | 'tags': json_data.get('tags', []), | |
615 | 'is_live': is_live, | |
616 | } | |
617 | ||
618 | ||
619 | class BrightcoveNewIE(BrightcoveNewBaseIE): | |
e721d857 | 620 | IE_NAME = 'brightcove:new' |
3031b7c4 | 621 | _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*(?P<content_type>video|playlist)Id=(?P<video_id>\d+|ref:[^&]+)' |
cb33d389 | 622 | _TESTS = [{ |
ed126900 | 623 | 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', |
c01e1a96 | 624 | 'md5': 'c8100925723840d4b0d243f7025703be', |
ed126900 | 625 | 'info_dict': { |
626 | 'id': '4463358922001', | |
c01e1a96 | 627 | 'ext': 'mp4', |
ed126900 | 628 | 'title': 'Meet the man behind Popcorn Time', |
c01e1a96 | 629 | 'description': 'md5:eac376a4fe366edc70279bfb681aea16', |
cb33d389 | 630 | 'duration': 165.768, |
c01e1a96 | 631 | 'timestamp': 1441391203, |
632 | 'upload_date': '20150904', | |
c01e1a96 | 633 | 'uploader_id': '929656772001', |
4b7dd170 | 634 | 'formats': 'mincount:20', |
cb33d389 | 635 | }, |
f5a9e9df | 636 | 'skip': '404 Not Found', |
cb33d389 S |
637 | }, { |
638 | # with rtmp streams | |
639 | 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001', | |
640 | 'info_dict': { | |
641 | 'id': '4279049078001', | |
642 | 'ext': 'mp4', | |
643 | 'title': 'Titansgrave: Chapter 0', | |
644 | 'description': 'Titansgrave: Chapter 0', | |
645 | 'duration': 1242.058, | |
646 | 'timestamp': 1433556729, | |
647 | 'upload_date': '20150606', | |
648 | 'uploader_id': '4036320279001', | |
4b7dd170 | 649 | 'formats': 'mincount:39', |
cb33d389 S |
650 | }, |
651 | 'params': { | |
08136dc1 | 652 | # m3u8 download |
cb33d389 | 653 | 'skip_download': True, |
add96eb9 | 654 | }, |
3031b7c4 K |
655 | }, { |
656 | # playlist stream | |
657 | 'url': 'https://players.brightcove.net/1752604059001/S13cJdUBz_default/index.html?playlistId=5718313430001', | |
658 | 'info_dict': { | |
659 | 'id': '5718313430001', | |
660 | 'title': 'No Audio Playlist', | |
661 | }, | |
662 | 'playlist_count': 7, | |
663 | 'params': { | |
664 | # m3u8 download | |
665 | 'skip_download': True, | |
add96eb9 | 666 | }, |
3031b7c4 K |
667 | }, { |
668 | 'url': 'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=5743160747001', | |
669 | 'only_matching': True, | |
4f29fa99 S |
670 | }, { |
671 | # ref: prefixed video id | |
672 | 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442', | |
673 | 'only_matching': True, | |
8ac73bdb | 674 | }, { |
675 | # non numeric ref: prefixed video id | |
676 | 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356', | |
677 | 'only_matching': True, | |
b219f5e5 S |
678 | }, { |
679 | # unavailable video without message but with error_code | |
680 | 'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001', | |
681 | 'only_matching': True, | |
cb33d389 | 682 | }] |
ed126900 | 683 | |
f5a9e9df | 684 | _WEBPAGE_TESTS = [{ |
685 | # brightcove player url embed | |
686 | 'url': 'https://nbc-2.com/weather/forecast/2022/11/16/forecast-warmest-day-of-the-week/', | |
687 | 'md5': '2934d5372b354d27083ccf8575dbfee2', | |
688 | 'info_dict': { | |
689 | 'id': '6315650313112', | |
690 | 'title': 'First Alert Forecast: November 15, 2022', | |
691 | 'ext': 'mp4', | |
692 | 'tags': ['nbc2', 'forecast'], | |
693 | 'uploader_id': '6146886170001', | |
694 | 'thumbnail': r're:^https?://.*\.jpg$', | |
695 | 'timestamp': 1668574571, | |
696 | 'duration': 233.375, | |
697 | 'upload_date': '20221116', | |
698 | }, | |
699 | }, { | |
700 | # embedded with video tag only | |
701 | 'url': 'https://www.gooddishtv.com/tiktok-rapping-chef-mr-pyrex', | |
702 | 'info_dict': { | |
703 | 'id': 'tiktok-rapping-chef-mr-pyrex', | |
704 | 'title': 'TikTok\'s Rapping Chef Makes Jambalaya for the Hosts', | |
705 | 'thumbnail': r're:^https?://.*\.jpg$', | |
706 | 'age_limit': 0, | |
707 | 'description': 'Just in time for Mardi Gras', | |
708 | }, | |
709 | 'playlist': [{ | |
710 | 'info_dict': { | |
711 | 'id': '6299189544001', | |
712 | 'ext': 'mp4', | |
713 | 'title': 'TGD_01-032_5', | |
714 | 'thumbnail': r're:^https?://.*\.jpg$', | |
715 | 'tags': [], | |
716 | 'timestamp': 1646078943, | |
717 | 'uploader_id': '1569565978001', | |
718 | 'upload_date': '20220228', | |
719 | 'duration': 217.195, | |
720 | }, | |
721 | }, { | |
722 | 'info_dict': { | |
723 | 'id': '6305565995112', | |
724 | 'ext': 'mp4', | |
725 | 'title': 'TGD 01-087 (Airs 05.25.22)_Segment 5', | |
726 | 'thumbnail': r're:^https?://.*\.jpg$', | |
727 | 'tags': [], | |
728 | 'timestamp': 1651604591, | |
729 | 'uploader_id': '1569565978001', | |
730 | 'upload_date': '20220503', | |
731 | 'duration': 310.421, | |
732 | }, | |
733 | }], | |
734 | }, { | |
735 | # Brightcove:new type [2]. | |
736 | 'url': 'http://www.delawaresportszone.com/video-st-thomas-more-earns-first-trip-to-basketball-semis', | |
737 | 'md5': '2b35148fcf48da41c9fb4591650784f3', | |
738 | 'info_dict': { | |
739 | 'id': '5348741021001', | |
740 | 'ext': 'mp4', | |
741 | 'upload_date': '20170306', | |
742 | 'uploader_id': '4191638492001', | |
743 | 'timestamp': 1488769918, | |
744 | 'title': 'VIDEO: St. Thomas More earns first trip to basketball semis', | |
745 | }, | |
746 | 'skip': '404 Not Found', | |
747 | }, { | |
748 | # Alternative brightcove <video> attributes | |
749 | 'url': 'http://www.programme-tv.net/videos/extraits/81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche/', | |
750 | 'info_dict': { | |
751 | 'id': '81095-guillaume-canet-evoque-les-rumeurs-d-infidelite-de-marion-cotillard-avec-brad-pitt-dans-vivement-dimanche', | |
752 | 'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche, Extraits : toutes les vidéos avec Télé-Loisirs", | |
753 | }, | |
754 | 'playlist': [{ | |
755 | 'md5': '732d22ba3d33f2f3fc253c39f8f36523', | |
756 | 'info_dict': { | |
757 | 'id': '5311302538001', | |
758 | 'ext': 'mp4', | |
759 | 'title': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche", | |
760 | 'description': "Guillaume Canet évoque les rumeurs d'infidélité de Marion Cotillard avec Brad Pitt dans Vivement Dimanche (France 2, 5 février 2017)", | |
761 | 'timestamp': 1486321708, | |
762 | 'upload_date': '20170205', | |
763 | 'uploader_id': '800000640001', | |
764 | }, | |
765 | 'only_matching': True, | |
766 | }], | |
767 | 'skip': '404 Not Found', | |
768 | }, { | |
769 | # Brightcove URL in single quotes | |
770 | 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/', | |
771 | 'md5': '4ae374f1f8b91c889c4b9203c8c752af', | |
772 | 'info_dict': { | |
773 | 'id': '4255764656001', | |
774 | 'ext': 'mp4', | |
775 | 'title': 'SN Presents: Russell Martin, World Citizen', | |
776 | 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.', | |
777 | 'uploader': 'Rogers Sportsnet', | |
778 | 'uploader_id': '1704050871', | |
779 | 'upload_date': '20150525', | |
780 | 'timestamp': 1432570283, | |
781 | }, | |
782 | 'skip': 'Page no longer has URL, now has javascript', | |
783 | }] | |
784 | ||
7234d1d9 | 785 | @staticmethod |
f631b557 | 786 | def _extract_url(ie, webpage): |
8f97a15d | 787 | urls = BrightcoveNewIE._extract_brightcove_urls(ie, webpage) |
7234d1d9 S |
788 | return urls[0] if urls else None |
789 | ||
24af8529 | 790 | @staticmethod |
8f97a15d | 791 | def _extract_brightcove_urls(ie, webpage): |
536f819e S |
792 | # Reference: |
793 | # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe | |
40158f55 JH |
794 | # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#tag |
795 | # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript | |
796 | # 4. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/in-page-embed-player-implementation.html | |
797 | # 5. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player | |
536f819e S |
798 | |
799 | entries = [] | |
800 | ||
801 | # Look for iframe embeds [1] | |
802 | for _, url in re.findall( | |
354dbbd8 | 803 | r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage): |
804 | entries.append(url if url.startswith('http') else 'http:' + url) | |
d8a1caf0 | 805 | |
40158f55 JH |
806 | # Look for <video> tags [2] and embed_in_page embeds [3] |
807 | # [2] looks like: | |
40158f55 | 808 | for video, script_tag, account_id, player_id, embed in re.findall( |
0254f93b | 809 | r'''(?isx) |
2391941f | 810 | (<video(?:-js)?\s+[^>]*\bdata-video-id\s*=\s*['"]?[^>]+>) |
0254f93b S |
811 | (?:.*? |
812 | (<script[^>]+ | |
813 | src=["\'](?:https?:)?//players\.brightcove\.net/ | |
814 | (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js | |
815 | ) | |
816 | )? | |
817 | ''', webpage): | |
40158f55 JH |
818 | attrs = extract_attributes(video) |
819 | ||
820 | # According to examples from [4] it's unclear whether video id | |
821 | # may be optional and what to do when it is | |
822 | video_id = attrs.get('data-video-id') | |
0254f93b S |
823 | if not video_id: |
824 | continue | |
40158f55 | 825 | |
0254f93b | 826 | account_id = account_id or attrs.get('data-account') |
40158f55 | 827 | if not account_id: |
0254f93b S |
828 | continue |
829 | ||
830 | player_id = player_id or attrs.get('data-player') or 'default' | |
831 | embed = embed or attrs.get('data-embed') or 'default' | |
832 | ||
add96eb9 | 833 | bc_url = f'http://players.brightcove.net/{account_id}/{player_id}_{embed}/index.html?videoId={video_id}' |
0254f93b S |
834 | |
835 | # Some brightcove videos may be embedded with video tag only and | |
836 | # without script tag or any mentioning of brightcove at all. Such | |
837 | # embeds are considered ambiguous since they are matched based only | |
838 | # on data-video-id and data-account attributes and in the wild may | |
839 | # not be brightcove embeds at all. Let's check reconstructed | |
840 | # brightcove URLs in case of such embeds and only process valid | |
841 | # ones. By this we ensure there is indeed a brightcove embed. | |
842 | if not script_tag and not ie._is_valid_url( | |
843 | bc_url, video_id, 'possible brightcove video'): | |
844 | continue | |
845 | ||
846 | entries.append(bc_url) | |
d8a1caf0 | 847 | |
536f819e | 848 | return entries |
ed126900 | 849 | |
f5a9e9df | 850 | def _extract_from_webpage(self, url, webpage): |
851 | bc_urls = self._extract_brightcove_urls(self, webpage) | |
852 | for bc_url in bc_urls: | |
853 | yield self.url_result(smuggle_url(bc_url, {'referrer': url}), BrightcoveNewIE) | |
4b7dd170 RA |
854 | |
855 | def _real_extract(self, url): | |
856 | url, smuggled_data = unsmuggle_url(url, {}) | |
5f95927a S |
857 | self._initialize_geo_bypass({ |
858 | 'countries': smuggled_data.get('geo_countries'), | |
859 | 'ip_blocks': smuggled_data.get('geo_ip_blocks'), | |
860 | }) | |
4b7dd170 | 861 | |
5ad28e7f | 862 | account_id, player_id, embed, content_type, video_id = self._match_valid_url(url).groups() |
4b7dd170 | 863 | |
add96eb9 | 864 | policy_key_id = f'{account_id}_{player_id}' |
9809740b | 865 | policy_key = self.cache.load('brightcove', policy_key_id) |
06068087 | 866 | policy_key_extracted = False |
9809740b | 867 | store_pk = lambda x: self.cache.store('brightcove', policy_key_id, x) |
06068087 RA |
868 | |
869 | def extract_policy_key(): | |
add96eb9 | 870 | base_url = f'http://players.brightcove.net/{account_id}/{player_id}_{embed}/' |
29f7c58a | 871 | config = self._download_json( |
872 | base_url + 'config.json', video_id, fatal=False) or {} | |
873 | policy_key = try_get( | |
874 | config, lambda x: x['video_cloud']['policy_key']) | |
875 | if not policy_key: | |
876 | webpage = self._download_webpage( | |
877 | base_url + 'index.min.js', video_id) | |
06068087 | 878 | |
29f7c58a | 879 | catalog = self._search_regex( |
880 | r'catalog\(({.+?})\);', webpage, 'catalog', default=None) | |
75ef77c1 | 881 | if catalog: |
29f7c58a | 882 | catalog = self._parse_json( |
883 | js_to_json(catalog), video_id, fatal=False) | |
884 | if catalog: | |
885 | policy_key = catalog.get('policyKey') | |
886 | ||
887 | if not policy_key: | |
888 | policy_key = self._search_regex( | |
889 | r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1', | |
890 | webpage, 'policy key', group='pk') | |
06068087 | 891 | |
2d30b92e | 892 | store_pk(policy_key) |
06068087 | 893 | return policy_key |
4b7dd170 | 894 | |
44369c9a | 895 | token = smuggled_data.get('token') |
896 | api_url = f'https://{"edge-auth" if token else "edge"}.api.brightcove.com/playback/v1/accounts/{account_id}/{content_type}s/{video_id}' | |
897 | headers = {'Authorization': f'Bearer {token}'} if token else {} | |
f5a9e9df | 898 | referrer = smuggled_data.get('referrer') # XXX: notice the spelling/case of the key |
5399ab3f S |
899 | if referrer: |
900 | headers.update({ | |
901 | 'Referer': referrer, | |
902 | 'Origin': re.search(r'https?://[^/]+', referrer).group(0), | |
4b7dd170 | 903 | }) |
06068087 RA |
904 | |
905 | for _ in range(2): | |
906 | if not policy_key: | |
907 | policy_key = extract_policy_key() | |
908 | policy_key_extracted = True | |
add96eb9 | 909 | headers['Accept'] = f'application/json;pk={policy_key}' |
06068087 RA |
910 | try: |
911 | json_data = self._download_json(api_url, video_id, headers=headers) | |
912 | break | |
913 | except ExtractorError as e: | |
3d2623a8 | 914 | if isinstance(e.cause, HTTPError) and e.cause.status in (401, 403): |
915 | json_data = self._parse_json(e.cause.response.read().decode(), video_id)[0] | |
06068087 RA |
916 | message = json_data.get('message') or json_data['error_code'] |
917 | if json_data.get('error_subcode') == 'CLIENT_GEO': | |
918 | self.raise_geo_restricted(msg=message) | |
919 | elif json_data.get('error_code') == 'INVALID_POLICY_KEY' and not policy_key_extracted: | |
920 | policy_key = None | |
2d30b92e | 921 | store_pk(None) |
06068087 RA |
922 | continue |
923 | raise ExtractorError(message, expected=True) | |
924 | raise | |
4b7dd170 RA |
925 | |
926 | errors = json_data.get('errors') | |
927 | if errors and errors[0].get('error_subcode') == 'TVE_AUTH': | |
928 | custom_fields = json_data['custom_fields'] | |
929 | tve_token = self._extract_mvpd_auth( | |
930 | smuggled_data['source_url'], video_id, | |
931 | custom_fields['bcadobepassrequestorid'], | |
932 | custom_fields['bcadobepassresourceid']) | |
933 | json_data = self._download_json( | |
934 | api_url, video_id, headers={ | |
add96eb9 | 935 | 'Accept': f'application/json;pk={policy_key}', |
4b7dd170 RA |
936 | }, query={ |
937 | 'tveToken': tve_token, | |
938 | }) | |
939 | ||
3031b7c4 K |
940 | if content_type == 'playlist': |
941 | return self.playlist_result( | |
942 | [self._parse_brightcove_metadata(vid, vid.get('id'), headers) | |
943 | for vid in json_data.get('videos', []) if vid.get('id')], | |
944 | json_data.get('id'), json_data.get('name'), | |
945 | json_data.get('description')) | |
946 | ||
18d7aa6e S |
947 | return self._parse_brightcove_metadata( |
948 | json_data, video_id, headers=headers) |