]>
Commit | Line | Data |
---|---|---|
dcdb292f | 1 | # coding: utf-8 |
400e5810 | 2 | from __future__ import unicode_literals |
592882aa | 3 | |
fbaaad49 JMF |
4 | import re |
5 | import json | |
6 | ||
7 | from .common import InfoExtractor | |
1cc79574 | 8 | from ..compat import ( |
36e6f62c | 9 | compat_etree_fromstring, |
7b0817e8 | 10 | compat_parse_qs, |
1cc79574 | 11 | compat_str, |
37aab278 | 12 | compat_urllib_parse_urlparse, |
1cc79574 | 13 | compat_urlparse, |
e20d0c1e | 14 | compat_xml_parse_error, |
d13bdc38 | 15 | compat_HTTPError, |
1cc79574 PH |
16 | ) |
17 | from ..utils import ( | |
5524b242 | 18 | determine_ext, |
592882aa | 19 | ExtractorError, |
40158f55 | 20 | extract_attributes, |
1cc79574 PH |
21 | find_xpath_attr, |
22 | fix_xml_ampersands, | |
a6624898 | 23 | float_or_none, |
fd91257c | 24 | js_to_json, |
ed126900 | 25 | int_or_none, |
26 | parse_iso8601, | |
fd91257c S |
27 | unescapeHTML, |
28 | unsmuggle_url, | |
48254c3f | 29 | update_url_query, |
f1b4af7d | 30 | clean_html, |
e910fe2f | 31 | mimetype2ext, |
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>.*)' |
48254c3f | 38 | _FEDERATED_URL = 'http://c.brightcove.com/services/viewer/htmlFederated' |
592882aa JMF |
39 | |
40 | _TESTS = [ | |
41 | { | |
4de1994b | 42 | # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ |
400e5810 | 43 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', |
400e5810 PH |
44 | 'md5': '5423e113865d26e40624dce2e4b45d95', |
45 | 'note': 'Test Brightcove downloads and detection in GenericIE', | |
46 | 'info_dict': { | |
96bef88f JMF |
47 | 'id': '2371591881001', |
48 | 'ext': 'mp4', | |
400e5810 PH |
49 | 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', |
50 | 'uploader': '8TV', | |
51 | 'description': 'md5:a950cc4285c43e44d763d036710cd9cd', | |
41f5492f | 52 | 'timestamp': 1368213670, |
53 | 'upload_date': '20130510', | |
4d4cd35f | 54 | 'uploader_id': '1589608506001', |
592882aa JMF |
55 | } |
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 JMF |
69 | }, |
70 | }, | |
fc4a0c2a JMF |
71 | { |
72 | # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/ | |
400e5810 PH |
73 | 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001', |
74 | 'info_dict': { | |
75 | 'id': '2750934548001', | |
76 | 'ext': 'mp4', | |
77 | 'title': 'This Bracelet Acts as a Personal Thermostat', | |
78 | 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0', | |
79 | 'uploader': 'Mashable', | |
41f5492f | 80 | 'timestamp': 1382041798, |
81 | 'upload_date': '20131017', | |
4d4cd35f | 82 | 'uploader_id': '1130468786001', |
fc4a0c2a JMF |
83 | }, |
84 | }, | |
77526143 JMF |
85 | { |
86 | # test that the default referer works | |
87 | # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ | |
400e5810 PH |
88 | 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', |
89 | 'info_dict': { | |
90 | 'id': '2878862109001', | |
91 | 'ext': 'mp4', | |
92 | 'title': 'Lost in Motion II', | |
93 | 'description': 'md5:363109c02998fee92ec02211bd8000df', | |
94 | 'uploader': 'National Ballet of Canada', | |
77526143 | 95 | }, |
c65aa4e9 | 96 | 'skip': 'Video gone', |
5524b242 JMF |
97 | }, |
98 | { | |
99 | # test flv videos served by akamaihd.net | |
100 | # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william | |
41f5492f | 101 | '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 |
102 | # The md5 checksum changes on each download |
103 | 'info_dict': { | |
41f5492f | 104 | 'id': '3750436379001', |
5524b242 JMF |
105 | 'ext': 'flv', |
106 | 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', | |
41f5492f | 107 | 'uploader': 'RBTV Old (do not use)', |
5524b242 | 108 | 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', |
41f5492f | 109 | 'timestamp': 1409122195, |
110 | 'upload_date': '20140827', | |
4d4cd35f | 111 | 'uploader_id': '710858724001', |
5524b242 | 112 | }, |
40158f55 | 113 | 'skip': 'Video gone', |
5524b242 | 114 | }, |
bd4e40df | 115 | { |
c65aa4e9 | 116 | # playlist with 'videoList' |
bd4e40df JMF |
117 | # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players |
118 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL', | |
119 | 'info_dict': { | |
120 | 'title': 'Sealife', | |
d47c26e1 | 121 | 'id': '3550319591001', |
bd4e40df JMF |
122 | }, |
123 | 'playlist_mincount': 7, | |
124 | }, | |
c65aa4e9 YCH |
125 | { |
126 | # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965) | |
127 | 'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg', | |
128 | 'info_dict': { | |
129 | 'id': '1522758701001', | |
130 | 'title': 'Lesson 08', | |
131 | }, | |
132 | 'playlist_mincount': 10, | |
133 | }, | |
592882aa | 134 | ] |
41f5492f | 135 | FLV_VCODECS = { |
136 | 1: 'SORENSON', | |
137 | 2: 'ON2', | |
138 | 3: 'H264', | |
139 | 4: 'VP8', | |
140 | } | |
cfe50f04 JMF |
141 | |
142 | @classmethod | |
143 | def _build_brighcove_url(cls, object_str): | |
144 | """ | |
145 | Build a Brightcove url from a xml string containing | |
146 | <object class="BrightcoveExperience">{params}</object> | |
147 | """ | |
46e28a84 PH |
148 | |
149 | # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553 | |
c1147c05 | 150 | object_str = re.sub(r'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>', |
46e28a84 | 151 | lambda m: m.group(1) + '/>', object_str) |
2d0efe70 | 152 | # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608 |
400e5810 | 153 | object_str = object_str.replace('<--', '<!--') |
d6f714f3 JMF |
154 | # remove namespace to simplify extraction |
155 | object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str) | |
7b0817e8 | 156 | object_str = fix_xml_ampersands(object_str) |
46e28a84 | 157 | |
94c12557 | 158 | try: |
36e6f62c | 159 | object_doc = compat_etree_fromstring(object_str.encode('utf-8')) |
e20d0c1e | 160 | except compat_xml_parse_error: |
94c12557 | 161 | return |
7b0817e8 PH |
162 | |
163 | fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars') | |
47917f24 JMF |
164 | if fv_el is not None: |
165 | flashvars = dict( | |
166 | (k, v[0]) | |
167 | for k, v in compat_parse_qs(fv_el.attrib['value']).items()) | |
168 | else: | |
169 | flashvars = {} | |
7b0817e8 | 170 | |
5964b598 YCH |
171 | data_url = object_doc.attrib.get('data', '') |
172 | data_url_params = compat_parse_qs(compat_urllib_parse_urlparse(data_url).query) | |
173 | ||
36de0a0e | 174 | def find_param(name): |
7b0817e8 PH |
175 | if name in flashvars: |
176 | return flashvars[name] | |
d214fdb8 JMF |
177 | node = find_xpath_attr(object_doc, './param', 'name', name) |
178 | if node is not None: | |
179 | return node.attrib['value'] | |
5964b598 | 180 | return data_url_params.get(name) |
7b0817e8 PH |
181 | |
182 | params = {} | |
183 | ||
16e2c8f7 | 184 | playerID = find_param('playerID') or find_param('playerId') |
7b0817e8 PH |
185 | if playerID is None: |
186 | raise ExtractorError('Cannot find player ID') | |
187 | params['playerID'] = playerID | |
188 | ||
36de0a0e | 189 | playerKey = find_param('playerKey') |
cfe50f04 JMF |
190 | # Not all pages define this value |
191 | if playerKey is not None: | |
d214fdb8 | 192 | params['playerKey'] = playerKey |
48254c3f | 193 | # These fields hold the id of the video |
2cab4870 | 194 | videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList') |
abb285fb | 195 | if videoPlayer is not None: |
049a0f4d RA |
196 | if isinstance(videoPlayer, list): |
197 | videoPlayer = videoPlayer[0] | |
2e76bdc8 S |
198 | videoPlayer = videoPlayer.strip() |
199 | # UUID is also possible for videoPlayer (e.g. | |
200 | # http://www.popcornflix.com/hoodies-vs-hooligans/7f2d2b87-bbf2-4623-acfb-ea942b4f01dd | |
201 | # or http://www8.hp.com/cn/zh/home.html) | |
202 | if not (re.match( | |
203 | r'^(?:\d+|[\da-fA-F]{8}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{4}-?[\da-fA-F]{12})$', | |
204 | videoPlayer) or videoPlayer.startswith('ref:')): | |
049a0f4d | 205 | return None |
d214fdb8 | 206 | params['@videoPlayer'] = videoPlayer |
36de0a0e | 207 | linkBase = find_param('linkBaseURL') |
dd5bcdc4 | 208 | if linkBase is not None: |
d214fdb8 | 209 | params['linkBaseURL'] = linkBase |
af9cdee9 S |
210 | return cls._make_brightcove_url(params) |
211 | ||
212 | @classmethod | |
213 | def _build_brighcove_url_from_js(cls, object_js): | |
214 | # The layout of JS is as follows: | |
215 | # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) { | |
216 | # // build Brightcove <object /> XML | |
217 | # } | |
218 | m = re.search( | |
4f66c16f | 219 | r'''(?x)customBC\.createVideo\( |
af9cdee9 S |
220 | .*? # skipping width and height |
221 | ["\'](?P<playerID>\d+)["\']\s*,\s* # playerID | |
222 | ["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters | |
223 | # in length, however it's appended to itself | |
224 | # in places, so truncate | |
225 | ["\'](?P<videoID>\d+)["\'] # @videoPlayer | |
226 | ''', object_js) | |
227 | if m: | |
228 | return cls._make_brightcove_url(m.groupdict()) | |
229 | ||
230 | @classmethod | |
231 | def _make_brightcove_url(cls, params): | |
48254c3f | 232 | return update_url_query(cls._FEDERATED_URL, params) |
fbaaad49 | 233 | |
eeb165e6 JMF |
234 | @classmethod |
235 | def _extract_brightcove_url(cls, webpage): | |
99877772 | 236 | """Try to extract the brightcove url from the webpage, returns None |
eeb165e6 JMF |
237 | if it can't be found |
238 | """ | |
99877772 PH |
239 | urls = cls._extract_brightcove_urls(webpage) |
240 | return urls[0] if urls else None | |
241 | ||
242 | @classmethod | |
243 | def _extract_brightcove_urls(cls, webpage): | |
244 | """Return a list of all Brightcove URLs from the webpage """ | |
117bec93 | 245 | |
e1ab5000 | 246 | url_m = re.search( |
53a664ed S |
247 | r'''(?x) |
248 | <meta\s+ | |
249 | (?:property|itemprop)=([\'"])(?:og:video|embedURL)\1[^>]+ | |
250 | content=([\'"])(?P<url>https?://(?:secure|c)\.brightcove.com/(?:(?!\2).)+)\2 | |
251 | ''', webpage) | |
117bec93 | 252 | if url_m: |
53a664ed | 253 | url = unescapeHTML(url_m.group('url')) |
381640e3 JMF |
254 | # Some sites don't add it, we can't download with this url, for example: |
255 | # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/ | |
53a664ed | 256 | if 'playerKey' in url or 'videoId' in url or 'idVideo' in url: |
381640e3 | 257 | return [url] |
117bec93 | 258 | |
99877772 | 259 | matches = re.findall( |
7b0817e8 PH |
260 | r'''(?sx)<object |
261 | (?: | |
99877772 | 262 | [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] | |
7b0817e8 | 263 | [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/ |
476e1095 | 264 | ).+?>\s*</object>''', |
7b0817e8 | 265 | webpage) |
b4e1576a | 266 | if matches: |
267 | return list(filter(None, [cls._build_brighcove_url(m) for m in matches])) | |
268 | ||
af9cdee9 S |
269 | return list(filter(None, [ |
270 | cls._build_brighcove_url_from_js(custom_bc) | |
271 | for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)])) | |
eeb165e6 | 272 | |
fbaaad49 | 273 | def _real_extract(self, url): |
79f82953 PH |
274 | url, smuggled_data = unsmuggle_url(url, {}) |
275 | ||
51040b72 | 276 | # Change the 'videoId' and others field to '@videoPlayer' |
53a664ed | 277 | url = re.sub(r'(?<=[?&])(videoI(d|D)|idVideo|bctid)', '%40videoPlayer', url) |
51040b72 JMF |
278 | # Change bckey (used by bcove.me urls) to playerKey |
279 | url = re.sub(r'(?<=[?&])bckey', 'playerKey', url) | |
fbaaad49 | 280 | mobj = re.match(self._VALID_URL, url) |
6543f0dc JMF |
281 | query_str = mobj.group('query') |
282 | query = compat_urlparse.parse_qs(query_str) | |
fbaaad49 | 283 | |
6543f0dc JMF |
284 | videoPlayer = query.get('@videoPlayer') |
285 | if videoPlayer: | |
79f82953 PH |
286 | # We set the original url as the default 'Referer' header |
287 | referer = smuggled_data.get('Referer', url) | |
288 | return self._get_video_info( | |
48254c3f | 289 | videoPlayer[0], query, referer=referer) |
68575900 | 290 | elif 'playerKey' in query: |
6543f0dc JMF |
291 | player_key = query['playerKey'] |
292 | return self._get_playlist_info(player_key[0]) | |
68575900 PH |
293 | else: |
294 | raise ExtractorError( | |
295 | 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?', | |
296 | expected=True) | |
abb285fb | 297 | |
48254c3f | 298 | def _get_video_info(self, video_id, query, referer=None): |
299 | headers = {} | |
dd5bcdc4 JMF |
300 | linkBase = query.get('linkBaseURL') |
301 | if linkBase is not None: | |
77526143 JMF |
302 | referer = linkBase[0] |
303 | if referer is not None: | |
48254c3f | 304 | headers['Referer'] = referer |
305 | webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query) | |
fbaaad49 | 306 | |
a5ed3e57 | 307 | error_msg = self._html_search_regex( |
4d46c1c6 | 308 | r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage, |
a5ed3e57 PH |
309 | 'error message', default=None) |
310 | if error_msg is not None: | |
311 | raise ExtractorError( | |
312 | 'brightcove said: %s' % error_msg, expected=True) | |
313 | ||
fbaaad49 | 314 | self.report_extraction(video_id) |
a013eba6 | 315 | info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json') |
fbaaad49 JMF |
316 | info = json.loads(info)['data'] |
317 | video_info = info['programmedContent']['videoPlayer']['mediaDTO'] | |
7b0817e8 | 318 | video_info['_youtubedl_adServerURL'] = info.get('adServerURL') |
abb285fb JMF |
319 | |
320 | return self._extract_video_info(video_info) | |
321 | ||
322 | def _get_playlist_info(self, player_key): | |
117bec93 PH |
323 | info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key |
324 | playlist_info = self._download_webpage( | |
325 | info_url, player_key, 'Downloading playlist information') | |
abb285fb | 326 | |
59145479 | 327 | json_data = json.loads(playlist_info) |
c65aa4e9 YCH |
328 | if 'videoList' in json_data: |
329 | playlist_info = json_data['videoList'] | |
330 | playlist_dto = playlist_info['mediaCollectionDTO'] | |
331 | elif 'playlistTabs' in json_data: | |
332 | playlist_info = json_data['playlistTabs'] | |
333 | playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0] | |
334 | else: | |
400e5810 | 335 | raise ExtractorError('Empty playlist') |
c65aa4e9 YCH |
336 | |
337 | videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']] | |
abb285fb | 338 | |
d47c26e1 | 339 | return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'], |
c65aa4e9 | 340 | playlist_title=playlist_dto['displayName']) |
abb285fb JMF |
341 | |
342 | def _extract_video_info(self, video_info): | |
ad55e101 | 343 | video_id = compat_str(video_info['id']) |
23576edb | 344 | publisher_id = video_info.get('publisherId') |
592882aa | 345 | info = { |
ad55e101 | 346 | 'id': video_id, |
066f6a06 | 347 | 'title': video_info['displayName'].strip(), |
592882aa JMF |
348 | 'description': video_info.get('shortDescription'), |
349 | 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'), | |
350 | 'uploader': video_info.get('publisherName'), | |
23576edb | 351 | 'uploader_id': compat_str(publisher_id) if publisher_id else None, |
41f5492f | 352 | 'duration': float_or_none(video_info.get('length'), 1000), |
353 | 'timestamp': int_or_none(video_info.get('creationDate'), 1000), | |
592882aa | 354 | } |
abb285fb | 355 | |
d6c340ca | 356 | renditions = video_info.get('renditions', []) + video_info.get('IOSRenditions', []) |
592882aa | 357 | if renditions: |
5524b242 JMF |
358 | formats = [] |
359 | for rend in renditions: | |
360 | url = rend['defaultURL'] | |
b5af6fcd NJ |
361 | if not url: |
362 | continue | |
233d37fb | 363 | ext = None |
5524b242 | 364 | if rend['remote']: |
37aab278 NJ |
365 | url_comp = compat_urllib_parse_urlparse(url) |
366 | if url_comp.path.endswith('.m3u8'): | |
367 | formats.extend( | |
ad55e101 | 368 | self._extract_m3u8_formats( |
369 | url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) | |
37aab278 NJ |
370 | continue |
371 | elif 'akamaihd.net' in url_comp.netloc: | |
372 | # This type of renditions are served through | |
373 | # akamaihd.net, but they don't use f4m manifests | |
374 | url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB' | |
375 | ext = 'flv' | |
233d37fb | 376 | if ext is None: |
5524b242 | 377 | ext = determine_ext(url) |
8334637f | 378 | tbr = int_or_none(rend.get('encodingRate'), 1000) |
d6c340ca | 379 | a_format = { |
41f5492f | 380 | 'format_id': 'http%s' % ('-%s' % tbr if tbr else ''), |
5524b242 JMF |
381 | 'url': url, |
382 | 'ext': ext, | |
41f5492f | 383 | 'filesize': int_or_none(rend.get('size')) or None, |
384 | 'tbr': tbr, | |
d6c340ca | 385 | } |
41f5492f | 386 | if rend.get('audioOnly'): |
387 | a_format.update({ | |
388 | 'vcodec': 'none', | |
389 | }) | |
390 | else: | |
391 | a_format.update({ | |
392 | 'height': int_or_none(rend.get('frameHeight')), | |
393 | 'width': int_or_none(rend.get('frameWidth')), | |
394 | 'vcodec': rend.get('videoCodec'), | |
395 | }) | |
d6c340ca YCH |
396 | |
397 | # m3u8 manifests with remote == false are media playlists | |
398 | # Not calling _extract_m3u8_formats here to save network traffic | |
399 | if ext == 'm3u8': | |
400 | a_format.update({ | |
41f5492f | 401 | 'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''), |
d6c340ca | 402 | 'ext': 'mp4', |
ad55e101 | 403 | 'protocol': 'm3u8_native', |
d6c340ca YCH |
404 | }) |
405 | ||
406 | formats.append(a_format) | |
35eacd0d | 407 | self._sort_formats(formats) |
5524b242 | 408 | info['formats'] = formats |
592882aa JMF |
409 | elif video_info.get('FLVFullLengthURL') is not None: |
410 | info.update({ | |
411 | 'url': video_info['FLVFullLengthURL'], | |
41f5492f | 412 | 'vcodec': self.FLV_VCODECS.get(video_info.get('FLVFullCodec')), |
413 | 'filesize': int_or_none(video_info.get('FLVFullSize')), | |
592882aa | 414 | }) |
7b0817e8 PH |
415 | |
416 | if self._downloader.params.get('include_ads', False): | |
417 | adServerURL = video_info.get('_youtubedl_adServerURL') | |
418 | if adServerURL: | |
419 | ad_info = { | |
420 | '_type': 'url', | |
421 | 'url': adServerURL, | |
422 | } | |
423 | if 'url' in info: | |
424 | return { | |
425 | '_type': 'playlist', | |
426 | 'title': info['title'], | |
427 | 'entries': [ad_info, info], | |
428 | } | |
429 | else: | |
430 | return ad_info | |
431 | ||
d614aa40 | 432 | if 'url' not in info and not info.get('formats'): |
ad55e101 | 433 | raise ExtractorError('Unable to extract video url for %s' % video_id) |
592882aa | 434 | return info |
ed126900 | 435 | |
436 | ||
5c17f0a6 | 437 | class BrightcoveNewIE(InfoExtractor): |
e721d857 | 438 | IE_NAME = 'brightcove:new' |
8ac73bdb | 439 | _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)' |
cb33d389 | 440 | _TESTS = [{ |
ed126900 | 441 | 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', |
c01e1a96 | 442 | 'md5': 'c8100925723840d4b0d243f7025703be', |
ed126900 | 443 | 'info_dict': { |
444 | 'id': '4463358922001', | |
c01e1a96 | 445 | 'ext': 'mp4', |
ed126900 | 446 | 'title': 'Meet the man behind Popcorn Time', |
c01e1a96 | 447 | 'description': 'md5:eac376a4fe366edc70279bfb681aea16', |
cb33d389 | 448 | 'duration': 165.768, |
c01e1a96 | 449 | 'timestamp': 1441391203, |
450 | 'upload_date': '20150904', | |
c01e1a96 | 451 | 'uploader_id': '929656772001', |
cb33d389 S |
452 | 'formats': 'mincount:22', |
453 | }, | |
454 | }, { | |
455 | # with rtmp streams | |
456 | 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001', | |
457 | 'info_dict': { | |
458 | 'id': '4279049078001', | |
459 | 'ext': 'mp4', | |
460 | 'title': 'Titansgrave: Chapter 0', | |
461 | 'description': 'Titansgrave: Chapter 0', | |
462 | 'duration': 1242.058, | |
463 | 'timestamp': 1433556729, | |
464 | 'upload_date': '20150606', | |
465 | 'uploader_id': '4036320279001', | |
466 | 'formats': 'mincount:41', | |
467 | }, | |
468 | 'params': { | |
08136dc1 | 469 | # m3u8 download |
cb33d389 | 470 | 'skip_download': True, |
ed126900 | 471 | } |
4f29fa99 S |
472 | }, { |
473 | # ref: prefixed video id | |
474 | 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442', | |
475 | 'only_matching': True, | |
8ac73bdb | 476 | }, { |
477 | # non numeric ref: prefixed video id | |
478 | 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356', | |
479 | 'only_matching': True, | |
b219f5e5 S |
480 | }, { |
481 | # unavailable video without message but with error_code | |
482 | 'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001', | |
483 | 'only_matching': True, | |
cb33d389 | 484 | }] |
ed126900 | 485 | |
7234d1d9 S |
486 | @staticmethod |
487 | def _extract_url(webpage): | |
488 | urls = BrightcoveNewIE._extract_urls(webpage) | |
489 | return urls[0] if urls else None | |
490 | ||
24af8529 | 491 | @staticmethod |
0254f93b | 492 | def _extract_urls(ie, webpage): |
536f819e S |
493 | # Reference: |
494 | # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe | |
40158f55 JH |
495 | # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#tag |
496 | # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript | |
497 | # 4. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/in-page-embed-player-implementation.html | |
498 | # 5. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player | |
536f819e S |
499 | |
500 | entries = [] | |
501 | ||
502 | # Look for iframe embeds [1] | |
503 | for _, url in re.findall( | |
354dbbd8 | 504 | r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage): |
505 | entries.append(url if url.startswith('http') else 'http:' + url) | |
d8a1caf0 | 506 | |
40158f55 JH |
507 | # Look for <video> tags [2] and embed_in_page embeds [3] |
508 | # [2] looks like: | |
40158f55 | 509 | for video, script_tag, account_id, player_id, embed in re.findall( |
0254f93b S |
510 | r'''(?isx) |
511 | (<video\s+[^>]+>) | |
512 | (?:.*? | |
513 | (<script[^>]+ | |
514 | src=["\'](?:https?:)?//players\.brightcove\.net/ | |
515 | (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js | |
516 | ) | |
517 | )? | |
518 | ''', webpage): | |
40158f55 JH |
519 | attrs = extract_attributes(video) |
520 | ||
521 | # According to examples from [4] it's unclear whether video id | |
522 | # may be optional and what to do when it is | |
523 | video_id = attrs.get('data-video-id') | |
0254f93b S |
524 | if not video_id: |
525 | continue | |
40158f55 | 526 | |
0254f93b | 527 | account_id = account_id or attrs.get('data-account') |
40158f55 | 528 | if not account_id: |
0254f93b S |
529 | continue |
530 | ||
531 | player_id = player_id or attrs.get('data-player') or 'default' | |
532 | embed = embed or attrs.get('data-embed') or 'default' | |
533 | ||
534 | bc_url = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' % ( | |
535 | account_id, player_id, embed, video_id) | |
536 | ||
537 | # Some brightcove videos may be embedded with video tag only and | |
538 | # without script tag or any mentioning of brightcove at all. Such | |
539 | # embeds are considered ambiguous since they are matched based only | |
540 | # on data-video-id and data-account attributes and in the wild may | |
541 | # not be brightcove embeds at all. Let's check reconstructed | |
542 | # brightcove URLs in case of such embeds and only process valid | |
543 | # ones. By this we ensure there is indeed a brightcove embed. | |
544 | if not script_tag and not ie._is_valid_url( | |
545 | bc_url, video_id, 'possible brightcove video'): | |
546 | continue | |
547 | ||
548 | entries.append(bc_url) | |
d8a1caf0 | 549 | |
536f819e | 550 | return entries |
ed126900 | 551 | |
552 | def _real_extract(self, url): | |
d16f27ca S |
553 | url, smuggled_data = unsmuggle_url(url, {}) |
554 | self._initialize_geo_bypass(smuggled_data.get('geo_countries')) | |
555 | ||
9550ca50 | 556 | account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups() |
ed126900 | 557 | |
a6624898 S |
558 | webpage = self._download_webpage( |
559 | 'http://players.brightcove.net/%s/%s_%s/index.min.js' | |
560 | % (account_id, player_id, embed), video_id) | |
ed126900 | 561 | |
a6624898 S |
562 | policy_key = None |
563 | ||
564 | catalog = self._search_regex( | |
565 | r'catalog\(({.+?})\);', webpage, 'catalog', default=None) | |
566 | if catalog: | |
567 | catalog = self._parse_json( | |
568 | js_to_json(catalog), video_id, fatal=False) | |
569 | if catalog: | |
570 | policy_key = catalog.get('policyKey') | |
571 | ||
572 | if not policy_key: | |
573 | policy_key = self._search_regex( | |
574 | r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1', | |
575 | webpage, 'policy key', group='pk') | |
ed126900 | 576 | |
48254c3f | 577 | api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id) |
d13bdc38 | 578 | try: |
48254c3f | 579 | json_data = self._download_json(api_url, video_id, headers={ |
580 | 'Accept': 'application/json;pk=%s' % policy_key | |
581 | }) | |
d13bdc38 | 582 | except ExtractorError as e: |
583 | if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: | |
b219f5e5 | 584 | json_data = self._parse_json(e.cause.read().decode(), video_id)[0] |
d02d4fa0 S |
585 | message = json_data.get('message') or json_data['error_code'] |
586 | if json_data.get('error_subcode') == 'CLIENT_GEO': | |
587 | self.raise_geo_restricted(msg=message) | |
588 | raise ExtractorError(message, expected=True) | |
d13bdc38 | 589 | raise |
ed126900 | 590 | |
e47d19e9 | 591 | title = json_data['name'].strip() |
ed126900 | 592 | |
593 | formats = [] | |
a6624898 | 594 | for source in json_data.get('sources', []): |
975b9c9a | 595 | container = source.get('container') |
e910fe2f | 596 | ext = mimetype2ext(source.get('type')) |
a6624898 | 597 | src = source.get('src') |
954529c1 | 598 | if ext == 'ism' or container == 'WVM': |
e910fe2f RA |
599 | continue |
600 | elif ext == 'm3u8' or container == 'M2TS': | |
a6624898 S |
601 | if not src: |
602 | continue | |
7e5edcfd | 603 | formats.extend(self._extract_m3u8_formats( |
ad55e101 | 604 | src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) |
e910fe2f | 605 | elif ext == 'mpd': |
89807d6a | 606 | if not src: |
607 | continue | |
608 | formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False)) | |
ed126900 | 609 | else: |
a6624898 S |
610 | streaming_src = source.get('streaming_src') |
611 | stream_name, app_name = source.get('stream_name'), source.get('app_name') | |
612 | if not src and not streaming_src and (not stream_name or not app_name): | |
613 | continue | |
614 | tbr = float_or_none(source.get('avg_bitrate'), 1000) | |
615 | height = int_or_none(source.get('height')) | |
89807d6a | 616 | width = int_or_none(source.get('width')) |
a6624898 S |
617 | f = { |
618 | 'tbr': tbr, | |
a6624898 | 619 | 'filesize': int_or_none(source.get('size')), |
975b9c9a | 620 | 'container': container, |
e910fe2f | 621 | 'ext': ext or container.lower(), |
a6624898 | 622 | } |
89807d6a | 623 | if width == 0 and height == 0: |
624 | f.update({ | |
625 | 'vcodec': 'none', | |
626 | }) | |
627 | else: | |
628 | f.update({ | |
629 | 'width': width, | |
630 | 'height': height, | |
631 | 'vcodec': source.get('codec'), | |
632 | }) | |
ed126900 | 633 | |
a6624898 S |
634 | def build_format_id(kind): |
635 | format_id = kind | |
636 | if tbr: | |
637 | format_id += '-%dk' % int(tbr) | |
638 | if height: | |
639 | format_id += '-%dp' % height | |
640 | return format_id | |
641 | ||
642 | if src or streaming_src: | |
643 | f.update({ | |
644 | 'url': src or streaming_src, | |
645 | 'format_id': build_format_id('http' if src else 'http-streaming'), | |
08136dc1 | 646 | 'source_preference': 0 if src else -1, |
a6624898 S |
647 | }) |
648 | else: | |
649 | f.update({ | |
650 | 'url': app_name, | |
651 | 'play_path': stream_name, | |
652 | 'format_id': build_format_id('rtmp'), | |
653 | }) | |
654 | formats.append(f) | |
dd801bbe RA |
655 | |
656 | errors = json_data.get('errors') | |
657 | if not formats and errors: | |
658 | error = errors[0] | |
659 | raise ExtractorError( | |
660 | error.get('message') or error.get('error_subcode') or error['error_code'], expected=True) | |
661 | ||
ed126900 | 662 | self._sort_formats(formats) |
663 | ||
e47d19e9 | 664 | subtitles = {} |
665 | for text_track in json_data.get('text_tracks', []): | |
666 | if text_track.get('src'): | |
667 | subtitles.setdefault(text_track.get('srclang'), []).append({ | |
668 | 'url': text_track['src'], | |
669 | }) | |
a6624898 | 670 | |
27e99078 RA |
671 | is_live = False |
672 | duration = float_or_none(json_data.get('duration'), 1000) | |
673 | if duration and duration < 0: | |
674 | is_live = True | |
675 | ||
ed126900 | 676 | return { |
677 | 'id': video_id, | |
27e99078 | 678 | 'title': self._live_title(title) if is_live else title, |
f1b4af7d | 679 | 'description': clean_html(json_data.get('description')), |
e47d19e9 | 680 | 'thumbnail': json_data.get('thumbnail') or json_data.get('poster'), |
27e99078 | 681 | 'duration': duration, |
e47d19e9 | 682 | 'timestamp': parse_iso8601(json_data.get('published_at')), |
c01e1a96 | 683 | 'uploader_id': account_id, |
a6624898 | 684 | 'formats': formats, |
e47d19e9 | 685 | 'subtitles': subtitles, |
686 | 'tags': json_data.get('tags', []), | |
27e99078 | 687 | 'is_live': is_live, |
ed126900 | 688 | } |