2 from __future__ import unicode_literals
7 from .common import InfoExtractor
9 compat_etree_fromstring,
12 compat_urllib_parse_urlparse,
14 compat_xml_parse_error,
33 class BrightcoveLegacyIE(InfoExtractor):
34 IE_NAME = 'brightcove:legacy'
35 _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
36 _FEDERATED_URL = 'http://c.brightcove.com/services/viewer/htmlFederated'
40 # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
41 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
42 'md5': '5423e113865d26e40624dce2e4b45d95',
43 'note': 'Test Brightcove downloads and detection in GenericIE',
45 'id': '2371591881001',
47 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
49 'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
50 'timestamp': 1368213670,
51 'upload_date': '20130510',
52 'uploader_id': '1589608506001',
56 # From http://medianetwork.oracle.com/video/player/1785452137001
57 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
59 'id': '1785452137001',
61 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
62 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
64 'timestamp': 1344975024,
65 'upload_date': '20120814',
66 'uploader_id': '1460825906',
70 # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
71 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
73 'id': '2750934548001',
75 'title': 'This Bracelet Acts as a Personal Thermostat',
76 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
77 'uploader': 'Mashable',
78 'timestamp': 1382041798,
79 'upload_date': '20131017',
80 'uploader_id': '1130468786001',
84 # test that the default referer works
85 # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
86 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
88 'id': '2878862109001',
90 'title': 'Lost in Motion II',
91 'description': 'md5:363109c02998fee92ec02211bd8000df',
92 'uploader': 'National Ballet of Canada',
97 # test flv videos served by akamaihd.net
98 # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william
99 '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',
100 # The md5 checksum changes on each download
102 'id': '3750436379001',
104 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
105 'uploader': 'RBTV Old (do not use)',
106 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
107 'timestamp': 1409122195,
108 'upload_date': '20140827',
109 'uploader_id': '710858724001',
113 # playlist with 'videoList'
114 # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
115 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
118 'id': '3550319591001',
120 'playlist_mincount': 7,
123 # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965)
124 'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg',
126 'id': '1522758701001',
127 'title': 'Lesson 08',
129 'playlist_mincount': 10,
140 def _build_brighcove_url(cls, object_str):
142 Build a Brightcove url from a xml string containing
143 <object class="BrightcoveExperience">{params}</object>
146 # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
147 object_str = re.sub(r'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>',
148 lambda m: m.group(1) + '/>', object_str)
149 # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
150 object_str = object_str.replace('<--', '<!--')
151 # remove namespace to simplify extraction
152 object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
153 object_str = fix_xml_ampersands(object_str)
156 object_doc = compat_etree_fromstring(object_str.encode('utf-8'))
157 except compat_xml_parse_error:
160 fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
161 if fv_el is not None:
164 for k, v in compat_parse_qs(fv_el.attrib['value']).items())
168 data_url = object_doc.attrib.get('data', '')
169 data_url_params = compat_parse_qs(compat_urllib_parse_urlparse(data_url).query)
171 def find_param(name):
172 if name in flashvars:
173 return flashvars[name]
174 node = find_xpath_attr(object_doc, './param', 'name', name)
176 return node.attrib['value']
177 return data_url_params.get(name)
181 playerID = find_param('playerID')
183 raise ExtractorError('Cannot find player ID')
184 params['playerID'] = playerID
186 playerKey = find_param('playerKey')
187 # Not all pages define this value
188 if playerKey is not None:
189 params['playerKey'] = playerKey
190 # These fields hold the id of the video
191 videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
192 if videoPlayer is not None:
193 params['@videoPlayer'] = videoPlayer
194 linkBase = find_param('linkBaseURL')
195 if linkBase is not None:
196 params['linkBaseURL'] = linkBase
197 return cls._make_brightcove_url(params)
200 def _build_brighcove_url_from_js(cls, object_js):
201 # The layout of JS is as follows:
202 # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) {
203 # // build Brightcove <object /> XML
206 r'''(?x)customBC.\createVideo\(
207 .*? # skipping width and height
208 ["\'](?P<playerID>\d+)["\']\s*,\s* # playerID
209 ["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters
210 # in length, however it's appended to itself
211 # in places, so truncate
212 ["\'](?P<videoID>\d+)["\'] # @videoPlayer
215 return cls._make_brightcove_url(m.groupdict())
218 def _make_brightcove_url(cls, params):
219 return update_url_query(cls._FEDERATED_URL, params)
222 def _extract_brightcove_url(cls, webpage):
223 """Try to extract the brightcove url from the webpage, returns None
226 urls = cls._extract_brightcove_urls(webpage)
227 return urls[0] if urls else None
230 def _extract_brightcove_urls(cls, webpage):
231 """Return a list of all Brightcove URLs from the webpage """
234 r'<meta\s+property=[\'"]og:video[\'"]\s+content=[\'"](https?://(?:secure|c)\.brightcove.com/[^\'"]+)[\'"]',
237 url = unescapeHTML(url_m.group(1))
238 # Some sites don't add it, we can't download with this url, for example:
239 # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
240 if 'playerKey' in url or 'videoId' in url:
243 matches = re.findall(
246 [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
247 [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
248 ).+?>\s*</object>''',
251 return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
253 return list(filter(None, [
254 cls._build_brighcove_url_from_js(custom_bc)
255 for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)]))
257 def _real_extract(self, url):
258 url, smuggled_data = unsmuggle_url(url, {})
260 # Change the 'videoId' and others field to '@videoPlayer'
261 url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
262 # Change bckey (used by bcove.me urls) to playerKey
263 url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)
264 mobj = re.match(self._VALID_URL, url)
265 query_str = mobj.group('query')
266 query = compat_urlparse.parse_qs(query_str)
268 videoPlayer = query.get('@videoPlayer')
270 # We set the original url as the default 'Referer' header
271 referer = smuggled_data.get('Referer', url)
272 return self._get_video_info(
273 videoPlayer[0], query, referer=referer)
274 elif 'playerKey' in query:
275 player_key = query['playerKey']
276 return self._get_playlist_info(player_key[0])
278 raise ExtractorError(
279 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
282 def _get_video_info(self, video_id, query, referer=None):
284 linkBase = query.get('linkBaseURL')
285 if linkBase is not None:
286 referer = linkBase[0]
287 if referer is not None:
288 headers['Referer'] = referer
289 webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
291 error_msg = self._html_search_regex(
292 r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
293 'error message', default=None)
294 if error_msg is not None:
295 raise ExtractorError(
296 'brightcove said: %s' % error_msg, expected=True)
298 self.report_extraction(video_id)
299 info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
300 info = json.loads(info)['data']
301 video_info = info['programmedContent']['videoPlayer']['mediaDTO']
302 video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
304 return self._extract_video_info(video_info)
306 def _get_playlist_info(self, player_key):
307 info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
308 playlist_info = self._download_webpage(
309 info_url, player_key, 'Downloading playlist information')
311 json_data = json.loads(playlist_info)
312 if 'videoList' in json_data:
313 playlist_info = json_data['videoList']
314 playlist_dto = playlist_info['mediaCollectionDTO']
315 elif 'playlistTabs' in json_data:
316 playlist_info = json_data['playlistTabs']
317 playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0]
319 raise ExtractorError('Empty playlist')
321 videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']]
323 return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
324 playlist_title=playlist_dto['displayName'])
326 def _extract_video_info(self, video_info):
327 video_id = compat_str(video_info['id'])
328 publisher_id = video_info.get('publisherId')
331 'title': video_info['displayName'].strip(),
332 'description': video_info.get('shortDescription'),
333 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
334 'uploader': video_info.get('publisherName'),
335 'uploader_id': compat_str(publisher_id) if publisher_id else None,
336 'duration': float_or_none(video_info.get('length'), 1000),
337 'timestamp': int_or_none(video_info.get('creationDate'), 1000),
340 renditions = video_info.get('renditions', []) + video_info.get('IOSRenditions', [])
343 for rend in renditions:
344 url = rend['defaultURL']
349 url_comp = compat_urllib_parse_urlparse(url)
350 if url_comp.path.endswith('.m3u8'):
352 self._extract_m3u8_formats(
353 url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
355 elif 'akamaihd.net' in url_comp.netloc:
356 # This type of renditions are served through
357 # akamaihd.net, but they don't use f4m manifests
358 url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
361 ext = determine_ext(url)
362 tbr = int_or_none(rend.get('encodingRate'), 1000)
364 'format_id': 'http%s' % ('-%s' % tbr if tbr else ''),
367 'filesize': int_or_none(rend.get('size')) or None,
370 if rend.get('audioOnly'):
376 'height': int_or_none(rend.get('frameHeight')),
377 'width': int_or_none(rend.get('frameWidth')),
378 'vcodec': rend.get('videoCodec'),
381 # m3u8 manifests with remote == false are media playlists
382 # Not calling _extract_m3u8_formats here to save network traffic
385 'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''),
387 'protocol': 'm3u8_native',
390 formats.append(a_format)
391 self._sort_formats(formats)
392 info['formats'] = formats
393 elif video_info.get('FLVFullLengthURL') is not None:
395 'url': video_info['FLVFullLengthURL'],
396 'vcodec': self.FLV_VCODECS.get(video_info.get('FLVFullCodec')),
397 'filesize': int_or_none(video_info.get('FLVFullSize')),
400 if self._downloader.params.get('include_ads', False):
401 adServerURL = video_info.get('_youtubedl_adServerURL')
410 'title': info['title'],
411 'entries': [ad_info, info],
416 if 'url' not in info and not info.get('formats'):
417 raise ExtractorError('Unable to extract video url for %s' % video_id)
421 class BrightcoveNewIE(InfoExtractor):
422 IE_NAME = 'brightcove:new'
423 _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)'
425 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
426 'md5': 'c8100925723840d4b0d243f7025703be',
428 'id': '4463358922001',
430 'title': 'Meet the man behind Popcorn Time',
431 'description': 'md5:eac376a4fe366edc70279bfb681aea16',
433 'timestamp': 1441391203,
434 'upload_date': '20150904',
435 'uploader_id': '929656772001',
436 'formats': 'mincount:22',
440 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
442 'id': '4279049078001',
444 'title': 'Titansgrave: Chapter 0',
445 'description': 'Titansgrave: Chapter 0',
446 'duration': 1242.058,
447 'timestamp': 1433556729,
448 'upload_date': '20150606',
449 'uploader_id': '4036320279001',
450 'formats': 'mincount:41',
454 'skip_download': True,
457 # ref: prefixed video id
458 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
459 'only_matching': True,
461 # non numeric ref: prefixed video id
462 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
463 'only_matching': True,
465 # unavailable video without message but with error_code
466 'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001',
467 'only_matching': True,
471 def _extract_url(webpage):
472 urls = BrightcoveNewIE._extract_urls(webpage)
473 return urls[0] if urls else None
476 def _extract_urls(webpage):
478 # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
479 # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
480 # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
481 # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
485 # Look for iframe embeds [1]
486 for _, url in re.findall(
487 r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
488 entries.append(url if url.startswith('http') else 'http:' + url)
490 # Look for embed_in_page embeds [2]
491 for video_id, account_id, player_id, embed in re.findall(
492 # According to examples from [3] it's unclear whether video id
493 # may be optional and what to do when it is
494 # According to [4] data-video-id may be prefixed with ref:
497 data-video-id=["\'](\d+|ref:[^"\']+)["\'][^>]*>.*?
500 src=["\'](?:https?:)?//players\.brightcove\.net/
501 (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
504 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
505 % (account_id, player_id, embed, video_id))
509 def _real_extract(self, url):
510 account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
512 webpage = self._download_webpage(
513 'http://players.brightcove.net/%s/%s_%s/index.min.js'
514 % (account_id, player_id, embed), video_id)
518 catalog = self._search_regex(
519 r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
521 catalog = self._parse_json(
522 js_to_json(catalog), video_id, fatal=False)
524 policy_key = catalog.get('policyKey')
527 policy_key = self._search_regex(
528 r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
529 webpage, 'policy key', group='pk')
531 api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id)
533 json_data = self._download_json(api_url, video_id, headers={
534 'Accept': 'application/json;pk=%s' % policy_key
536 except ExtractorError as e:
537 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
538 json_data = self._parse_json(e.cause.read().decode(), video_id)[0]
539 raise ExtractorError(
540 json_data.get('message') or json_data['error_code'], expected=True)
543 title = json_data['name'].strip()
546 for source in json_data.get('sources', []):
547 container = source.get('container')
548 source_type = source.get('type')
549 src = source.get('src')
550 if source_type == 'application/x-mpegURL' or container == 'M2TS':
553 formats.extend(self._extract_m3u8_formats(
554 src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
555 elif source_type == 'application/dash+xml':
558 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
560 streaming_src = source.get('streaming_src')
561 stream_name, app_name = source.get('stream_name'), source.get('app_name')
562 if not src and not streaming_src and (not stream_name or not app_name):
564 tbr = float_or_none(source.get('avg_bitrate'), 1000)
565 height = int_or_none(source.get('height'))
566 width = int_or_none(source.get('width'))
569 'filesize': int_or_none(source.get('size')),
570 'container': container,
571 'ext': container.lower(),
573 if width == 0 and height == 0:
581 'vcodec': source.get('codec'),
584 def build_format_id(kind):
587 format_id += '-%dk' % int(tbr)
589 format_id += '-%dp' % height
592 if src or streaming_src:
594 'url': src or streaming_src,
595 'format_id': build_format_id('http' if src else 'http-streaming'),
596 'source_preference': 0 if src else -1,
601 'play_path': stream_name,
602 'format_id': build_format_id('rtmp'),
606 errors = json_data.get('errors')
607 if not formats and errors:
609 raise ExtractorError(
610 error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
612 self._sort_formats(formats)
615 for text_track in json_data.get('text_tracks', []):
616 if text_track.get('src'):
617 subtitles.setdefault(text_track.get('srclang'), []).append({
618 'url': text_track['src'],
624 'description': clean_html(json_data.get('description')),
625 'thumbnail': json_data.get('thumbnail') or json_data.get('poster'),
626 'duration': float_or_none(json_data.get('duration'), 1000),
627 'timestamp': parse_iso8601(json_data.get('published_at')),
628 'uploader_id': account_id,
630 'subtitles': subtitles,
631 'tags': json_data.get('tags', []),