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,
32 class BrightcoveLegacyIE(InfoExtractor):
33 IE_NAME = 'brightcove:legacy'
34 _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
35 _FEDERATED_URL = 'http://c.brightcove.com/services/viewer/htmlFederated'
39 # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
40 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
41 'md5': '5423e113865d26e40624dce2e4b45d95',
42 'note': 'Test Brightcove downloads and detection in GenericIE',
44 'id': '2371591881001',
46 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
48 'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
52 # From http://medianetwork.oracle.com/video/player/1785452137001
53 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
55 'id': '1785452137001',
57 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
58 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
63 # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
64 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
66 'id': '2750934548001',
68 'title': 'This Bracelet Acts as a Personal Thermostat',
69 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
70 'uploader': 'Mashable',
74 # test that the default referer works
75 # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
76 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
78 'id': '2878862109001',
80 'title': 'Lost in Motion II',
81 'description': 'md5:363109c02998fee92ec02211bd8000df',
82 'uploader': 'National Ballet of Canada',
86 # test flv videos served by akamaihd.net
87 # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william
88 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?%40videoPlayer=ref%3ABC2996102916001&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',
89 # The md5 checksum changes on each download
91 'id': '2996102916001',
93 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
94 'uploader': 'Red Bull TV',
95 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
100 # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
101 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
104 'id': '3550319591001',
106 'playlist_mincount': 7,
111 def _build_brighcove_url(cls, object_str):
113 Build a Brightcove url from a xml string containing
114 <object class="BrightcoveExperience">{params}</object>
117 # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
118 object_str = re.sub(r'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>',
119 lambda m: m.group(1) + '/>', object_str)
120 # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
121 object_str = object_str.replace('<--', '<!--')
122 # remove namespace to simplify extraction
123 object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
124 object_str = fix_xml_ampersands(object_str)
127 object_doc = compat_etree_fromstring(object_str.encode('utf-8'))
128 except compat_xml_parse_error:
131 fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
132 if fv_el is not None:
135 for k, v in compat_parse_qs(fv_el.attrib['value']).items())
139 def find_param(name):
140 if name in flashvars:
141 return flashvars[name]
142 node = find_xpath_attr(object_doc, './param', 'name', name)
144 return node.attrib['value']
149 playerID = find_param('playerID')
151 raise ExtractorError('Cannot find player ID')
152 params['playerID'] = playerID
154 playerKey = find_param('playerKey')
155 # Not all pages define this value
156 if playerKey is not None:
157 params['playerKey'] = playerKey
158 # These fields hold the id of the video
159 videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
160 if videoPlayer is not None:
161 params['@videoPlayer'] = videoPlayer
162 linkBase = find_param('linkBaseURL')
163 if linkBase is not None:
164 params['linkBaseURL'] = linkBase
165 return cls._make_brightcove_url(params)
168 def _build_brighcove_url_from_js(cls, object_js):
169 # The layout of JS is as follows:
170 # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) {
171 # // build Brightcove <object /> XML
174 r'''(?x)customBC.\createVideo\(
175 .*? # skipping width and height
176 ["\'](?P<playerID>\d+)["\']\s*,\s* # playerID
177 ["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters
178 # in length, however it's appended to itself
179 # in places, so truncate
180 ["\'](?P<videoID>\d+)["\'] # @videoPlayer
183 return cls._make_brightcove_url(m.groupdict())
186 def _make_brightcove_url(cls, params):
187 return update_url_query(cls._FEDERATED_URL, params)
190 def _extract_brightcove_url(cls, webpage):
191 """Try to extract the brightcove url from the webpage, returns None
194 urls = cls._extract_brightcove_urls(webpage)
195 return urls[0] if urls else None
198 def _extract_brightcove_urls(cls, webpage):
199 """Return a list of all Brightcove URLs from the webpage """
202 r'<meta\s+property=[\'"]og:video[\'"]\s+content=[\'"](https?://(?:secure|c)\.brightcove.com/[^\'"]+)[\'"]',
205 url = unescapeHTML(url_m.group(1))
206 # Some sites don't add it, we can't download with this url, for example:
207 # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
208 if 'playerKey' in url or 'videoId' in url:
211 matches = re.findall(
214 [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
215 [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
216 ).+?>\s*</object>''',
219 return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
221 return list(filter(None, [
222 cls._build_brighcove_url_from_js(custom_bc)
223 for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)]))
225 def _real_extract(self, url):
226 url, smuggled_data = unsmuggle_url(url, {})
228 # Change the 'videoId' and others field to '@videoPlayer'
229 url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
230 # Change bckey (used by bcove.me urls) to playerKey
231 url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)
232 mobj = re.match(self._VALID_URL, url)
233 query_str = mobj.group('query')
234 query = compat_urlparse.parse_qs(query_str)
236 videoPlayer = query.get('@videoPlayer')
238 # We set the original url as the default 'Referer' header
239 referer = smuggled_data.get('Referer', url)
240 return self._get_video_info(
241 videoPlayer[0], query, referer=referer)
242 elif 'playerKey' in query:
243 player_key = query['playerKey']
244 return self._get_playlist_info(player_key[0])
246 raise ExtractorError(
247 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
250 def _get_video_info(self, video_id, query, referer=None):
252 linkBase = query.get('linkBaseURL')
253 if linkBase is not None:
254 referer = linkBase[0]
255 if referer is not None:
256 headers['Referer'] = referer
257 webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
259 error_msg = self._html_search_regex(
260 r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
261 'error message', default=None)
262 if error_msg is not None:
263 raise ExtractorError(
264 'brightcove said: %s' % error_msg, expected=True)
266 self.report_extraction(video_id)
267 info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
268 info = json.loads(info)['data']
269 video_info = info['programmedContent']['videoPlayer']['mediaDTO']
270 video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
272 return self._extract_video_info(video_info)
274 def _get_playlist_info(self, player_key):
275 info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
276 playlist_info = self._download_webpage(
277 info_url, player_key, 'Downloading playlist information')
279 json_data = json.loads(playlist_info)
280 if 'videoList' not in json_data:
281 raise ExtractorError('Empty playlist')
282 playlist_info = json_data['videoList']
283 videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
285 return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
286 playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
288 def _extract_video_info(self, video_info):
290 'id': compat_str(video_info['id']),
291 'title': video_info['displayName'].strip(),
292 'description': video_info.get('shortDescription'),
293 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
294 'uploader': video_info.get('publisherName'),
297 renditions = video_info.get('renditions')
300 for rend in renditions:
301 url = rend['defaultURL']
306 url_comp = compat_urllib_parse_urlparse(url)
307 if url_comp.path.endswith('.m3u8'):
309 self._extract_m3u8_formats(url, info['id'], 'mp4'))
311 elif 'akamaihd.net' in url_comp.netloc:
312 # This type of renditions are served through
313 # akamaihd.net, but they don't use f4m manifests
314 url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
317 ext = determine_ext(url)
318 size = rend.get('size')
322 'height': rend.get('frameHeight'),
323 'width': rend.get('frameWidth'),
324 'filesize': size if size != 0 else None,
326 self._sort_formats(formats)
327 info['formats'] = formats
328 elif video_info.get('FLVFullLengthURL') is not None:
330 'url': video_info['FLVFullLengthURL'],
333 if self._downloader.params.get('include_ads', False):
334 adServerURL = video_info.get('_youtubedl_adServerURL')
343 'title': info['title'],
344 'entries': [ad_info, info],
349 if 'url' not in info and not info.get('formats'):
350 raise ExtractorError('Unable to extract video url for %s' % info['id'])
354 class BrightcoveNewIE(InfoExtractor):
355 IE_NAME = 'brightcove:new'
356 _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)'
358 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
359 'md5': 'c8100925723840d4b0d243f7025703be',
361 'id': '4463358922001',
363 'title': 'Meet the man behind Popcorn Time',
364 'description': 'md5:eac376a4fe366edc70279bfb681aea16',
366 'timestamp': 1441391203,
367 'upload_date': '20150904',
368 'uploader_id': '929656772001',
369 'formats': 'mincount:22',
373 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
375 'id': '4279049078001',
377 'title': 'Titansgrave: Chapter 0',
378 'description': 'Titansgrave: Chapter 0',
379 'duration': 1242.058,
380 'timestamp': 1433556729,
381 'upload_date': '20150606',
382 'uploader_id': '4036320279001',
383 'formats': 'mincount:41',
386 'skip_download': True,
389 # ref: prefixed video id
390 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
391 'only_matching': True,
393 # non numeric ref: prefixed video id
394 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
395 'only_matching': True,
399 def _extract_url(webpage):
400 urls = BrightcoveNewIE._extract_urls(webpage)
401 return urls[0] if urls else None
404 def _extract_urls(webpage):
406 # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
407 # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
408 # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
409 # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
413 # Look for iframe embeds [1]
414 for _, url in re.findall(
415 r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
416 entries.append(url if url.startswith('http') else 'http:' + url)
418 # Look for embed_in_page embeds [2]
419 for video_id, account_id, player_id, embed in re.findall(
420 # According to examples from [3] it's unclear whether video id
421 # may be optional and what to do when it is
422 # According to [4] data-video-id may be prefixed with ref:
425 data-video-id=["\'](\d+|ref:[^"\']+)["\'][^>]*>.*?
428 src=["\'](?:https?:)?//players\.brightcove\.net/
429 (\d+)/([\da-f-]+)_([^/]+)/index(?:\.min)?\.js
432 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
433 % (account_id, player_id, embed, video_id))
437 def _real_extract(self, url):
438 account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
440 webpage = self._download_webpage(
441 'http://players.brightcove.net/%s/%s_%s/index.min.js'
442 % (account_id, player_id, embed), video_id)
446 catalog = self._search_regex(
447 r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
449 catalog = self._parse_json(
450 js_to_json(catalog), video_id, fatal=False)
452 policy_key = catalog.get('policyKey')
455 policy_key = self._search_regex(
456 r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
457 webpage, 'policy key', group='pk')
459 api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id)
461 json_data = self._download_json(api_url, video_id, headers={
462 'Accept': 'application/json;pk=%s' % policy_key
464 except ExtractorError as e:
465 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
466 json_data = self._parse_json(e.cause.read().decode(), video_id)
467 raise ExtractorError(json_data[0]['message'], expected=True)
470 title = json_data['name']
473 for source in json_data.get('sources', []):
474 container = source.get('container')
475 source_type = source.get('type')
476 src = source.get('src')
477 if source_type == 'application/x-mpegURL' or container == 'M2TS':
480 formats.extend(self._extract_m3u8_formats(
481 src, video_id, 'mp4', m3u8_id='hls', fatal=False))
482 elif source_type == 'application/dash+xml':
485 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
487 streaming_src = source.get('streaming_src')
488 stream_name, app_name = source.get('stream_name'), source.get('app_name')
489 if not src and not streaming_src and (not stream_name or not app_name):
491 tbr = float_or_none(source.get('avg_bitrate'), 1000)
492 height = int_or_none(source.get('height'))
493 width = int_or_none(source.get('width'))
496 'filesize': int_or_none(source.get('size')),
497 'container': container,
498 'ext': container.lower(),
500 if width == 0 and height == 0:
508 'vcodec': source.get('codec'),
511 def build_format_id(kind):
514 format_id += '-%dk' % int(tbr)
516 format_id += '-%dp' % height
519 if src or streaming_src:
521 'url': src or streaming_src,
522 'format_id': build_format_id('http' if src else 'http-streaming'),
523 'preference': 2 if src else 1,
528 'play_path': stream_name,
529 'format_id': build_format_id('rtmp'),
532 self._sort_formats(formats)
534 description = json_data.get('description')
535 thumbnail = json_data.get('thumbnail')
536 timestamp = parse_iso8601(json_data.get('published_at'))
537 duration = float_or_none(json_data.get('duration'), 1000)
538 tags = json_data.get('tags', [])
543 'description': description,
544 'thumbnail': thumbnail,
545 'duration': duration,
546 'timestamp': timestamp,
547 'uploader_id': account_id,