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 data_url = object_doc.attrib.get('data', '')
140 data_url_params = compat_parse_qs(compat_urllib_parse_urlparse(data_url).query)
142 def find_param(name):
143 if name in flashvars:
144 return flashvars[name]
145 node = find_xpath_attr(object_doc, './param', 'name', name)
147 return node.attrib['value']
148 return data_url_params.get(name)
152 playerID = find_param('playerID')
154 raise ExtractorError('Cannot find player ID')
155 params['playerID'] = playerID
157 playerKey = find_param('playerKey')
158 # Not all pages define this value
159 if playerKey is not None:
160 params['playerKey'] = playerKey
161 # These fields hold the id of the video
162 videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
163 if videoPlayer is not None:
164 params['@videoPlayer'] = videoPlayer
165 linkBase = find_param('linkBaseURL')
166 if linkBase is not None:
167 params['linkBaseURL'] = linkBase
168 return cls._make_brightcove_url(params)
171 def _build_brighcove_url_from_js(cls, object_js):
172 # The layout of JS is as follows:
173 # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) {
174 # // build Brightcove <object /> XML
177 r'''(?x)customBC.\createVideo\(
178 .*? # skipping width and height
179 ["\'](?P<playerID>\d+)["\']\s*,\s* # playerID
180 ["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters
181 # in length, however it's appended to itself
182 # in places, so truncate
183 ["\'](?P<videoID>\d+)["\'] # @videoPlayer
186 return cls._make_brightcove_url(m.groupdict())
189 def _make_brightcove_url(cls, params):
190 return update_url_query(cls._FEDERATED_URL, params)
193 def _extract_brightcove_url(cls, webpage):
194 """Try to extract the brightcove url from the webpage, returns None
197 urls = cls._extract_brightcove_urls(webpage)
198 return urls[0] if urls else None
201 def _extract_brightcove_urls(cls, webpage):
202 """Return a list of all Brightcove URLs from the webpage """
205 r'<meta\s+property=[\'"]og:video[\'"]\s+content=[\'"](https?://(?:secure|c)\.brightcove.com/[^\'"]+)[\'"]',
208 url = unescapeHTML(url_m.group(1))
209 # Some sites don't add it, we can't download with this url, for example:
210 # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
211 if 'playerKey' in url or 'videoId' in url:
214 matches = re.findall(
217 [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
218 [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
219 ).+?>\s*</object>''',
222 return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
224 return list(filter(None, [
225 cls._build_brighcove_url_from_js(custom_bc)
226 for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)]))
228 def _real_extract(self, url):
229 url, smuggled_data = unsmuggle_url(url, {})
231 # Change the 'videoId' and others field to '@videoPlayer'
232 url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
233 # Change bckey (used by bcove.me urls) to playerKey
234 url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)
235 mobj = re.match(self._VALID_URL, url)
236 query_str = mobj.group('query')
237 query = compat_urlparse.parse_qs(query_str)
239 videoPlayer = query.get('@videoPlayer')
241 # We set the original url as the default 'Referer' header
242 referer = smuggled_data.get('Referer', url)
243 return self._get_video_info(
244 videoPlayer[0], query, referer=referer)
245 elif 'playerKey' in query:
246 player_key = query['playerKey']
247 return self._get_playlist_info(player_key[0])
249 raise ExtractorError(
250 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
253 def _get_video_info(self, video_id, query, referer=None):
255 linkBase = query.get('linkBaseURL')
256 if linkBase is not None:
257 referer = linkBase[0]
258 if referer is not None:
259 headers['Referer'] = referer
260 webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
262 error_msg = self._html_search_regex(
263 r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
264 'error message', default=None)
265 if error_msg is not None:
266 raise ExtractorError(
267 'brightcove said: %s' % error_msg, expected=True)
269 self.report_extraction(video_id)
270 info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
271 info = json.loads(info)['data']
272 video_info = info['programmedContent']['videoPlayer']['mediaDTO']
273 video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
275 return self._extract_video_info(video_info)
277 def _get_playlist_info(self, player_key):
278 info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
279 playlist_info = self._download_webpage(
280 info_url, player_key, 'Downloading playlist information')
282 json_data = json.loads(playlist_info)
283 if 'videoList' not in json_data:
284 raise ExtractorError('Empty playlist')
285 playlist_info = json_data['videoList']
286 videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
288 return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
289 playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
291 def _extract_video_info(self, video_info):
293 'id': compat_str(video_info['id']),
294 'title': video_info['displayName'].strip(),
295 'description': video_info.get('shortDescription'),
296 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
297 'uploader': video_info.get('publisherName'),
300 renditions = video_info.get('renditions', []) + video_info.get('IOSRenditions', [])
303 for rend in renditions:
304 url = rend['defaultURL']
309 url_comp = compat_urllib_parse_urlparse(url)
310 if url_comp.path.endswith('.m3u8'):
312 self._extract_m3u8_formats(url, info['id'], 'mp4'))
314 elif 'akamaihd.net' in url_comp.netloc:
315 # This type of renditions are served through
316 # akamaihd.net, but they don't use f4m manifests
317 url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
320 ext = determine_ext(url)
321 size = rend.get('size')
325 'height': rend.get('frameHeight'),
326 'width': rend.get('frameWidth'),
327 'filesize': size if size != 0 else None,
330 # m3u8 manifests with remote == false are media playlists
331 # Not calling _extract_m3u8_formats here to save network traffic
338 formats.append(a_format)
339 self._sort_formats(formats)
340 info['formats'] = formats
341 elif video_info.get('FLVFullLengthURL') is not None:
343 'url': video_info['FLVFullLengthURL'],
346 if self._downloader.params.get('include_ads', False):
347 adServerURL = video_info.get('_youtubedl_adServerURL')
356 'title': info['title'],
357 'entries': [ad_info, info],
362 if 'url' not in info and not info.get('formats'):
363 raise ExtractorError('Unable to extract video url for %s' % info['id'])
367 class BrightcoveNewIE(InfoExtractor):
368 IE_NAME = 'brightcove:new'
369 _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)'
371 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
372 'md5': 'c8100925723840d4b0d243f7025703be',
374 'id': '4463358922001',
376 'title': 'Meet the man behind Popcorn Time',
377 'description': 'md5:eac376a4fe366edc70279bfb681aea16',
379 'timestamp': 1441391203,
380 'upload_date': '20150904',
381 'uploader_id': '929656772001',
382 'formats': 'mincount:22',
386 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
388 'id': '4279049078001',
390 'title': 'Titansgrave: Chapter 0',
391 'description': 'Titansgrave: Chapter 0',
392 'duration': 1242.058,
393 'timestamp': 1433556729,
394 'upload_date': '20150606',
395 'uploader_id': '4036320279001',
396 'formats': 'mincount:41',
399 'skip_download': True,
402 # ref: prefixed video id
403 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
404 'only_matching': True,
406 # non numeric ref: prefixed video id
407 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
408 'only_matching': True,
412 def _extract_url(webpage):
413 urls = BrightcoveNewIE._extract_urls(webpage)
414 return urls[0] if urls else None
417 def _extract_urls(webpage):
419 # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
420 # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
421 # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
422 # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
426 # Look for iframe embeds [1]
427 for _, url in re.findall(
428 r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
429 entries.append(url if url.startswith('http') else 'http:' + url)
431 # Look for embed_in_page embeds [2]
432 for video_id, account_id, player_id, embed in re.findall(
433 # According to examples from [3] it's unclear whether video id
434 # may be optional and what to do when it is
435 # According to [4] data-video-id may be prefixed with ref:
438 data-video-id=["\'](\d+|ref:[^"\']+)["\'][^>]*>.*?
441 src=["\'](?:https?:)?//players\.brightcove\.net/
442 (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
445 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
446 % (account_id, player_id, embed, video_id))
450 def _real_extract(self, url):
451 account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
453 webpage = self._download_webpage(
454 'http://players.brightcove.net/%s/%s_%s/index.min.js'
455 % (account_id, player_id, embed), video_id)
459 catalog = self._search_regex(
460 r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
462 catalog = self._parse_json(
463 js_to_json(catalog), video_id, fatal=False)
465 policy_key = catalog.get('policyKey')
468 policy_key = self._search_regex(
469 r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
470 webpage, 'policy key', group='pk')
472 api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id)
474 json_data = self._download_json(api_url, video_id, headers={
475 'Accept': 'application/json;pk=%s' % policy_key
477 except ExtractorError as e:
478 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
479 json_data = self._parse_json(e.cause.read().decode(), video_id)
480 raise ExtractorError(json_data[0]['message'], expected=True)
483 title = json_data['name']
486 for source in json_data.get('sources', []):
487 container = source.get('container')
488 source_type = source.get('type')
489 src = source.get('src')
490 if source_type == 'application/x-mpegURL' or container == 'M2TS':
493 formats.extend(self._extract_m3u8_formats(
494 src, video_id, 'mp4', m3u8_id='hls', fatal=False))
495 elif source_type == 'application/dash+xml':
498 formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
500 streaming_src = source.get('streaming_src')
501 stream_name, app_name = source.get('stream_name'), source.get('app_name')
502 if not src and not streaming_src and (not stream_name or not app_name):
504 tbr = float_or_none(source.get('avg_bitrate'), 1000)
505 height = int_or_none(source.get('height'))
506 width = int_or_none(source.get('width'))
509 'filesize': int_or_none(source.get('size')),
510 'container': container,
511 'ext': container.lower(),
513 if width == 0 and height == 0:
521 'vcodec': source.get('codec'),
524 def build_format_id(kind):
527 format_id += '-%dk' % int(tbr)
529 format_id += '-%dp' % height
532 if src or streaming_src:
534 'url': src or streaming_src,
535 'format_id': build_format_id('http' if src else 'http-streaming'),
536 'preference': 2 if src else 1,
541 'play_path': stream_name,
542 'format_id': build_format_id('rtmp'),
545 self._sort_formats(formats)
547 description = json_data.get('description')
548 thumbnail = json_data.get('thumbnail')
549 timestamp = parse_iso8601(json_data.get('published_at'))
550 duration = float_or_none(json_data.get('duration'), 1000)
551 tags = json_data.get('tags', [])
556 'description': description,
557 'thumbnail': thumbnail,
558 'duration': duration,
559 'timestamp': timestamp,
560 'uploader_id': account_id,