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',
49 'timestamp': 1368213670,
50 'upload_date': '20130510',
51 'uploader_id': '1589608506001',
55 # From http://medianetwork.oracle.com/video/player/1785452137001
56 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
58 'id': '1785452137001',
60 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
61 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
63 'timestamp': 1344975024,
64 'upload_date': '20120814',
65 'uploader_id': '1460825906',
69 # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
70 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
72 'id': '2750934548001',
74 'title': 'This Bracelet Acts as a Personal Thermostat',
75 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
76 'uploader': 'Mashable',
77 'timestamp': 1382041798,
78 'upload_date': '20131017',
79 'uploader_id': '1130468786001',
83 # test that the default referer works
84 # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
85 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
87 'id': '2878862109001',
89 'title': 'Lost in Motion II',
90 'description': 'md5:363109c02998fee92ec02211bd8000df',
91 'uploader': 'National Ballet of Canada',
96 # test flv videos served by akamaihd.net
97 # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william
98 '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',
99 # The md5 checksum changes on each download
101 'id': '3750436379001',
103 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
104 'uploader': 'RBTV Old (do not use)',
105 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
106 'timestamp': 1409122195,
107 'upload_date': '20140827',
108 'uploader_id': '710858724001',
112 # playlist with 'videoList'
113 # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
114 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
117 'id': '3550319591001',
119 'playlist_mincount': 7,
122 # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965)
123 'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg',
125 'id': '1522758701001',
126 'title': 'Lesson 08',
128 'playlist_mincount': 10,
139 def _build_brighcove_url(cls
, object_str
):
141 Build a Brightcove url from a xml string containing
142 <object class="BrightcoveExperience">{params}</object>
145 # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553
146 object_str
= re
.sub(r
'(<param(?:\s+[a-zA-Z0-9_]+="[^"]*")*)>',
147 lambda m
: m
.group(1) + '/>', object_str
)
148 # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
149 object_str
= object_str
.replace('<--', '<!--')
150 # remove namespace to simplify extraction
151 object_str
= re
.sub(r
'(<object[^>]*)(xmlns=".*?")', r
'\1', object_str
)
152 object_str
= fix_xml_ampersands(object_str
)
155 object_doc
= compat_etree_fromstring(object_str
.encode('utf-8'))
156 except compat_xml_parse_error
:
159 fv_el
= find_xpath_attr(object_doc
, './param', 'name', 'flashVars')
160 if fv_el
is not None:
163 for k
, v
in compat_parse_qs(fv_el
.attrib
['value']).items())
167 data_url
= object_doc
.attrib
.get('data', '')
168 data_url_params
= compat_parse_qs(compat_urllib_parse_urlparse(data_url
).query
)
170 def find_param(name
):
171 if name
in flashvars
:
172 return flashvars
[name
]
173 node
= find_xpath_attr(object_doc
, './param', 'name', name
)
175 return node
.attrib
['value']
176 return data_url_params
.get(name
)
180 playerID
= find_param('playerID')
182 raise ExtractorError('Cannot find player ID')
183 params
['playerID'] = playerID
185 playerKey
= find_param('playerKey')
186 # Not all pages define this value
187 if playerKey
is not None:
188 params
['playerKey'] = playerKey
189 # These fields hold the id of the video
190 videoPlayer
= find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
191 if videoPlayer
is not None:
192 params
['@videoPlayer'] = videoPlayer
193 linkBase
= find_param('linkBaseURL')
194 if linkBase
is not None:
195 params
['linkBaseURL'] = linkBase
196 return cls
._make
_brightcove
_url
(params
)
199 def _build_brighcove_url_from_js(cls
, object_js
):
200 # The layout of JS is as follows:
201 # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) {
202 # // build Brightcove <object /> XML
205 r
'''(?x)customBC.\createVideo\(
206 .*? # skipping width and height
207 ["\'](?P
<playerID
>\d
+)["\']\s*,\s* # playerID
208 ["\'](?P
<playerKey
>AQ
[^
"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters
209 # in length, however it's appended to itself
210 # in places, so truncate
211 ["\'](?P
<videoID
>\d
+)["\'] # @videoPlayer
214 return cls._make_brightcove_url(m.groupdict())
217 def _make_brightcove_url(cls, params):
218 return update_url_query(cls._FEDERATED_URL, params)
221 def _extract_brightcove_url(cls, webpage):
222 """Try to extract the brightcove url from the webpage, returns None
225 urls = cls._extract_brightcove_urls(webpage)
226 return urls[0] if urls else None
229 def _extract_brightcove_urls(cls, webpage):
230 """Return a list of all Brightcove URLs from the webpage """
233 r'<meta\s+property=[\'"]og
:video
[\'"]\s+content=[\'"](https?
://(?
:secure|c
)\
.brightcove
.com
/[^
\'"]+)[\'"]',
236 url = unescapeHTML(url_m.group(1))
237 # Some sites don't add it
, we can
't download with this url, for example:
238 # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
239 if 'playerKey
' in url or 'videoId
' in url:
242 matches = re.findall(
245 [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
246 [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
247 ).+?>\s*</object>''',
250 return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
252 return list(filter(None, [
253 cls._build_brighcove_url_from_js(custom_bc)
254 for custom_bc in re.findall(r'(customBC\
.createVideo\
(.+?\
);)', webpage)]))
256 def _real_extract(self, url):
257 url, smuggled_data = unsmuggle_url(url, {})
259 # Change the 'videoId
' and others field to '@videoPlayer'
260 url = re.sub(r'(?
<=[?
&])(videoI(d|D
)|bctid
)', '%40videoPlayer
', url)
261 # Change bckey (used by bcove.me urls) to playerKey
262 url = re.sub(r'(?
<=[?
&])bckey
', 'playerKey
', url)
263 mobj = re.match(self._VALID_URL, url)
264 query_str = mobj.group('query
')
265 query = compat_urlparse.parse_qs(query_str)
267 videoPlayer = query.get('@videoPlayer')
269 # We set the original url as the default 'Referer
' header
270 referer = smuggled_data.get('Referer
', url)
271 return self._get_video_info(
272 videoPlayer[0], query, referer=referer)
273 elif 'playerKey
' in query:
274 player_key = query['playerKey
']
275 return self._get_playlist_info(player_key[0])
277 raise ExtractorError(
278 'Cannot find playerKey
= variable
. Did you forget quotes
in a shell invocation?
',
281 def _get_video_info(self, video_id, query, referer=None):
283 linkBase = query.get('linkBaseURL
')
284 if linkBase is not None:
285 referer = linkBase[0]
286 if referer is not None:
287 headers['Referer
'] = referer
288 webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
290 error_msg = self._html_search_regex(
291 r"<h1>We're sorry
.</h1
>([\s
\n]*<p
>.*?
</p
>)+", webpage,
292 'error message', default=None)
293 if error_msg is not None:
294 raise ExtractorError(
295 'brightcove said: %s' % error_msg, expected=True)
297 self.report_extraction(video_id)
298 info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
299 info = json.loads(info)['data']
300 video_info = info['programmedContent']['videoPlayer']['mediaDTO']
301 video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
303 return self._extract_video_info(video_info)
305 def _get_playlist_info(self, player_key):
306 info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
307 playlist_info = self._download_webpage(
308 info_url, player_key, 'Downloading playlist information')
310 json_data = json.loads(playlist_info)
311 if 'videoList' in json_data:
312 playlist_info = json_data['videoList']
313 playlist_dto = playlist_info['mediaCollectionDTO']
314 elif 'playlistTabs' in json_data:
315 playlist_info = json_data['playlistTabs']
316 playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0]
318 raise ExtractorError('Empty playlist')
320 videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']]
322 return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
323 playlist_title=playlist_dto['displayName'])
325 def _extract_video_info(self, video_info):
326 video_id = compat_str(video_info['id'])
327 publisher_id = video_info.get('publisherId')
330 'title': video_info['displayName'].strip(),
331 'description': video_info.get('shortDescription'),
332 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
333 'uploader': video_info.get('publisherName'),
334 'uploader_id': compat_str(publisher_id) if publisher_id else None,
335 'duration': float_or_none(video_info.get('length'), 1000),
336 'timestamp': int_or_none(video_info.get('creationDate'), 1000),
339 renditions = video_info.get('renditions', []) + video_info.get('IOSRenditions', [])
342 for rend in renditions:
343 url = rend['defaultURL']
348 url_comp = compat_urllib_parse_urlparse(url)
349 if url_comp.path.endswith('.m3u8'):
351 self._extract_m3u8_formats(
352 url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
354 elif 'akamaihd.net' in url_comp.netloc:
355 # This type of renditions are served through
356 # akamaihd.net, but they don't use f4m manifests
357 url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
360 ext = determine_ext(url)
361 tbr = int_or_none(rend.get('encodingRate'), 1000)
363 'format_id': 'http%s' % ('-%s' % tbr if tbr else ''),
366 'filesize': int_or_none(rend.get('size')) or None,
369 if rend.get('audioOnly'):
375 'height': int_or_none(rend.get('frameHeight')),
376 'width': int_or_none(rend.get('frameWidth')),
377 'vcodec': rend.get('videoCodec'),
380 # m3u8 manifests with remote == false are media playlists
381 # Not calling _extract_m3u8_formats here to save network traffic
384 'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''),
386 'protocol': 'm3u8_native',
389 formats.append(a_format)
390 self._sort_formats(formats)
391 info['formats'] = formats
392 elif video_info.get('FLVFullLengthURL') is not None:
394 'url': video_info['FLVFullLengthURL'],
395 'vcodec': self.FLV_VCODECS.get(video_info.get('FLVFullCodec')),
396 'filesize': int_or_none(video_info.get('FLVFullSize')),
399 if self._downloader.params.get('include_ads', False):
400 adServerURL = video_info.get('_youtubedl_adServerURL')
409 'title': info['title'],
410 'entries': [ad_info, info],
415 if 'url' not in info and not info.get('formats'):
416 raise ExtractorError('Unable to extract video url for %s' % video_id)
420 class BrightcoveNewIE(InfoExtractor):
421 IE_NAME = 'brightcove:new'
422 _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)'
424 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
425 'md5': 'c8100925723840d4b0d243f7025703be',
427 'id': '4463358922001',
429 'title': 'Meet the man behind Popcorn Time',
430 'description': 'md5:eac376a4fe366edc70279bfb681aea16',
432 'timestamp': 1441391203,
433 'upload_date': '20150904',
434 'uploader_id': '929656772001',
435 'formats': 'mincount:22',
439 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
441 'id': '4279049078001',
443 'title': 'Titansgrave: Chapter 0',
444 'description': 'Titansgrave: Chapter 0',
445 'duration': 1242.058,
446 'timestamp': 1433556729,
447 'upload_date': '20150606',
448 'uploader_id': '4036320279001',
449 'formats': 'mincount:41',
453 'skip_download': True,
456 # ref: prefixed video id
457 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
458 'only_matching': True,
460 # non numeric ref: prefixed video id
461 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
462 'only_matching': True,
464 # unavailable video without message but with error_code
465 'url': 'http://players.brightcove.net/1305187701/c832abfb-641b-44eb-9da0-2fe76786505f_default/index.html?videoId=4377407326001',
466 'only_matching': True,
470 def _extract_url(webpage):
471 urls = BrightcoveNewIE._extract_urls(webpage)
472 return urls[0] if urls else None
475 def _extract_urls(webpage):
477 # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
478 # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
479 # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
480 # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
484 # Look for iframe embeds [1]
485 for _, url in re.findall(
486 r'<iframe[^>]+src=(["\'])((?
:https?
:)?
//players\
.brightcove\
.net
/\d
+/[^
/]+/index\
.html
.+?
)\
1', webpage):
487 entries.append(url if url.startswith('http
') else 'http
:' + url)
489 # Look for embed_in_page embeds [2]
490 for video_id, account_id, player_id, embed in re.findall(
491 # According to examples from [3] it's unclear whether video
id
492 # may be optional and what to do when it is
493 # According to [4] data-video-id may be prefixed with ref:
496 data-video-id=["\'](\d
+|ref
:[^
"\']+)["\'][^
>]*>.*?
499 src
=["\'](?:https?:)?//players\.brightcove\.net/
500 (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
503 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
504 % (account_id, player_id, embed, video_id))
508 def _real_extract(self, url):
509 account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
511 webpage = self._download_webpage(
512 'http://players.brightcove.net/%s/%s_%s/index.min.js'
513 % (account_id, player_id, embed), video_id)
517 catalog = self._search_regex(
518 r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
520 catalog = self._parse_json(
521 js_to_json(catalog), video_id, fatal=False)
523 policy_key = catalog.get('policyKey')
526 policy_key = self._search_regex(
527 r'policyKey\s*:\s*(["\'])(?P
<pk
>.+?
)\
1',
528 webpage, 'policy key
', group='pk
')
530 api_url = 'https
://edge
.api
.brightcove
.com
/playback
/v1
/accounts
/%s/videos
/%s' % (account_id, video_id)
532 json_data = self._download_json(api_url, video_id, headers={
533 'Accept
': 'application
/json
;pk
=%s' % policy_key
535 except ExtractorError as e:
536 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
537 json_data = self._parse_json(e.cause.read().decode(), video_id)[0]
538 raise ExtractorError(
539 json_data.get('message
') or json_data['error_code
'], expected=True)
542 title = json_data['name
'].strip()
545 for source in json_data.get('sources
', []):
546 container = source.get('container
')
547 source_type = source.get('type')
548 src = source.get('src
')
549 if source_type == 'application
/x
-mpegURL
' or container == 'M2TS
':
552 formats.extend(self._extract_m3u8_formats(
553 src, video_id, 'mp4
', 'm3u8_native
', m3u8_id='hls
', fatal=False))
554 elif source_type == 'application
/dash
+xml
':
557 formats.extend(self._extract_mpd_formats(src, video_id, 'dash
', fatal=False))
559 streaming_src = source.get('streaming_src
')
560 stream_name, app_name = source.get('stream_name
'), source.get('app_name
')
561 if not src and not streaming_src and (not stream_name or not app_name):
563 tbr = float_or_none(source.get('avg_bitrate
'), 1000)
564 height = int_or_none(source.get('height
'))
565 width = int_or_none(source.get('width
'))
568 'filesize
': int_or_none(source.get('size
')),
569 'container
': container,
570 'ext
': container.lower(),
572 if width == 0 and height == 0:
580 'vcodec
': source.get('codec
'),
583 def build_format_id(kind):
586 format_id += '-%dk
' % int(tbr)
588 format_id += '-%dp
' % height
591 if src or streaming_src:
593 'url
': src or streaming_src,
594 'format_id
': build_format_id('http
' if src else 'http
-streaming
'),
595 'source_preference
': 0 if src else -1,
600 'play_path
': stream_name,
601 'format_id
': build_format_id('rtmp
'),
605 errors = json_data.get('errors
')
606 if not formats and errors:
608 raise ExtractorError(
609 error.get('message
') or error.get('error_subcode
') or error['error_code
'], expected=True)
611 self._sort_formats(formats)
614 for text_track in json_data.get('text_tracks
', []):
615 if text_track.get('src
'):
616 subtitles.setdefault(text_track.get('srclang
'), []).append({
617 'url
': text_track['src
'],
623 'description
': json_data.get('description
'),
624 'thumbnail
': json_data.get('thumbnail
') or json_data.get('poster
'),
625 'duration
': float_or_none(json_data.get('duration
'), 1000),
626 'timestamp
': parse_iso8601(json_data.get('published_at
')),
627 'uploader_id
': account_id,
629 'subtitles
': subtitles,
630 'tags
': json_data.get('tags
', []),