]>
Commit | Line | Data |
---|---|---|
592882aa | 1 | # encoding: utf-8 |
400e5810 | 2 | from __future__ import unicode_literals |
592882aa | 3 | |
fbaaad49 JMF |
4 | import re |
5 | import json | |
cfe50f04 | 6 | import xml.etree.ElementTree |
fbaaad49 JMF |
7 | |
8 | from .common import InfoExtractor | |
cfe50f04 JMF |
9 | from ..utils import ( |
10 | compat_urllib_parse, | |
45ff2d51 | 11 | find_xpath_attr, |
7b0817e8 | 12 | fix_xml_ampersands, |
6543f0dc | 13 | compat_urlparse, |
eeb165e6 | 14 | compat_str, |
dd5bcdc4 | 15 | compat_urllib_request, |
7b0817e8 | 16 | compat_parse_qs, |
37aab278 | 17 | compat_urllib_parse_urlparse, |
592882aa | 18 | |
5524b242 | 19 | determine_ext, |
592882aa | 20 | ExtractorError, |
79f82953 | 21 | unsmuggle_url, |
ac6e4ca1 | 22 | unescapeHTML, |
cfe50f04 | 23 | ) |
fbaaad49 | 24 | |
dd5bcdc4 | 25 | |
fbaaad49 | 26 | class BrightcoveIE(InfoExtractor): |
b5af6fcd | 27 | _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*?\?(?P<query>.*)' |
cfe50f04 | 28 | _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' |
592882aa JMF |
29 | |
30 | _TESTS = [ | |
31 | { | |
4de1994b | 32 | # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ |
400e5810 | 33 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', |
400e5810 PH |
34 | 'md5': '5423e113865d26e40624dce2e4b45d95', |
35 | 'note': 'Test Brightcove downloads and detection in GenericIE', | |
36 | 'info_dict': { | |
96bef88f JMF |
37 | 'id': '2371591881001', |
38 | 'ext': 'mp4', | |
400e5810 PH |
39 | 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', |
40 | 'uploader': '8TV', | |
41 | 'description': 'md5:a950cc4285c43e44d763d036710cd9cd', | |
592882aa JMF |
42 | } |
43 | }, | |
44 | { | |
4de1994b | 45 | # From http://medianetwork.oracle.com/video/player/1785452137001 |
400e5810 | 46 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001', |
400e5810 | 47 | 'info_dict': { |
96bef88f JMF |
48 | 'id': '1785452137001', |
49 | 'ext': 'flv', | |
400e5810 PH |
50 | 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges', |
51 | 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.', | |
52 | 'uploader': 'Oracle', | |
592882aa JMF |
53 | }, |
54 | }, | |
fc4a0c2a JMF |
55 | { |
56 | # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/ | |
400e5810 PH |
57 | 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001', |
58 | 'info_dict': { | |
59 | 'id': '2750934548001', | |
60 | 'ext': 'mp4', | |
61 | 'title': 'This Bracelet Acts as a Personal Thermostat', | |
62 | 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0', | |
63 | 'uploader': 'Mashable', | |
fc4a0c2a JMF |
64 | }, |
65 | }, | |
77526143 JMF |
66 | { |
67 | # test that the default referer works | |
68 | # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ | |
400e5810 PH |
69 | 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', |
70 | 'info_dict': { | |
71 | 'id': '2878862109001', | |
72 | 'ext': 'mp4', | |
73 | 'title': 'Lost in Motion II', | |
74 | 'description': 'md5:363109c02998fee92ec02211bd8000df', | |
75 | 'uploader': 'National Ballet of Canada', | |
77526143 | 76 | }, |
5524b242 JMF |
77 | }, |
78 | { | |
79 | # test flv videos served by akamaihd.net | |
80 | # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william | |
81 | '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', | |
82 | # The md5 checksum changes on each download | |
83 | 'info_dict': { | |
84 | 'id': '2996102916001', | |
85 | 'ext': 'flv', | |
86 | 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', | |
87 | 'uploader': 'Red Bull TV', | |
88 | 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals', | |
89 | }, | |
90 | }, | |
bd4e40df JMF |
91 | { |
92 | # playlist test | |
93 | # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players | |
94 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL', | |
95 | 'info_dict': { | |
96 | 'title': 'Sealife', | |
97 | }, | |
98 | 'playlist_mincount': 7, | |
99 | }, | |
592882aa | 100 | ] |
cfe50f04 JMF |
101 | |
102 | @classmethod | |
103 | def _build_brighcove_url(cls, object_str): | |
104 | """ | |
105 | Build a Brightcove url from a xml string containing | |
106 | <object class="BrightcoveExperience">{params}</object> | |
107 | """ | |
46e28a84 PH |
108 | |
109 | # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553 | |
110 | object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>', | |
111 | lambda m: m.group(1) + '/>', object_str) | |
2d0efe70 | 112 | # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608 |
400e5810 | 113 | object_str = object_str.replace('<--', '<!--') |
7b0817e8 | 114 | object_str = fix_xml_ampersands(object_str) |
46e28a84 | 115 | |
0479c625 | 116 | object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8')) |
7b0817e8 PH |
117 | |
118 | fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars') | |
47917f24 JMF |
119 | if fv_el is not None: |
120 | flashvars = dict( | |
121 | (k, v[0]) | |
122 | for k, v in compat_parse_qs(fv_el.attrib['value']).items()) | |
123 | else: | |
124 | flashvars = {} | |
7b0817e8 | 125 | |
36de0a0e | 126 | def find_param(name): |
7b0817e8 PH |
127 | if name in flashvars: |
128 | return flashvars[name] | |
d214fdb8 JMF |
129 | node = find_xpath_attr(object_doc, './param', 'name', name) |
130 | if node is not None: | |
131 | return node.attrib['value'] | |
132 | return None | |
7b0817e8 PH |
133 | |
134 | params = {} | |
135 | ||
136 | playerID = find_param('playerID') | |
137 | if playerID is None: | |
138 | raise ExtractorError('Cannot find player ID') | |
139 | params['playerID'] = playerID | |
140 | ||
36de0a0e | 141 | playerKey = find_param('playerKey') |
cfe50f04 JMF |
142 | # Not all pages define this value |
143 | if playerKey is not None: | |
d214fdb8 | 144 | params['playerKey'] = playerKey |
36de0a0e JMF |
145 | # The three fields hold the id of the video |
146 | videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') | |
abb285fb | 147 | if videoPlayer is not None: |
d214fdb8 | 148 | params['@videoPlayer'] = videoPlayer |
36de0a0e | 149 | linkBase = find_param('linkBaseURL') |
dd5bcdc4 | 150 | if linkBase is not None: |
d214fdb8 | 151 | params['linkBaseURL'] = linkBase |
cfe50f04 JMF |
152 | data = compat_urllib_parse.urlencode(params) |
153 | return cls._FEDERATED_URL_TEMPLATE % data | |
fbaaad49 | 154 | |
eeb165e6 JMF |
155 | @classmethod |
156 | def _extract_brightcove_url(cls, webpage): | |
99877772 | 157 | """Try to extract the brightcove url from the webpage, returns None |
eeb165e6 JMF |
158 | if it can't be found |
159 | """ | |
99877772 PH |
160 | urls = cls._extract_brightcove_urls(webpage) |
161 | return urls[0] if urls else None | |
162 | ||
163 | @classmethod | |
164 | def _extract_brightcove_urls(cls, webpage): | |
165 | """Return a list of all Brightcove URLs from the webpage """ | |
117bec93 | 166 | |
e1ab5000 PH |
167 | url_m = re.search( |
168 | r'<meta\s+property="og:video"\s+content="(https?://(?:secure|c)\.brightcove.com/[^"]+)"', | |
169 | webpage) | |
117bec93 | 170 | if url_m: |
381640e3 JMF |
171 | url = unescapeHTML(url_m.group(1)) |
172 | # Some sites don't add it, we can't download with this url, for example: | |
173 | # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/ | |
e1ab5000 | 174 | if 'playerKey' in url or 'videoId' in url: |
381640e3 | 175 | return [url] |
117bec93 | 176 | |
99877772 | 177 | matches = re.findall( |
7b0817e8 PH |
178 | r'''(?sx)<object |
179 | (?: | |
99877772 | 180 | [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] | |
7b0817e8 PH |
181 | [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/ |
182 | ).+?</object>''', | |
183 | webpage) | |
99877772 | 184 | return [cls._build_brighcove_url(m) for m in matches] |
eeb165e6 | 185 | |
fbaaad49 | 186 | def _real_extract(self, url): |
79f82953 PH |
187 | url, smuggled_data = unsmuggle_url(url, {}) |
188 | ||
51040b72 JMF |
189 | # Change the 'videoId' and others field to '@videoPlayer' |
190 | url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url) | |
191 | # Change bckey (used by bcove.me urls) to playerKey | |
192 | url = re.sub(r'(?<=[?&])bckey', 'playerKey', url) | |
fbaaad49 | 193 | mobj = re.match(self._VALID_URL, url) |
6543f0dc JMF |
194 | query_str = mobj.group('query') |
195 | query = compat_urlparse.parse_qs(query_str) | |
fbaaad49 | 196 | |
6543f0dc JMF |
197 | videoPlayer = query.get('@videoPlayer') |
198 | if videoPlayer: | |
79f82953 PH |
199 | # We set the original url as the default 'Referer' header |
200 | referer = smuggled_data.get('Referer', url) | |
201 | return self._get_video_info( | |
202 | videoPlayer[0], query_str, query, referer=referer) | |
68575900 | 203 | elif 'playerKey' in query: |
6543f0dc JMF |
204 | player_key = query['playerKey'] |
205 | return self._get_playlist_info(player_key[0]) | |
68575900 PH |
206 | else: |
207 | raise ExtractorError( | |
208 | 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?', | |
209 | expected=True) | |
abb285fb | 210 | |
77526143 | 211 | def _get_video_info(self, video_id, query_str, query, referer=None): |
dd5bcdc4 JMF |
212 | request_url = self._FEDERATED_URL_TEMPLATE % query_str |
213 | req = compat_urllib_request.Request(request_url) | |
214 | linkBase = query.get('linkBaseURL') | |
215 | if linkBase is not None: | |
77526143 JMF |
216 | referer = linkBase[0] |
217 | if referer is not None: | |
218 | req.add_header('Referer', referer) | |
dd5bcdc4 | 219 | webpage = self._download_webpage(req, video_id) |
fbaaad49 | 220 | |
a5ed3e57 PH |
221 | error_msg = self._html_search_regex( |
222 | r"<h1>We're sorry.</h1>\s*<p>(.*?)</p>", webpage, | |
223 | 'error message', default=None) | |
224 | if error_msg is not None: | |
225 | raise ExtractorError( | |
226 | 'brightcove said: %s' % error_msg, expected=True) | |
227 | ||
fbaaad49 | 228 | self.report_extraction(video_id) |
a013eba6 | 229 | info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json') |
fbaaad49 JMF |
230 | info = json.loads(info)['data'] |
231 | video_info = info['programmedContent']['videoPlayer']['mediaDTO'] | |
7b0817e8 | 232 | video_info['_youtubedl_adServerURL'] = info.get('adServerURL') |
abb285fb JMF |
233 | |
234 | return self._extract_video_info(video_info) | |
235 | ||
236 | def _get_playlist_info(self, player_key): | |
117bec93 PH |
237 | info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key |
238 | playlist_info = self._download_webpage( | |
239 | info_url, player_key, 'Downloading playlist information') | |
abb285fb | 240 | |
59145479 PH |
241 | json_data = json.loads(playlist_info) |
242 | if 'videoList' not in json_data: | |
400e5810 | 243 | raise ExtractorError('Empty playlist') |
59145479 | 244 | playlist_info = json_data['videoList'] |
abb285fb JMF |
245 | videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']] |
246 | ||
247 | return self.playlist_result(videos, playlist_id=playlist_info['id'], | |
248 | playlist_title=playlist_info['mediaCollectionDTO']['displayName']) | |
249 | ||
250 | def _extract_video_info(self, video_info): | |
592882aa | 251 | info = { |
eeb165e6 | 252 | 'id': compat_str(video_info['id']), |
066f6a06 | 253 | 'title': video_info['displayName'].strip(), |
592882aa JMF |
254 | 'description': video_info.get('shortDescription'), |
255 | 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'), | |
256 | 'uploader': video_info.get('publisherName'), | |
257 | } | |
abb285fb | 258 | |
592882aa JMF |
259 | renditions = video_info.get('renditions') |
260 | if renditions: | |
5524b242 JMF |
261 | formats = [] |
262 | for rend in renditions: | |
263 | url = rend['defaultURL'] | |
b5af6fcd NJ |
264 | if not url: |
265 | continue | |
5524b242 | 266 | if rend['remote']: |
37aab278 NJ |
267 | url_comp = compat_urllib_parse_urlparse(url) |
268 | if url_comp.path.endswith('.m3u8'): | |
269 | formats.extend( | |
270 | self._extract_m3u8_formats(url, info['id'], 'mp4')) | |
271 | continue | |
272 | elif 'akamaihd.net' in url_comp.netloc: | |
273 | # This type of renditions are served through | |
274 | # akamaihd.net, but they don't use f4m manifests | |
275 | url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB' | |
276 | ext = 'flv' | |
5524b242 JMF |
277 | else: |
278 | ext = determine_ext(url) | |
35eacd0d | 279 | size = rend.get('size') |
5524b242 JMF |
280 | formats.append({ |
281 | 'url': url, | |
282 | 'ext': ext, | |
283 | 'height': rend.get('frameHeight'), | |
284 | 'width': rend.get('frameWidth'), | |
35eacd0d | 285 | 'filesize': size if size != 0 else None, |
5524b242 | 286 | }) |
35eacd0d | 287 | self._sort_formats(formats) |
5524b242 | 288 | info['formats'] = formats |
592882aa JMF |
289 | elif video_info.get('FLVFullLengthURL') is not None: |
290 | info.update({ | |
291 | 'url': video_info['FLVFullLengthURL'], | |
592882aa | 292 | }) |
7b0817e8 PH |
293 | |
294 | if self._downloader.params.get('include_ads', False): | |
295 | adServerURL = video_info.get('_youtubedl_adServerURL') | |
296 | if adServerURL: | |
297 | ad_info = { | |
298 | '_type': 'url', | |
299 | 'url': adServerURL, | |
300 | } | |
301 | if 'url' in info: | |
302 | return { | |
303 | '_type': 'playlist', | |
304 | 'title': info['title'], | |
305 | 'entries': [ad_info, info], | |
306 | } | |
307 | else: | |
308 | return ad_info | |
309 | ||
d614aa40 | 310 | if 'url' not in info and not info.get('formats'): |
400e5810 | 311 | raise ExtractorError('Unable to extract video url for %s' % info['id']) |
592882aa | 312 | return info |