]>
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, |
592882aa JMF |
17 | |
18 | ExtractorError, | |
79f82953 | 19 | unsmuggle_url, |
ac6e4ca1 | 20 | unescapeHTML, |
cfe50f04 | 21 | ) |
fbaaad49 | 22 | |
dd5bcdc4 | 23 | |
fbaaad49 | 24 | class BrightcoveIE(InfoExtractor): |
abb285fb | 25 | _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)' |
cfe50f04 | 26 | _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s' |
592882aa JMF |
27 | |
28 | _TESTS = [ | |
29 | { | |
4de1994b | 30 | # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/ |
400e5810 PH |
31 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001', |
32 | 'file': '2371591881001.mp4', | |
33 | 'md5': '5423e113865d26e40624dce2e4b45d95', | |
34 | 'note': 'Test Brightcove downloads and detection in GenericIE', | |
35 | 'info_dict': { | |
36 | 'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', | |
37 | 'uploader': '8TV', | |
38 | 'description': 'md5:a950cc4285c43e44d763d036710cd9cd', | |
592882aa JMF |
39 | } |
40 | }, | |
41 | { | |
4de1994b | 42 | # From http://medianetwork.oracle.com/video/player/1785452137001 |
400e5810 PH |
43 | 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001', |
44 | 'file': '1785452137001.flv', | |
45 | 'info_dict': { | |
46 | 'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges', | |
47 | 'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.', | |
48 | 'uploader': 'Oracle', | |
592882aa JMF |
49 | }, |
50 | }, | |
fc4a0c2a JMF |
51 | { |
52 | # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/ | |
400e5810 PH |
53 | 'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001', |
54 | 'info_dict': { | |
55 | 'id': '2750934548001', | |
56 | 'ext': 'mp4', | |
57 | 'title': 'This Bracelet Acts as a Personal Thermostat', | |
58 | 'description': 'md5:547b78c64f4112766ccf4e151c20b6a0', | |
59 | 'uploader': 'Mashable', | |
fc4a0c2a JMF |
60 | }, |
61 | }, | |
77526143 JMF |
62 | { |
63 | # test that the default referer works | |
64 | # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/ | |
400e5810 PH |
65 | 'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001', |
66 | 'info_dict': { | |
67 | 'id': '2878862109001', | |
68 | 'ext': 'mp4', | |
69 | 'title': 'Lost in Motion II', | |
70 | 'description': 'md5:363109c02998fee92ec02211bd8000df', | |
71 | 'uploader': 'National Ballet of Canada', | |
77526143 | 72 | }, |
117bec93 | 73 | } |
592882aa | 74 | ] |
cfe50f04 JMF |
75 | |
76 | @classmethod | |
77 | def _build_brighcove_url(cls, object_str): | |
78 | """ | |
79 | Build a Brightcove url from a xml string containing | |
80 | <object class="BrightcoveExperience">{params}</object> | |
81 | """ | |
46e28a84 PH |
82 | |
83 | # Fix up some stupid HTML, see https://github.com/rg3/youtube-dl/issues/1553 | |
84 | object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>', | |
85 | lambda m: m.group(1) + '/>', object_str) | |
2d0efe70 | 86 | # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608 |
400e5810 | 87 | object_str = object_str.replace('<--', '<!--') |
7b0817e8 | 88 | object_str = fix_xml_ampersands(object_str) |
46e28a84 | 89 | |
0479c625 | 90 | object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8')) |
7b0817e8 PH |
91 | |
92 | fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars') | |
47917f24 JMF |
93 | if fv_el is not None: |
94 | flashvars = dict( | |
95 | (k, v[0]) | |
96 | for k, v in compat_parse_qs(fv_el.attrib['value']).items()) | |
97 | else: | |
98 | flashvars = {} | |
7b0817e8 | 99 | |
36de0a0e | 100 | def find_param(name): |
7b0817e8 PH |
101 | if name in flashvars: |
102 | return flashvars[name] | |
d214fdb8 JMF |
103 | node = find_xpath_attr(object_doc, './param', 'name', name) |
104 | if node is not None: | |
105 | return node.attrib['value'] | |
106 | return None | |
7b0817e8 PH |
107 | |
108 | params = {} | |
109 | ||
110 | playerID = find_param('playerID') | |
111 | if playerID is None: | |
112 | raise ExtractorError('Cannot find player ID') | |
113 | params['playerID'] = playerID | |
114 | ||
36de0a0e | 115 | playerKey = find_param('playerKey') |
cfe50f04 JMF |
116 | # Not all pages define this value |
117 | if playerKey is not None: | |
d214fdb8 | 118 | params['playerKey'] = playerKey |
36de0a0e JMF |
119 | # The three fields hold the id of the video |
120 | videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') | |
abb285fb | 121 | if videoPlayer is not None: |
d214fdb8 | 122 | params['@videoPlayer'] = videoPlayer |
36de0a0e | 123 | linkBase = find_param('linkBaseURL') |
dd5bcdc4 | 124 | if linkBase is not None: |
d214fdb8 | 125 | params['linkBaseURL'] = linkBase |
cfe50f04 JMF |
126 | data = compat_urllib_parse.urlencode(params) |
127 | return cls._FEDERATED_URL_TEMPLATE % data | |
fbaaad49 | 128 | |
eeb165e6 JMF |
129 | @classmethod |
130 | def _extract_brightcove_url(cls, webpage): | |
99877772 | 131 | """Try to extract the brightcove url from the webpage, returns None |
eeb165e6 JMF |
132 | if it can't be found |
133 | """ | |
99877772 PH |
134 | urls = cls._extract_brightcove_urls(webpage) |
135 | return urls[0] if urls else None | |
136 | ||
137 | @classmethod | |
138 | def _extract_brightcove_urls(cls, webpage): | |
139 | """Return a list of all Brightcove URLs from the webpage """ | |
117bec93 PH |
140 | |
141 | url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage) | |
142 | if url_m: | |
381640e3 JMF |
143 | url = unescapeHTML(url_m.group(1)) |
144 | # Some sites don't add it, we can't download with this url, for example: | |
145 | # http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/ | |
146 | if 'playerKey' in url: | |
147 | return [url] | |
117bec93 | 148 | |
99877772 | 149 | matches = re.findall( |
7b0817e8 PH |
150 | r'''(?sx)<object |
151 | (?: | |
99877772 | 152 | [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] | |
7b0817e8 PH |
153 | [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/ |
154 | ).+?</object>''', | |
155 | webpage) | |
99877772 | 156 | return [cls._build_brighcove_url(m) for m in matches] |
eeb165e6 | 157 | |
fbaaad49 | 158 | def _real_extract(self, url): |
79f82953 PH |
159 | url, smuggled_data = unsmuggle_url(url, {}) |
160 | ||
51040b72 JMF |
161 | # Change the 'videoId' and others field to '@videoPlayer' |
162 | url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url) | |
163 | # Change bckey (used by bcove.me urls) to playerKey | |
164 | url = re.sub(r'(?<=[?&])bckey', 'playerKey', url) | |
fbaaad49 | 165 | mobj = re.match(self._VALID_URL, url) |
6543f0dc JMF |
166 | query_str = mobj.group('query') |
167 | query = compat_urlparse.parse_qs(query_str) | |
fbaaad49 | 168 | |
6543f0dc JMF |
169 | videoPlayer = query.get('@videoPlayer') |
170 | if videoPlayer: | |
79f82953 PH |
171 | # We set the original url as the default 'Referer' header |
172 | referer = smuggled_data.get('Referer', url) | |
173 | return self._get_video_info( | |
174 | videoPlayer[0], query_str, query, referer=referer) | |
abb285fb | 175 | else: |
6543f0dc JMF |
176 | player_key = query['playerKey'] |
177 | return self._get_playlist_info(player_key[0]) | |
abb285fb | 178 | |
77526143 | 179 | def _get_video_info(self, video_id, query_str, query, referer=None): |
dd5bcdc4 JMF |
180 | request_url = self._FEDERATED_URL_TEMPLATE % query_str |
181 | req = compat_urllib_request.Request(request_url) | |
182 | linkBase = query.get('linkBaseURL') | |
183 | if linkBase is not None: | |
77526143 JMF |
184 | referer = linkBase[0] |
185 | if referer is not None: | |
186 | req.add_header('Referer', referer) | |
dd5bcdc4 | 187 | webpage = self._download_webpage(req, video_id) |
fbaaad49 JMF |
188 | |
189 | self.report_extraction(video_id) | |
190 | info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json') | |
191 | info = json.loads(info)['data'] | |
192 | video_info = info['programmedContent']['videoPlayer']['mediaDTO'] | |
7b0817e8 | 193 | video_info['_youtubedl_adServerURL'] = info.get('adServerURL') |
abb285fb JMF |
194 | |
195 | return self._extract_video_info(video_info) | |
196 | ||
197 | def _get_playlist_info(self, player_key): | |
117bec93 PH |
198 | info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key |
199 | playlist_info = self._download_webpage( | |
200 | info_url, player_key, 'Downloading playlist information') | |
abb285fb | 201 | |
59145479 PH |
202 | json_data = json.loads(playlist_info) |
203 | if 'videoList' not in json_data: | |
400e5810 | 204 | raise ExtractorError('Empty playlist') |
59145479 | 205 | playlist_info = json_data['videoList'] |
abb285fb JMF |
206 | videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']] |
207 | ||
208 | return self.playlist_result(videos, playlist_id=playlist_info['id'], | |
209 | playlist_title=playlist_info['mediaCollectionDTO']['displayName']) | |
210 | ||
211 | def _extract_video_info(self, video_info): | |
592882aa | 212 | info = { |
eeb165e6 | 213 | 'id': compat_str(video_info['id']), |
066f6a06 | 214 | 'title': video_info['displayName'].strip(), |
592882aa JMF |
215 | 'description': video_info.get('shortDescription'), |
216 | 'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'), | |
217 | 'uploader': video_info.get('publisherName'), | |
218 | } | |
abb285fb | 219 | |
592882aa JMF |
220 | renditions = video_info.get('renditions') |
221 | if renditions: | |
222 | renditions = sorted(renditions, key=lambda r: r['size']) | |
b0759f0c JMF |
223 | info['formats'] = [{ |
224 | 'url': rend['defaultURL'], | |
225 | 'height': rend.get('frameHeight'), | |
226 | 'width': rend.get('frameWidth'), | |
227 | } for rend in renditions] | |
592882aa JMF |
228 | elif video_info.get('FLVFullLengthURL') is not None: |
229 | info.update({ | |
230 | 'url': video_info['FLVFullLengthURL'], | |
592882aa | 231 | }) |
7b0817e8 PH |
232 | |
233 | if self._downloader.params.get('include_ads', False): | |
234 | adServerURL = video_info.get('_youtubedl_adServerURL') | |
235 | if adServerURL: | |
236 | ad_info = { | |
237 | '_type': 'url', | |
238 | 'url': adServerURL, | |
239 | } | |
240 | if 'url' in info: | |
241 | return { | |
242 | '_type': 'playlist', | |
243 | 'title': info['title'], | |
244 | 'entries': [ad_info, info], | |
245 | } | |
246 | else: | |
247 | return ad_info | |
248 | ||
d614aa40 | 249 | if 'url' not in info and not info.get('formats'): |
400e5810 | 250 | raise ExtractorError('Unable to extract video url for %s' % info['id']) |
592882aa | 251 | return info |