1 from __future__ import unicode_literals
11 from .subtitles import SubtitlesInfoExtractor
12 from ..compat import (
22 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
25 class ThePlatformIE(SubtitlesInfoExtractor):
27 (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
28 (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
29 |theplatform:)(?P<id>[^/\?&]+)'''
32 # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
33 'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
37 'title': 'Blackberry\'s big, bold Z30',
38 'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
43 'skip_download': True,
48 def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
49 flags = '10' if include_qs else '00'
50 expiration_date = '%x' % (int(time.time()) + life)
53 return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
56 return binascii.a2b_hex(hex)
58 relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0]
59 clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
60 checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
61 sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
62 return '%s&sig=%s' % (url, sig)
64 def _real_extract(self, url):
65 url, smuggled_data = unsmuggle_url(url, {})
67 mobj = re.match(self._VALID_URL, url)
68 provider_id = mobj.group('provider_id')
69 video_id = mobj.group('id')
72 provider_id = 'dJ5BDC'
74 if mobj.group('config'):
75 config_url = url + '&form=json'
76 config_url = config_url.replace('swf/', 'config/')
77 config_url = config_url.replace('onsite/', 'onsite/config/')
78 config = self._download_json(config_url, video_id, 'Downloading config')
79 smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
81 smil_url = ('http://link.theplatform.com/s/{0}/{1}/meta.smil?'
82 'format=smil&mbr=true'.format(provider_id, video_id))
84 sig = smuggled_data.get('sig')
86 smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
88 meta = self._download_xml(smil_url, video_id)
92 for n in meta.findall(_x('.//smil:ref'))
93 if n.attrib.get('title') == 'Geographic Restriction')
97 raise ExtractorError(error_msg, expected=True)
99 info_url = 'http://link.theplatform.com/s/{0}/{1}?format=preview'.format(provider_id, video_id)
100 info_json = self._download_webpage(info_url, video_id)
101 info = json.loads(info_json)
104 captions = info.get('captions')
105 if isinstance(captions, list):
106 for caption in captions:
107 lang, src = caption.get('lang'), caption.get('src')
109 subtitles[lang] = src
111 if self._downloader.params.get('listsubtitles', False):
112 self._list_available_subtitles(video_id, subtitles)
115 subtitles = self.extract_subtitles(video_id, subtitles)
117 head = meta.find(_x('smil:head'))
118 body = meta.find(_x('smil:body'))
120 f4m_node = body.find(_x('smil:seq//smil:video'))
121 if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
122 f4m_url = f4m_node.attrib['src']
123 if 'manifest.f4m?' not in f4m_url:
125 # the parameters are from syfy.com, other sites may use others,
126 # they also work for nbc.com
127 f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
128 formats = self._extract_f4m_formats(f4m_url, video_id)
131 switch = body.find(_x('smil:switch'))
132 if switch is not None:
133 base_url = head.find(_x('smil:meta')).attrib['base']
134 for f in switch.findall(_x('smil:video')):
136 width = int(attr['width'])
137 height = int(attr['height'])
138 vbr = int(attr['system-bitrate']) // 1000
139 format_id = '%dx%d_%dk' % (width, height, vbr)
141 'format_id': format_id,
143 'play_path': 'mp4:' + attr['src'],
150 switch = body.find(_x('smil:seq//smil:switch'))
151 for f in switch.findall(_x('smil:video')):
153 vbr = int(attr['system-bitrate']) // 1000
154 ext = determine_ext(attr['src'])
158 'format_id': compat_str(vbr),
163 self._sort_formats(formats)
167 'title': info['title'],
168 'subtitles': subtitles,
170 'description': info['description'],
171 'thumbnail': info['defaultThumbnailUrl'],
172 'duration': info['duration'] // 1000,