2 from __future__
import unicode_literals
11 from .once
import OnceIE
12 from .adobepass
import AdobePassIE
13 from ..compat
import (
15 compat_urllib_parse_urlparse
,
30 default_ns
= 'http://www.w3.org/2005/SMIL21/Language'
31 _x
= lambda p
: xpath_with_ns(p
, {'smil': default_ns}
)
34 class ThePlatformBaseIE(OnceIE
):
35 def _extract_theplatform_smil(self
, smil_url
, video_id
, note
='Downloading SMIL data'):
36 meta
= self
._download
_xml
(
37 smil_url
, video_id
, note
=note
, query
={'format': 'SMIL'}
,
38 headers
=self
.geo_verification_headers())
39 error_element
= find_xpath_attr(meta
, _x('.//smil:ref'), 'src')
40 if error_element
is not None and error_element
.attrib
['src'].startswith(
41 'http://link.theplatform.com/s/errorFiles/Unavailable.'):
42 raise ExtractorError(error_element
.attrib
['abstract'], expected
=True)
44 smil_formats
= self
._parse
_smil
_formats
(
45 meta
, smil_url
, video_id
, namespace
=default_ns
,
46 # the parameters are from syfy.com, other sites may use others,
47 # they also work for nbc.com
48 f4m_params
={'g': 'UXWGVKRWHFSP', 'hdcore': '3.0.3'}
,
49 transform_rtmp_url
=lambda streamer
, src
: (streamer
, 'mp4:' + src
))
52 for _format
in smil_formats
:
53 if OnceIE
.suitable(_format
['url']):
54 formats
.extend(self
._extract
_once
_formats
(_format
['url']))
56 media_url
= _format
['url']
57 if determine_ext(media_url
) == 'm3u8':
58 hdnea2
= self
._get
_cookies
(media_url
).get('hdnea2')
60 _format
['url'] = update_url_query(media_url
, {'hdnea3': hdnea2.value}
)
62 formats
.append(_format
)
64 subtitles
= self
._parse
_smil
_subtitles
(meta
, default_ns
)
66 return formats
, subtitles
68 def _download_theplatform_metadata(self
, path
, video_id
):
69 info_url
= 'http://link.theplatform.com/s/%s?format=preview' % path
70 return self
._download
_json
(info_url
, video_id
)
72 def _parse_theplatform_metadata(self
, info
):
74 captions
= info
.get('captions')
75 if isinstance(captions
, list):
76 for caption
in captions
:
77 lang
, src
, mime
= caption
.get('lang', 'en'), caption
.get('src'), caption
.get('type')
78 subtitles
.setdefault(lang
, []).append({
79 'ext': mimetype2ext(mime
),
84 'title': info
['title'],
85 'subtitles': subtitles
,
86 'description': info
['description'],
87 'thumbnail': info
['defaultThumbnailUrl'],
88 'duration': int_or_none(info
.get('duration'), 1000),
89 'timestamp': int_or_none(info
.get('pubDate'), 1000) or None,
90 'uploader': info
.get('billingCode'),
93 def _extract_theplatform_metadata(self
, path
, video_id
):
94 info
= self
._download
_theplatform
_metadata
(path
, video_id
)
95 return self
._parse
_theplatform
_metadata
(info
)
98 class ThePlatformIE(ThePlatformBaseIE
, AdobePassIE
):
100 (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
101 (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)?|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
102 |theplatform:)(?P<id>[^/\?&]+)'''
105 # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
106 'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
108 'id': 'e9I_cZgTgIPd',
110 'title': 'Blackberry\'s big, bold Z30',
111 'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
113 'timestamp': 1383239700,
114 'upload_date': '20131031',
115 'uploader': 'CBSI-NEW',
119 'skip_download': True,
121 'skip': '404 Not Found',
123 # from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/
124 'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT',
126 'id': '22d_qsQ6MIRT',
128 'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
129 'title': 'Tesla Model S: A second step towards a cleaner motoring future',
130 'timestamp': 1426176191,
131 'upload_date': '20150312',
132 'uploader': 'CBSI-NEW',
136 'skip_download': True,
139 'url': 'https://player.theplatform.com/p/D6x-PC/pulse_preview/embed/select/media/yMBg9E8KFxZD',
141 'id': 'yMBg9E8KFxZD',
143 'description': 'md5:644ad9188d655b742f942bf2e06b002d',
144 'title': 'HIGHLIGHTS: USA bag first ever series Cup win',
148 'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7',
149 'only_matching': True,
151 'url': 'http://player.theplatform.com/p/2E2eJC/nbcNewsOffsite?guid=tdy_or_siri_150701',
152 'md5': 'fb96bb3d85118930a5b055783a3bd992',
154 'id': 'tdy_or_siri_150701',
156 'title': 'iPhone Siri’s sassy response to a math question has people talking',
157 'description': 'md5:a565d1deadd5086f3331d57298ec6333',
159 'thumbnail': r
're:^https?://.*\.jpg$',
160 'timestamp': 1435752600,
161 'upload_date': '20150701',
162 'uploader': 'NBCU-NEWS',
165 # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1
166 # geo-restricted (US), HLS encrypted with AES-128
167 'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781',
168 'only_matching': True,
172 def _extract_urls(cls
, webpage
):
176 property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+
177 content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2
180 return [m
.group('url')]
182 # Are whitesapces ignored in URLs?
183 # https://github.com/rg3/youtube-dl/issues/12044
184 matches
= re
.findall(
185 r
'(?s)<(?:iframe|script)[^>]+src=(["\'])((?
:https?
:)?
//player\
.theplatform\
.com
/p
/.+?
)\
1', webpage)
187 return [re.sub(r'\s
', '', list(zip(*matches))[1][0])]
190 def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
191 flags = '10' if include_qs else '00'
192 expiration_date = '%x' % (int(time.time()) + life)
195 return binascii.b2a_hex(str.encode('ascii
')).decode('ascii
')
197 def hex_to_bytes(hex):
198 return binascii.a2b_hex(hex.encode('ascii
'))
200 relative_path = re.match(r'https?
://link
.theplatform
.com
/s
/([^?
]+)', url).group(1)
201 clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
202 checksum = hmac.new(sig_key.encode('ascii
'), clear_text, hashlib.sha1).hexdigest()
203 sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
204 return '%s&sig
=%s' % (url, sig)
206 def _real_extract(self, url):
207 url, smuggled_data = unsmuggle_url(url, {})
209 mobj = re.match(self._VALID_URL, url)
210 provider_id = mobj.group('provider_id
')
211 video_id = mobj.group('id')
214 provider_id = 'dJ5BDC
'
216 path = provider_id + '/'
217 if mobj.group('media
'):
218 path += mobj.group('media
')
221 qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
222 if 'guid
' in qs_dict:
223 webpage = self._download_webpage(url, video_id)
224 scripts = re.findall(r'<script
[^
>]+src
="([^"]+)"', webpage)
226 # feed id usually locates in the last script.
227 # Seems there's no pattern for the interested script filename, so
229 for script in reversed(scripts):
230 feed_script = self._download_webpage(
231 self._proto_relative_url(script, 'http:'),
232 video_id, 'Downloading feed script')
233 feed_id = self._search_regex(
234 r'defaultFeedId\s*:\s*"([^
"]+)"', feed_script,
235 'default feed
id', default=None)
236 if feed_id is not None:
239 raise ExtractorError('Unable to find feed
id')
240 return self.url_result('http
://feed
.theplatform
.com
/f
/%s/%s?byGuid
=%s' % (
241 provider_id, feed_id, qs_dict['guid
'][0]))
243 if smuggled_data.get('force_smil_url
', False):
245 # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385)
246 elif '/guid
/' in url:
248 source_url = smuggled_data.get('source_url
')
250 headers['Referer
'] = source_url
251 request = sanitized_Request(url, headers=headers)
252 webpage = self._download_webpage(request, video_id)
253 smil_url = self._search_regex(
254 r'<link
[^
>]+href
=(["\'])(?P<url>.+?)\1[^>]+type=["\']application
/smil\
+xml
',
255 webpage, 'smil url
', group='url
')
256 path = self._search_regex(
257 r'link\
.theplatform\
.com
/s
/((?
:[^
/?
#&]+/)+[^/?#&]+)', smil_url, 'path')
258 smil_url
+= '?' if '?' not in smil_url
else '&' + 'formats=m3u,mpeg4'
259 elif mobj
.group('config'):
260 config_url
= url
+ '&form=json'
261 config_url
= config_url
.replace('swf/', 'config/')
262 config_url
= config_url
.replace('onsite/', 'onsite/config/')
263 config
= self
._download
_json
(config_url
, video_id
, 'Downloading config')
264 if 'releaseUrl' in config
:
265 release_url
= config
['releaseUrl']
267 release_url
= 'http://link.theplatform.com/s/%s?mbr=true' % path
268 smil_url
= release_url
+ '&formats=MPEG4&manifest=f4m'
270 smil_url
= 'http://link.theplatform.com/s/%s?mbr=true' % path
272 sig
= smuggled_data
.get('sig')
274 smil_url
= self
._sign
_url
(smil_url
, sig
['key'], sig
['secret'])
276 formats
, subtitles
= self
._extract
_theplatform
_smil
(smil_url
, video_id
)
277 self
._sort
_formats
(formats
)
279 ret
= self
._extract
_theplatform
_metadata
(path
, video_id
)
280 combined_subtitles
= self
._merge
_subtitles
(ret
.get('subtitles', {}), subtitles
)
284 'subtitles': combined_subtitles
,
290 class ThePlatformFeedIE(ThePlatformBaseIE
):
291 _URL_TEMPLATE
= '%s//feed.theplatform.com/f/%s/%s?form=json&%s'
292 _VALID_URL
= r
'https?://feed\.theplatform\.com/f/(?P<provider_id>[^/]+)/(?P<feed_id>[^?/]+)\?(?:[^&]+&)*(?P<filter>by(?:Gui|I)d=(?P<id>[\w-]+))'
294 # From http://player.theplatform.com/p/7wvmTC/MSNBCEmbeddedOffSite?guid=n_hardball_5biden_140207
295 'url': 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207',
296 'md5': '6e32495b5073ab414471b615c5ded394',
298 'id': 'n_hardball_5biden_140207',
300 'title': 'The Biden factor: will Joe run in 2016?',
301 'description': 'Could Vice President Joe Biden be preparing a 2016 campaign? Mark Halperin and Sam Stein weigh in.',
302 'thumbnail': r
're:^https?://.*\.jpg$',
303 'upload_date': '20140208',
304 'timestamp': 1391824260,
306 'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
307 'uploader': 'NBCU-NEWS',
311 def _extract_feed_info(self
, provider_id
, feed_id
, filter_query
, video_id
, custom_fields
=None, asset_types_query
={}, account_id
=None):
312 real_url
= self
._URL
_TEMPLATE
% (self
.http_scheme(), provider_id
, feed_id
, filter_query
)
313 entry
= self
._download
_json
(real_url
, video_id
)['entries'][0]
314 main_smil_url
= 'http://link.theplatform.com/s/%s/media/guid/%d/%s' % (provider_id
, account_id
, entry
['guid']) if account_id
else None
318 first_video_id
= None
321 for item
in entry
['media$content']:
322 smil_url
= item
['plfile$url']
323 cur_video_id
= ThePlatformIE
._match
_id
(smil_url
)
324 if first_video_id
is None:
325 first_video_id
= cur_video_id
326 duration
= float_or_none(item
.get('plfile$duration'))
327 for asset_type
in item
['plfile$assetTypes']:
328 if asset_type
in asset_types
:
330 asset_types
.append(asset_type
)
333 'formats': item
['plfile$format'],
334 'assetTypes': asset_type
,
336 if asset_type
in asset_types_query
:
337 query
.update(asset_types_query
[asset_type
])
338 cur_formats
, cur_subtitles
= self
._extract
_theplatform
_smil
(update_url_query(
339 main_smil_url
or smil_url
, query
), video_id
, 'Downloading SMIL data for %s' % asset_type
)
340 formats
.extend(cur_formats
)
341 subtitles
= self
._merge
_subtitles
(subtitles
, cur_subtitles
)
343 self
._sort
_formats
(formats
)
346 'url': thumbnail
['plfile$url'],
347 'width': int_or_none(thumbnail
.get('plfile$width')),
348 'height': int_or_none(thumbnail
.get('plfile$height')),
349 } for thumbnail
in entry
.get('media$thumbnails', [])]
351 timestamp
= int_or_none(entry
.get('media$availableDate'), scale
=1000)
352 categories
= [item
['media$name'] for item
in entry
.get('media$categories', [])]
354 ret
= self
._extract
_theplatform
_metadata
('%s/%s' % (provider_id
, first_video_id
), video_id
)
355 subtitles
= self
._merge
_subtitles
(subtitles
, ret
['subtitles'])
359 'subtitles': subtitles
,
360 'thumbnails': thumbnails
,
361 'duration': duration
,
362 'timestamp': timestamp
,
363 'categories': categories
,
366 ret
.update(custom_fields(entry
))
370 def _real_extract(self
, url
):
371 mobj
= re
.match(self
._VALID
_URL
, url
)
373 video_id
= mobj
.group('id')
374 provider_id
= mobj
.group('provider_id')
375 feed_id
= mobj
.group('feed_id')
376 filter_query
= mobj
.group('filter')
378 return self
._extract
_feed
_info
(provider_id
, feed_id
, filter_query
, video_id
)