class ThePlatformBaseIE(OnceIE):
def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
- meta = self._download_xml(smil_url, video_id, note=note)
- error_element = find_xpath_attr(
- meta, _x('.//smil:ref'), 'src',
- 'http://link.theplatform.com/s/errorFiles/Unavailable.mp4')
- if error_element is not None:
+ meta = self._download_xml(smil_url, video_id, note=note, query={'format': 'SMIL'})
+ error_element = find_xpath_attr(meta, _x('.//smil:ref'), 'src')
+ if error_element is not None and error_element.attrib['src'].startswith(
+ 'http://link.theplatform.com/s/errorFiles/Unavailable.'):
raise ExtractorError(error_element.attrib['abstract'], expected=True)
smil_formats = self._parse_smil_formats(
else:
formats.append(_format)
- self._sort_formats(formats)
-
subtitles = self._parse_smil_subtitles(meta, default_ns)
return formats, subtitles
'description': info['description'],
'thumbnail': info['defaultThumbnailUrl'],
'duration': int_or_none(info.get('duration'), 1000),
+ 'timestamp': int_or_none(info.get('pubDate'), 1000) or None,
+ 'uploader': info.get('billingCode'),
}
class ThePlatformIE(ThePlatformBaseIE):
_VALID_URL = r'''(?x)
(?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
- (?:(?P<media>(?:(?:[^/]+/)+select/)?media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
+ (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
|theplatform:)(?P<id>[^/\?&]+)'''
_TESTS = [{
'title': 'Blackberry\'s big, bold Z30',
'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
'duration': 247,
+ 'timestamp': 1383239700,
+ 'upload_date': '20131031',
+ 'uploader': 'CBSI-NEW',
},
'params': {
# rtmp download
'ext': 'flv',
'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
'title': 'Tesla Model S: A second step towards a cleaner motoring future',
+ 'timestamp': 1426176191,
+ 'upload_date': '20150312',
+ 'uploader': 'CBSI-NEW',
},
'params': {
# rtmp download
'ext': 'mp4',
'description': 'md5:644ad9188d655b742f942bf2e06b002d',
'title': 'HIGHLIGHTS: USA bag first ever series Cup win',
+ 'uploader': 'EGSM',
}
}, {
'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7',
'thumbnail': 're:^https?://.*\.jpg$',
'timestamp': 1435752600,
'upload_date': '20150701',
+ 'uploader': 'NBCU-NEWS',
},
}, {
# From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1
'only_matching': True,
}]
+ @classmethod
+ def _extract_urls(cls, webpage):
+ m = re.search(
+ r'''(?x)
+ <meta\s+
+ property=(["'])(?:og:video(?::(?:secure_)?url)?|twitter:player)\1\s+
+ content=(["'])(?P<url>https?://player\.theplatform\.com/p/.+?)\2
+ ''', webpage)
+ if m:
+ return [m.group('url')]
+
+ matches = re.findall(
+ r'<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
+ if matches:
+ return list(zip(*matches))[1]
+
@staticmethod
def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
flags = '10' if include_qs else '00'
def str_to_hex(str):
return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
- def hex_to_str(hex):
- return binascii.a2b_hex(hex)
+ def hex_to_bytes(hex):
+ return binascii.a2b_hex(hex.encode('ascii'))
- relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0]
- clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
+ relative_path = re.match(r'https?://link.theplatform.com/s/([^?]+)', url).group(1)
+ clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
return '%s&sig=%s' % (url, sig)
if not provider_id:
provider_id = 'dJ5BDC'
- path = provider_id
+ path = provider_id + '/'
if mobj.group('media'):
- path += '/media'
- path += '/' + video_id
+ path += mobj.group('media')
+ path += video_id
qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
if 'guid' in qs_dict:
webpage, 'smil url', group='url')
path = self._search_regex(
r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path')
- smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4&format=SMIL'
+ smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4'
elif mobj.group('config'):
config_url = url + '&form=json'
config_url = config_url.replace('swf/', 'config/')
release_url = config['releaseUrl']
else:
release_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
- smil_url = release_url + '&format=SMIL&formats=MPEG4&manifest=f4m'
+ smil_url = release_url + '&formats=MPEG4&manifest=f4m'
else:
- smil_url = 'http://link.theplatform.com/s/%s/meta.smil?format=smil&mbr=true' % path
+ smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
sig = smuggled_data.get('sig')
if sig:
smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
formats, subtitles = self._extract_theplatform_smil(smil_url, video_id)
+ self._sort_formats(formats)
ret = self.get_metadata(path, video_id)
combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles)
'timestamp': 1391824260,
'duration': 467.0,
'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
+ 'uploader': 'NBCU-NEWS',
},
}
first_video_id = None
duration = None
for item in entry['media$content']:
- smil_url = item['plfile$url'] + '&format=SMIL&mbr=true'
+ smil_url = item['plfile$url'] + '&mbr=true'
cur_video_id = ThePlatformIE._match_id(smil_url)
if first_video_id is None:
first_video_id = cur_video_id
projects / youtube-dl / blobdiff