2013-12-04 22:28:40 +00:00
|
|
|
import re
|
|
|
|
import json
|
|
|
|
|
|
|
|
from .common import InfoExtractor
|
|
|
|
from ..utils import (
|
2013-12-16 02:34:46 +00:00
|
|
|
ExtractorError,
|
2013-12-04 22:28:40 +00:00
|
|
|
xpath_with_ns,
|
|
|
|
)
|
|
|
|
|
|
|
|
_x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
|
|
|
|
|
|
|
|
|
|
|
|
class ThePlatformIE(InfoExtractor):
|
2013-12-23 15:41:35 +00:00
|
|
|
_VALID_URL = r'''(?x)
|
|
|
|
(?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
|
2014-02-25 22:57:54 +00:00
|
|
|
(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
|
2013-12-23 15:41:35 +00:00
|
|
|
|theplatform:)(?P<id>[^/\?&]+)'''
|
2013-12-04 22:28:40 +00:00
|
|
|
|
|
|
|
_TEST = {
|
|
|
|
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
|
|
|
|
u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
|
|
|
|
u'info_dict': {
|
|
|
|
u'id': u'e9I_cZgTgIPd',
|
|
|
|
u'ext': u'flv',
|
|
|
|
u'title': u'Blackberry\'s big, bold Z30',
|
|
|
|
u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
|
|
|
|
u'duration': 247,
|
|
|
|
},
|
|
|
|
u'params': {
|
|
|
|
# rtmp download
|
|
|
|
u'skip_download': True,
|
|
|
|
},
|
|
|
|
}
|
|
|
|
|
2013-12-23 15:41:35 +00:00
|
|
|
def _get_info(self, video_id, smil_url):
|
2013-12-04 22:28:40 +00:00
|
|
|
meta = self._download_xml(smil_url, video_id)
|
2013-12-16 02:34:46 +00:00
|
|
|
|
|
|
|
try:
|
|
|
|
error_msg = next(
|
|
|
|
n.attrib['abstract']
|
|
|
|
for n in meta.findall(_x('.//smil:ref'))
|
2013-12-16 02:43:45 +00:00
|
|
|
if n.attrib.get('title') == u'Geographic Restriction')
|
2013-12-16 02:34:46 +00:00
|
|
|
except StopIteration:
|
|
|
|
pass
|
|
|
|
else:
|
|
|
|
raise ExtractorError(error_msg, expected=True)
|
|
|
|
|
2013-12-04 22:28:40 +00:00
|
|
|
info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
|
|
|
|
info_json = self._download_webpage(info_url, video_id)
|
|
|
|
info = json.loads(info_json)
|
|
|
|
|
|
|
|
head = meta.find(_x('smil:head'))
|
|
|
|
body = meta.find(_x('smil:body'))
|
2013-12-26 20:18:18 +00:00
|
|
|
|
2013-12-23 15:41:35 +00:00
|
|
|
f4m_node = body.find(_x('smil:seq/smil:video'))
|
|
|
|
if f4m_node is not None:
|
2014-02-25 22:57:54 +00:00
|
|
|
f4m_url = f4m_node.attrib['src']
|
|
|
|
if 'manifest.f4m?' not in f4m_url:
|
|
|
|
f4m_url += '?'
|
|
|
|
# the parameters are from syfy.com, other sites may use others,
|
|
|
|
# they also work for nbc.com
|
|
|
|
f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
|
2013-12-23 15:41:35 +00:00
|
|
|
formats = [{
|
|
|
|
'ext': 'flv',
|
2014-02-25 22:57:54 +00:00
|
|
|
'url': f4m_url,
|
2013-12-23 15:41:35 +00:00
|
|
|
}]
|
|
|
|
else:
|
|
|
|
base_url = head.find(_x('smil:meta')).attrib['base']
|
|
|
|
switch = body.find(_x('smil:switch'))
|
|
|
|
formats = []
|
|
|
|
for f in switch.findall(_x('smil:video')):
|
|
|
|
attr = f.attrib
|
|
|
|
width = int(attr['width'])
|
|
|
|
height = int(attr['height'])
|
|
|
|
vbr = int(attr['system-bitrate']) // 1000
|
|
|
|
format_id = '%dx%d_%dk' % (width, height, vbr)
|
|
|
|
formats.append({
|
|
|
|
'format_id': format_id,
|
|
|
|
'url': base_url,
|
|
|
|
'play_path': 'mp4:' + attr['src'],
|
|
|
|
'ext': 'flv',
|
|
|
|
'width': width,
|
|
|
|
'height': height,
|
|
|
|
'vbr': vbr,
|
|
|
|
})
|
|
|
|
self._sort_formats(formats)
|
2013-12-04 22:28:40 +00:00
|
|
|
|
|
|
|
return {
|
|
|
|
'id': video_id,
|
|
|
|
'title': info['title'],
|
|
|
|
'formats': formats,
|
|
|
|
'description': info['description'],
|
|
|
|
'thumbnail': info['defaultThumbnailUrl'],
|
|
|
|
'duration': info['duration']//1000,
|
|
|
|
}
|
|
|
|
|
|
|
|
def _real_extract(self, url):
|
|
|
|
mobj = re.match(self._VALID_URL, url)
|
|
|
|
video_id = mobj.group('id')
|
2013-12-23 15:41:35 +00:00
|
|
|
if mobj.group('config'):
|
|
|
|
config_url = url+ '&form=json'
|
|
|
|
config_url = config_url.replace('swf/', 'config/')
|
2014-02-25 22:57:54 +00:00
|
|
|
config_url = config_url.replace('onsite/', 'onsite/config/')
|
2013-12-23 15:41:35 +00:00
|
|
|
config_json = self._download_webpage(config_url, video_id, u'Downloading config')
|
|
|
|
config = json.loads(config_json)
|
2014-02-25 22:57:54 +00:00
|
|
|
smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
|
2013-12-23 15:41:35 +00:00
|
|
|
else:
|
|
|
|
smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
|
|
|
|
'format=smil&mbr=true'.format(video_id))
|
|
|
|
return self._get_info(video_id, smil_url)
|