youtube-dl/youtube_dl/extractor/theplatform.py

169 lines
6.0 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
2013-12-04 22:28:40 +00:00
import re
import json
2015-02-14 22:56:12 +00:00
import time
import hmac
import binascii
import hashlib
2013-12-04 22:28:40 +00:00
from .common import InfoExtractor
from ..utils import (
2014-10-27 00:14:17 +00:00
determine_ext,
ExtractorError,
2013-12-04 22:28:40 +00:00
xpath_with_ns,
2015-02-14 22:56:12 +00:00
unsmuggle_url,
2015-04-08 16:29:10 +00:00
int_or_none,
2013-12-04 22:28:40 +00:00
)
default_ns = 'http://www.w3.org/2005/SMIL21/Language'
_x = lambda p: xpath_with_ns(p, {'smil': default_ns})
2013-12-04 22:28:40 +00:00
class ThePlatformIE(InfoExtractor):
_VALID_URL = r'''(?x)
2015-02-14 22:56:12 +00:00
(?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
(?:(?P<media>(?:[^/]+/)+select/media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
|theplatform:)(?P<id>[^/\?&]+)'''
2013-12-04 22:28:40 +00:00
_TESTS = [{
2013-12-04 22:28:40 +00:00
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
'info_dict': {
'id': 'e9I_cZgTgIPd',
'ext': 'flv',
'title': 'Blackberry\'s big, bold Z30',
'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
'duration': 247,
2013-12-04 22:28:40 +00:00
},
'params': {
2013-12-04 22:28:40 +00:00
# rtmp download
'skip_download': True,
2013-12-04 22:28:40 +00:00
},
}, {
# from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/
'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT',
'info_dict': {
'id': '22d_qsQ6MIRT',
'ext': 'flv',
'description': 'md5:ac330c9258c04f9d7512cf26b9595409',
'title': 'Tesla Model S: A second step towards a cleaner motoring future',
},
'params': {
# rtmp download
'skip_download': True,
}
}, {
'url': 'https://player.theplatform.com/p/D6x-PC/pulse_preview/embed/select/media/yMBg9E8KFxZD',
'info_dict': {
'id': 'yMBg9E8KFxZD',
'ext': 'mp4',
'description': 'md5:644ad9188d655b742f942bf2e06b002d',
'title': 'HIGHLIGHTS: USA bag first ever series Cup win',
}
}, {
'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7',
'only_matching': True,
}]
2014-11-23 19:41:03 +00:00
2015-02-14 22:56:12 +00:00
@staticmethod
def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
flags = '10' if include_qs else '00'
expiration_date = '%x' % (int(time.time()) + life)
def str_to_hex(str):
return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
def hex_to_str(hex):
return binascii.a2b_hex(hex)
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))
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)
2014-10-27 00:14:17 +00:00
def _real_extract(self, url):
2015-02-14 22:56:12 +00:00
url, smuggled_data = unsmuggle_url(url, {})
2014-10-27 00:14:17 +00:00
mobj = re.match(self._VALID_URL, url)
2015-02-14 22:56:12 +00:00
provider_id = mobj.group('provider_id')
2014-10-27 00:14:17 +00:00
video_id = mobj.group('id')
2015-02-14 22:56:12 +00:00
if not provider_id:
provider_id = 'dJ5BDC'
path = provider_id
if mobj.group('media'):
path += '/media'
path += '/' + video_id
if smuggled_data.get('force_smil_url', False):
smil_url = url
elif mobj.group('config'):
2014-11-23 19:41:03 +00:00
config_url = url + '&form=json'
2014-10-27 00:14:17 +00:00
config_url = config_url.replace('swf/', 'config/')
config_url = config_url.replace('onsite/', 'onsite/config/')
config = self._download_json(config_url, video_id, 'Downloading config')
if 'releaseUrl' in 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'
2014-10-27 00:14:17 +00:00
else:
smil_url = 'http://link.theplatform.com/s/%s/meta.smil?format=smil&mbr=true' % path
2015-02-14 22:56:12 +00:00
sig = smuggled_data.get('sig')
if sig:
smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])
2013-12-04 22:28:40 +00:00
2014-10-27 00:14:17 +00:00
meta = self._download_xml(smil_url, video_id)
try:
error_msg = next(
n.attrib['abstract']
for n in meta.findall(_x('.//smil:ref'))
if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired')
except StopIteration:
pass
else:
raise ExtractorError(error_msg, expected=True)
info_url = 'http://link.theplatform.com/s/%s?format=preview' % path
2013-12-04 22:28:40 +00:00
info_json = self._download_webpage(info_url, video_id)
info = json.loads(info_json)
subtitles = {}
captions = info.get('captions')
if isinstance(captions, list):
for caption in captions:
lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
subtitles[lang] = [{
'ext': 'srt' if mime == 'text/srt' else 'ttml',
'url': src,
}]
formats = self._parse_smil_formats(
meta, smil_url, video_id, namespace=default_ns,
# the parameters are from syfy.com, other sites may use others,
# they also work for nbc.com
f4m_params={'g': 'UXWGVKRWHFSP', 'hdcore': '3.0.3'},
transform_rtmp_url=lambda streamer, src: (streamer, 'mp4:' + src))
for _format in formats:
ext = determine_ext(_format['url'])
if ext == 'once':
_format['ext'] = 'mp4'
self._sort_formats(formats)
2013-12-04 22:28:40 +00:00
return {
'id': video_id,
'title': info['title'],
'subtitles': subtitles,
2013-12-04 22:28:40 +00:00
'formats': formats,
'description': info['description'],
'thumbnail': info['defaultThumbnailUrl'],
2015-04-08 16:29:10 +00:00
'duration': int_or_none(info.get('duration'), 1000),
2013-12-04 22:28:40 +00:00
}