[cbsinteractive] extract formats with CBSIE

This commit is contained in:
Tithen-Firion 2017-04-27 20:23:52 +02:00
parent e095109da1
commit 96820c1c6b
2 changed files with 22 additions and 53 deletions

View file

@ -49,13 +49,13 @@ class CBSIE(CBSBaseIE):
'only_matching': True, 'only_matching': True,
}] }]
def _extract_video_info(self, content_id): def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
items_data = self._download_xml( items_data = self._download_xml(
'http://can.cbs.com/thunder/player/videoPlayerService.php', 'http://can.cbs.com/thunder/player/videoPlayerService.php',
content_id, query={'partner': 'cbs', 'contentId': content_id}) content_id, query={'partner': site, 'contentId': content_id})
video_data = xpath_element(items_data, './/item') video_data = xpath_element(items_data, './/item')
title = xpath_text(video_data, 'videoTitle', 'title', True) title = xpath_text(video_data, 'videoTitle', 'title', True)
tp_path = 'dJ5BDC/media/guid/2198311517/%s' % content_id tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
tp_release_url = 'http://link.theplatform.com/s/' + tp_path tp_release_url = 'http://link.theplatform.com/s/' + tp_path
asset_types = [] asset_types = []

View file

@ -3,18 +3,17 @@
import re import re
from .theplatform import ThePlatformIE from .cbs import CBSIE
from ..compat import compat_urllib_parse
from ..utils import int_or_none from ..utils import int_or_none
class CBSInteractiveIE(ThePlatformIE): class CBSInteractiveIE(CBSIE):
_VALID_URL = r'https?://(?:www\.)?(?P<site>cnet|zdnet)\.com/(?:videos|video/share)/(?P<id>[^/?]+)' _VALID_URL = r'https?://(?:www\.)?(?P<site>cnet|zdnet)\.com/(?:videos|video/share)/(?P<id>[^/?]+)'
_TESTS = [{ _TESTS = [{
'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/', 'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/',
'md5': '041233212a0d06b179c87cbcca1577b8',
'info_dict': { 'info_dict': {
'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60', 'id': 'R49SYt__yAfmlXR85z4f7gNmCBDcN_00',
'display_id': 'hands-on-with-microsofts-windows-8-1-update',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Hands-on with Microsoft Windows 8.1 Update', 'title': 'Hands-on with Microsoft Windows 8.1 Update',
'description': 'The new update to the Windows 8 OS brings improved performance for mouse and keyboard users.', 'description': 'The new update to the Windows 8 OS brings improved performance for mouse and keyboard users.',
@ -25,45 +24,46 @@ class CBSInteractiveIE(ThePlatformIE):
'upload_date': '20140402', 'upload_date': '20140402',
}, },
'params': { 'params': {
'format': 'mp4', # m3u8 download
'skip_download': True,
}, },
}, { }, {
'url': 'http://www.cnet.com/videos/whiny-pothole-tweets-at-local-government-when-hit-by-cars-tomorrow-daily-187/', 'url': 'http://www.cnet.com/videos/whiny-pothole-tweets-at-local-government-when-hit-by-cars-tomorrow-daily-187/',
'md5': 'f2b16d73e08d69591dd9e25564695c0c', 'md5': 'f11d27b2fa18597fbf92444d2a9ed386',
'info_dict': { 'info_dict': {
'id': '56527b93-d25d-44e3-b738-f989ce2e49ba', 'id': 'kjOJd_OoVJqbg_ZD8MZCOk8Wekb9QccK',
'display_id': 'whiny-pothole-tweets-at-local-government-when-hit-by-cars-tomorrow-daily-187',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Whiny potholes tweet at local government when hit by cars (Tomorrow Daily 187)', 'title': 'Whiny potholes tweet at local government when hit by cars (Tomorrow Daily 187)',
'description': 'Khail and Ashley wonder what other civic woes can be solved by self-tweeting objects, investigate a new kind of VR camera and watch an origami robot self-assemble, walk, climb, dig and dissolve. #TDPothole', 'description': 'md5:d2b9a95a5ffe978ae6fbd4cf944d618f',
'uploader_id': 'b163284d-6b73-44fc-b3e6-3da66c392d40', 'uploader_id': 'b163284d-6b73-44fc-b3e6-3da66c392d40',
'uploader': 'Ashley Esqueda', 'uploader': 'Ashley Esqueda',
'duration': 1482, 'duration': 1482,
'timestamp': 1433289889, 'timestamp': 1433289889,
'upload_date': '20150603', 'upload_date': '20150603',
}, },
'params': {
'format': 'mp4',
},
}, { }, {
'url': 'http://www.zdnet.com/video/share/video-keeping-android-smartphones-and-tablets-secure/', 'url': 'http://www.zdnet.com/video/share/video-keeping-android-smartphones-and-tablets-secure/',
'info_dict': { 'info_dict': {
'id': 'bc1af9f0-a2b5-4e54-880d-0d95525781c0', 'id': 'k0r4T_ehht4xW_hAOqiVQPuBDPZ8SRjt',
'display_id': 'video-keeping-android-smartphones-and-tablets-secure',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Video: Keeping Android smartphones and tablets secure', 'title': 'Video: Keeping Android smartphones and tablets secure',
'description': 'Here\'s the best way to keep Android devices secure, and what you do when they\'ve come to the end of their lives.', 'description': 'Here\'s the best way to keep Android devices secure, and what you do when they\'ve come to the end of their lives.',
'uploader_id': 'f2d97ea2-8175-11e2-9d12-0018fe8a00b0', 'uploader_id': 'f2d97ea2-8175-11e2-9d12-0018fe8a00b0',
'uploader': 'Adrian Kingsley-Hughes', 'uploader': 'Adrian Kingsley-Hughes',
'duration': 731,
'timestamp': 1449129925, 'timestamp': 1449129925,
'upload_date': '20151203', 'upload_date': '20151203',
}, },
'params': { 'params': {
# m3u8 download # m3u8 download
'skip_download': True, 'skip_download': True,
} },
}] }]
TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/kYEXFC/%s?mbr=true'
MPX_ACCOUNTS = { MPX_ACCOUNTS = {
'cnet': 2288573011, 'cnet': 2198311517,
'zdnet': 2387448114, 'zdnet': 2387448114,
} }
@ -77,7 +77,8 @@ def _real_extract(self, url):
data = self._parse_json(data_json, display_id) data = self._parse_json(data_json, display_id)
vdata = data.get('video') or data['videos'][0] vdata = data.get('video') or data['videos'][0]
video_id = vdata['id'] video_id = vdata['mpxRefId']
title = vdata['title'] title = vdata['title']
author = vdata.get('author') author = vdata.get('author')
if author: if author:
@ -87,37 +88,7 @@ def _real_extract(self, url):
uploader = None uploader = None
uploader_id = None uploader_id = None
media_guid_path = 'media/guid/%d/%s' % (self.MPX_ACCOUNTS[site], vdata['mpxRefId']) info = self._extract_video_info(video_id, site, self.MPX_ACCOUNTS[site])
formats, subtitles = [], {}
for (fkey, vid) in vdata.get('files', {}).items():
if fkey == 'hls_phone' and 'hls_tablet' in vdata['files']:
continue
release_url = self.TP_RELEASE_URL_TEMPLATE % vid
if fkey == 'hds':
release_url += '&manifest=f4m'
tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % fkey)
formats.extend(tp_formats)
subtitles = self._merge_subtitles(subtitles, tp_subtitles)
if 'm3u8' in vdata:
parsed_url = compat_urllib_parse.urlparse(url)
m3u8_url = ('%s://%s%s'
% (parsed_url.scheme, parsed_url.netloc, vdata['m3u8']))
m3u8_formats = self._extract_m3u8_formats(m3u8_url, video_id)
for format in m3u8_formats:
format['url'] = format['url'].replace('https://', 'http://')
formats.extend(m3u8_formats)
if 'mp4' in vdata:
formats.append({
'url': vdata['mp4'],
'format_id': 'mp4',
'ext': 'mp4',
})
self._sort_formats(formats)
info = self._extract_theplatform_metadata('kYEXFC/%s' % media_guid_path, video_id)
info.update({ info.update({
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
@ -125,7 +96,5 @@ def _real_extract(self, url):
'duration': int_or_none(vdata.get('duration')), 'duration': int_or_none(vdata.get('duration')),
'uploader': uploader, 'uploader': uploader,
'uploader_id': uploader_id, 'uploader_id': uploader_id,
'subtitles': subtitles,
'formats': formats,
}) })
return info return info