[cnet] fix extraction and extract more formats

This commit is contained in:
remitamine 2015-09-23 13:28:05 +01:00
parent f102819463
commit b306c439d7

View File

@ -4,9 +4,7 @@ from __future__ import unicode_literals
import json import json
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from .theplatform import ThePlatformIE
ExtractorError,
)
class CNETIE(InfoExtractor): class CNETIE(InfoExtractor):
@ -15,29 +13,22 @@ class CNETIE(InfoExtractor):
'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/',
'info_dict': { 'info_dict': {
'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60', 'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
'ext': 'flv', '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.',
'thumbnail': 're:^http://.*/flmswindows8.jpg$',
'uploader_id': '6085384d-619e-11e3-b231-14feb5ca9861', 'uploader_id': '6085384d-619e-11e3-b231-14feb5ca9861',
'uploader': 'Sarah Mitroff', 'uploader': 'Sarah Mitroff',
}, },
'params': {
'skip_download': 'requires rtmpdump',
}
}, { }, {
'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/',
'info_dict': { 'info_dict': {
'id': '56527b93-d25d-44e3-b738-f989ce2e49ba', 'id': '56527b93-d25d-44e3-b738-f989ce2e49ba',
'ext': 'flv', 'ext': 'mp4',
'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': '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',
'uploader_id': 'b163284d-6b73-44fc-b3e6-3da66c392d40', 'uploader_id': 'b163284d-6b73-44fc-b3e6-3da66c392d40',
'uploader': 'Ashley Esqueda', 'uploader': 'Ashley Esqueda',
'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)',
}, },
'params': {
'skip_download': True, # requires rtmpdump
},
}] }]
def _real_extract(self, url): def _real_extract(self, url):
@ -45,26 +36,13 @@ class CNETIE(InfoExtractor):
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
data_json = self._html_search_regex( data_json = self._html_search_regex(
r"<div class=\"cnetVideoPlayer\"\s+.*?data-cnet-video-options='([^']+)'", r"<div class=\"videoPlayer\"\s+.*?data-cnet-video-uvp-options='([^']+)'",
webpage, 'data json') webpage, 'data json')
data = json.loads(data_json) data = json.loads(data_json)
vdata = data['video'] vdata = data['videos'][0]
if not vdata:
vdata = data['videos'][0]
if not vdata:
raise ExtractorError('Cannot find video data')
mpx_account = data['config']['players']['default']['mpx_account']
vid = vdata['files'].get('rtmp', vdata['files']['hds'])
tp_link = 'http://link.theplatform.com/s/%s/%s' % (mpx_account, vid)
video_id = vdata['id'] video_id = vdata['id']
title = vdata.get('headline') title = vdata['title']
if title is None:
title = vdata.get('title')
if title is None:
raise ExtractorError('Cannot find title!')
thumbnail = vdata.get('image', {}).get('path')
author = vdata.get('author') author = vdata.get('author')
if author: if author:
uploader = '%s %s' % (author['firstName'], author['lastName']) uploader = '%s %s' % (author['firstName'], author['lastName'])
@ -73,13 +51,27 @@ class CNETIE(InfoExtractor):
uploader = None uploader = None
uploader_id = None uploader_id = None
mpx_account = data['config']['uvpConfig']['default']['mpx_account']
tp = ThePlatformIE(self._downloader)
formats = []
subtitles = {}
description = vdata.get('description')
for vid in vdata['files'].values():
result = tp.extract(('http://link.theplatform.com/s/%s/%s' % (mpx_account, vid)))
formats.extend(result['formats'])
subtitles = self._merge_subtitles(subtitles, result['subtitles'])
description = description or result.get('description')
self._sort_formats(formats)
return { return {
'_type': 'url_transparent',
'url': tp_link,
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': title, 'title': title,
'description': description,
'uploader': uploader, 'uploader': uploader,
'uploader_id': uploader_id, 'uploader_id': uploader_id,
'thumbnail': thumbnail, 'subtitles': subtitles,
'formats': formats,
} }