commit 9e35298f973b25c513e30bbc1fe2a05a1c6b7fab
parent 0551f1b07b41bf64b4b8c1150abec2903e37a81f
Author: remitamine <remitamine@gmail.com>
Date: Wed, 24 May 2017 10:21:33 +0200
Merge pull request #12861 from Tithen-Firion/cbsinteractive-fix
[cbsinteractive] update extractor and test cases
Diffstat:
2 files changed, 28 insertions(+), 33 deletions(-)
diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py
@@ -49,13 +49,13 @@ class CBSIE(CBSBaseIE):
'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(
'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')
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
asset_types = []
diff --git a/youtube_dl/extractor/cbsinteractive.py b/youtube_dl/extractor/cbsinteractive.py
@@ -3,17 +3,18 @@ from __future__ import unicode_literals
import re
-from .theplatform import ThePlatformIE
+from .cbs import CBSIE
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>[^/?]+)'
_TESTS = [{
'url': 'http://www.cnet.com/videos/hands-on-with-microsofts-windows-8-1-update/',
'info_dict': {
- 'id': '56f4ea68-bd21-4852-b08c-4de5b8354c60',
- 'ext': 'flv',
+ 'id': 'R49SYt__yAfmlXR85z4f7gNmCBDcN_00',
+ 'display_id': 'hands-on-with-microsofts-windows-8-1-update',
+ 'ext': 'mp4',
'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.',
'uploader_id': '6085384d-619e-11e3-b231-14feb5ca9861',
@@ -22,13 +23,19 @@ class CBSInteractiveIE(ThePlatformIE):
'timestamp': 1396479627,
'upload_date': '20140402',
},
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ },
}, {
'url': 'http://www.cnet.com/videos/whiny-pothole-tweets-at-local-government-when-hit-by-cars-tomorrow-daily-187/',
+ 'md5': 'f11d27b2fa18597fbf92444d2a9ed386',
'info_dict': {
- 'id': '56527b93-d25d-44e3-b738-f989ce2e49ba',
- 'ext': 'flv',
+ 'id': 'kjOJd_OoVJqbg_ZD8MZCOk8Wekb9QccK',
+ 'display_id': 'whiny-pothole-tweets-at-local-government-when-hit-by-cars-tomorrow-daily-187',
+ 'ext': 'mp4',
'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': 'Ashley Esqueda',
'duration': 1482,
@@ -38,23 +45,25 @@ class CBSInteractiveIE(ThePlatformIE):
}, {
'url': 'http://www.zdnet.com/video/share/video-keeping-android-smartphones-and-tablets-secure/',
'info_dict': {
- 'id': 'bc1af9f0-a2b5-4e54-880d-0d95525781c0',
+ 'id': 'k0r4T_ehht4xW_hAOqiVQPuBDPZ8SRjt',
+ 'display_id': 'video-keeping-android-smartphones-and-tablets-secure',
'ext': 'mp4',
'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.',
'uploader_id': 'f2d97ea2-8175-11e2-9d12-0018fe8a00b0',
'uploader': 'Adrian Kingsley-Hughes',
- 'timestamp': 1448961720,
- 'upload_date': '20151201',
+ 'duration': 731,
+ 'timestamp': 1449129925,
+ 'upload_date': '20151203',
},
'params': {
# m3u8 download
'skip_download': True,
- }
+ },
}]
- TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/kYEXFC/%s?mbr=true'
+
MPX_ACCOUNTS = {
- 'cnet': 2288573011,
+ 'cnet': 2198311517,
'zdnet': 2387448114,
}
@@ -68,7 +77,8 @@ class CBSInteractiveIE(ThePlatformIE):
data = self._parse_json(data_json, display_id)
vdata = data.get('video') or data['videos'][0]
- video_id = vdata['id']
+ video_id = vdata['mpxRefId']
+
title = vdata['title']
author = vdata.get('author')
if author:
@@ -78,20 +88,7 @@ class CBSInteractiveIE(ThePlatformIE):
uploader = None
uploader_id = None
- media_guid_path = 'media/guid/%d/%s' % (self.MPX_ACCOUNTS[site], vdata['mpxRefId'])
- formats, subtitles = [], {}
- for (fkey, vid) in vdata['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)
- self._sort_formats(formats)
-
- info = self._extract_theplatform_metadata('kYEXFC/%s' % media_guid_path, video_id)
+ info = self._extract_video_info(video_id, site, self.MPX_ACCOUNTS[site])
info.update({
'id': video_id,
'display_id': display_id,
@@ -99,7 +96,5 @@ class CBSInteractiveIE(ThePlatformIE):
'duration': int_or_none(vdata.get('duration')),
'uploader': uploader,
'uploader_id': uploader_id,
- 'subtitles': subtitles,
- 'formats': formats,
})
return info
