X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcbs.py;h=ac2c7dced6f3561bb90f679947cfe35d6a31e2b1;hb=daa0df9e8beac1325e5fb55d828e7a3a38e74bf6;hp=ac0315853626279c5cdd1257ec16ff659946e622;hpb=1bb2fc98e05b7cc556aa63aa226b5349e8a39159;p=youtube-dl

diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py
index ac0315853..ac2c7dced 100644
--- a/youtube_dl/extractor/cbs.py
+++ b/youtube_dl/extractor/cbs.py
@@ -1,30 +1,111 @@
+from __future__ import unicode_literals
+
 import re
 
-from .common import InfoExtractor
+from .theplatform import ThePlatformIE
+from ..utils import (
+    xpath_text,
+    xpath_element,
+    int_or_none,
+    find_xpath_attr,
+)
+
 
+class CBSBaseIE(ThePlatformIE):
+    def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
+        closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
+        return {
+            'en': [{
+                'ext': 'ttml',
+                'url': closed_caption_e.attrib['value'],
+            }]
+        } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
-class CBSIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cbs\.com/shows/[^/]+/video/(?P<id>[^/]+)/.*'
 
-    _TEST = {
-        u'url': u'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
-        u'file': u'4JUVEwq3wUT7.flv',
-        u'info_dict': {
-            u'title': u'Connect Chat feat. Garth Brooks',
-            u'description': u'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
-            u'duration': 1495,
+class CBSIE(CBSBaseIE):
+    _VALID_URL = r'(?:cbs:(?P<content_id>\w+)|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<display_id>[^/]+))'
+
+    _TESTS = [{
+        'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
+        'info_dict': {
+            'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
+            'display_id': 'connect-chat-feat-garth-brooks',
+            'ext': 'mp4',
+            'title': 'Connect Chat feat. Garth Brooks',
+            'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
+            'duration': 1495,
+            'timestamp': 1385585425,
+            'upload_date': '20131127',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+        '_skip': 'Blocked outside the US',
+    }, {
+        'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/',
+        'info_dict': {
+            'id': 'WWF_5KqY3PK1',
+            'display_id': 'st-vincent',
+            'ext': 'flv',
+            'title': 'Live on Letterman - St. Vincent',
+            'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.',
+            'duration': 3221,
         },
-        u'params': {
+        'params': {
             # rtmp download
-            u'skip_download': True,
+            'skip_download': True,
         },
-    }
+        '_skip': 'Blocked outside the US',
+    }, {
+        'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
+        'only_matching': True,
+    }]
+    TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        real_id = self._search_regex(
-            r"video\.settings\.pid\s*=\s*'([^']+)';",
-            webpage, u'real video ID')
-        return self.url_result(u'theplatform:%s' % real_id)
+        content_id, display_id = re.match(self._VALID_URL, url).groups()
+        if not content_id:
+            webpage = self._download_webpage(url, display_id)
+            content_id = self._search_regex(
+                [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
+                webpage, 'content id')
+        items_data = self._download_xml(
+            'http://can.cbs.com/thunder/player/videoPlayerService.php',
+            content_id, query={'partner': 'cbs', 'contentId': content_id})
+        video_data = xpath_element(items_data, './/item')
+        title = xpath_text(video_data, 'videoTitle', 'title', True)
+
+        subtitles = {}
+        formats = []
+        for item in items_data.findall('.//item'):
+            pid = xpath_text(item, 'pid')
+            if not pid:
+                continue
+            tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid
+            if '.m3u8' in xpath_text(item, 'contentUrl', default=''):
+                tp_release_url += '&manifest=m3u'
+            tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                tp_release_url, content_id, 'Downloading %s SMIL data' % pid)
+            formats.extend(tp_formats)
+            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+        self._sort_formats(formats)
+
+        info = self.get_metadata('dJ5BDC/media/guid/2198311517/%s' % content_id, content_id)
+        info.update({
+            'id': content_id,
+            'display_id': display_id,
+            'title': title,
+            'series': xpath_text(video_data, 'seriesTitle'),
+            'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
+            'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
+            'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
+            'thumbnail': xpath_text(video_data, 'previewImageURL'),
+            'formats': formats,
+            'subtitles': subtitles,
+        })
+        return info