Merge branch 'charlierose' of https://github.com/TRox1972/youtube-dl into TRox1972...

[youtube-dl] / youtube_dl / extractor / cbs.py

diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py
index 43f05d278060c47215122dc3b5a86d6480b5d891..c72ed2dbb7d2f6c97d1cca1e8652451db91e49ee 100644 (file)
--- a/youtube_dl/extractor/cbs.py
+++ b/youtube_dl/extractor/cbs.py
@@ -1,40 +1,41 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..compat import compat_urllib_request
-from ..utils import smuggle_url
+from .theplatform import ThePlatformFeedIE
+from ..utils import (
+    int_or_none,
+    find_xpath_attr,
+    ExtractorError,
+)
 
 
-class CBSIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<id>[^/]+)'
+class CBSBaseIE(ThePlatformFeedIE):
+    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(CBSBaseIE):
+    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
 
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
         'info_dict': {
-            'id': '4JUVEwq3wUT7',
-            'display_id': 'connect-chat-feat-garth-brooks',
-            'ext': 'flv',
+            'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
+            '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,
-        },
-        'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
         '_skip': 'Blocked outside the US',
@@ -46,21 +47,30 @@ class CBSIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _extract_video_info(self, guid):
+        path = 'dJ5BDC/media/guid/2198311517/' + guid
+        smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
+        formats, subtitles = self._extract_theplatform_smil(smil_url + '&manifest=m3u', guid)
+        for r in ('HLS&formats=M3U', 'RTMP', 'WIFI', '3G'):
+            try:
+                tp_formats, _ = self._extract_theplatform_smil(smil_url + '&assetTypes=' + r, guid, 'Downloading %s SMIL data' % r.split('&')[0])
+                formats.extend(tp_formats)
+            except ExtractorError:
+                continue
+        self._sort_formats(formats)
+        metadata = self._download_theplatform_metadata(path, guid)
+        info = self._parse_theplatform_metadata(metadata)
+        info.update({
+            'id': guid,
+            'formats': formats,
+            'subtitles': subtitles,
+            'series': metadata.get('cbs$SeriesTitle'),
+            'season_number': int_or_none(metadata.get('cbs$SeasonNumber')),
+            'episode': metadata.get('cbs$EpisodeTitle'),
+            'episode_number': int_or_none(metadata.get('cbs$EpisodeNumber')),
+        })
+        return info
+
     def _real_extract(self, url):
-        display_id = self._match_id(url)
-        request = compat_urllib_request.Request(url)
-        # Android UA is served with higher quality (720p) streams (see
-        # https://github.com/rg3/youtube-dl/issues/7490)
-        request.add_header('User-Agent', 'Mozilla/5.0 (Linux; Android 4.4; Nexus 5)')
-        webpage = self._download_webpage(request, display_id)
-        real_id = self._search_regex(
-            [r"video\.settings\.pid\s*=\s*'([^']+)';", r"cbsplayer\.pid\s*=\s*'([^']+)';"],
-            webpage, 'real video ID')
-        return {
-            '_type': 'url_transparent',
-            'ie_key': 'ThePlatform',
-            'url': smuggle_url(
-                'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true&manifest=m3u' % real_id,
-                {'force_smil_url': True}),
-            'display_id': display_id,
-        }
+        content_id = self._match_id(url)
+        return self._extract_video_info(content_id)