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 ac2c7dced6f3561bb90f679947cfe35d6a31e2b1..c72ed2dbb7d2f6c97d1cca1e8652451db91e49ee 100644 (file)
--- a/youtube_dl/extractor/cbs.py
+++ b/youtube_dl/extractor/cbs.py
@@ -1,17 +1,14 @@
 from __future__ import unicode_literals
 
-import re
-
-from .theplatform import ThePlatformIE
+from .theplatform import ThePlatformFeedIE
 from ..utils import (
-    xpath_text,
-    xpath_element,
     int_or_none,
     find_xpath_attr,
+    ExtractorError,
 )
 
 
-class CBSBaseIE(ThePlatformIE):
+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 {
@@ -23,13 +20,12 @@ class CBSBaseIE(ThePlatformIE):
 
 
 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>[^/]+))'
+    _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': '_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!',
@@ -39,22 +35,7 @@ class CBSIE(CBSBaseIE):
             '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',
@@ -65,47 +46,31 @@ class CBSIE(CBSBaseIE):
         '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):
-        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:
+    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
-            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)
+        metadata = self._download_theplatform_metadata(path, guid)
+        info = self._parse_theplatform_metadata(metadata)
         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'),
+            '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):
+        content_id = self._match_id(url)
+        return self._extract_video_info(content_id)