[ign] improve extraction and extract uploader_id

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

diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py
index 7377ac7b9a143c0587c5a715e0e89c99f1c5480b..fbefd37d09a98bb19c82b4c09b7b08c99d147d35 100644 (file)
--- a/youtube_dl/extractor/cspan.py
+++ b/youtube_dl/extractor/cspan.py
@@ -8,6 +8,7 @@ from ..utils import (
     unescapeHTML,
     find_xpath_attr,
     smuggle_url,
+    determine_ext,
 )
 from .senateisvp import SenateISVPIE
 
@@ -37,15 +38,17 @@ class CSpanIE(InfoExtractor):
         }
     }, {
         'url': 'http://www.c-span.org/video/?318608-1/gm-ignition-switch-recall',
+        'md5': '446562a736c6bf97118e389433ed88d4',
         'info_dict': {
             'id': '342759',
+            'ext': 'mp4',
             'title': 'General Motors Ignition Switch Recall',
+            'duration': 14848,
+            'description': 'md5:70c7c3b8fa63fa60d42772440596034c'
         },
-        'playlist_duration_sum': 14855,
     }, {
         # Video from senate.gov
         'url': 'http://www.c-span.org/video/?104517-1/immigration-reforms-needed-protect-skilled-american-workers',
-        'md5': '7314c4b96dad66dd8e63dc3518ceaa6f',
         'info_dict': {
             'id': 'judiciary031715',
             'ext': 'flv',
@@ -85,6 +88,10 @@ class CSpanIE(InfoExtractor):
             return self.url_result(surl, 'SenateISVP', video_id, title)
 
         files = data['video']['files']
+        try:
+            capfile = data['video']['capfile']['#text']
+        except KeyError:
+            capfile = None
 
         entries = [{
             'id': '%s_%d' % (video_id, partnum + 1),
@@ -95,11 +102,22 @@ class CSpanIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'duration': int_or_none(f.get('length', {}).get('#text')),
+            'subtitles': {
+                'en': [{
+                    'url': capfile,
+                    'ext': determine_ext(capfile, 'dfxp')
+                }],
+            } if capfile else None,
         } for partnum, f in enumerate(files)]
 
-        return {
-            '_type': 'playlist',
-            'entries': entries,
-            'title': title,
-            'id': video_id,
-        }
+        if len(entries) == 1:
+            entry = dict(entries[0])
+            entry['id'] = video_id
+            return entry
+        else:
+            return {
+                '_type': 'playlist',
+                'entries': entries,
+                'title': title,
+                'id': video_id,
+            }