[generic] Prefer enclosures over links in RSS feeds

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

diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index af1322e0085befa144605f16c22a52fcca5a3bcf..252f97c2617c2322ca454263536e6839b996cefc 100644 (file)
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -23,6 +23,7 @@ from ..utils import (
     is_html,
     js_to_json,
     KNOWN_EXTENSIONS,
+    merge_dicts,
     mimetype2ext,
     orderedSet,
     sanitized_Request,
@@ -190,6 +191,16 @@ class GenericIE(InfoExtractor):
                 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
             }
         },
+        # RSS feed with enclosures and unsupported link URLs
+        {
+            'url': 'http://www.hellointernet.fm/podcast?format=rss',
+            'info_dict': {
+                'id': 'http://www.hellointernet.fm/podcast?format=rss',
+                'description': 'CGP Grey and Brady Haran talk about YouTube, life, work, whatever.',
+                'title': 'Hello Internet',
+            },
+            'playlist_mincount': 100,
+        },
         # SMIL from http://videolectures.net/promogram_igor_mekjavic_eng
         {
             'url': 'http://videolectures.net/promogram_igor_mekjavic_eng/video/1/smil.xml',
@@ -2025,13 +2036,15 @@ class GenericIE(InfoExtractor):
 
         entries = []
         for it in doc.findall('./channel/item'):
-            next_url = xpath_text(it, 'link', fatal=False)
+            next_url = None
+            enclosure_nodes = it.findall('./enclosure')
+            for e in enclosure_nodes:
+                next_url = e.attrib.get('url')
+                if next_url:
+                    break
+
             if not next_url:
-                enclosure_nodes = it.findall('./enclosure')
-                for e in enclosure_nodes:
-                    next_url = e.attrib.get('url')
-                    if next_url:
-                        break
+                next_url = xpath_text(it, 'link', fatal=False)
 
             if not next_url:
                 continue
@@ -3002,21 +3015,6 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 sharevideos_urls, video_id, video_title)
 
-        def merge_dicts(dict1, dict2):
-            merged = {}
-            for k, v in dict1.items():
-                if v is not None:
-                    merged[k] = v
-            for k, v in dict2.items():
-                if v is None:
-                    continue
-                if (k not in merged or
-                        (isinstance(v, compat_str) and v and
-                            isinstance(merged[k], compat_str) and
-                            not merged[k])):
-                    merged[k] = v
-            return merged
-
         # Look for HTML5 media
         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
         if entries: