Grab RSS media:content as Entry thumbnail

coldsweat/fetcher.py

@@ -241,6 +241,7 @@ def _parse_feed(self, data):
 
             timestamp = t.get_timestamp(default=self.instant)
             content_type, content = t.get_content(('text/plain', ''))
+            thumbnail_url = t.get_thumbnail_url()
 
             entry = {
                 'feed_id': self.feed.id,
@@ -250,6 +251,7 @@ def _parse_feed(self, data):
                 'author': t.get_author() or feed_author,
                 'content': content,
                 'content_type': content_type,
+                'thumbnail_url': thumbnail_url,
                 'published_on': timestamp
             }
             new_entries.append(entry)
@@ -270,7 +272,7 @@ def _parse_feed(self, data):
                     # MySQL doesn't support conlict targets, see:
                     # https://stackoverflow.com/questions/74691515/python-peewee-using-excluded-to-resolve-conflict-resolution
                     count += (Entry.insert_many(batch).on_conflict(
-                        # Pass down these new values only for certain values
+                        # Pass down these new values only for certain fields
                         preserve=[Entry.title, Entry.author, Entry.content, Entry.content_type])
                         .as_rowcount()
                         .execute())

coldsweat/translators.py

@@ -34,6 +34,7 @@ def get_title(self):
                             Feed.MAX_TITLE_LENGTH)
         return ''
 
+IMAGE_TYPES = ['image/jpeg', 'image/png', 'image/gif']
 
 class EntryTranslator(object):
 
@@ -87,7 +88,21 @@ def get_content(self, default):
         # app.logger.debug(u'no entry content found, using default')
         return default
 
-    # Nullable fields
+    def get_thumbnail_url(self):
+        #See https://www.rssboard.org/media-rss
+        if 'media_content' in self.entry_dict:
+            media_content = self.entry_dict['media_content'][0]
+            try:
+                image_type = media_content['type']
+            except KeyError:
+                image_type = None
+            if image_type in IMAGE_TYPES:
+                try:
+                    return media_content['url']
+                except KeyError:
+                    pass
+        # @@TODO: Try to get thumbnail[0] instead 
+        return ''
 
     def get_link(self):
         # Special case for FeedBurner entries