Improve detection of podcast mime type

newswire.py

@@ -482,27 +482,33 @@ def xml_podcast_to_dict(xml_str: str) -> {}:
     return podcast_properties
 
 
-def get_link_from_rss_item(rss_item: str) -> str:
+def get_link_from_rss_item(rss_item: str) -> (str, str):
     """Extracts rss link from rss item string
     """
+    mime_type = None
+
     if '<enclosure ' in rss_item:
         # get link from audio or video enclosure
         enclosure = rss_item.split('<enclosure ')[1]
         if '>' in enclosure:
             enclosure = enclosure.split('>')[0]
+            if ' type="' in enclosure:
+                mime_type = enclosure.split(' type="')[1]
+                if '"' in mime_type:
+                    mime_type = mime_type.split('"')[0]
             if 'url="' in enclosure and \
                ('"audio/' in enclosure or '"video/' in enclosure):
                 link_str = enclosure.split('url="')[1]
                 if '"' in link_str:
                     link = link_str.split('"')[0]
                     if '://' in link:
-                        return link
+                        return link, mime_type
 
     link = rss_item.split('<link>')[1]
     link = link.split('</link>')[0]
     if '://' not in link:
-        return None
+        return None, None
-    return link
+    return link, mime_type
 
 
 def _xml2str_to_dict(base_dir: str, domain: str, xml_str: str,
@@ -560,7 +566,7 @@ def _xml2str_to_dict(base_dir: str, domain: str, xml_str: str,
                 description = description.split('</media:description>')[0]
                 description = remove_html(description)
 
-        link = get_link_from_rss_item(rss_item)
+        link, link_mime_type = get_link_from_rss_item(rss_item)
         if not link:
             continue
 
@@ -579,6 +585,8 @@ def _xml2str_to_dict(base_dir: str, domain: str, xml_str: str,
                 post_filename = ''
                 votes_status = []
                 podcast_properties = xml_podcast_to_dict(rss_item)
+                if podcast_properties:
+                    podcast_properties['linkMimeType'] = link_mime_type
                 _add_newswire_dict_entry(base_dir, domain,
                                          result, pub_date_str,
                                          title, link,
@@ -651,7 +659,7 @@ def _xml1str_to_dict(base_dir: str, domain: str, xml_str: str,
                 description = description.split('</media:description>')[0]
                 description = remove_html(description)
 
-        link = get_link_from_rss_item(rss_item)
+        link, link_mime_type = get_link_from_rss_item(rss_item)
         if not link:
             continue
 
@@ -670,6 +678,8 @@ def _xml1str_to_dict(base_dir: str, domain: str, xml_str: str,
                 post_filename = ''
                 votes_status = []
                 podcast_properties = xml_podcast_to_dict(rss_item)
+                if podcast_properties:
+                    podcast_properties['linkMimeType'] = link_mime_type
                 _add_newswire_dict_entry(base_dir, domain,
                                          result, pub_date_str,
                                          title, link,
@@ -730,7 +740,7 @@ def _atom_feed_to_dict(base_dir: str, domain: str, xml_str: str,
                 description = description.split('</media:description>')[0]
                 description = remove_html(description)
 
-        link = get_link_from_rss_item(atom_item)
+        link, link_mime_type = get_link_from_rss_item(atom_item)
         if not link:
             continue
 
@@ -749,6 +759,8 @@ def _atom_feed_to_dict(base_dir: str, domain: str, xml_str: str,
                 post_filename = ''
                 votes_status = []
                 podcast_properties = xml_podcast_to_dict(atom_item)
+                if podcast_properties:
+                    podcast_properties['linkMimeType'] = link_mime_type
                 _add_newswire_dict_entry(base_dir, domain,
                                          result, pub_date_str,
                                          title, link,

tests.py

@@ -6460,9 +6460,11 @@ def _test_get_link_from_rss_item() -> None:
         'play/46054222/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net' + \
         '%2Fstaging%2F2022-0-12%2F7352f28c-a928-ea7a-65ae-' + \
         'ccb5edffbac1.mp3" length="67247880" type="audio/mpeg"/>'
-    link = get_link_from_rss_item(rss_item)
+    link, mime_type = get_link_from_rss_item(rss_item)
     assert link
     assert link.endswith('.mp3')
+    assert mime_type
+    assert mime_type == 'audio/mpeg'
 
     rss_item = \
         '<link>' + \
@@ -6470,9 +6472,10 @@ def _test_get_link_from_rss_item() -> None:
         'Hessel-van-Oorschot-of-Tribe-of-Noise--Free-Music-Archive-e1crvce' + \
         '</link>' + \
         '<pubDate>Wed, 12 Jan 2022 14:28:46 GMT</pubDate>'
-    link = get_link_from_rss_item(rss_item)
+    link, mime_type = get_link_from_rss_item(rss_item)
     assert link
     assert link.startswith('https://anchor.fm')
+    assert not mime_type
 
 
 def run_all_tests():

webapp_column_right.py

@@ -270,8 +270,7 @@ def _html_newswire(base_dir: str, newswire: {}, nickname: str, moderator: bool,
             # change the link url to a podcast episode screen
             podcast_properties = item[8]
             if podcast_properties:
-                if podcast_properties.get('image') and \
+                if podcast_properties.get('image'):
-                   'explicit' in podcast_properties:
                     episode_id = date_str.replace(' ', '__')
                     episode_id = episode_id.replace(':', 'aa')
                     link_url = \

webapp_podcast.py

@@ -156,13 +156,20 @@ def html_podcast_episode(css_cache: {}, translate: {},
             '" alt="" ' + get_broken_link_substitute() + '/></a>\n'
     podcast_str += '  </div>\n'
 
+    audio_extension = None
     if path_is_audio(link_url):
         if '.mp3' in link_url:
-            extension = 'mp3'
+            audio_extension = 'mpeg'
         else:
-            extension = 'ogg'
+            audio_extension = 'ogg'
-
+    else:
-        podcast_str += _html_podcast_soundbites(link_url, extension,
+        if podcast_properties.get('linkMimeType'):
+            if 'audio' in podcast_properties['linkMimeType']:
+                audio_extension = \
+                    podcast_properties['linkMimeType'].split('/')[1]
+    # show widgets for soundbites
+    if audio_extension:
+        podcast_str += _html_podcast_soundbites(link_url, audio_extension,
                                                 podcast_properties,
                                                 translate)
 
@@ -170,7 +177,7 @@ def html_podcast_episode(css_cache: {}, translate: {},
         podcast_str += \
             '<audio controls>\n' + \
             '<source src="' + link_url + '" type="audio/' + \
-            extension.replace('.', '') + '">' + \
+            audio_extension.replace('.', '') + '">' + \
             translate['Your browser does not support the audio element.'] + \
             '</audio>\n'