Upload 4 files

app.py

@@ -172,10 +172,6 @@ def parse_file(file_obj):
             with zipfile.ZipFile(file_obj.name, 'r') as z:
                 z.extractall(extract_root)
             
-            # Check for existing media (rough check)
-            media_dir = os.path.join(extract_root, "media") 
-            has_media = os.path.exists(media_dir) or any(f.isdigit() for f in os.listdir(extract_root))
-
             col_path = os.path.join(extract_root, "collection.anki2")
             if not os.path.exists(col_path):
                 shutil.rmtree(extract_root)
@@ -187,16 +183,23 @@ def parse_file(file_obj):
             rows = cur.fetchall()
             
             data = []
+            audio_count = 0  # Count cards with existing audio
             for r in rows:
                 flds = r[0].split('\x1f')
                 tags = r[1].strip() if len(r) > 1 else ""
                 q = flds[0] if len(flds) > 0 else ""
                 a = flds[1] if len(flds) > 1 else ""
+                # Check if either field has audio tags
+                if re.search(r'\[sound:.*?\]', q) or re.search(r'\[sound:.*?\]', a):
+                    audio_count += 1
                 data.append([q, a, tags])
             
             df = pd.DataFrame(data, columns=["Question", "Answer", "Tags"])
             conn.close()
             
+            # has_media means existing AUDIO, not images
+            has_media = audio_count > 0
+            
         else:
             return None, None, None, "Unsupported file type", "", None
         
@@ -204,7 +207,7 @@ def parse_file(file_obj):
         
         msg = f"✅ Loaded {len(df)} cards."
         if has_media:
-            msg += " 🎵 Existing media detected."
+            msg += f" 🎵 {audio_count} cards have existing audio."
             
         return df, has_media, df.head(PREVIEW_LIMIT), msg, estimate_time(len(df), has_media), extract_root
         
@@ -473,6 +476,24 @@ with gr.Blocks(title="Pocket TTS Anki") as app:
     file_input.upload(on_upload, inputs=file_input, 
                       outputs=[full_df_state, has_media_state, preview_table, status, eta_box, extract_root_state, tag_dropdown])
 
+    def on_clear():
+        """Reset all fields when file is cleared."""
+        return (
+            None,                                    # full_df_state
+            False,                                   # has_media_state
+            None,                                    # preview_table
+            "",                                      # status
+            "",                                      # eta_box
+            None,                                    # extract_root_state
+            gr.Dropdown(choices=["All"], value="All"),  # tag_dropdown
+            "",                                      # search_box
+            None,                                    # dl (download file)
+            ""                                       # result_lbl
+        )
+
+    file_input.clear(on_clear, inputs=[], 
+                     outputs=[full_df_state, has_media_state, preview_table, status, eta_box, extract_root_state, tag_dropdown, search_box, dl, result_lbl])
+
     def on_search(term, df, has_media, mode, search_in, selected_tag):
         if df is None: return None, "No data"