fixing download

src/app.py

@@ -118,8 +118,10 @@ def load_rag_engine():
             st.stop()
 
         # Ensure dataset files are downloaded from HF Datasets if needed
-        # First check without UI to see if download is needed
-        success, files_downloaded = ensure_dataset_files(show_ui=True)
+        # Create a container for download progress that will be cleared after completion
+        download_container = st.container()
+        success, files_downloaded = ensure_dataset_files(progress_container=download_container)
+
         if not success:
             st.error("⚠️ Failed to load dataset files from Hugging Face Datasets.")
             st.info("Please check your internet connection and try again.")

src/dataset_loader.py

@@ -24,14 +24,14 @@ DATASET_FILES = [
     "extraction_progress.json"
 ]
 
-def ensure_dataset_files(dataset_dir: str = "dataset/wikipedia_ireland", show_ui: bool = False) -> tuple:
+def ensure_dataset_files(dataset_dir: str = "dataset/wikipedia_ireland", progress_container=None) -> tuple:
     """
     Ensure all dataset files are available locally.
     Downloads from HF Datasets if missing.
 
     Args:
         dataset_dir: Local directory for dataset files
-        show_ui: Whether to show Streamlit UI indicators
+        progress_container: Streamlit container for progress updates (optional)
 
     Returns:
         Tuple of (success: bool, files_downloaded: bool)
@@ -50,18 +50,30 @@ def ensure_dataset_files(dataset_dir: str = "dataset/wikipedia_ireland", show_ui
         return True, False  # Success, no files downloaded
 
     print(f"[INFO] Missing {len(missing_files)} files, downloading from HF Datasets...")
-    if show_ui:
-        st.info(f"📥 Downloading {len(missing_files)} missing dataset files from Hugging Face...")
 
     # Download missing files
     import shutil
+    import time
+
+    # Create UI elements if container provided
+    status_placeholder = None
+    progress_bar_placeholder = None
+    progress_text_placeholder = None
+
+    if progress_container:
+        status_placeholder = progress_container.empty()
+        progress_text_placeholder = progress_container.empty()
+        progress_bar_placeholder = progress_container.empty()
+        status_placeholder.info(f"📥 Downloading {len(missing_files)} missing dataset files from Hugging Face...")
+
     try:
         for idx, filename in enumerate(missing_files, 1):
             print(f"[INFO] Downloading {filename} ({idx}/{len(missing_files)})...")
 
-            # Only show UI progress if show_ui is True
-            if show_ui:
-                st.progress((idx - 1) / len(missing_files), text=f"Downloading {filename}...")
+            # Update progress if container provided
+            if progress_text_placeholder and progress_bar_placeholder:
+                progress_text_placeholder.text(f"⬇️ {filename} ({idx}/{len(missing_files)})")
+                progress_bar_placeholder.progress((idx - 1) / len(missing_files))
 
             downloaded_path = hf_hub_download(
                 repo_id=DATASET_REPO,
@@ -73,17 +85,36 @@ def ensure_dataset_files(dataset_dir: str = "dataset/wikipedia_ireland", show_ui
             shutil.copy2(downloaded_path, target_path)
             print(f"[SUCCESS] Downloaded {filename}")
 
-        if show_ui:
-            st.progress(1.0, text="All files downloaded!")
-            st.success("✅ Dataset files ready!")
+        # Update to show completion
+        if progress_bar_placeholder:
+            progress_bar_placeholder.progress(1.0)
+
+        # Small delay to show completion, then clear
+        time.sleep(0.5)
+
+        # Clear ALL UI elements after successful download
+        if status_placeholder:
+            status_placeholder.empty()
+        if progress_text_placeholder:
+            progress_text_placeholder.empty()
+        if progress_bar_placeholder:
+            progress_bar_placeholder.empty()
 
         print("[SUCCESS] All dataset files downloaded successfully!")
         return True, True  # Success, files were downloaded
 
     except Exception as e:
         print(f"[ERROR] Failed to download dataset files: {e}")
-        if show_ui:
-            st.error(f"Failed to download dataset files: {e}")
+        if progress_container:
+            # Clear progress indicators
+            if status_placeholder:
+                status_placeholder.empty()
+            if progress_text_placeholder:
+                progress_text_placeholder.empty()
+            if progress_bar_placeholder:
+                progress_bar_placeholder.empty()
+            # Show error
+            progress_container.error(f"❌ Failed to download dataset files: {e}")
         return False, False  # Failure, no files downloaded