Update app.py

app.py

@@ -194,23 +194,31 @@ for url in md_files_url:
 print(f"Fetched {len(documents)} documents.")  
 
 text_chunk = chunk_text(documents)
+
 # Define paths
 FAISS_LOCAL_PATH = "/tmp/faiss_index"
+os.makedirs(FAISS_LOCAL_PATH, exist_ok=True)  # Ensure directory exists
 
-# 🔹 Ensure FAISS directory exists
-os.makedirs(FAISS_LOCAL_PATH, exist_ok=True)
+FAISS_INDEX_FILE = os.path.join(FAISS_LOCAL_PATH, "index.faiss")  # FAISS index file
 
-try:
-    # 🔹 Create FAISS index
-    faiss_index = FAISS.from_documents(text_chunk, download_hugging_face_embeddings())
-    
-    # 🔹 Save FAISS locally
-    faiss_index.save_local(FAISS_LOCAL_PATH)
-    print(f"✅ FAISS index successfully saved to {FAISS_LOCAL_PATH}")
+# 🔹 Check if FAISS index exists before creating a new one
+if not os.path.exists(FAISS_INDEX_FILE):
+    print("🔴 FAISS index not found! Creating a new FAISS index...")
 
-except Exception as e:
-    logging.error(f"🚨 Error creating or saving FAISS index: {e}")
+    try:
+        # 🔹 Create FAISS index
+        faiss_index = FAISS.from_documents(text_chunk, download_hugging_face_embeddings())
+
+        # 🔹 Save FAISS locally
+        faiss_index.save_local(FAISS_LOCAL_PATH)
+        print(f"✅ FAISS index successfully saved to {FAISS_LOCAL_PATH}")
 
+    except Exception as e:
+        logging.error(f"🚨 Error creating or saving FAISS index: {e}")
+else:
+    print(f"🟢 FAISS index already exists at {FAISS_LOCAL_PATH}, skipping creation.")
+
+# 🔹 Check if dataset exists before uploading FAISS
 try:
     api.repo_info(repo_id, repo_type="dataset", token=HF_TOKEN)
     print(f"✅ Dataset '{repo_id}' already exists in the organization.")
@@ -228,33 +236,47 @@ except Exception:
     print(f"✅ Dataset '{repo_id}' created successfully in the organization.")
 
 # 🔹 Upload FAISS to the organization dataset
-api.upload_folder(
-    folder_path=FAISS_LOCAL_PATH,
-    repo_id=repo_id,
-    repo_type="dataset",
-    token=HF_TOKEN
-)
+try:
+    print("📤 Uploading FAISS index to Hugging Face...")
+    api.upload_folder(
+        folder_path=FAISS_LOCAL_PATH,
+        repo_id=repo_id,
+        repo_type="dataset",
+        token=HF_TOKEN
+    )
+    print("✅ FAISS index successfully uploaded to the organization's private dataset.")
 
-print("✅ FAISS index successfully uploaded to the organization's private dataset.")
+except Exception as e:
+    logging.error(f"🚨 Error uploading FAISS index: {e}")
 
-# 🔹 Download FAISS index from the private organization dataset
-faiss_download_folder = snapshot_download(
-    repo_id=repo_id,
-    allow_patterns=["faiss_index/*"],
-    use_auth_token=HF_TOKEN
-)
+FAISS_DOWNLOAD_PATH = "/tmp/faiss_index_download"
+
+# 🔹 Check if FAISS exists before downloading
+if not os.path.exists(FAISS_DOWNLOAD_PATH):
+    print("📥 Downloading FAISS index from Hugging Face...")
+
+    try:
+        faiss_download_folder = snapshot_download(
+            repo_id=repo_id,
+            repo_type="dataset",  # ✅ Ensure it's a dataset repo
+            allow_patterns=["faiss_index/*"],
+            use_auth_token=HF_TOKEN
+        )
+
+        print(f"✅ FAISS index downloaded at {faiss_download_folder}")
 
-FAISS_PATH = os.path.join(faiss_download_folder, "faiss_index")
+    except Exception as e:
+        raise FileNotFoundError(f"🚨 Failed to download FAISS index: {e}")
 
-if os.path.exists(FAISS_PATH):
-    print(f"✅ FAISS index found at {FAISS_PATH}, loading it now...")
 else:
-    raise FileNotFoundError("🚨 FAISS index not found in the organization's dataset.")
+    print("🟢 FAISS index already exists locally, skipping download.")
+
 try:
     # 🔹 Load FAISS index with LangChain
+    embedding_function = download_hugging_face_embeddings()
     docsearch = FAISS.load_local(
-        FAISS_PATH,
-        download_hugging_face_embeddings(),
+        FAISS_DOWNLOAD_PATH,
+        embedding_function,
         allow_dangerous_deserialization=True
     )