Update app.py

moved metadata and faiss indexing to flask backend

app.py

@@ -32,45 +32,19 @@ embedding_dim = 768  # Adjust according to model
 def store_document_data(PDF_FILE, METADATA_FILE, INDEX_FILE):
     print(" Storing document...")
 
-    if PDF_FILE:
+   if PDF_FILE:
         # Extract text from the PDF
         text = extract_text_from_pdf(PDF_FILE)
         if not text:
             return "Could not extract any text from the PDF."
-    
-    if METADATA_FILE:
-         # extract metadata 
-        print(" Metadata file exists")
-        with open(METADATA_FILE, "r") as f:
-            metadata = json.load(f)
+
+        # Generate and return embedding
+        embedding = embedding_model.encode([text]).astype(np.float32)
+        print("Embeddings generated")
+
+        return embedding
     else:
-        print("metadata_file is empty")
-        metadata = {}
-
-    if INDEX_FILE:
-        # extract Faiss
-        print("index_file recieved")
-        index = faiss.read_index(INDEX_FILE)
-    else: 
-        print(" No FAISS index found. Creating a new one.")
-        index = faiss.IndexFlatL2(embedding_dim)  # Empty FAISS index
-
-    # Generate and store embedding
-    embedding = embedding_model.encode([text]).astype(np.float32)
-    index.add(embedding)  # Add to FAISS index
-    print(" Embeddings generated")
-
-    # Get FAISS index for the new document
-    doc_index = index.ntotal - 1
-
-    # Update metadata with FAISS index
-    metadata[str(doc_index)] = PDF_FILE
-    print(" Saved Metadata")
-
-    return json.dumps({
-        "metadata": metadata, 
-        "index": index
-    })
+        return "No PDF file provided."
 
 def retrieve_document(query):
     print(f"Retrieving document based on:\n{query}")