Update app.py

app.py

@@ -13,39 +13,31 @@ from llama_index.core.chat_engine.condense_plus_context import CondensePlusConte
 from llama_index.core.schema import Document
 
 # ===================================
-# 1️⃣ Fungsi untuk Membaca Google Spreadsheet
+# 1️⃣ Fungsi untuk Membaca Google Spreadsheet dari Beberapa Worksheet
 # ===================================
-def read_google_sheet():
+def read_google_sheets():
     try:
-        # Tentukan scope akses ke Google Sheets & Drive
         scope = ["https://www.googleapis.com/auth/spreadsheets", "https://www.googleapis.com/auth/drive"]
-        
-        # Load kredensial dari file credentials.json
         creds = ServiceAccountCredentials.from_json_keyfile_name("credentials.json", scope)
         client = gspread.authorize(creds)
         
-        # 📌 Gunakan ID Spreadsheet (lebih aman)
-        SPREADSHEET_ID = "1e_cNMhwF-QYpyYUpqQh-XCw-OdhWS6EuYsoBUsVtdNg"  # 🔹 Ganti dengan ID spreadsheet Anda
-        SHEET_NAME = "datatarget"  # 🔹 Ganti dengan nama sheet
+        SPREADSHEET_ID = "1e_cNMhwF-QYpyYUpqQh-XCw-OdhWS6EuYsoBUsVtdNg"
+        sheet_names = ["datatarget", "dataabsen", "datacuti", "datalembur"]  # 🔹 Daftar sheet yang akan dibaca
         
-        # Buka spreadsheet dan worksheet
+        combined_text = ""
         spreadsheet = client.open_by_key(SPREADSHEET_ID)
-        sheet = spreadsheet.worksheet(SHEET_NAME)
-
-        # Ambil semua data dalam bentuk list (baris & kolom)
-        data = sheet.get_all_values()
-
-        # Format ulang data menjadi satu teks panjang (dapat disesuaikan)
-        formatted_text = "\n".join([" | ".join(row) for row in data])
         
-        return formatted_text
-    
-    except gspread.exceptions.SpreadsheetNotFound:
-        return "❌ ERROR: Spreadsheet tidak ditemukan. Pastikan ID/nama benar!"
-    
-    except gspread.exceptions.WorksheetNotFound:
-        return "❌ ERROR: Worksheet tidak ditemukan. Periksa kembali nama sheet!"
-    
+        for sheet_name in sheet_names:
+            try:
+                sheet = spreadsheet.worksheet(sheet_name)
+                data = sheet.get_all_values()
+                formatted_text = f"\n=== Data dari {sheet_name} ===\n"
+                formatted_text += "\n".join([" | ".join(row) for row in data])
+                combined_text += formatted_text + "\n"
+            except gspread.exceptions.WorksheetNotFound:
+                combined_text += f"\n❌ ERROR: Worksheet '{sheet_name}' tidak ditemukan!\n"
+        
+        return combined_text.strip()
     except Exception as e:
         return f"❌ ERROR: {str(e)}"
 
@@ -54,8 +46,8 @@ def read_google_sheet():
 # ===================================
 def initialize_llama_model():
     model_path = hf_hub_download(
-        repo_id="TheBLoke/zephyr-7b-beta-GGUF",  # 📌 Repo model HuggingFace
-        filename="zephyr-7b-beta.Q4_K_M.gguf",  # 📌 Nama file model
+        repo_id="TheBLoke/zephyr-7b-beta-GGUF",
+        filename="zephyr-7b-beta.Q4_K_M.gguf",
         cache_dir="./models"
     )
     return model_path
@@ -73,22 +65,16 @@ def initialize_settings(model_path):
 # 4️⃣ Inisialisasi Index dari Data Spreadsheet
 # ===================================
 def initialize_index():
-    # 🔹 Ambil teks dari Google Spreadsheet
-    text_data = read_google_sheet()
-
-    # 🔹 Konversi teks ke dalam format dokumen yang benar
-    document = Document(text=text_data)  # 🔹 Ubah teks menjadi objek `Document`
-    documents = [document]  # 🔹 Masukkan ke dalam list
-
-    # 🔹 Proses data menjadi node untuk vektor embedding
+    text_data = read_google_sheets()
+    document = Document(text=text_data)
+    documents = [document]
+    
     parser = SentenceSplitter(chunk_size=150, chunk_overlap=10)
-    nodes = parser.get_nodes_from_documents(documents)  # ✅ Sekarang `documents` adalah list of `Document`
-
-    # 🔹 Gunakan model embedding
+    nodes = parser.get_nodes_from_documents(documents)
+    
     embedding = HuggingFaceEmbedding("sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2")
     Settings.embed_model = embedding
-
-    # 🔹 Buat index vektor
+    
     index = VectorStoreIndex(nodes)
     return index
 
@@ -110,23 +96,20 @@ def generate_response(message, history, chat_engine):
     if history is None:
         history = []
     
-    # 🔹 Ambil data terbaru dari Google Spreadsheet setiap kali ada pertanyaan
-    text_data = read_google_sheet()
+    text_data = read_google_sheets()
     document = Document(text=text_data)
     documents = [document]
-
-    # 🔹 Perbarui index dengan data terbaru
+    
     parser = SentenceSplitter(chunk_size=150, chunk_overlap=10)
     nodes = parser.get_nodes_from_documents(documents)
     index = VectorStoreIndex(nodes)
     retriever = index.as_retriever(similarity_top_k=3)
-
-    # 🔹 Buat ulang chat engine dengan index yang diperbarui
+    
     chat_engine = CondensePlusContextChatEngine.from_defaults(
         retriever=retriever,
         verbose=True,
     )
-
+    
     chat_messages = [
         ChatMessage(
             role="system",
@@ -138,11 +121,10 @@ def generate_response(message, history, chat_engine):
             "Fokuslah memberikan jawaban yang akurat dan relevan sesuai dengan dokumen yang tersedia."
         ),
     ]
-
-    # 🔹 Gunakan chat engine baru untuk menjawab pertanyaan
+    
     response = chat_engine.stream_chat(message)
     text = "".join(response.response_gen)
-
+    
     history.append((message, text))
     return history