init project

app.py

@@ -1,3 +1,4 @@
+import os
 import gradio as gr
 import logging
 from fastapi import FastAPI, Request
@@ -6,20 +7,14 @@ from rag_core.embedder import get_embedding
 from rag_core.retriever import Retriever
 from rag_core.llm import generate_answer
 
-app = FastAPI()
 retriever = Retriever()
+app = FastAPI()
+ready = retriever.index is not None
 
-# Khởi tạo nếu chưa có index
-if retriever.index is None:
-    logging.info("Không tìm thấy FAISS index, bắt đầu xử lý...")
-    with open("data/raw_law.txt", "r", encoding="utf-8") as f:
-        text = f.read()
-    chunks = chunk_legal_text(text)
-    retriever.build(chunks, get_embedding)
-
-# API endpoint
 @app.post("/ask")
 async def ask_api(req: Request):
+    if not ready:
+        return {"error": "Index chưa sẵn sàng. Vui lòng thử lại sau."}
     data = await req.json()
     query = data.get("query")
     docs = retriever.query(query, get_embedding)
@@ -29,25 +24,42 @@ async def ask_api(req: Request):
 
 @app.post("/rescan")
 async def rescan_api():
+    if not ready:
+        return {"error": "Index chưa sẵn sàng."}
     with open("data/raw_law.txt", "r", encoding="utf-8") as f:
         text = f.read()
     chunks = chunk_legal_text(text)
     retriever.rescan_and_append(chunks, get_embedding)
     return {"status": "Rescan & update thành công."}
 
-# Gradio UI
-iface = gr.Interface(
-    fn=lambda q: generate_answer("\n\n".join(retriever.query(q, get_embedding)) + f"\n\nCâu hỏi: {q}\nTrả lời:"),
-    inputs=gr.Textbox(label="Nhập câu hỏi"),
-    outputs=gr.Textbox(label="Trả lời"),
-    title="Luật Giao Thông RAG"
-)
+def build_index_ui():
+    global ready
+    with gr.Textbox(visible=False):
+        pass  # trigger
+    with open("data/raw_law.txt", "r", encoding="utf-8") as f:
+        text = f.read()
+    chunks = chunk_legal_text(text)
+    retriever.build(chunks, get_embedding)
+    ready = True
+    return gr.update(visible=False), gr.update(visible=True), gr.update(visible=True)
+
+def answer_fn(query):
+    if not ready:
+        return "Index chưa sẵn sàng. Vui lòng chờ hoàn tất xử lý."
+    docs = retriever.query(query, get_embedding)
+    prompt = "\n\n".join(docs) + f"\n\nCâu hỏi: {query}\nTrả lời:"
+    return generate_answer(prompt)
 
-import uvicorn
-import threading
+with gr.Blocks() as iface:
+    build_btn = gr.Button("🔄 Xây Index", visible=not ready)
+    query_box = gr.Textbox(label="Nhập câu hỏi", visible=ready)
+    output_box = gr.Textbox(label="Trả lời", visible=ready)
+    query_box.submit(fn=answer_fn, inputs=query_box, outputs=output_box)
+    build_btn.click(fn=build_index_ui, outputs=[build_btn, query_box, output_box])
 
-def start_fastapi():
-    uvicorn.run(app, host="0.0.0.0", port=7861)
+# mount Gradio lên FastAPI
+app = gr.mount_gradio_app(app, iface, path="/")
 
-threading.Thread(target=start_fastapi).start()
-iface.launch()
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=7860)

rag_core/retriever.py

@@ -20,6 +20,8 @@ class Retriever:
 
     @log_timed("xây FAISS index")
     def build(self, texts: list, embed_fn):
+        os.makedirs(os.path.dirname(INDEX_PATH), exist_ok=True)
+
         embeddings = []
         valid_texts = []
         for i, t in enumerate(texts):