Phase 8 Complete: Sync Audit Log + Frontend Integration

- Thêm audit/sync_audit.py: Ghi lịch sử sync vào audit/sync_log.json
- Thêm API endpoints: /sync/history, /sync/history/{run_id}
- Frontend: Nút 'Lịch sử đồng bộ' + panel expand chi tiết từng file
- Sửa frontend served từ backend (http://localhost:8000)
- Cập nhật DEPLOYMENT_GUIDE với hướng dẫn chạy localhost
- Cập nhật ARCHITECTURE_MAP: Phase 8 hoàn thành

api/main.py

@@ -5,7 +5,8 @@ import secrets
 from enum import Enum
 from typing import List, Optional, Dict, Any
 from fastapi import FastAPI, HTTPException, BackgroundTasks, Request, status
-from fastapi.responses import RedirectResponse
+from fastapi.responses import RedirectResponse, FileResponse
+from fastapi.staticfiles import StaticFiles
 from fastapi.middleware.cors import CORSMiddleware
 from pydantic import BaseModel, Field, validator
 import uvicorn
@@ -24,6 +25,7 @@ from extraction.ocr_service import OCRService
 from extraction.text_extractor import TextExtractor
 from chunking.markdown_chunker import MarkdownChunker
 from indexing.vector_store import VectorStore
+from audit import sync_audit
 
 # --- Cấu hình Logging chuyên nghiệp ---
 logging.basicConfig(
@@ -50,6 +52,15 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+# Serve frontend static files
+FRONTEND_DIR = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "frontend")
+app.mount("/static", StaticFiles(directory=FRONTEND_DIR), name="static")
+
+@app.get("/", tags=["Frontend"])
+async def serve_frontend():
+    """Serve frontend index.html."""
+    return FileResponse(os.path.join(FRONTEND_DIR, "index.html"))
+
 # --- Singleton Engine Instance ---
 rag_engine = None
 sync_status = {"running": False, "last_run": None, "processed": 0, "skipped": 0, "errors": []}
@@ -270,6 +281,8 @@ def run_sync_background():
     global sync_status
     sync_status = {"running": True, "last_run": None, "processed": 0, "skipped": 0, "errors": []}
     
+    run_id = sync_audit.start_run()
+    
     try:
         provider = SharePointProvider()
         dce = DocumentClassificationEngine(provider=provider)
@@ -304,16 +317,22 @@ def run_sync_background():
             
             if classification.processing_policy in (ProcessingPolicy.UNSUPPORTED, ProcessingPolicy.METADATA_ONLY, ProcessingPolicy.REQUIRES_REVIEW):
                 sync_status["skipped"] += 1
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value, 
+                                   classification.doc_type.value, 0, "skipped", classification.reason)
                 continue
             
             try:
                 file_bytes = provider.download_file(item)
             except Exception as e:
                 sync_status["errors"].append(f"{name}: download failed")
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value,
+                                   classification.doc_type.value, 0, "error", str(e))
                 continue
             
             if not file_bytes:
                 sync_status["errors"].append(f"{name}: empty file")
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value,
+                                   classification.doc_type.value, 0, "error", "empty file")
                 continue
             
             pages = []
@@ -336,6 +355,8 @@ def run_sync_background():
             
             if not pages:
                 sync_status["skipped"] += 1
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value,
+                                   classification.doc_type.value, 0, "skipped", "no content extracted")
                 continue
             
             metadata = {
@@ -352,16 +373,22 @@ def run_sync_background():
                 vector_db.delete_by_file_id(item_id)
                 vector_db.embed_and_index(chunks)
                 sync_status["processed"] += 1
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value,
+                                   classification.doc_type.value, len(chunks), "indexed")
                 logger.info(f"Sync: Indexed {name} → {len(chunks)} chunks")
             else:
                 sync_status["skipped"] += 1
+                sync_audit.log_file(run_id, name, item_id, classification.processing_policy.value,
+                                   classification.doc_type.value, 0, "skipped", "no chunks generated")
         
         sync_status["last_run"] = "completed"
+        sync_audit.finish_run(run_id, "completed")
         logger.info(f"Sync completed: {sync_status['processed']} processed, {sync_status['skipped']} skipped")
         
     except Exception as e:
         sync_status["last_run"] = "failed"
         sync_status["errors"].append(str(e))
+        sync_audit.finish_run(run_id, "failed")
         logger.error(f"Sync failed: {e}")
     finally:
         sync_status["running"] = False
@@ -382,8 +409,30 @@ async def sync_endpoint(background_tasks: BackgroundTasks):
 
 @app.get("/sync/status", tags=["Ingestion"])
 async def sync_status_endpoint():
-    """Kiểm tra trạng thái đồng bộ."""
+    """Kiểm tra trạng thái đồng bộ hiện tại."""
     return sync_status
 
+
+@app.get("/sync/history", tags=["Ingestion"])
+async def sync_history_endpoint(limit: int = 10):
+    """
+    Lấy lịch sử đồng bộ (audit log).
+    Args:
+        limit: Số lần sync gần nhất (mặc định 10)
+    """
+    return {"runs": sync_audit.get_history(limit)}
+
+
+@app.get("/sync/history/{run_id}", tags=["Ingestion"])
+async def sync_run_detail_endpoint(run_id: str):
+    """
+    Lấy chi tiết của 1 lần đồng bộ.
+    """
+    run = sync_audit.get_run_detail(run_id)
+    if not run:
+        raise HTTPException(status_code=404, detail="Run not found")
+    return run
+
+
 if __name__ == "__main__":
     uvicorn.run("main:app", host="0.0.0.0", port=8000, reload=True)