logger_service.py

import os
import pandas as pd
from huggingface_hub import hf_hub_download, upload_file, login
from dotenv import load_dotenv
from config import DATASET_REPO_ID

# Load .env at import time so HF_TOKEN works in local dev
load_dotenv()

# --- 配置區 ---
# DATASET_REPO_ID 已從 config 匯入
CSV_FILENAME = "baseline_logs.csv"
HF_TOKEN = os.getenv("HF_TOKEN")

# 🌟 批次上傳門檻 (每累積幾筆資料才備份到雲端一次)
BATCH_SIZE = 30  
unsynced_count = 0  
pending_logs = []  # 用於存放尚未寫入 CSV 的紀錄

if HF_TOKEN:
    login(token=HF_TOKEN)

def sync_from_hf():
    """從雲端下載最新的 CSV 並回傳 DataFrame (系統啟動時只會呼叫一次)"""
    try:
        from config import MODEL_FILES
        local_path = MODEL_FILES["baseline_logs"]
        path = hf_hub_download(
            repo_id=DATASET_REPO_ID,
            filename=CSV_FILENAME,
            repo_type="dataset",
            token=HF_TOKEN
        )
        # 同步回本地
        df = pd.read_csv(path)
        df.to_csv(local_path, index=False)
        print(f"✅ 已從 Hugging Face 同步最新資料庫 ({len(df)} 筆)")
        return df
    except Exception as e:
        print(f"⚠️ 無法下載雲端資料，可能為首次運行或權限不足: {e}")
        return None

def upload_to_hf():
    """將本地的 CSV 檔案同步回雲端"""
    try:
        upload_file(
            path_or_fileobj=CSV_FILENAME,
            path_in_repo=CSV_FILENAME,
            repo_id=DATASET_REPO_ID,
            repo_type="dataset",
            token=HF_TOKEN,
            commit_message=f"System auto-log batch update ({BATCH_SIZE} logs)"
        )
        return True
    except Exception as e:
        print(f"❌ 雲端同步失敗: {e}")
        return False

def push_new_log(new_log_dict):
    """
    接收新紀錄並計數。達到 BATCH_SIZE 門檻時才執行「本地寫入」與「雲端上傳」。
    有效避免頻繁 IO 與 429 Rate Limit。
    """
    global unsynced_count, pending_logs
    
    pending_logs.append(new_log_dict)
    unsynced_count += 1
    
    if unsynced_count >= BATCH_SIZE:
        print(f"📦 已累積 {BATCH_SIZE} 筆操作，正在執行批次寫入與同步...")
        
        try:
            from config import MODEL_FILES
            local_path = MODEL_FILES["baseline_logs"]
            
            # 執行本地批次寫入
            new_df = pd.DataFrame(pending_logs)
            new_df.to_csv(local_path, mode='a', header=not os.path.exists(local_path), index=False)
            
            # 執行雲端同步
            success = upload_to_hf()
            
            if success:
                print("✅ 批次寫入與雲端同步成功！")
                unsynced_count = 0  
                pending_logs = []  # 清空緩存
            return success
        except Exception as e:
            print(f"❌ 批次寫入失敗: {e}")
            return False
        
    return True