Update Dockerfile

Dockerfile

@@ -140,11 +140,14 @@ from pathlib import Path
 from watchdog.observers import Observer
 from watchdog.events import FileSystemEventHandler
 from huggingface_hub import HfApi
-from huggingface_hub.utils.tqdm_utils import disable_progress_bars, enable_progress_bars  # 关键：全局禁用进度条
+from huggingface_hub import utils  # 关键：正确导入
 import base64
 from datetime import datetime
 import pytz
 
+# 全局禁用所有 Hugging Face 进度条（启动时立即执行）
+utils.disable_progress_bars()
+
 # 配置日志，使用北京时间
 beijing_tz = pytz.timezone('Asia/Shanghai')
 logging.basicConfig(
@@ -154,7 +157,7 @@ logging.basicConfig(
 )
 logger = logging.getLogger(__name__)
 
-# 自定义日志格式化器以使用北京时间
+# 自定义日志格式化器
 class BeijingTimeFormatter(logging.Formatter):
     def formatTime(self, record, datefmt=None):
         dt = datetime.fromtimestamp(record.created, tz=beijing_tz)
@@ -162,7 +165,6 @@ class BeijingTimeFormatter(logging.Formatter):
             return dt.strftime(datefmt)
         return dt.strftime("%Y-%m-%d %H:%M:%S %Z")
 
-# 应用自定义格式化器
 for handler in logging.getLogger().handlers:
     handler.setFormatter(BeijingTimeFormatter(
         fmt='%(asctime)s - %(levelname)s - %(message)s',
@@ -170,47 +172,39 @@ for handler in logging.getLogger().handlers:
     ))
 
 class DataDirectoryHandler(FileSystemEventHandler):
-    """处理 /data 目录文件变化的监控器"""
-    
     def __init__(self, repo_id, hf_token, data_directory="/data"):
         self.repo_id = repo_id
         self.hf_token = hf_token
         self.data_directory = data_directory
         self.api = HfApi(token=hf_token)
         self.last_commit_time = 0
-        self.commit_delay = 1  # 防抖延迟 1 秒
-        self.pending_changes = []  # 缓冲待上传变更
+        self.commit_delay = 1
+        self.pending_changes = []
         logger.info(f"初始化监控器，监控目录: {data_directory}，目标仓库: {repo_id}")
     
     def on_any_event(self, event):
-        """捕获所有文件系统事件"""
         if event.is_directory:
             return
         self.pending_changes.append((event.event_type, event.src_path))
         self.schedule_commit(f"文件{event.event_type}")
 
     def schedule_commit(self, change_type):
-        """安排提交任务，带有防抖机制"""
         current_time = time.time()
         if current_time - self.last_commit_time > self.commit_delay:
             self.last_commit_time = current_time
             asyncio.run(self.commit_changes(change_type))
 
     async def commit_changes(self, change_type):
-        """异步提交变更到 Hugging Face Hub，带重试机制 + 全局禁用进度条"""
         max_retries = 3
         retry_delay = 5
         change_summary = f"{change_type} ({len(self.pending_changes)} 文件)"
-        self.pending_changes = []  # 清空缓冲区
+        self.pending_changes = []
         
         for attempt in range(max_retries):
             try:
                 commit_message = f"自动提交: {change_summary} - {time.strftime('%Y-%m-%d %H:%M:%S')}"
                 logger.info(f"开始上传: {commit_message}")
 
-                # 全局禁用所有 Hugging Face 进度条
-                disable_progress_bars()
-
                 await asyncio.to_thread(
                     self.api.upload_folder,
                     folder_path=self.data_directory,
@@ -220,27 +214,18 @@ class DataDirectoryHandler(FileSystemEventHandler):
                     ignore_patterns=["*.tmp", "*.log", "*.temp", ".git/*"]
                 )
 
-                # 恢复进度条（防止影响其他库）
-                enable_progress_bars()
-
                 logger.info(f"成功提交变更到 Hugging Face Hub: {commit_message}")
                 return
 
             except Exception as e:
-                # 出错时也恢复进度条
-                enable_progress_bars()
                 logger.error(f"提交失败 (尝试 {attempt + 1}/{max_retries}): {e}")
                 if attempt < max_retries - 1:
                     logger.info(f"将在 {retry_delay} 秒后重试...")
                     await asyncio.sleep(retry_delay)
         
-        # 最终失败也恢复
-        enable_progress_bars()
         logger.error(f"达到最大重试次数，上传失败")
 
 def start_directory_monitoring(data_directory="/data", repo_id=None, hf_token=None):
-    """启动目录监控服务"""
-    # 硬编码的 base64 编码的 HF_TOKEN
     if not hf_token:
         hf_token_encoded = "aGZfcXllTEJnUUtPb2FUbHBMZ0FuTGFGTmJPV2xjUUtJT0VycQ=="
         hf_token = base64.b64decode(hf_token_encoded).decode('utf-8')
@@ -267,7 +252,6 @@ def start_directory_monitoring(data_directory="/data", repo_id=None, hf_token=No
     return observer
 
 def start_gitea_server(port=7860):
-    """启动 Gitea 服务器"""
     def run_gitea():
         try:
             gitea_process = subprocess.Popen(
@@ -306,8 +290,6 @@ def start_gitea_server(port=7860):
     return gitea_thread
 
 async def main():
-    """主函数 - 启动 Gitea 和目录监控服务，带崩溃重启功能"""
-    # 配置参数
     CONFIG = {
         "data_directory": "/data",
         "repo_id": os.getenv("REPO_ID", "02engine/02gitea"),
@@ -322,17 +304,13 @@ async def main():
         try:
             logger.info(f"启动集成服务 (尝试 {retry_count + 1}/{max_retries})...")
             
-            # 先运行 pullhf.py 拉取最新数据集
             logger.info("运行 pullhf.py 拉取最新数据集")
             pull_result = subprocess.run(["python3", "/pullhf.py"], check=True)
             if pull_result.returncode != 0:
                 logger.error("pullhf.py 执行失败，退出")
                 exit(1)
             
-            # 启动 Gitea 服务器
             gitea_thread = start_gitea_server(CONFIG["gitea_port"])
-            
-            # 启动目录监控服务
             observer = start_directory_monitoring(
                 data_directory=CONFIG["data_directory"],
                 repo_id=CONFIG["repo_id"],
@@ -340,7 +318,7 @@ async def main():
             )
             
             logger.info("所有服务已启动完成！")
-            logger.info("目录监控: /data → Hugging Face Hub")
+            logger.info("目录监控: /data to Hugging Face Hub")
             logger.info("Gitea 服务: http://localhost:7860")
             logger.info("按 Ctrl+C 停止所有服务")