Update sync_data.sh

sync_data.sh

@@ -14,29 +14,50 @@ fi
 # 创建数据目录
 mkdir -p ./data
 
-# 创建 HuggingFace 同步脚本
+# 创建 Hugging Face 同步脚本
 cat > /tmp/hf_sync.py << 'EOL'
 from huggingface_hub import HfApi
 import sys
 import os
 
 def manage_backups(api, repo_id, max_files=50):
+    """管理备份文件，保留最新的max_files个文件"""
     files = api.list_repo_files(repo_id=repo_id, repo_type="dataset")
     backup_files = [f for f in files if f.startswith('webui_backup_') and f.endswith('.db')]
-    backup_files.sort()
     
-    if len(backup_files) >= max_files:
-        files_to_delete = backup_files[:(len(backup_files) - max_files + 1)]
-        for file_to_delete in files_to_delete:
+    # 按日期分组文件（从文件名中提取日期）
+    backup_by_date = {}
+    for file in backup_files:
+        try:
+            # 从文件名格式 webui_backup_YYYYMMDD.db 中提取日期部分
+            date_part = file.split('_')[2].split('.')[0]  # 提取YYYYMMDD部分
+            backup_by_date[date_part] = file
+        except:
+            # 如果文件名格式不符，则跳过
+            continue
+    
+    # 日期排序并仅保留最新max_files个文件
+    sorted_dates = sorted(backup_by_date.keys())
+    if len(sorted_dates) > max_files:
+        dates_to_delete = sorted_dates[:(len(sorted_dates) - max_files)]
+        for date in dates_to_delete:
             try:
-                api.delete_file(path_in_repo=file_to_delete, repo_id=repo_id, repo_type="dataset")
-                print(f'已删除旧备份: {file_to_delete}')
+                api.delete_file(path_in_repo=backup_by_date[date], repo_id=repo_id, repo_type="dataset")
+                print(f'已删除旧备份: {backup_by_date[date]}')
             except Exception as e:
-                print(f'删除 {file_to_delete} 时出错: {str(e)}')
+                print(f'删除 {backup_by_date[date]} 时出错: {str(e)}')
 
 def upload_backup(file_path, file_name, token, repo_id):
+    """上传备份文件到Hugging Face"""
     api = HfApi(token=token)
     try:
+        # 检查同名文件是否已存在
+        files = api.list_repo_files(repo_id=repo_id, repo_type="dataset")
+        if file_name in files:
+            # 上传前删除同名文件
+            api.delete_file(path_in_repo=file_name, repo_id=repo_id, repo_type="dataset")
+            print(f"已删除同名文件: {file_name}")
+            
         api.upload_file(
             path_or_fileobj=file_path,
             path_in_repo=file_name,
@@ -44,7 +65,6 @@ def upload_backup(file_path, file_name, token, repo_id):
             repo_type="dataset"
         )
         print(f"成功上传 {file_name}")
-        
         manage_backups(api, repo_id)
     except Exception as e:
         print(f"文件上传出错: {str(e)}")
@@ -55,26 +75,24 @@ def download_latest_backup(token, repo_id):
         api = HfApi(token=token)
         files = api.list_repo_files(repo_id=repo_id, repo_type="dataset")
         backup_files = [f for f in files if f.startswith('webui_backup_') and f.endswith('.db')]
-        
         if not backup_files:
             print("未找到备份文件")
             return False
-            
-        latest_backup = sorted(backup_files)[-1]
         
-        filepath = api.hf_hub_download(
+        latest_backup = sorted(backup_files)[-1]
+        file = api.hf_hub_download(
             repo_id=repo_id,
             filename=latest_backup,
             repo_type="dataset"
         )
         
-        if filepath and os.path.exists(filepath):
+        if file and os.path.exists(file):
             os.makedirs('./data', exist_ok=True)
-            os.system(f'cp "{filepath}" ./data/webui.db')
+            os.system(f'cp "{file}" ./data/webui.db')
             print(f"成功从 {latest_backup} 恢复备份")
             return True
+        
         return False
-                
     except Exception as e:
         print(f"下载备份时出错: {str(e)}")
         return False
@@ -103,10 +121,10 @@ echo "WebDAV 密码: $masked_password"
 
 # 首先尝试从 WebDAV 恢复最新文件
 echo "正在尝试从 WebDAV 获取文件列表..."
-echo "PROPFIND 请求: $WEBDAV_URL/openwebui/"
+echo "PROPFIND 请求: $WEBDAV_URL/open/"
 
 # 获取并打印 WebDAV 目录内容
-webdav_list_output=$(curl -v -X PROPFIND --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" -H "Depth: 1" "$WEBDAV_URL/openwebui/" 2>&1)
+webdav_list_output=$(curl -v -X PROPFIND --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" -H "Depth: 1" "$WEBDAV_URL/open/" 2>&1)
 echo "WebDAV 目录内容响应:"
 echo "$webdav_list_output"
 
@@ -125,32 +143,34 @@ if [ -n "$webdav_files" ]; then
     # 下载最新文件
     curl -v -L --fail --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$download_url" -o "./data/webui.db" && {
         echo "成功从 WebDAV 下载最新数据库文件: $latest_file"
+        
         # 上传到 HuggingFace
-        timestamp=$(date +%Y%m%d_%H%M%S)
-        backup_file="webui_backup_${timestamp}.db"
+        DATE_FORMAT=$(date +%Y%m%d)
+        backup_file="webui_backup_${DATE_FORMAT}.db"
         cp ./data/webui.db "/tmp/${backup_file}"
-        echo "正在上传初始备份到 HuggingFace..."
-        python3 /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
+        echo "正在上传初始备份到 Hugging Face..."
+        python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
         rm -f "/tmp/${backup_file}"
     } || {
         echo "从 WebDAV 下载最新文件失败，尝试下载 WebDAV 的 webui.db..."
         
         # 打印 webui.db 下载链接
-        main_db_url="$WEBDAV_URL/openwebui/webui.db"
+        main_db_url="$WEBDAV_URL/open/webui.db"
         echo "尝试下载主数据库文件 URL: $main_db_url"
         
         curl -v -L --fail --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$main_db_url" -o "./data/webui.db" && {
             echo "成功从 WebDAV 下载主数据库文件"
+            
             # 上传到 HuggingFace
-            timestamp=$(date +%Y%m%d_%H%M%S)
-            backup_file="webui_backup_${timestamp}.db"
+            DATE_FORMAT=$(date +%Y%m%d)
+            backup_file="webui_backup_${DATE_FORMAT}.db"
             cp ./data/webui.db "/tmp/${backup_file}"
-            echo "正在上传初始备份到 HuggingFace..."
-            python3 /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
+            echo "正在上传初始备份到 Hugging Face..."
+            python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
             rm -f "/tmp/${backup_file}"
         } || {
-            echo "从 WebDAV 下载失败，尝试从 HuggingFace 恢复..."
-            python3 /tmp/hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" || {
+            echo "从 WebDAV 下载失败，尝试从 Hugging Face 恢复..."
+            python /tmp/hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" || {
                 echo "所有恢复方式均失败，将使用空数据库开始"
             }
         }
@@ -159,50 +179,50 @@ else
     echo "WebDAV 中没有找到符合格式的备份文件，尝试下载 webui.db..."
     
     # 打印 webui.db 下载链接
-    main_db_url="$WEBDAV_URL/openwebui/webui.db"
+    main_db_url="$WEBDAV_URL/open/webui.db"
     echo "尝试下载主数据库文件 URL: $main_db_url"
     
     curl -v -L --fail --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$main_db_url" -o "./data/webui.db" && {
         echo "成功从 WebDAV 下载主数据库文件"
+        
         # 上传到 HuggingFace
-        timestamp=$(date +%Y%m%d_%H%M%S)
-        backup_file="webui_backup_${timestamp}.db"
+        DATE_FORMAT=$(date +%Y%m%d)
+        backup_file="webui_backup_${DATE_FORMAT}.db"
         cp ./data/webui.db "/tmp/${backup_file}"
-        echo "正在上传初始备份到 HuggingFace..."
-        python3 /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
+        echo "正在上传初始备份到 Hugging Face..."
+        python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
         rm -f "/tmp/${backup_file}"
     } || {
-        echo "从 WebDAV 下载失败，尝试从 HuggingFace 恢复..."
-        python3 /tmp/hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" || {
+        echo "从 WebDAV 下载失败，尝试从 Hugging Face 恢复..."
+        python /tmp/hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" || {
             echo "所有恢复方式均失败，将使用空数据库开始"
         }
     }
 fi
 
 # 定义 WebDAV 同步函数
-webdav_sync() {
+webdav_sync_() {
     SYNC_INTERVAL=${SYNC_INTERVAL:-7200}  # 默认间隔时间为 7200 秒
     echo "WebDAV 同步进程启动，等待 ${SYNC_INTERVAL} 秒后开始同步..."
     sleep $SYNC_INTERVAL
-
+    
     while true; do
         echo "开始 WebDAV 同步 $(date)"
         
         # 检查数据库文件是否存在
         if [ -f "./data/webui.db" ]; then
-            # 生成当前时间的文件名
-            FILENAME="webui_$(date +'%m-%d_%H%M').db"
-            
+            # 生成仅含日期的文件名（月-日）
+            FILE="webui_$(date +'%m-%d').db"
             echo "同步到 WebDAV..."
-            upload_url="$WEBDAV_URL/openwebui/$FILENAME"
+            upload_url="$WEBDAV_URL/open/$FILE"
             echo "上传 URL: $upload_url"
-
+            
             # 上传以日期命名的数据库文件
             curl -v -L -T "./data/webui.db" --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$upload_url" && {
-                echo "WebDAV 上传成功: $FILENAME"
+                echo "WebDAV 上传成功: $FILE"
                 
                 # 覆盖Webdav目录下默认的webui.db文件
-                main_file_url="$WEBDAV_URL/openwebui/webui.db"
+                main_file_url="$WEBDAV_URL/open/webui.db"
                 echo "更新主文件 URL: $main_file_url"
                 curl -v -L -T "./data/webui.db" --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$main_file_url" && {
                     echo "WebDAV 更新主文件成功"
@@ -220,42 +240,41 @@ webdav_sync() {
         else
             echo "未找到 webui.db 文件，跳过 WebDAV 同步"
         fi
-
+        
         # 等待下一次同步间隔
         echo "WebDAV 同步完成，下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
         sleep $SYNC_INTERVAL
     done
 }
 
-# 定义 HuggingFace 同步函数
-hf_sync() {
+# 定义 Hugging Face 同步函数
+hf_sync_() {
     SYNC_INTERVAL=${SYNC_INTERVAL:-7200}
-    echo "HuggingFace 同步进程启动，等待 ${SYNC_INTERVAL} 秒后开始同步..."
+    echo "Hugging Face 同步进程启动，等待 ${SYNC_INTERVAL} 秒后开始同步..."
     sleep $(($SYNC_INTERVAL / 2))  # 错开与 WebDAV 同步的时间
-
+    
     while true; do
-        echo "开始 HuggingFace 同步 $(date)"
+        echo "开始 Hugging Face 同步 $(date)"
         
         if [ -f "./data/webui.db" ]; then
-            timestamp=$(date +%Y%m%d_%H%M%S)
-            backup_file="webui_backup_${timestamp}.db"
+            # 使用仅含日期的格式（年月日）
+            DATE_FORMAT=$(date +%Y%m%d)
+            backup_file="webui_backup_${DATE_FORMAT}.db"
             
             # 复制数据库文件
             cp ./data/webui.db "/tmp/${backup_file}"
-            
-            echo "正在上传备份到 HuggingFace..."
-            python3 /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
-            
+            echo "正在上传备份到 Hugging Face..."
+            python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
             rm -f "/tmp/${backup_file}"
         else
-            echo "数据库文件不存在，跳过 HuggingFace 同步"
+            echo "数据库文件不存在，跳过 Hugging Face 同步"
         fi
         
-        echo "HuggingFace 同步完成，下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
+        echo "Hugging Face 同步完成，下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
         sleep $SYNC_INTERVAL
     done
 }
 
 # 后台启动同步进程
-webdav_sync &
-hf_sync &
+webdav_sync_ &
+hf_sync_ &