Update sync_data.sh

sync_data.sh

@@ -21,77 +21,77 @@ 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")
+  """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')]
-    
+
     # 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
-    
+    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=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 {backup_by_date[date]} 时出错: {str(e)}')
+     dates_to_delete = sorted_dates[:(len(sorted_dates) - max_files)]
+     for date in dates_to_delete:
+   try:
+    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 {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,
-            repo_id=repo_id,
-            repo_type="dataset"
-        )
-        print(f"成功上传 {file_name}")
-        manage_backups(api, repo_id)
+     # 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,
+      repo_id=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)}")
+     print(f"文件上传出错: {str(e)}")
 
 # 下载最新备份
 def download_latest_backup(token, repo_id):
     try:
-        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]
-        file = api.hf_hub_download(
-            repo_id=repo_id,
-            filename=latest_backup,
-            repo_type="dataset"
-        )
-        
-        if file and os.path.exists(file):
-            os.makedirs('./data', exist_ok=True)
-            os.system(f'cp "{file}" ./data/webui.db')
-            print(f"成功从 {latest_backup} 恢复备份")
-            return True
-        
+     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]
+    file = api.hf_hub_download(
+      repo_id=repo_id,
+      filename=latest_backup,
+      repo_type="dataset"
+     )
+   
+     if file and os.path.exists(file):
+        os.makedirs('./data', exist_ok=True)
+        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)}")
@@ -101,13 +101,13 @@ if __name__ == "__main__":
     action = sys.argv[1]
     token = sys.argv[2]
     repo_id = sys.argv[3]
-    
+  
     if action == "upload":
-        file_path = sys.argv[4]
-        file_name = sys.argv[5]
-        upload_backup(file_path, file_name, token, repo_id)
+     file_path = sys.argv[4]
+     file_name = sys.argv[5]
+     upload_backup(file_path, file_name, token, repo_id)
     elif action == "download":
-        download_latest_backup(token, repo_id)
+      download_latest_backup(token, repo_id)
 EOL
 
 # 首次启动时的数据恢复策略
@@ -121,82 +121,46 @@ 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/open/"
+echo "PROPFIND 请求: $WEBDAV_URL/openwebui/"
 
 # 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/open/" 2>&1)
+webdav_list_output=$(curl -v -X PROPFIND --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" -H "Depth: 1" "$WEBDAV_URL/openwebui/" 2>&1)
 echo "WebDAV 目录内容响应:"
 echo "$webdav_list_output"
 
 # 获取 WebDAV 目录中的文件列表并找出最新的备份文件
-webdav_files=$(echo "$webdav_list_output" | grep -o '<d:href>[^<]*webui_[0-9][0-9]-[0-9][0-9]\.db</d:href>' | sed 's/<d:href>//g' | sed 's/<\/d:href>//g')
+webdav_files=$(echo "$webdav_list_output" | grep -o '<d:href>[^<]*webui_[0-9]\{4\}[0-9]\{2\}[0-9]\{2\}\.db</d:href>' | sed 's/<d:href>//g' | sed 's/<\/d:href>//g')
 
 if [ -n "$webdav_files" ]; then
-    # 找出最新的文件 (按文件名排序，取最后一个)
-    latest_file=$(echo "$webdav_files" | sort | tail -n 1)
+  # 找出最新的文件 (按文件名排序，取最后一个)
+  latest_file=$(echo "$webdav_files" | sort | tail -n 1)
     echo "找到最新的 WebDAV 备份文件: $latest_file"
-    
-    # 构建并打印完整下载链接
-    download_url="$WEBDAV_URL$latest_file"
+  
+  # 构建并打印完整下载链接
+  download_url="$WEBDAV_URL$latest_file"
     echo "下载 URL: $download_url"
-    
+  
     # 下载最新文件
     curl -v -L --fail --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$download_url" -o "./data/webui.db" && {
-        echo "成功从 WebDAV 下载最新数据库文件: $latest_file"
-        
-        # 上传到 HuggingFace
-        DATE_FORMAT=$(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 Hugging Face..."
-        python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
-        rm -f "/tmp/${backup_file}"
+     echo "成功从 WebDAV 下载最新数据库文件: $latest_file"
+   
+     # 上传到 HuggingFace
+     DATE_FORMAT=$(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 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/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
-            DATE_FORMAT=$(date +%Y%m%d)
-            backup_file="webui_backup_${DATE_FORMAT}.db"
-            cp ./data/webui.db "/tmp/${backup_file}"
-            echo "正在上传初始备份到 Hugging Face..."
-            python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
-            rm -f "/tmp/${backup_file}"
-        } || {
-            echo "从 WebDAV 下载失败，尝试从 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"
-            }
-        }
+     echo "从 WebDAV 下载最新文件失败，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"
+      }
     }
 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/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
-        DATE_FORMAT=$(date +%Y%m%d)
-        backup_file="webui_backup_${DATE_FORMAT}.db"
-        cp ./data/webui.db "/tmp/${backup_file}"
-        echo "正在上传初始备份到 Hugging Face..."
-        python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
-        rm -f "/tmp/${backup_file}"
-    } || {
-        echo "从 WebDAV 下载失败，尝试从 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"
-        }
+  echo "WebDAV 中没有找到符合格式的备份文件，尝试从 Hugging Face 恢复..."
+  python /tmp/hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" || {
+    echo "从 Hugging Face 恢复失败，将使用空数据库开始"
     }
 fi
 
@@ -205,45 +169,55 @@ 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="$WEBDAV_URL/open/$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: $FILE"
-                
-                # 覆盖Webdav目录下默认的webui.db文件
-                main_file_url="$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"
-                } || {
-                    echo "WebDAV 更新主文件失败"
-                }
-            } || {
-                echo "WebDAV 上传失败，等待重试..."
-                sleep 10
-                echo "重试上传 URL: $upload_url"
-                curl -v -L -T "./data/webui.db" --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$upload_url" || {
-                    echo "重试失败，放弃上传。"
-                }
-            }
-        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
+    echo "开始 WebDAV 同步 $(date)"
+   
+     # 检查数据库文件是否存在
+     if [ -f "./data/webui.db" ]; then
+   # 生成当天的日期文件名（YYYYMMDD格式）
+     DATE_FORMAT=$(date +%Y%m%d)
+     FILE="webui_${DATE_FORMAT}.db"
+      echo "同步到 WebDAV..."
+      upload_url="$WEBDAV_URL/openwebui/$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: $FILE"
+  
+       # 删除较旧的文件以管理空间
+    echo "获取 WebDAV 目录以清理旧文件..."
+       webdav_files_list=$(curl -s -X PROPFIND --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" -H "Depth: 1" "$WEBDAV_URL/openwebui/" | grep -o '<d:href>[^<]*webui_[0-9]\{8\}\.db</d:href>' | sed 's/<d:href>//g' | sed 's/<\/d:href>//g')
+     
+    # 保留最新的10个文件
+    if [ -n "$webdav_files_list" ]; then
+        files_to_keep=$(echo "$webdav_files_list" | sort | tail -n 10)
+   
+     # 删除较旧的文件
+     for old_file in $webdav_files_list; do
+   if ! echo "$files_to_keep" | grep -q "$old_file"; then
+    echo "删除旧文件: $old_file"
+       curl -X DELETE --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$WEBDAV_URL$old_file"
+      fi
+     done
+    fi
+  
+         } || {
+    echo "WebDAV 上传失败，等待重试..."
+        sleep 10
+        echo "重试上传 URL: $upload_url"
+       curl -v -L -T "./data/webui.db" --user "$WEBDAV_USERNAME:$WEBDAV_PASSWORD" "$upload_url" || {
+     echo "重试失败，放弃上传。"
+    }
+   }
+     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
 }
 
@@ -252,29 +226,29 @@ hf_sync_() {
     SYNC_INTERVAL=${SYNC_INTERVAL:-7200}
     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 Hugging Face 同步 $(date)"
-        
-        if [ -f "./data/webui.db" ]; then
-            # 使用仅含日期的格式（年月日）
-            DATE_FORMAT=$(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 Hugging Face..."
-            python /tmp/hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
-            rm -f "/tmp/${backup_file}"
-        else
-            echo "数据库文件不存在，跳过 Hugging Face 同步"
-        fi
-        
+   
+     if [ -f "./data/webui.db" ]; then
+      # 使用仅含日期的格式（年月日）
+      DATE_FORMAT=$(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 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 Hugging Face 同步"
+     fi
+   
         echo "Hugging Face 同步完成，下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
-        sleep $SYNC_INTERVAL
+     sleep $SYNC_INTERVAL
     done
 }
 
 # 后台启动同步进程
 webdav_sync_ &
-hf_sync_ &
+hf_sync_ &