sync_data.sh

#!/bin/sh
set -e

# 必须设置的环境变量
if [ -z "$HF_TOKEN" ] || [ -z "$DATASET_ID" ]; then
    echo "请设置环境变量 HF_TOKEN 和 DATASET_ID"
    exit 1
fi

# 激活 Python 虚拟环境（请根据实际路径修改）
. "$HOME/venv/bin/activate"

STORAGE_PATH="$HOME/app/data"
FLAG_FILE="$HOME/.hf_backup_first_done"

# 生成 hf_sync.py 脚本
cat > hf_sync.py << 'EOL'
from huggingface_hub import HfApi
import sys, os, tarfile, tempfile

def manage_backups(api, repo_id, max_files=50):
    files = api.list_repo_files(repo_id=repo_id, repo_type="dataset")
    backup_files = [f for f in files if f.startswith('backup_') and f.endswith('.tar.gz')]
    backup_files.sort()
    if len(backup_files) >= max_files:
        for file_to_delete in backup_files[:(len(backup_files) - max_files + 1)]:
            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}')
            except Exception as e:
                print(f'删除 {file_to_delete} 时出错: {str(e)}')

def upload_backup(file_path, file_name, token, repo_id):
    api = HfApi(token=token)
    try:
        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"上传文件出错: {str(e)}")

def download_latest_backup(token, repo_id, extract_path):
    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('backup_') and f.endswith('.tar.gz')]
        if not backup_files:
            print("未找到任何备份文件")
            return
        latest_backup = sorted(backup_files)[-1]
        with tempfile.TemporaryDirectory() as temp_dir:
            filepath = api.hf_hub_download(
                repo_id=repo_id,
                filename=latest_backup,
                repo_type="dataset",
                local_dir=temp_dir
            )
            if filepath and os.path.exists(filepath):
                with tarfile.open(filepath, 'r:gz') as tar:
                    tar.extractall(extract_path)
                print(f"已成功恢复备份: {latest_backup}")
    except Exception as e:
        print(f"下载备份出错: {str(e)}")

def super_squash_history(token, repo_id):
    try:
        api = HfApi(token=token)
        api.super_squash_history(repo_id=repo_id, repo_type="dataset")
        print("历史合并完成。")
    except Exception as e:
        print(f"合并历史出错: {str(e)}")

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)
    elif action == "download":
        extract_path = sys.argv[4] if len(sys.argv) > 4 else '.'
        download_latest_backup(token, repo_id, extract_path)
    elif action == "super_squash":
        super_squash_history(token, repo_id)
EOL

upload_backup() {
    if [ -d "${STORAGE_PATH}" ]; then
        echo "上传备份数据..."
        timestamp=$(date +%Y%m%d_%H%M%S)
        backup_file="backup_${timestamp}.tar.gz"
        tar -czf "/tmp/${backup_file}" -C "$(dirname "${STORAGE_PATH}")" "$(basename "${STORAGE_PATH}")"
        python hf_sync.py upload "${HF_TOKEN}" "${DATASET_ID}" "/tmp/${backup_file}" "${backup_file}"
        rm -f "/tmp/${backup_file}"
    else
        echo "数据目录不存在，无法上传备份"
    fi
}

download_restore() {
    echo "从 HuggingFace 下载最新备份并恢复..."
    python hf_sync.py download "${HF_TOKEN}" "${DATASET_ID}" "$HOME/app"
}

if [ ! -f "$FLAG_FILE" ]; then
    echo "首次运行，开始上传备份..."
    upload_backup
    touch "$FLAG_FILE"
    echo "首次备份完成，程序退出。"
    exit 0
fi

sync_data() {
    while true; do
        echo "同步进程启动于 $(date)"
        upload_backup
        download_restore

        SQUASH_FLAG_FILE="/tmp/last_squash_time"
        NOW=$(date +%s)
        SEVEN_DAYS=$((7*24*60*60))

        if [ ! -f "$SQUASH_FLAG_FILE" ]; then
            echo $NOW > "$SQUASH_FLAG_FILE"
            echo "首次合并历史提交..."
            python hf_sync.py super_squash "${HF_TOKEN}" "${DATASET_ID}"
        else
            LAST=$(cat "$SQUASH_FLAG_FILE")
            DIFF=$((NOW - LAST))
            if [ $DIFF -ge $SEVEN_DAYS ]; then
                echo $NOW > "$SQUASH_FLAG_FILE"
                echo "超过7天，合并历史提交..."
                python hf_sync.py super_squash "${HF_TOKEN}" "${DATASET_ID}"
            else
                echo "未满7天，跳过合并历史提交。"
            fi
        fi

        SYNC_INTERVAL=${SYNC_INTERVAL:-7200}
        echo "下次同步将在 ${SYNC_INTERVAL} 秒后进行..."
        sleep $SYNC_INTERVAL
    done
}

sync_data &