如何用Microsoft Graph API的/lists端点获取SharePoint指定路径的文件和文件夹

核心思路 #

1. 先通过你熟悉的/drives端点获取指定路径文件夹的唯一ID
2. 在/lists请求中添加$filter条件，筛选ParentFolderId等于目标文件夹ID的列表项，精准定位指定路径下的内容
3. 利用列表项的FileSystemObjectType字段区分文件（值为0）和文件夹（值为1）

修改后的完整代码 #

import asyncio
import aiohttp

async def get_target_folder_id(session, accessToken, siteId, driveId, target_path):
    """通过/drives端点获取指定路径文件夹的ID"""
    url = f"https://graph.microsoft.com/v1.0/sites/{siteId}/drives/{driveId}/root:/{target_path}:"
    headers = {
        'Authorization': f'Bearer {accessToken}',
        'Accept': 'application/json'
    }
    async with session.get(url, headers=headers) as response:
        if response.status != 200:
            print(f"获取目标文件夹ID失败: {response.status}")
            return None
        folder_data = await response.json()
        return folder_data.get('id')

async def get_all_items_in_target_path(accessToken, siteId, libraryId, driveId, target_path, batch_size=15):
    """获取指定路径下的文件和文件夹"""
    async with aiohttp.ClientSession() as session:
        # 第一步：获取目标文件夹ID
        folder_id = await get_target_folder_id(session, accessToken, siteId, driveId, target_path)
        if not folder_id:
            print("无法获取目标文件夹ID，终止操作")
            return

        # 构造带过滤条件的lists请求URL
        filter_str = f"ParentFolderId eq '{folder_id}'"
        url = f"https://graph.microsoft.com/v1.0/sites/{siteId}/lists/{libraryId}/items?top={batch_size}&$filter={filter_str}&$expand=fields"
        headers = {
            'Authorization': f'Bearer {accessToken}',
            'Accept': 'application/json',
            'Prefer': 'HonorNonIndexedQueriesWarningMayFailRandomly'
        }

        while url:
            async with session.get(url, headers=headers) as response:
                if response.status != 200:
                    print(f"请求失败: {response.status}")
                    # 处理节流信息
                    retry_after = response.headers.get('Retry-After')
                    throttle_info = {
                        'Retry-After': retry_after,
                        'Throttle Limit Percentage': response.headers.get('x-ms-throttle-limit-percentage'),
                        'Throttle Scope': response.headers.get('x-ms-throttle-scope'),
                        'Throttle Reason': response.headers.get('x-ms-throttle-reason')
                    }
                    for k, v in throttle_info.items():
                        if v:
                            print(f"{k}: {v}")
                    break

                data = await response.json()

            items = data.get('value', [])
            if not items:
                break

            # 处理当前批次的项：区分文件/文件夹，过滤图片
            processed_items = []
            for item in items:
                fields = item.get('fields', {})
                item_type = fields.get('FileSystemObjectType')
                web_url = item.get('webUrl', '')
                item_info = {
                    'id': item['id'],
                    'name': fields.get('FileLeafRef'),
                    'webUrl': web_url,
                    'is_folder': item_type == 1,
                    'is_file': item_type == 0
                }
                # 保留图片文件（如果需要）
                if item_type == 0 and web_url.lower().endswith(('.jpg', '.jpeg', '.png', '.gif')):
                    processed_items.append(item_info)
                # 如果需要保留文件夹，取消下面的注释
                # elif item_type == 1:
                #     processed_items.append(item_info)

            if processed_items:
                yield processed_items

            await asyncio.sleep(0.1)
            url = data.get('@odata.nextLink')

# 使用示例（取消注释即可运行）
# async def main():
#     access_token = "你的access token"
#     site_id = "你的site ID"
#     library_id = "你的文档库列表ID"
#     drive_id = "你的文档库drive ID"
#     target_path = "General/index/images/animals"
#     async for batch in get_all_items_in_target_path(access_token, site_id, library_id, drive_id, target_path):
#         print(batch)

# if __name__ == "__main__":
#     asyncio.run(main())

关键修改说明 #

• 新增文件夹ID获取逻辑：get_target_folder_id函数复用你熟悉的/drives端点，快速定位目标文件夹的ID
• 精准过滤路径：通过ParentFolderId eq '{folder_id}'的filter条件，确保只获取指定路径下的直接子项
• 字段复用优化：请求时添加$expand=fields，一次性获取FileSystemObjectType（区分文件/文件夹）、FileLeafRef（文件名/文件夹名）等字段，无需额外调用API获取详情
• 灵活筛选：保留了原有的图片格式过滤逻辑，同时可通过注释开启文件夹的返回

内容的提问来源于stack exchange，提问作者the programmer