POST方法从Blob Storage下载Buffer生成空白PDF求助

AI 小北

2026-6-16

问题描述

使用POST方法从Azure Blob Storage下载PDF到Buffer时，先出现Content-Length不匹配错误（写入216523字节，预期215706字节），修复后能成功触发下载，但打开PDF后内容为空白。因隐私要求代码已脱敏，且无法使用pipe（Postman请求耗时过长），寻求可行解决方案。

相关代码

app.post(
    "/api/**client-endpoint**/download/attachments",
    async (req, res, next) => {
        try {
            let claims = await authentication.authJwtToken(req, req.context);
            req.context.log("Validated claims: ", JSON.stringify(claims));
            let allowedPermissions = [permissionConstants.PERMISSION_ADMIN_CUSTOMER_VIEW];
            let auth = await authorization.authorize(req.context, claims, permissionConstants.AUTH_MODE_CLIENT, allowedPermissions, req.body.clientId, undefined, undefined, constants.AUTH_TOKEN_TYPE_ADMIN);

            let client = await coreClientDao.mssqlGetClientByIdAuthObject(constants.**CLIENTNAME**_CLIENT_ID, req.context);

            const { clientId, customerId, fileUrl } = req.body;
            
            if (!fileUrl) {
                return res.status(400).send({ message: 'fileUrl is required.' });
            }

            req.context.log("Processing file for clientId: ", clientId);

            let blobUrl = fileUrl;

            if (!blobUrl) {
                req.context.log(`Skipping attachment, URL is not yet ready.`, new Date().toString());
                return res.status(400).send({ message: 'Attachment is not ready.' });
            } else {
                let sharedKeyCredential = new StorageSharedKeyCredential(utility.dec(client.storageName), utility.dec(client.storageKey));
                let blobServiceClient = new BlobServiceClient(
                    `https://${utility.dec(client.storageName)}.blob.core.windows.net`,
                    sharedKeyCredential
                );

                let urlParts = new URL(blobUrl);
                let pathParts = urlParts.pathname.split('/');
                let blobName = pathParts.slice(2).join('/'); 

                let sourceBlob = blobServiceClient.getContainerClient(constants.BLOB_CONTAINER_ATTACHMENT).getBlobClient(blobName);

                req.context.log('Downloading blob:', blobName);

                const pdfFile = await sourceBlob.downloadToBuffer({
                    onProgress: (progress) => {
                      req.context.log(`Downloaded ${progress.loadedBytes} bytes`);
                    }
                  });
                  
                res.setHeader('Content-Type', 'application/pdf');
                res.setHeader('Content-Disposition', `attachment; filename="${blobName}"`);
                res.set('Transfer-Encoding', 'identity');
                res.send(pdfFile);
            }
        } catch (error) {
            utility.catchErrorRespnse(res, req.context, error);
        }
    }
);

解决方案建议

验证Buffer完整性：在downloadToBuffer后添加日志，打印pdfFile.length，对比Blob存储中目标文件的实际大小。如果长度不匹配，说明下载的Buffer不完整或指向错误文件。
修正Blob名称解析：URL路径可能包含编码字符（如空格转%20），需解码后再拼接Blob名称：
```
let blobName = decodeURIComponent(pathParts.slice(2).join('/'));
```
移除强制传输编码设置：注释掉res.set('Transfer-Encoding', 'identity');，让Express自动处理响应头，避免手动设置导致内容传输异常。
确认Blob文件属性：通过sourceBlob.getProperties()获取文件的contentType，确保存储的PDF文件类型为application/pdf，避免内容格式不匹配。

优化下载参数：给downloadToBuffer添加重试参数，提升下载可靠性：

const pdfFile = await sourceBlob.downloadToBuffer({
    onProgress: (progress) => {
      req.context.log(`Downloaded ${progress.loadedBytes} bytes`);
    },
    maxRetryRequests: 3
  });

替换响应方法：将res.send(pdfFile)改为res.end(pdfFile)，直接输出Buffer，避免Express对内容做额外处理导致损坏。

内容的提问来源于stack exchange，提问作者hellblazer