.NET6 WebAPI下载PDF返回JSON未自动下载，需如何修改？

我正在学习使用C# .NET6 WebAPI下载PDF文件的教程，已完全按照教程步骤操作，但得到了与教程不同的响应。点击链接后没有自动下载PDF，而是返回了一个JSON对象：

{"version":"1.1","content":{"headers":[{"key":"Content-Disposition","value":["attachment; filename=FileTest.pdf"]},{"key":"Content-Type","value":["application/pdf"]},{"key":"Content-Length","value":["362671"]}]},"statusCode":200,"reasonPhrase":"OK","headers":[],"trailingHeaders":[],"requestMessage":null,"isSuccessStatusCode":true}

原代码如下：

[HttpGet]
[Route("GetPdf")]
public HttpResponseMessage Get()
{
    HttpResponseMessage result = new HttpResponseMessage(System.Net.HttpStatusCode.OK);

    string pdfLocation = _webHostEnvironment.WebRootPath + "\\PDF\\Biletas.pdf";

    var stream = new MemoryStream(System.IO.File.ReadAllBytes(pdfLocation));
    stream.Position = 0;

    if (stream == null)
    {
        result = new HttpResponseMessage(System.Net.HttpStatusCode.BadRequest);
        return result;
    }

    result.Content = new StreamContent(stream);
    result.Content.Headers.ContentDisposition = new System.Net.Http.Headers.ContentDispositionHeaderValue("attachment");

    result.Content.Headers.ContentDisposition.FileName = "FileTest.pdf";
    result.Content.Headers.ContentType = new System.Net.Http.Headers.MediaTypeHeaderValue("application/pdf");

    result.Content.Headers.ContentLength = stream.Length;


    return result;
}

问题根源 #

在.NET Core/.NET 6+中，直接返回HttpResponseMessage会被框架自动序列化为JSON，而非作为原生HTTP响应发送给客户端——这是ASP.NET Core与旧版ASP.NET Web API的核心行为差异。

两种修改方案 #

方案1：使用FileStreamResult返回（推荐） #

ASP.NET Core提供了专门的文件返回类型，更贴合框架设计，代码更简洁：

using Microsoft.AspNetCore.Mvc;
using System.IO;

[HttpGet]
[Route("GetPdf")]
public IActionResult Get()
{
    string pdfLocation = Path.Combine(_webHostEnvironment.WebRootPath, "PDF", "Biletas.pdf");

    if (!System.IO.File.Exists(pdfLocation))
    {
        return BadRequest();
    }

    var stream = new FileStream(pdfLocation, FileMode.Open, FileAccess.Read);
    return File(stream, "application/pdf", "FileTest.pdf");
}

• 用Path.Combine替代硬编码的路径拼接，避免跨平台路径分隔符问题
• 提前检查文件是否存在，比判断流是否为空更可靠
• File()方法会自动配置Content-Disposition、Content-Type等响应头，无需手动设置

方案2：适配HttpResponseMessage（兼容旧代码） #

如果需要保留HttpResponseMessage的写法，需要用ResponseMessageResult包装返回：

using Microsoft.AspNetCore.Mvc;
using Microsoft.AspNetCore.Mvc.WebApiCompatShim;
using System.IO;

[HttpGet]
[Route("GetPdf")]
public IActionResult Get()
{
    HttpResponseMessage result = new HttpResponseMessage(System.Net.HttpStatusCode.OK);

    string pdfLocation = Path.Combine(_webHostEnvironment.WebRootPath, "PDF", "Biletas.pdf");

    if (!System.IO.File.Exists(pdfLocation))
    {
        result = new HttpResponseMessage(System.Net.HttpStatusCode.BadRequest);
        return new ResponseMessageResult(result);
    }

    var stream = new MemoryStream(System.IO.File.ReadAllBytes(pdfLocation));
    stream.Position = 0;

    result.Content = new StreamContent(stream);
    result.Content.Headers.ContentDisposition = new System.Net.Http.Headers.ContentDispositionHeaderValue("attachment");
    result.Content.Headers.ContentDisposition.FileName = "FileTest.pdf";
    result.Content.Headers.ContentType = new System.Net.Http.Headers.MediaTypeHeaderValue("application/pdf");
    result.Content.Headers.ContentLength = stream.Length;

    return new ResponseMessageResult(result);
}

• 需要确保项目已引用Microsoft.AspNetCore.Mvc.WebApiCompatShim NuGet包
• ResponseMessageResult会告诉框架将HttpResponseMessage转换为原生HTTP响应，而非序列化

内容的提问来源于stack exchange，提问作者Modestas Vacerskas