文件很大时如何在WebAPI中读取文件块

Question

我有一个大文件，我想将它发送到 Web API，Web API 将它发送到亚马逊。由于文件很大，我想将文件分块发送到亚马逊。 因此，如果我有 1 GB 的文件，我希望我的 API 以 20 MB 的块接收文件，以便我可以将其发送到 Amazon，然后再次接收 20 MB 的块。这怎么可能。以下是我的尝试。

public async Task<bool> Upload()
{
    var fileuploadPath = ConfigurationManager.AppSettings["FileUploadLocation"];
    var provider = new MultipartFormDataStreamProvider(fileuploadPath);
    var content = new StreamContent(HttpContext.Current.Request.GetBufferlessInputStream(true));
    // Now code below writes to a folder, but I want to make sure I read it as soon as I receive some chunk
    await content.ReadAsMultipartAsync(provider);
    return true;
}

伪代码：

While (await content.ReadAsMultipartAsync(provider) == 20 MB chunk)
{
  //Do something
  // Then again do something with rest of chunk and so on.
}

文件最大为 1 GB。

到目前为止，这行代码正在发送整个文件：

await content.ReadAsMultipartAsync(provider);

我在这里迷路了，请帮助我。我想要的只是接收小块文件并处理它。

P.S：我正在从 Postman 发送文件作为 MultiPart/Form-Data 进行测试。

第 2 次尝试：

var filesReadToProvider = await Request.Content.ReadAsMultipartAsync();
foreach (var content in filesReadToProvider.Contents)
{
   var stream = await content.ReadAsStreamAsync();
   using (StreamReader sr = new StreamReader(stream))
   {
       string line = "";
       while ((line = sr.ReadLine()) != null)
       {
           using (MemoryStream outputStream = new MemoryStream())
           using (StreamWriter sw = new StreamWriter(outputStream))
           {
               sw.WriteLine(line);
               sw.Flush();
               // Do Something
           }
       }
   }
}

Answer 1

没有时间对此进行测试，但 ReadBlock 方法似乎是您想要使用的。 应该看起来像我下面的内容，但它假设您的所有其他代码都很好，并且您只需要一些缓冲帮助。这是一个“阻塞”读取操作，但还有一个返回 Task 的 ReadBlockAsync 方法。

const int bufferSize= 1024;
var filesReadToProvider = await Request.Content.ReadAsMultipartAsync();
foreach (var content in filesReadToProvider.Contents)
{
    var stream = await content.ReadAsStreamAsync();
    using (StreamReader sr = new StreamReader(stream))
    {
        int bytesRead;
        char[] buffer = new char[bufferSize];
        while ((bytesRead = sr.ReadBlock(buffer, 0, bufferSize)) > 0)
        {
              // Do something with the first <bytesRead> of buffer and
              // not with <bufferSize> as <bytesRead> will contain the
              // number of bytes actually read by the call to ReadBlock
        }
    }
}