C#断点续传的实现示例

断点续传是一种可以在文件传输过程中出现断电、网络故障等情况时，能够保证传输内容不会全部丢失，而是可以从已传输的位置继续传输的机制。在文件传输较大、较复杂的情况下，使用断点续传可以提高传输质量、稳定性和效率。

在C#中，可以使用HTTP协议的Range头部域来实现断点续传。使用HTTP Range头部域，可以控制取哪个字节范围内的字节。具体实现方法，在HTTP请求头中填写Range头部信息，指明下载区间：

Range: bytes=[start]-[end]
start和end的值为0和文件大小减1，表示下载全部数据；若要实现断点续传，则start的值为当前已下载的数据大小，end的值不变。

在本篇文章中，我们将详细介绍如何使用C#实现HTTP协议的断点续传功能，并提供了完整的代码示例。

实现步骤

C#实现断点续传功能的步骤，简要描述如下：

1.定义HTTP请求，并填写Range头部信息，指明下载区间信息。

2.执行HTTP请求，接收服务端返回的字节流，并将流写入本地文件。

3.检查最终下载文件的大小，与服务端的文件大小是否一致，若不一致则下载失败。

4.上传文件时，同样需制定Range信息，然后发送PUT请求进行上传。

代码实现

我们将使用HttpClient来执行请求，使用FileStream来读写文件。下面是代码实现的详细过程。

1.下载文件

下载文件时，首先需要判断本地是否已经存在相同的文件，如果存在，则需要计算出当前已下载数据的大小（即起始位置startPosition），否则从头开始下载。

下载时，需要在HTTP请求头中填写Range头部信息，指明下载区间。同时，需要注意控制下载缓冲区大小，以避免内存不足的情况。

最后，需要检查下载完成后文件的大小是否与服务端的文件大小一致，若不一致，则下载失败。

代码示例：

private static async Task DownloadFileAsync(Uri uri, string filename, CancellationToken cancellationToken = default)
{
    long startPosition;
    var fileInfo = new FileInfo(filename);
 
    if (fileInfo.Exists)
    {
        startPosition = fileInfo.Length;
        if (startPosition == uri.GetFileSize())
        {
            Console.WriteLine($"The file '{filename}' has already been downloaded.");
            return;
        }
    }
    else
    {
        startPosition = 0;
    }
 
    using var fs = new FileStream(filename, startPosition == 0 ? FileMode.Create : FileMode.Append);
 
    var rangeHeader = new RangeHeaderValue(startPosition, null);
 
    var request = new HttpRequestMessage(HttpMethod.Get, uri);
    request.Headers.Range = rangeHeader;
 
    using var response = await HttpClient.SendAsync(request, HttpCompletionOption.ResponseHeadersRead, cancellationToken);
    var contentLength = response.Content.Headers.ContentLength;
    if (!contentLength.HasValue)
    {
        throw new InvalidOperationException("The server did not provide the content length.");
    }
 
    var totalSize = contentLength.Value + startPosition;
 
    using var stream = await response.Content.ReadAsStreamAsync(cancellationToken);
    await stream.CopyToAsync(fs);
 
    if (fs.Length != totalSize)
    {
        fileInfo.Refresh();
        if (fileInfo.Length < totalSize)
        {
            throw new InvalidOperationException($"The file '{filename}' was not downloaded correctly.");
        }
    }
 
    Console.WriteLine($"The file '{filename}' has been downloaded.");
}

2.上传文件

上传文件与下载文件相似，同样需要在HTTP请求头中填写Range头部信息，以限制上传的范围。同时，需要指定Content-Type，以明确上传数据的类型。

上传文件需要注意的一点是，如果文件较大，则需要分多次上传。可以将文件分割成多个大小相同的片段，逐个上传，确保操作的稳定性和效率。

上传完成后，会收到服务端的响应。如果响应码为2xx，则表示上传成功；否则，表示上传失败。

代码示例：

public static async Task UploadFileAsync(Uri uri, string filename, int bufferSize = 4096, CancellationToken cancellationToken = default)
{
    long startPosition;
    var fileInfo = new FileInfo(filename);
 
    if (fileInfo.Exists)
    {
        startPosition = fileInfo.Length;
    }
    else
    {
        throw new FileNotFoundException("The file was not found.", filename);
    }
 
    using var fs = new FileStream(filename, FileMode.Open);
 
    var rangeHeader = new RangeHeaderValue(startPosition, fs.Length - 1);
 
    var request = new HttpRequestMessage(HttpMethod.Put, uri);
    request.Headers.Range = rangeHeader;
    request.Headers.TryAddWithoutValidation("Content-Type", "application/octet-stream");
 
    var content = new StreamContent(fs, bufferSize);
    request.Content = content;
 
    using var response = await HttpClient.SendAsync(request, cancellationToken);
 
    if (!response.IsSuccessStatusCode)
    {
        var responseMessage = await response.Content.ReadAsStringAsync();
        throw new InvalidOperationException($"Failed to upload file: {response.StatusCode} {responseMessage}");
    }
 
    Console.WriteLine($"The file '{filename}' has been uploaded.");
}

完整代码

上述代码仅为示例，仍然需要加入部分边界检查、异常处理等逻辑，以保证代码的健壮性。下面是完整的实现代码，包含了断点续传功能的完整实现。

using System;
using System.IO;
using System.Net.Http;
using System.Net.Http.Headers;
using System.Threading;
using System.Threading.Tasks;
 
namespace ConsoleApp
{
    internal static class Program
    {
        private static readonly HttpClient HttpClient = new HttpClient();
 
        private static async Task Main(string[] args)
        {
            var uri = new Uri("https://download.visualstudio.microsoft.com/download/pr/26246709-5c10-4383-ad1a-f22f3e8e5e15/23e2d41d2e57b81fc0f9c72068994e70/vc_redist.x64.exe");
 
            var filename = Path.Combine(Path.GetTempPath(), "vc_redist.x64.exe");
 
            Console.WriteLine("Start downloading the file...");
 
            try
            {
                await DownloadFileAsync(uri, filename, CancellationToken.None);
            }
            catch (Exception ex)
            {
                Console.WriteLine($"Failed to download the file: {ex.Message}");
                return;
            }
 
            Console.WriteLine("\nStart uploading the file...\n");
 
            try
            {
                await UploadFileAsync(uri, filename, 4096, CancellationToken.None);
            }
            catch (Exception ex)
            {
                Console.WriteLine($"Failed to upload the file: {ex.Message}");
                return;
            }
 
            Console.WriteLine("Done.");
        }
 
        private static async Task DownloadFileAsync(Uri uri, string filename, CancellationToken cancellationToken = default)
        {
            long startPosition;
            var fileInfo = new FileInfo(filename);
 
            if (fileInfo.Exists)
            {
                startPosition = fileInfo.Length;
                if (startPosition == uri.GetFileSize())
                {
                    Console.WriteLine($"The file '{filename}' has already been downloaded.");
                    return;
                }
            }
            else
            {
                startPosition = 0;
            }
 
            using var fs = new FileStream(filename, startPosition == 0 ? FileMode.Create : FileMode.Append);
 
            var rangeHeader = new RangeHeaderValue(startPosition, null);
 
            var request = new HttpRequestMessage(HttpMethod.Get, uri);
            request.Headers.Range = rangeHeader;
 
            using var response = await HttpClient.SendAsync(request, HttpCompletionOption.ResponseHeadersRead, cancellationToken);
            var contentLength = response.Content.Headers.ContentLength;
            if (!contentLength.HasValue)
            {
                throw new InvalidOperationException("The server did not provide the content length.");
            }
 
            var totalSize = contentLength.Value + startPosition;
 
            using var stream = await response.Content.ReadAsStreamAsync(cancellationToken);
            await stream.CopyToAsync(fs);
 
            if (fs.Length != totalSize)
            {
                fileInfo.Refresh();
                if (fileInfo.Length < totalSize)
                {
                    throw new InvalidOperationException($"The file '{filename}' was not downloaded correctly.");
                }
            }
 
            Console.WriteLine($"The file '{filename}' has been downloaded.");
        }
 
        public static async Task UploadFileAsync(Uri uri, string filename, int bufferSize = 4096, CancellationToken cancellationToken = default)
        {
            long startPosition;
            var fileInfo = new FileInfo(filename);
 
            if (fileInfo.Exists)
            {
                startPosition = fileInfo.Length;
            }
            else
            {
                throw new FileNotFoundException("The file was not found.", filename);
            }
 
            using var fs = new FileStream(filename, FileMode.Open);
 
            var rangeHeader = new RangeHeaderValue(startPosition, fs.Length - 1);
 
            var request = new HttpRequestMessage(HttpMethod.Put, uri);
            request.Headers.Range = rangeHeader;
            request.Headers.TryAddWithoutValidation("Content-Type", "application/octet-stream");
 
            var content = new StreamContent(fs, bufferSize);
            request.Content = content;
 
            using var response = await HttpClient.SendAsync(request, cancellationToken);
 
            if (!response.IsSuccessStatusCode)
            {
                var responseMessage = await response.Content.ReadAsStringAsync();
                throw new InvalidOperationException($"Failed to upload file: {response.StatusCode} {responseMessage}");
            }
 
            Console.WriteLine($"The file '{filename}' has been uploaded.");
        }
    }
 
    public static class UriExtensions
    {
        public static long GetFileSize(this Uri uri)
        {
            using var client = new HttpClient();
            using var response = client.Send(new HttpRequestMessage(HttpMethod.Head, uri));
            var contentLength = response.Content.Headers.ContentLength;
            if (!contentLength.HasValue)
            {
                throw new InvalidOperationException("The server did not provide the content length.");
            }
 
            return contentLength.Value;
        }
    }
}

总结

断点续传功能可以在文件传输的过程中，提高传输质量和效率，确保数据传输的安全性和稳定性。在本文中，我们介绍了C#中实现HTTP协议断点续传的方法，并提供了完整的代码示例。希望读者通过本文的介绍，能够成功实现断点续传功能，并在实际工作中应用到相应的场景中去。

参考文章：http://blog.ncmem.com/wordpress/2023/11/11/c%e6%96%ad%e7%82%b9%e7%bb%ad%e4%bc%a0%e7%9a%84%e5%ae%9e%e7%8e%b0%e7%a4%ba%e4%be%8b/
欢迎入群一起讨论