将.NET6进程内Azure函数迁移至.NET7独立进程：IBinder替代方案

首先明确：.NET7独立进程（Isolated Worker）模型不支持依赖注入IBinder，这是你运行时binder始终为null的核心原因。进程外模型更依赖强类型绑定或手动调用Azure服务客户端，以下是两种可行实现方案：

方案一：使用Blob输入绑定+自定义转换器 #

通过自定义输入转换器自动将文件名转为大写，贴合Azure函数的绑定模型：

1. 实现文件名转大写转换器 #

public class UpperCaseBlobNameConverter : IConverter<string, BlobClient>
{
    private readonly BlobServiceClient _blobServiceClient;

    public UpperCaseBlobNameConverter(BlobServiceClient blobServiceClient)
    {
        _blobServiceClient = blobServiceClient;
    }

    public BlobClient Convert(string input)
    {
        var upperCaseFilename = input?.ToUpperInvariant() ?? string.Empty;
        var containerClient = _blobServiceClient.GetBlobContainerClient("your-container-name");
        return containerClient.GetBlobClient(upperCaseFilename);
    }
}

2. 编写HttpTrigger函数 #

[Function("GetBlob")]
public static async Task<HttpResponseData> Run(
    [HttpTrigger(AuthorizationLevel.Function, "get", Route = "getblob/{filename}")] HttpRequestData req,
    [Blob("{filename}", FileAccess.Read, ConverterType = typeof(UpperCaseBlobNameConverter))] BlobClient blobClient,
    ILogger log)
{
    var response = req.CreateResponse(HttpStatusCode.OK);
    
    if (!await blobClient.ExistsAsync())
    {
        response.StatusCode = HttpStatusCode.NotFound;
        await response.WriteStringAsync("Blob不存在");
        return response;
    }

    using var stream = await blobClient.OpenReadAsync();
    await stream.CopyToAsync(response.Body);
    return response;
}

3. 注册Blob服务客户端到DI #

在Program.cs中添加：

var host = new HostBuilder()
    .ConfigureFunctionsWorkerDefaults()
    .ConfigureServices(services =>
    {
        services.AddAzureClients(clientBuilder =>
        {
            clientBuilder.AddBlobServiceClient(Environment.GetEnvironmentVariable("AzureWebJobsStorage"));
        });
    })
    .Build();

host.Run();

方案二：手动注入BlobServiceClient（更灵活） #

直接通过DI注入Blob客户端，手动处理文件名大写和Blob操作，适合动态逻辑较多的场景：

1. 编写函数类（构造注入BlobServiceClient） #

public class GetBlobFunction
{
    private readonly BlobServiceClient _blobServiceClient;

    public GetBlobFunction(BlobServiceClient blobServiceClient)
    {
        _blobServiceClient = blobServiceClient;
    }

    [Function("GetBlob")]
    public async Task<HttpResponseData> Run(
        [HttpTrigger(AuthorizationLevel.Function, "get")] HttpRequestData req,
        ILogger log)
    {
        // 从查询参数获取文件名
        if (!req.Query.TryGetValue("filename", out var filenameValues) || string.IsNullOrEmpty(filenameValues.FirstOrDefault()))
        {
            var badResponse = req.CreateResponse(HttpStatusCode.BadRequest);
            await badResponse.WriteStringAsync("请提供filename查询参数");
            return badResponse;
        }

        // 文件名转为大写
        var upperCaseFilename = filenameValues.First().ToUpperInvariant();
        var containerClient = _blobServiceClient.GetBlobContainerClient("your-container-name");
        var blobClient = containerClient.GetBlobClient(upperCaseFilename);

        if (!await blobClient.ExistsAsync())
        {
            var notFoundResponse = req.CreateResponse(HttpStatusCode.NotFound);
            await notFoundResponse.WriteStringAsync("Blob不存在");
            return notFoundResponse;
        }

        var response = req.CreateResponse(HttpStatusCode.OK);
        using var stream = await blobClient.OpenReadAsync();
        await stream.CopyToAsync(response.Body);
        return response;
    }
}

2. 同样需要在Program.cs注册BlobServiceClient #

和方案一的DI注册代码一致。

关于你的疑问解答 #

• 自定义属性：无需额外创建，使用内置Blob绑定属性配合自定义转换器即可满足需求。
• URL转大写中间件：没必要针对整个URL处理，仅需单独转换文件名参数即可，上述方案已覆盖该逻辑。
• 注册DI服务：必须注册BlobServiceClient，这是进程外模型访问Azure Blob存储的标准方式。

内容的提问来源于stack exchange，提问作者Paul Meems