WCF "Raw" programming model (Web) - receiving arbitrary data

本文介绍如何使用WCF服务接收任意格式的数据文件,包括配置WebContentTypeMapper以正确解析请求。

The previous post mentioned how to return arbitrary data from WCF services. To receive data, however, there is one extra step, which I'll try to explain here.

Like returning arbitrary data, the key for accepting arbitrary (in any format) data is for a method to have a parameter with type System.IO.Stream. This parameter needs to be the single parameter which is passed in the body of the request. By that we mean that the operation can have other parameters beside the Stream one, as long as they're used in the address (UriTemplate) for the operation. For example, this program below will simulate an UploadFile operation:

public class BlogPostRaw2
{
    [
ServiceContract]
    public interface ITest
    {
        [
OperationContractWebInvoke(UriTemplate = "UploadFile/{fileName}")]
        void UploadFile(string fileName, Stream fileContents);
    }
    public class Service : ITest
    {
        public void UploadFile(string fileName, Stream fileContents)
        {
            byte[] buffer = new byte[10000];
            int bytesRead, totalBytesRead = 0;
            do
            {
                bytesRead = fileContents.Read(buffer, 0, buffer.Length);
                totalBytesRead += bytesRead;
            } 
while (bytesRead > 0);
            Console.WriteLine("Uploaded file {0} with {1} bytes", fileName, totalBytesRead);
        }
    }
    public static void Test()
    {
        string baseAddress = "http://" + Environment.MachineName + ":8000/Service";
        ServiceHost host = new ServiceHost(typeof(Service), new Uri(baseAddress));
        host.AddServiceEndpoint(
typeof(ITest), new WebHttpBinding(), "").Behaviors.Add(new WebHttpBehavior());
        host.Open();
        Console.WriteLine("Host opened");

        
HttpWebRequest
 req = (HttpWebRequest)HttpWebRequest.Create(baseAddress + "/UploadFile/Test.txt");
        req.Method = 
"POST";
        req.ContentType = 
"text/plain";
        Stream reqStream = req.GetRequestStream();
        byte[] fileToSend = new byte[12345];
        for (int i = 0; i < fileToSend.Length; i++)
        {
            fileToSend[i] = (
byte)('a' + (i % 26));
        }
        reqStream.Write(fileToSend, 0, fileToSend.Length);
        reqStream.Close();
        HttpWebResponse resp = (HttpWebResponse)req.GetResponse();
        Console.WriteLine("HTTP/{0} {1} {2}", resp.ProtocolVersion, (int)resp.StatusCode, resp.StatusDescription);
        host.Close();
    }
}

Notice that the (POST) HTTP request was sent to http://machine_name:8000/Service/UploadFile/<name_of_the_file_to_be_uploaded>; on the body of the request were the file contents.

One important note about the line in bold about Content-Type: when returning arbitrary data, specifying the content type is advisable, but not (necessarily) required. When sending arbitrary data to WCF, it is required. That is because the WebMessageEncoder (the inner piece of WCF, which is created when the endpoint uses the WebHttpBinding), needs it to be able to decode the message.

So this appears to works fine, until I decided to send a XML file to the server (showing only part of the client code):

HttpWebRequest req = (HttpWebRequest)HttpWebRequest.Create(baseAddress + "/UploadFile/Test.xml");
req.Method = 
"POST";
req.ContentType = 
"text/xml";
Stream reqStream = req.GetRequestStream();
string fileContents = "<hello>world</hello>";
byte[] fileToSend = Encoding.UTF8.GetBytes(fileContents);
reqStream.Write(fileToSend, 0, fileToSend.Length);
reqStream.Close();
HttpWebResponse resp = (HttpWebResponse)req.GetResponse();
Console.WriteLine("HTTP/{0} {1} {2}", resp.ProtocolVersion, (int)resp.StatusCode, resp.StatusDescription);

When this request is sent to the server, the client gets a "400 Bad Request" response. A look at the server traces (enabled via instructions athttp://msdn2.microsoft.com/en-us/library/ms733025.aspx) shows the following error:

System.InvalidOperationException: Incoming message for operation 'UploadFile' (contract 'ITest' with namespace 'http://tempuri.org/') contains an unrecognized http body format value 'Xml'. The expected body format value is 'Raw'. This can be because a WebContentTypeMapper has not been configured on the binding. See the documentation of WebContentTypeMapper for more details.

This is right on target. Basically, as I mentioned before, the WebMessageEncoder is actually composed of three "inner" encoders: XML, JSON and Raw. For content-types which map to the first two, the requests will be processed by then; only if neither the XML or the JSON encoder can process the content-type, the Raw will be used. For the first example it worked fine, since text/plain content cannot be processed by XML or JSON. So we need a way to "force" the encoder to always use Raw. As mentioned in the exception, the WebContentTypeMapper is the solution. The code below can now handle all content-types on request (modified parts in bold). Notice that we need to use a custom binding, since the mapper is only accessible via the WebMessageEncodingBindingElment directly, not in the standard WebHttpBinding.

public class BlogPostRaw2
{
    [
ServiceContract]
    public interface ITest
    {
        [
OperationContractWebInvoke(UriTemplate = "UploadFile/{fileName}")]
        void UploadFile(string fileName, Stream fileContents);
    }
    public class Service : ITest
    {
        public void UploadFile(string fileName, Stream fileContents)
        {
            byte[] buffer = new byte[10000];
            int bytesRead, totalBytesRead = 0;
            do
            {
                bytesRead = fileContents.Read(buffer, 0, buffer.Length);
                totalBytesRead += bytesRead;
            } 
while (bytesRead > 0);
            Console.WriteLine("Uploaded file {0} with {1} bytes", fileName, totalBytesRead);
        }
    }
    public class MyMapper : WebContentTypeMapper
    {
        public override WebContentFormat GetMessageFormatForContentType(string
 contentType)
        {
            return WebContentFormat.Raw; 
// always
        }
    }
    static Binding
 GetBinding()
    {
        CustomBinding result = new CustomBinding(new WebHttpBinding
());
        WebMessageEncodingBindingElement webMEBE = result.Elements.Find<WebMessageEncodingBindingElement
>();
        webMEBE.ContentTypeMapper = 
new MyMapper
();
        return
 result;
    }
    public static void Test()
    {
        string baseAddress = "http://" + Environment.MachineName + ":8000/Service";
        ServiceHost host = new ServiceHost(typeof(Service), new Uri(baseAddress));
        host.AddServiceEndpoint(
typeof(ITest), GetBinding()"").Behaviors.Add(new WebHttpBehavior());
        host.Open();
        Console.WriteLine("Host opened");

        HttpWebRequest req = (HttpWebRequest)HttpWebRequest.Create(baseAddress + "/UploadFile/Test.xml");
        req.Method = 
"POST";
        req.ContentType = 
"text/xml";
        Stream reqStream = req.GetRequestStream();
        string fileContents = "<hello>world</hello>";
        byte[] fileToSend = Encoding.UTF8.GetBytes(fileContents);
        reqStream.Write(fileToSend, 0, fileToSend.Length);
        reqStream.Close();
        HttpWebResponse resp = (HttpWebResponse)req.GetResponse();
        Console.WriteLine("HTTP/{0} {1} {2}", resp.ProtocolVersion, (int)resp.StatusCode, resp.StatusDescription);
        host.Close();
    }
}

【直流微电网】径向直流微电网的状态空间建模与线性化:一种耦合DC-DC变换器状态空间平均模型的方法 (Matlab代码实现)内容概要:本文介绍了径向直流微电网的状态空间建模与线性化方法,重点提出了一种基于耦合DC-DC变换器状态空间平均模型的建模策略。该方法通过对系统中多个相互耦合的DC-DC变换器进行统一建模,构建出整个微电网的集中状态空间模型,并在此基础上实施线性化处理,便于后续的小信号分析与稳定性研究。文中详细阐述了建模过程中的关键步骤,包括电路拓扑分析、状态变量选取、平均化处理以及雅可比矩阵的推导,最终通过Matlab代码实现模型仿真验证,展示了该方法在动态响应分析和控制器设计中的有效性。; 适合人群:具备电力电子、自动控制理论基础,熟悉Matlab/Simulink仿真工具,从事微电网、新能源系统建模与控制研究的研究生、科研人员及工程技术人员。; 使用场景及目标:①掌握直流微电网中多变换器系统的统一建模方法;②理解状态空间平均法在非线性电力电子系统中的应用;③实现系统线性化并用于稳定性分析与控制器设计;④通过Matlab代码复现和扩展模型,服务于科研仿真与教学实践。; 阅读建议:建议读者结合Matlab代码逐步理解建模流程,重点关注状态变量的选择与平均化处理的数学推导,同时可尝试修改系统参数或拓扑结构以加深对模型通用性和适应性的理解。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值