为什么在 XML 序列化期间将名称空间添加到子节点？

Question

我有一个 class 为 XML 序列化装饰，如下所示：

[XmlRoot(ElementName = "XXX", Namespace = "http://www.example.com/Schemas/xxx/2011/11")]
public class Xxx<T> where T: Shipment
{
    [XmlAttribute("version")]
    public string Version = "1.1";

    public T Shipment { get; set; }

    public Xxx(string dataTargetType)
    {
        Shipment = (T)Activator.CreateInstance(typeof(T));
        Shipment.DataContext = new DataContext
        {
            DataTargetCollection = new DataTargetCollection
            {
                DataTarget = new DataTarget
                {
                    Type = dataTargetType
                }
            }
        };
    }
}

[XmlType("Shipment")]
public class Shipment
{
    public DataContext DataContext { get; set; }
}

序列化后输出如下 XML:

<?xml version="1.0" encoding="utf-8"?>
<XXX xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema">
    <Shipment xmlns="http://www.example.com/Schemas/xxx/2011/11">
        <DataContext>
            <DataTargetCollection>
                <DataTarget>
                    <Type>WarehouseOrder</Type>
                </DataTarget>
            </DataTargetCollection>
        </DataContext>
    </Shipment>
</XXX>

为什么将 xmlns 名称空间属性添加到 Shipment 节点而不是根 XXX 节点？

它被继承和序列化的使用示例：（解决序列化问题时的人为示例）

public class XxxVariation: Xxx<Shipment>
{
    public const string DataTargetType = "Something";

    public XxxVariation() : base(DataTargetType) {}
}

public async Task<string> CreateXxxVariationAsync(string todo)
{
    var request = new XxxVariation();

    string xmlRequest = SerializeRequest(request);
    return await PostRequestAsync(xmlRequest);
}

private static string SerializeRequest<T>(T request)
{
    using (var stream = new MemoryStream())
    {
        var serializer = new XmlSerializer(typeof(T));
        serializer.Serialize(XmlWriter.Create(stream), request);

        using (var reader = new StreamReader(stream))
        {
            stream.Seek(0, SeekOrigin.Begin);
            string xml = reader.ReadToEnd();
            return xml;
        }
    }
}

Answer 1

基于 Xxx<T> 没有 public 无参数构造函数这一事实，我怀疑你实际拥有的是这样的：

public class FooXxx : Xxx<Foo> {
    public FooXxx() : base("abc") { }
}

并且您正在序列化 FooXxx 实例，并在创建 XmlSerializer 时使用 typeof(FooXxx)。那是......好吧，我猜，理论上 XmlRootAttribute 是可继承的，但我怀疑属性检查代码在检查属性时明确使用 "only on the declared type, not inherited" 选项。因此，您似乎需要在 FooXxx:

上再次添加属性

[XmlRoot(ElementName = "XXX", Namespace = "http://www.example.com/Schemas/xxx/2011/11")]
public class FooXxx : Xxx<Foo> {
    public FooXxx() : base("abc") { }
}

注意：我不得不在这里推断很多；如果这没有帮助，请 post 您用于序列化 的实际代码，包括您的 XmlSerializer 初始化，并显示您创建的对象传入序列化器。

为什么在 XML 序列化期间将名称空间添加到子节点？

Why is the namespace being added to a child node during XML serialization?

c#

xml

xml-serialization

xmlserializer