Question

我正在尝试使用Json.Net序列化二叉树：

class Node
{
    Node Left;
    Node Right;
    int Data;
}

我想在每次Node序列化之前和之后添加自定义代码，因此我尝试使用自定义JsonConverter：

class NodeConverter : JsonConverter
{
    public override bool CanConvert(Type objectType)
    {
        return objectType == typeof(Node);
    }

    public override object ReadJson(JsonReader reader, Type objectType, object existingValue, JsonSerializer serializer)
    {
        throw new NotImplementedException();
    }

    public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
    {
        Console.WriteLine("Before");
        serializer.serialize(writer, value); // Oops
        Console.WriteLine("After");
    }   
}

但是这不起作用，因为JsonSerializer将无休止地调用NodeConverter.WriteJson(..)方法。

这可以在Json.NET中完成吗？

编辑：这就是我使用自定义转换器的方式：

var root = CreateTree(); 
using (var ms = new MemoryStream())
using (var writer = new StreamWriter(ms))
{
    var serializer = JsonSerializer.Create();
    serializer.Converters.Add(new NodeConverter());
    serializer.serialize(writer, root);
}

Answer 1

为了打破循环，您可以直接在WriteJson方法中序列化属性，例如：

public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
{
    Console.WriteLine("Before");
    var n = (Node)value;
    writer.WriteStartObject();
    SerializeReferencedNode(writer, serializer, "Left", n.Left);
    SerializeReferencedNode(writer, serializer, "Right", n.Right);
    writer.WritePropertyName("Value");
    writer.WriteValue(n.Data);
    writer.WriteEndObject();
    Console.WriteLine("After");
}

private static void SerializeReferencedNode(JsonWriter writer, JsonSerializer serializer, 
                                            string propName, Node node)
{
    if (node != null)
    {
        writer.WritePropertyName(propName);
        serializer.Serialize(writer, node);
    }
}

这将导致一个简单的示例，显示以下输出：

{"Left":{"Value":123},"Right":{"Value":246},"Value":0}

在Json.NET中每个树节点序列化之前和之后的自定义代码

1 个答案: