首先说一下使用场景 WebAPI接口入参使用 object和 dynamic 后续解析和处理
在.NET中,object
和dynamic
也有一些区别:
object:object
是.NET中的顶级类,所有类都是object
的子类。在C#中,您可以使用object
关键字来声明变量,该变量可以存储任何类型的对象引用。
dynamic:dynamic
是.NET中的一种类型,它允许您在运行时执行动态绑定。这意味着您可以在编译时避免类型检查,而是在运行时确定变量的类型。这使得代码更加灵活,但也可能导致运行时错误。
在C#中,您可以像这样使用它们:
这里使用 Raccoon 代码助手 给出实例代码: (国产免费ai有时候真智障,仅供参考)
object myObject = "Hello, World!";
dynamic myDynamic = "Hello, Dynamic!";
// 以下代码将引发编译错误,因为myObject的类型在编译时被确定
// myObject = 42;
// 以下代码不会引发编译错误,因为myDynamic的类型在运行时被确定
myDynamic = 42;
由于WebAPI接口接收的参数,来自于前端提交,没有一定的类型检查机制(js弱类型), 再加上有些内容不定(参数里面有单个对象,也有数组之类的)
后端处理 就会显得有点棘手.
在考虑如何解析这些前端传过来的结构不定对象之前,先复习一下基础知识,(掌握必备知识,才能更好的解决问题)
下面代码默认使用 TestDriven 插件执行, 命名空间有
using System.Collections.Generic;
using System.Text;
using System.Text.RegularExpressions;
using Newtonsoft.Json;
using System.Linq;
using System.Linq.Expressions;
using Newtonsoft.Json.Linq;
public static void Test1()
{
var defaultStringValue = new List<string> { "" };
JArray jArrayString = JArray.FromObject(defaultStringValue);
Console.WriteLine(JsonConvert.SerializeObject(defaultStringValue)); // 输出 [""]
Console.WriteLine(JsonConvert.SerializeObject(jArrayString)); // 输出 [""]
var defaultObjValue = new List<object> { new object() };
JArray jArrayObj = JArray.FromObject(defaultObjValue);
Console.WriteLine(JsonConvert.SerializeObject(defaultObjValue)); // 输出 [{}]
Console.WriteLine(JsonConvert.SerializeObject(jArrayObj)); // 输出 [{}]
}
接下来看一个复杂的对象构造示例:
现有如下json格式字符串 ,构造 dynamic 类型
{"Guid":"3eab6205-6a05-4d09-9dd3-7a004369bb32","names":["张三","李四","王五"],"Params":[{"id_list":["0001"]}]}
说明一下, 上面这个对象象 ,有三个属性, Guid 是一个 guid字符串, names 是一个数组, Params 是一个复杂对象(集合)
public static void Test02()
{
string[] strings = new string[] { "张三", "李四", "王五" };
JArray jArray = JArray.FromObject(strings);
JObject jObj_idlist = new JObject();
JArray j_id_list = JArray.FromObject(new string[] { "0001" });
JProperty p_id_list = new JProperty("id_list", j_id_list);
jObj_idlist.Add(p_id_list);
dynamic obj = new
{
Guid = Guid.NewGuid().ToString(),
names = jArray,
Params = JArray.FromObject(new List<object>() { jObj_idlist })
};
Console.WriteLine(JsonConvert.SerializeObject(obj));
}
判断一个dynamic 的值类型是 数组(集合),还是单个的对象 ,这里继续使用上面的 dynamic obj 对象
Console.WriteLine(obj.Guid is JArray); // False
Console.WriteLine(obj.names is JArray); // True
Console.WriteLine(obj.Params is JArray); // True
public class StudentInfo
{
public int Id { get; set; }
public string Name { get; set; }
public DateTime Birthday { get; set; }
public int? Age
{
get
{
if (Birthday.HasValue)
{
int year = (DateTime.Now - this.Birthday.Value).Days;
return year / 365;
}
else
{
return null;
}
}
}
}
public class TestClass
{
public static void Test1()
{
StudentInfo stu1 = new StudentInfo() { Id = 1, Name = "张三", Birthday = new DateTime(2000,1,1)};
Console.WriteLine(JsonConvert.SerializeObject(stu1));
// 输出 {"Id":1,"Name":"张三","Birthday":"2000-01-01T00:00:00","Age":24}
}
}
加入这个参数 {"Id":1,"Name":"张三","Birthday":"2000-01-01T00:00:00","Age":24} 从前端传入, 那么通过 StudentInfo 可以直接接收,
但是如果 不知道这个人的出生年月日,传入的 字符串为 下面这几种情况
{"Id":1,"Name":"张三","Birthday":""} // 反序列化失败
或者 {"Id":1,"Name":"张三","Birthday":"null"} , // 反序列化失败
或者 {"Id":1,"Name":"张三","Birthday":null} // 可成功序列化
public static void Test2()
{
string json1 = "{\"Id\":1,\"Name\":\"张三\",\"Birthday\":\"\"} ";
StudentInfo stu1 = JsonConvert.DeserializeObject<StudentInfo>(json1); // 报错
Console.WriteLine(JsonConvert.SerializeObject(stu1));
}
public static void Test3()
{
string json1 = "{\"Id\":1,\"Name\":\"张三\",\"Birthday\":\"null\"} ";
StudentInfo stu1 = JsonConvert.DeserializeObject<StudentInfo>(json1); // 报错
Console.WriteLine(JsonConvert.SerializeObject(stu1));
}
public static void Test4()
{
string json1 = "{\"Id\":1,\"Name\":\"张三\",\"Birthday\":null} ";
StudentInfo stu1 = JsonConvert.DeserializeObject<StudentInfo>(json1); // 可正常解析
Console.WriteLine(JsonConvert.SerializeObject(stu1));
// 输出 {"Id":1,"Name":"张三","Birthday":null,"Age":null}
}
为了接口不报错, 这个先用 dynamic obj 接收这个入参, 然后近一步处理
using System;
using System.Reflection;
using Newtonsoft.Json.Linq;
namespace Common.Helper
{
/// <summary>
/// dynamic 数据映射
/// </summary>
/// <typeparam name="T">目标实体类</typeparam>
public class MapHelper<T> where T : new()
{
/// <summary>
/// 将数据源 转换成对应的 目标对象(处理可空类型, 以及 补全缺失的字段)
/// </summary>
/// <param name="obj">dynamic 数据源对象</param>
/// <returns></returns>
public static dynamic DynamicFormat(dynamic obj)
{
JObject jobj = (Newtonsoft.Json.Linq.JObject)obj;
T model = new T();//不关心对象T的值, 只关心对象T的字段的类型
Type ts = model.GetType();
//反射, 获取 T的所有 属性
PropertyInfo[] PropertyList = ts.GetProperties();
foreach (PropertyInfo item in PropertyList)
{
string colName = item.Name;
string typeName = "";
if (item.PropertyType.IsGenericType && item.PropertyType.GetGenericTypeDefinition() == typeof(Nullable<>))
{
//如果是 Nullable<T>, 直接返回 T, 例如 DateTime? 返回 DateTime
Type[] arr = item.PropertyType.GetGenericArguments();
var columnType = arr[0];
typeName = columnType.ToString();
// Console.WriteLine($"{colName}的 目标类型是Nullable<{typeName}>");
}
else
{
//普通类型, int,string,datetime 等
typeName = item.PropertyType.Name;
}
//Console.WriteLine("字段:" + colName);
//Console.WriteLine("类型:" + typeName);
#region 尝试类型转换
var s = jobj.GetValue(colName);// {ID:"10"}
if (s == null)
{
jobj[colName] = null;
continue;
}
object va = ((Newtonsoft.Json.Linq.JValue)s).Value;//"10"
if (va == null)
{
jobj[colName] = null;
continue;
}
else
{
if (typeName == "System.Int32")
{
int t;
bool parseRes = int.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
jobj[colName] = null;
}
}
else if (typeName == "System.Decimal")
{
Decimal t;
bool parseRes = decimal.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
jobj[colName] = null;
}
}
else if (typeName == "System.Double")
{
double t;
bool parseRes = double.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
jobj[colName] = null;
}
}
else if (typeName == "System.Int64")
{
long t;
bool parseRes = long.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
jobj[colName] = null;
}
}
else if (typeName == "System.DateTime")
{
DateTime t;
bool parseRes = DateTime.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
DateTime? init = null;
jobj[colName] = init;
}
}
else if (typeName == "System.Int16")
{
short t;
bool parseRes = short.TryParse(va.ToString(), out t);
if (parseRes)
{
jobj[colName] = t;
}
else
{
jobj[colName] = null;
}
}
}
#endregion
}
return jobj;
}
public static dynamic DynamicFormatList(dynamic obj)
{
JArray jArray = new JArray();
foreach (dynamic item in obj)
{
JObject jObject = JObject.FromObject(item);
jArray.Add(DynamicFormat(jObject));
}
return jArray;
}
}
}
public static void TestNew2()
{
JObject jobj_stu = new JObject();
JProperty ip_id = new JProperty("Id", 1);
JProperty ip_name = new JProperty("Name", "张三");
JProperty ip_birthday = new JProperty("Birthday", "");
jobj_stu.Add(ip_id);
jobj_stu.Add(ip_name);
jobj_stu.Add(ip_birthday);
string json_ori = JsonConvert.SerializeObject(jobj_stu);
Console.WriteLine("初始JSON:"+json_ori);
// 输出 初始JSON:{"Id":1,"Name":"张三","Birthday":""}
dynamic objNew = MapHelper<StudentInfo>.DynamicFormat(jobj_stu);
string json_new = JsonConvert.SerializeObject(objNew);
Console.WriteLine("处理后JSON:" + json_new);
// 输出 处理后JSON:{"Id":1,"Name":"张三","Birthday":null,"Age":null}
StudentInfo stu1 = JsonConvert.DeserializeObject<StudentInfo>(json_new);
Console.WriteLine("对象序列化后JSON:"+JsonConvert.SerializeObject(stu1));
// 输出 对象序列化后JSON:{"Id":1,"Name":"张三","Birthday":null,"Age":null}
}
public static void TestNew3()
{
JObject jobj_stu = new JObject();
JProperty ip_id = new JProperty("Id", 1);
JProperty ip_name = new JProperty("Name", "张三");
JProperty ip_birthday = new JProperty("Birthday", "null");
jobj_stu.Add(ip_id);
jobj_stu.Add(ip_name);
jobj_stu.Add(ip_birthday);
string json_ori = JsonConvert.SerializeObject(jobj_stu);
Console.WriteLine("初始JSON:"+json_ori);
// 输出 初始JSON:{"Id":1,"Name":"张三","Birthday":"null"}
dynamic objNew = MapHelper<StudentInfo>.DynamicFormat(jobj_stu);
string json_new = JsonConvert.SerializeObject(objNew);
Console.WriteLine("处理后JSON:" + json_new);
// 输出 处理后JSON:{"Id":1,"Name":"张三","Birthday":null,"Age":null}
StudentInfo stu1 = JsonConvert.DeserializeObject<StudentInfo>(json_new);
Console.WriteLine("对象序列化后JSON:"+JsonConvert.SerializeObject(stu1));
// 输出 对象序列化后JSON:{"Id":1,"Name":"张三","Birthday":null,"Age":null}
}
除了使用 dynamic 和 JObject 之类的处理方式, 还有 使用 Dictionary<string,object>的处理方式,
var dict = JsonConvert.DeserializeObject<Dictionary<string, object>>(json_ori);
比如有如下json结构的参数
{"Id":1,"Name":"张三","Birthday":"null","Extends":[[]],"Extends2":[[[]],[[]]]}
添加了 Extends 和 Extends2
一眼看上去 好像不是很好构造,下面给出 C#代码 示例
public static void TestNew5()
{
JObject jobj_stu = new JObject();
JProperty ip_id = new JProperty("Id", 1);
JProperty ip_name = new JProperty("Name", "张三");
JProperty ip_birthday = new JProperty("Birthday", "null");
var l1 = new List<object>() { new List<object>() { }};
JProperty jp_extends = new JProperty("Extends", JArray.FromObject(l1) );
dynamic p = new List<JArray>() { JArray.FromObject(l1), JArray.FromObject(l1) };
JProperty jp_extends2 = new JProperty("Extends2", p);
jobj_stu.Add(ip_id);
jobj_stu.Add(ip_name);
jobj_stu.Add(ip_birthday);
jobj_stu.Add(jp_extends);
jobj_stu.Add(jp_extends2);
string json_ori = JsonConvert.SerializeObject(jobj_stu);
Console.WriteLine("JSON:" + json_ori);
// 输出 {"Id":1,"Name":"张三","Birthday":"null","Extends":[[]],"Extends2":[[[]],[[]]]}
var dict = JsonConvert.DeserializeObject < Dictionary<string, object>>(json_ori);
Console.WriteLine("JSON:" + JsonConvert.SerializeObject(dict));
// 输出 {"Id":1,"Name":"张三","Birthday":"null","Extends":[[]],"Extends2":[[[]],[[]]]}
}
关于 object,dynamic ,以及 配合使用 Newtonsoft.Json 库 就先介绍到这里.