Home  >  Article  >  Backend Development  >  What is a constructor? What is the reference type?

What is a constructor? What is the reference type?

零下一度
零下一度Original
2017-06-23 15:14:452369browse
Reference type
The constructor is a special method to initialize an instance of a type to a good state. When creating an instance of a reference type, memory is first allocated for the data fields of the instance, and then the additional fields of the object (type object pointer and synchronized block index), and finally calls the type's instance constructor to set the initialization state of the object.
When constructing a reference type object, the memory allocated for the object is always reset to 0 before the instance constructor of the electrophoresis type. All fields that are not explicitly overridden by the constructor are guaranteed to obtain 0 or null values. .
Unlike other methods, the strength constructor can never be inherited, that is to say, the class only has the smooth constructor defined by the class itself. Since an instance constructor can never be inherited, the following modifiers cannot be used on an instance constructor: Virtual, new, override, sealed, and abstract. If the class does not explicitly define any constructor, the C# compiler will default to a default parameterless constructor. In her implementation, it simply calls the parameterless constructor of the base class.
If the class modifier is abstract, the accessibility of the default constructor generated by the compiler is product; otherwise, the constructor will be assigned the public accessible attribute. If the base class does not provide a parameterless constructor, the derived class must explicitly call a base class constructor, otherwise the compiler will report an error. If the class modifier is static (sealed and abstract), the compiler will not generate a default constructor in the class definition at all.
A type can define multiple instance constructors. Each constructor must have a different signature, and each can have different accessible properties. In order to make the code "verifiable", the class's initializer must call the base class's constructor before accessing any fields inherited from the base class. If the derived class constructor does not explicitly call a base class constructor, the C# compiler will automatically generate a call to the default base class constructor. Eventually, System.Object's public no-argument constructor will be called. The constructor does nothing and returns directly. Since System.Object has no instance data fields, its constructor has nothing to do.
In rare cases it is possible to create a type instance without calling the instance constructor. A typical example is the MemberwiseClone method of Object. What this method does is allocate memory, initialize the object's additional fields, and then copy the source object's own data into the new object. In addition, when deserializing an object using a runtime serializer, the same process does not need to call the constructor. Deserialization uses the GetUnnitalizedObject or GetSafeUninitailizedObject method of the System.Runtime.Serialization.FormatterServices type to allocate memory for the object, and a constructor will not be called during the process.
Tips:
Do not call virtual methods in the constructor. The reason is that if the instantiated type overrides the virtual method, the derived type's implementation of the virtual method will be executed, but at this time, the initialization of all fields in the inheritance hierarchy has not been completed (the constructor of the instantiated type not running yet). Therefore, calling virtual methods can lead to unpredictable behavior. Ultimately, this is because when a virtual method is called, the actual type on which the method is executed is not selected until runtime.
Value type (struct) constructor
The value type (struct) constructor works completely differently from the constructor of the reference type (class). The CLR always allows the creation of instances of value types, and there is no way to prevent the instantiation of value types. Therefore, value types do not actually need to define a constructor, and the C# compiler will not inline a default no-argument constructor for value types at all. Let’s look at the following code:
internal struct Point {
public int m_x, m_y;
}
internal sealed class Reactangel
{
public Point m_TopLeft, m_bottomRight;
}
In order to construct a Rectangle, the new operator must be used and the constructor must be specified. In this example, the default constructor automatically generated by the C# compiler is called. Allocate memory for Reatangle, which contains two instances of the Point value type. Considering performance, the CLR will not actively call the constructor for each value type field contained in the reference type. However, as mentioned above, the value type fields will be initialized to 0 or null.
The CLR does allow constructors to be defined for value types, but they must be explicitly called before they are executed.
internal struct Point {
public int m_x, m_y;
public Point(int x, int y)
{
m_x = x;
m_y = y;
}
}
internal sealed class Reactangel
{
public Point m_TopLeft, m_bottomRight;
public Reactangel()
{
this.m_TopLeft = new Point(1 ,2);
this.m_bottomRight = new Point(100,200);
}
}
Instance constructors of value types will only be executed when explicitly called. Therefore, if the constructor of Rectangle does not use the new operator to call the constructor of Point to initialize the m_TopLeft and m_bottomRight fields of Reatangle, then the m_x and m_y fields in both point fields will be 0.
Rewrite the above code:
internal struct Point {
public int m_x, m_y;
public Point()
{
m_x = 5;
m_y = 6;
}
}
internal sealed class Reactangel
{
public Point m_TopLeft, m_bottomRight;
public Reactangel()
{
}
}
Now, when constructing a new Rectangle class, how much will the m_x and m_y fields in the two Point fields be initialized? Is it 0 or 5?
You may think that the C# compiler will generate code in Reactangel's constructor and automatically call Point's default parameterless constructor for Reactangel's two fields. However, to enhance the runtime performance of your application, the C# compiler does not automatically generate such code. In fact, even if a value type provides a parameterless constructor, many compilers will never generate code to call it. In order to implement a value type parameterless constructor, the developer must add code that explicitly calls the value type constructor. But will the two fields of Point' be initialized to 0 for this reason? The result is:
The C# compiler deliberately does not allow value types to define parameterless constructors in order to prevent Developers are confused about when this constructor is called. Since a parameterless constructor cannot be defined, the compiler will never generate code that automatically calls it. Without a parameterless constructor, fields of value types are always initialized to 0 or null.
Type constructor:
Also known as static constructor, class constructor or type initializer. Type constructors can be used with reference types and value types. The purpose of an instance constructor is to set the initial state of an instance of a type. Correspondingly, the role of the type constructor is to set the initial state of the type. The type does not define a type constructor by default. If it is defined, there can only be one. Furthermore, type constructors never have parameters.
internal sealed class SomeRefType {
static SomeRefType()
{
//When accessing for the first time, execute the code here
}
}
internal struct SomeValType
{
static SomeValType()
{
//When accessing for the first time, execute the code here
}
}
It can be seen that defining a type constructor is similar to defining a parameterless instance constructor. The difference is that it must be marked static. Furthermore, type constructors are always private. The reason why it is private is to prevent any developer from writing code to call it, and the CLR is always responsible for his calls.
Tips:
Although you can define a type constructor in a value type, you should never actually do it because the CLR sometimes does not call the static constructor of the value type: for example
internal struct SomeValType
{
static SomeValType()
{
Console.WriteLine("This sentence will never be displayed") ;
}
public int m_x;
}
class Program
{
static void Main( string[] args)
{
SomeValType[] a = new SomeValType[10];
a[0].m_x = 123;
Console.WriteLine(a[0].m_x);
Console.ReadKey();
}
}
The code of the type constructor only Can access the static fields of the type, and its unconventional use is to initialize these fields.

The above is the detailed content of What is a constructor? What is the reference type?. For more information, please follow other related articles on the PHP Chinese website!

Statement:
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn