Destructors in C#


Introduction

Before reading this post I will recommend you to read Constructors in C# post. Dot Net (.NET) framework has an in built mechanism known as Garbage Collection to deallocate memory occupied by the unused objects. In other word, destructor is something like method and constructor except that the runtime calls it after the last reference to object has disappeared. A destructor same name as the name of the class but it starts with the character ~.

using System;
class ClassRoom
{
    private int boycount;       //field
    public ClassRoom()          //default constructor
    {
        boycount = 30;
    }
    ~ ClassRoom()
    {
        boycount = 0;
    }
    static void Main(string[] args)
    {
        ClassRoom r;                //creating a class variable, object
        r = new ClassRoom();        //initializing class variable
                                    //or direct use "ClassRoom r = new ClassRoom();"
        int x = r.boycount;
        Console.WriteLine(x);
        Console.ReadKey();
    }
}

In above example, we have a constructor and a destructor named ClassRoom. Remember class name, constructor name and destructor names are always same in general situations, they might be different sometimes.

There are some restrictions apply to destructors:

(i) Destructors cannot be defined in struct. They are only apply to reference types. We cannot declare a destructor in a value type such as struct. For example:

struct ClassRoom
{
     ~ClassRoom() {….}    //compile-time error will appear
}

(ii) A class can only have one destructor. A single destructor will flush all unused memory. For example:

class ClassRoom
{
    ....
    ~ ClassRoom()
    {
        //place something
    }
    ~ClassRoom()
    {
        //place something    //compile-time error will appear
    }
    static void Main(string[] args)
    {
        ....
    }
}

(iii) Destructors cannot be inherited or overloaded.
(iv) Destructors cannot be called. They are invoked automatically.
(v) A destructor does not take modifiers or have parameters.
(vi) We cannot specify an access modifier such as public for a destructor. We cannot call destructor in our code as it is part of runtime calls from garbage collector.
(vii) C# Compiler automatically translates a destructor into an override of the Object.Finalize method. For example:

class ClassRoom
{
    ....
    ~ ClassRoom()
    {
        //place something
    }
    ....
}

into

class ClassRoom
{
    protected override void Finalize()
    {
        try
        {
            //place something
        }
        finally
        {
            base.Finalize();
        }
    }
}

Why my destructor codes placed in try catch block automatically?

To answer this question think, C# provided us to write our own destructors code that may be wrong or they may contain some error. So, to ensure destructor always calls its base class destructor, even if an exception occurs during your destructor code, only compiler can make this translation, we do not have freedom to write our own method to override Finalize and even we cannot call Finalize.

I hope you love this. Please post your feedbacks on this post.

Comments

Post a Comment

Popular posts from this blog

Migrating database from ASP.NET Identity to ASP.NET Core Identity

Image
As you know ASP.NET Core Identity (table structure) is different from what we had earlier in ASP.NET Identity. Actually the identity system which we have today with .NET Core is very mature and continuously evolved be it ASP.NET Membership, ASP.NET Identity 1, ASP.NET Identity 2 and now ASP.NET Core Identity. Recently I had to migrate few application to ASP.NET Core and similar its identity database. Because the table schema is changed, i had to re-think and create migration script which I would like to share with you today. It is very simple and easy, just three step and I had everything ready: STEP 1 : Change name of existing tables STEP 2 : Create ASP.NET Core Identity tables STEP 3 : Migrate data from old tables (ASP.NET Identity) to new tables (ASP.NET Core Identity) Script:  https://gist.github.com/itorian/c699e8534b392a6c726ec66c48100072 You should also watch my video, where I demoed migration.
Read More »

Customize User's Profile in ASP.NET Identity System

Image
Note: You should read this post instead, I found below walkthrough will not work on ASP.NET Identity 2. I will re-work on this post soon. In this post you will learn how to customize User’s Profile and add some more fields like FirstName, LastName, EmailID etc with ASP.NET Identity System. In my last post I listed some of the great new features introduced with Identity system .
Read More »

Lambda two tables and three tables inner join code samples

In this blog post you will learn inner join using lambda queries. There will be two samples, in first sample you will see how you can join two tables and in second sample you will see how you can extend even further to join three tables and so on. I will sample in-memory data, but the same implementation will work with IQueryable (lazy query) too. Here's the code snippet:- namespace ConsoleApp1 {     class Program     {         static void Main()         {             var table1 = new List<Table1>();             table1.Add( new Table1 { Id = 1, Name = "Name 1" , Address = "Address 1" });             table1.Add( new Table1 { Id = 2, Name = "Name 2" , Address = "Address 2" });             table1.Add( new Table1 { Id = 3, Name = "Name 3" , Address = "Address 3" });             table1.Add( new Table1 { Id = 4, Name = "Name 4" , Address = "Address 4" });   
Read More »