C# the use of ref keyword

Taswar Bhatti December 6, 2013 No Comments

In C# when using the ref keyword it causes an argument to be passed by reference, not by value.
In order to use a ref as a parameter, both the method definition and the calling method must explicitly use the ref keyword, and also the variable must be initialized before passing in.

In the example below we have two methods, one using the ref keyword the other not.

void Increment(int x)
{
  x = x + 1;
}


void IncrementWithRef(ref int y)
{
  y = y + 1;
}

int x = 1;
Increment(x); 

Console.WriteLine(x);//x is still 1

Increment(ref x);
Console.WriteLine(x);//x is now 2

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

void Increment(int x)

{

x = x + 1;

}

void IncrementWithRef(ref int y)

{

y = y + 1;

}

int x = 1;

Increment(x);

Console.WriteLine(x);//x is still 1

Increment(ref x);

Console.WriteLine(x);//x is now 2

But what about objects that are reference types. Here is where the confusion starts that the concept of passing by reference with the concept of reference types, the concepts are not the same. A method parameter can pass in a ref type regardless of its type (value or reference).

Below is an example that shows when a Person object reference type is used.

public class Person
{
  public string Name { get; set; }
}

void NameChange(Person p)
{
   p.Name = "John";
}

void NameChangeRef(ref Person p)
{
 p = new Person{ Name = "Ref Person" };
}

Person p = new Person { Name = "Peter" };
	
Console.WriteLine(p.Name); //output is Peter
	
NameChange(p);
	
Console.WriteLine(p.Name); //output is John we have changed the reference type property but not the object
	
NameChangeRef(ref p); 
	
Console.WriteLine(p.Name); //output is Ref Person we have changed the object that p points to

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

public class Person

{

public string Name { get; set; }

}

void NameChange(Person p)

{

p.Name = "John";

}

void NameChangeRef(ref Person p)

{

p = new Person{ Name = "Ref Person" };

}

Person p = new Person { Name = "Peter" };

Console.WriteLine(p.Name); //output is Peter

NameChange(p);

Console.WriteLine(p.Name); //output is John we have changed the reference type property but not the object

NameChangeRef(ref p);

Console.WriteLine(p.Name); //output is Ref Person we have changed the object that p points to

Last but not least I personally do not recommend using ref keyword and also from
The Framework Design Guidelines it also recommend to avoid using ref.

.NET, C#

.net, C#, csharp, dotnet

C# tip: Use Path Combine for file or directory path information

Taswar Bhatti December 4, 2013 1 Comment

.NET provides in its System.IO namespace the Path class which performs operations on String instances that contain file or directory path information. These operations are performed in a cross-platform manner.

Most of the time we see develeopers writing code like

string path = somePath + "\\" + filename;

1	string path = somePath + "\\" + filename;

But by using Path.Combine we can provide a cross platform path

string somePath = @"C:\temp";
string filename = "dat.txt";
string path = Path.Combine(somePath,filename);

//produce an output of C:\temp\dat.txt
//linux: C:/temp/dat.txt

string[] paths = {@"c:\My Music", "2013", "media", "banner"};
string fullPath = Path.Combine(paths);
//output c:\My Music\2013\media\banner

string path1 = @"C:\Temp";
string path2 = "My Music";
fullPath = Path.Combine(path1, path2);

//output: C:\Temp\My Music

fullPath = Path.Combine(string.Empty, path2);
//output: My Music
	
fullPath = Path.Combine(path1, string.Empty);
//output: C:\Temp

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

string somePath = @"C:\temp";

string filename = "dat.txt";

string path = Path.Combine(somePath,filename);

//produce an output of C:\temp\dat.txt

//linux: C:/temp/dat.txt

string[] paths = {@"c:\My Music", "2013", "media", "banner"};

string fullPath = Path.Combine(paths);

//output c:\My Music\2013\media\banner

string path1 = @"C:\Temp";

string path2 = "My Music";

fullPath = Path.Combine(path1, path2);

//output: C:\Temp\My Music

fullPath = Path.Combine(string.Empty, path2);

//output: My Music

fullPath = Path.Combine(path1, string.Empty);

//output: C:\Temp

By using the Path and Combine method, one can at least eliminate some of the headache of porting to another platform.

.NET, C#

.net, C#, csharp, dotnet

Documentation comment your C# code use GhostDoc at least

Taswar Bhatti December 2, 2013 No Comments

Recently have been going through some old code to review the comments in them from other developers and what I find out is developers tend to have really bad comments & documentation in their code

Example:

public class Person 
{
   /// &lt;summary>
   /// Constructor
   /// &lt;/summary>
   public Person() {}
   
}

1

2

3

4

5

6

7

8

public class Person

{

/// <summary>

/// Constructor

/// </summary>

public Person() {}

}

From the above code it is obvious that it is the constructor but does the comment tell me anything? I understand that sometimes it may be hard to document or comment certain method, thus I really strongly suggest at least use GhostDoc

What is GhostDoc?
GhostDoc is a Visual Studio extension that automatically generates XML documentation comments for methods and properties based on their type, parameters, name, and other contextual information.

Here is the result of using GhostDoc on the Person Class

public class Person 
{
   /// &lt;summary>
   /// Initializes a new instance of the &lt;see cref="Person"/> class.
   /// &lt;/summary>
   public Person() {}   
}

1

2

3

4

5

6

7

public class Person

{

/// <summary>

/// Initializes a new instance of the <see cref="Person"/> class.

/// </summary>

public Person() {}

}

As one can see Ghostdoc does a better job, don’t get me wrong GhostDoc still requires one to go through the comments here is an example where it tries to make sense.

 /// &lt;summary>
 /// Riches the text selection changed.
 /// &lt;/summary>
 /// &lt;param name="richTextBox">The rich text box.&lt;/param>
 private void RichTextSelection_Changed(System.Windows.Controls.RichTextBox richTextBox)

1

2

3

4

5

/// <summary>

/// Riches the text selection changed.

/// </summary>

/// <param name="richTextBox">The rich text box.</param>

private void RichTextSelection_Changed(System.Windows.Controls.RichTextBox richTextBox)

Riches the text selection changed. ????? As you can see it didn’t do one of the best job for this comment, but at least one can modify it and it may become “RichTextBox selection has changed event handler”

Personally I would recommend using the tool but just take it with a grain of salt and always check the documentation that it generates.

.NET, C#, Misc

.net, C#, csharp, dotnet

LINQ tip – Ordering collection using existing array/collection using OrderBy

Taswar Bhatti November 29, 2013 5 Comments

Here is a LINQ tip where you may wish to order a collection with an existing ordering of another collection.

Example:

int[] displaySeq = new int[] { 1, 8, 5, 7, 13 }; //the ordering we want at the end

//collection that we will be sorting on
List<Person> people = new List<Person>();
people.Add(new Person { DisplaySeq = 5});
people.Add(new Person { DisplaySeq = 13});
people.Add(new Person { DisplaySeq = 7});
people.Add(new Person { DisplaySeq = 1});
people.Add(new Person { DisplaySeq = 8});

1

2

3

4

5

6

7

8

9

int[] displaySeq = new int[] { 1, 8, 5, 7, 13 }; //the ordering we want at the end

//collection that we will be sorting on

List<Person> people = new List<Person>();

people.Add(new Person { DisplaySeq = 5});

people.Add(new Person { DisplaySeq = 13});

people.Add(new Person { DisplaySeq = 7});

people.Add(new Person { DisplaySeq = 1});

people.Add(new Person { DisplaySeq = 8});

Currently as it stands our data is stored in the order of { 5, 13, 7, 1, 8 } and we wish to order them in terms of { 1, 8, 5, 7, 13 }, we can use LINQ OrderBy to do so.

//use the displaySeq array and find the index of it
var sorted = people.OrderBy(x => { return Array.IndexOf(displaySeq, x.DisplaySeq); }).ToList();

1 2	//use the displaySeq array and find the index of it var sorted = people.OrderBy(x => { return Array.IndexOf(displaySeq, x.DisplaySeq); }).ToList();

By doing so we end up with { 1, 8, 5, 7, 13 }, now if we wanted to put elements that dont have matching at the end of the list we can modify the OrderBy like this.

//add another person that is not part of the displaySeq array
people.Add(new Person { DisplaySeq = 99});

var sorted = people.OrderBy(x => { 
   var index = Array.IndexOf(displaySeq, x.DisplaySeq);
   return index < 0 ? int.MaxValue : index;
 }).ToList();

1

2

3

4

5

6

7

//add another person that is not part of the displaySeq array

people.Add(new Person { DisplaySeq = 99});

var sorted = people.OrderBy(x => {

var index = Array.IndexOf(displaySeq, x.DisplaySeq);

return index < 0 ? int.MaxValue : index;

}).ToList();

.NET, C#

.net, C#, csharp, dotnet, LINQ

C# tip – Marking method or class as Obsolete/Deprecated

Taswar Bhatti November 25, 2013 No Comments

Obsolete or Deprecated?

How do you mark a class or method as deprecated/obsolete?

By using the Obsolete attribute

[Obsolete]
public class Person { }

[Obsolete("Class is obsolete, use PersonImpl")]
public class Person { }

//compilation to fail methods calling this code
[Obsolete("Class is obsolete, use PersonImpl", true)]
public class Person { }

[Obsolete("Method is obsolete, use SomeMethod2", true)]
public void SomeMethod() { }

1

2

3

4

5

6

7

8

9

10

11

12

[Obsolete]

public class Person { }

[Obsolete("Class is obsolete, use PersonImpl")]

public class Person { }

//compilation to fail methods calling this code

[Obsolete("Class is obsolete, use PersonImpl", true)]

public class Person { }

[Obsolete("Method is obsolete, use SomeMethod2", true)]

public void SomeMethod() { }

.NET, C#

.net, C#, csharp, dotnet

C# Null Coalescing Operator

Taswar Bhatti November 21, 2013 No Comments

The ?? operator is called the null coalescing operator. It is used for providing a default value for Nullable types or reference types.

Example:

int? age = null;

//what ?? implies is possiblyNullValue ?? valueIfNull
int myAge = x ?? 100; //if x is not null then assign x else 100

string? person = null;
string? localDefault = null;
string globalDefault = "abc";

string anybody = person ?? localDefault ?? globalDefault; //chaining ??

//lazy loading/populating
private SomeObj _lazyField = null;
public SomeObj MyProperty
{
    get { return _lazyField ?? (_lazyField = new SomeObj()); }
}

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

int? age = null;

//what ?? implies is possiblyNullValue ?? valueIfNull

int myAge = x ?? 100; //if x is not null then assign x else 100

string? person = null;

string? localDefault = null;

string globalDefault = "abc";

string anybody = person ?? localDefault ?? globalDefault; //chaining ??

//lazy loading/populating

private SomeObj _lazyField = null;

public SomeObj MyProperty

{

get { return _lazyField ?? (_lazyField = new SomeObj()); }

}

One of the disadvantage of ?? is it can create code that is not that readable.
e.g a ?? b ?? c ?? d ?? e

.NET, C#, Uncategorized

.net, C#, csharp, dotnet

C# tip property with different accessibility scopes

Taswar Bhatti November 20, 2013 No Comments

In C# you can have property with different scope (property accessors). Properties in C# can be marked as public, private, protected, internal, or protected internal.

Example:

public int MyProperty { get; private set; }
public string Name { get; protected set;}
public int Age { get { return _age; } } // no set but still valid
private string Fullname { get; set; } //both are private only class can access it
internal string LName { get; set; } //only code in same assembly can access
protected internal string FName { get; set; } //only code that inherit and in same assembly
virtual string PersonName { get; set; } //use override keyword in derived class
static int Count { get; set; } // avaiable to caller at any time

1

2

3

4

5

6

7

8

public int MyProperty { get; private set; }

public string Name { get; protected set;}

public int Age { get { return _age; } } // no set but still valid

private string Fullname { get; set; } //both are private only class can access it

internal string LName { get; set; } //only code in same assembly can access

protected internal string FName { get; set; } //only code that inherit and in same assembly

virtual string PersonName { get; set; } //use override keyword in derived class

static int Count { get; set; } // avaiable to caller at any time

.NET, C#

.net, C#, csharp, dotnet

C# optional parameters

Taswar Bhatti November 18, 2013 No Comments

C# 4.0 allows one to use named and optional parameters.

Example:

int Calculate(int x = 1, int y = 2)
{
   int someNumber = 0;
   //some algorithm to calculate
    return someNumber * y * x + 5 * 0.5;
}

1

2

3

4

5

6

int Calculate(int x = 1, int y = 2)

{

int someNumber = 0;

//some algorithm to calculate

return someNumber * y * x + 5 * 0.5;

}

One can call the method in multiple ways

int result = Calculate(1, 2);
int result = Calculate(5); //value for x only, y is default to 2
int result = Calculate(); //value for x is 1 and y is 2
int result = Calculate(x: 123, y: 987); //named argument
int result = Calculate(y: 345 x: 678); //named argument in reverse 
int result = Calculate(x: 123, 456); //compiler error not allowed position argument cannot follow name argument
int result = Calculate(1, y: 456); //name argument can follow position argument

1

2

3

4

5

6

7

int result = Calculate(1, 2);

int result = Calculate(5); //value for x only, y is default to 2

int result = Calculate(); //value for x is 1 and y is 2

int result = Calculate(x: 123, y: 987); //named argument

int result = Calculate(y: 345 x: 678); //named argument in reverse

int result = Calculate(x: 123, 456); //compiler error not allowed position argument cannot follow name argument

int result = Calculate(1, y: 456); //name argument can follow position argument

Additional note, one cannot provide a gap in arguments.
Example:

int Calculate(int x, int y = 2, int z) //invalid
int Calculate(int x, int y = 1, int z= 5) //valid


int result = Calculate(1, , 3); //invalid
int result = Calculate(1, 8); //z takes the value of 5 as default

1

2

3

4

5

6

int Calculate(int x, int y = 2, int z) //invalid

int Calculate(int x, int y = 1, int z= 5) //valid

int result = Calculate(1, , 3); //invalid

int result = Calculate(1, 8); //z takes the value of 5 as default

.NET, C#

.net, C#, csharp, dotnet

The using statement in C#

Taswar Bhatti November 14, 2013 No Comments

The using statement in C# is a form of shortcut for try and finally block of code. Things to note is in order to use the using statement the object needs to implement the IDisposable interface, and using does not catch Exception, it just guarantees the call of Dispose.

Example:

try {    
  connection = new SqlConnection("");
}
finally {
  if(connection != null)
     connection.Close();
}

1

2

3

4

5

6

7

try {

connection = new SqlConnection("");

}

finally {

if(connection != null)

connection.Close();

}

Can be replaced with

using(var connection = new SqlConnection(""))
{

}

1

2

3

4

using(var connection = new SqlConnection(""))

{

}

One will not need to call the Close method of sql connection, since the dispose will do that.

.NET, C#

.net, C#, csharp, dotnet

I refuse to answer your stupid singleton question in interview

Taswar Bhatti November 13, 2013 No Comments

In interviews there is always the Singleton question, I personally refuse to answer it on interview. The reason is very simple I don’t code singletons anymore. I use my IoC Container to provide me with a singleton. Sorry if you dont know what an IoC (Inversion of Control Container) is, then you are still in the old ages of coding.

For once and for all here is the hand written code of a singleton in multi-thread environment.

using System;

public sealed class Singleton
{
   private static volatile Singleton instance;
   private static object syncRoot = new Object();

   private Singleton() {}

   public static Singleton Instance
   {
      get 
      {
         if (instance == null) 
         {
            lock (syncRoot) 
            {
               if (instance == null) 
                  instance = new Singleton();
            }
         }
         return instance;
      }
   }
}

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

using System;

public sealed class Singleton

{

private static volatile Singleton instance;

private static object syncRoot = new Object();

private Singleton() {}

public static Singleton Instance

{

get

{

if (instance == null)

{

lock (syncRoot)

{

if (instance == null)

instance = new Singleton();

}

return instance;

}

And here is in an IoC container Unity & StructureMap

//stucture map
 var container = new Container(x =>
            {
                x.ForRequestedType<IEmailService>().CacheBy(InstanceScope.Singleton).TheDefaultIsConcreteType
                    <EmailService>();
            });

//unity
myContainer.RegisterType<IEmailService, EmailService>(new ContainerControlledLifetimeManager());

1

2

3

4

5

6

7

8

9

//stucture map

var container = new Container(x =>

{

x.ForRequestedType<IEmailService>().CacheBy(InstanceScope.Singleton).TheDefaultIsConcreteType

<EmailService>();

});

//unity

myContainer.RegisterType<IEmailService, EmailService>(new ContainerControlledLifetimeManager());

As you can see one can have a singleton with one line of code in Unity and say 2 lines of code in StructureMap, so I really wonder if one needs to proof that a double check is needed in a multi thread environment for singleton.

End rant……

.NET, C#, Design

.net, C#, csharp, Design Pattern, dotnet

C# the use of ref keyword

C# tip: Use Path Combine for file or directory path information

Documentation comment your C# code use GhostDoc at least

LINQ tip – Ordering collection using existing array/collection using OrderBy

C# tip – Marking method or class as Obsolete/Deprecated

Obsolete or Deprecated?

C# Null Coalescing Operator

C# tip property with different accessibility scopes

C# optional parameters

The using statement in C#

I refuse to answer your stupid singleton question in interview

Opinions are my own and not the views of my employer

LinkedIn

Archives