Enumerating a collection consists of visiting each member of the list, for any reason judged necessary. For example, you can enumerate a collection to display a list of its members. You can enumerate a collection when looking for a member that responds to a certain criterion. Besides, or instead of, a for loop, the .NET Framework provides another and better support for enumeration. In the C# language, you can enumerate a collection using the foreach operator, but the collection must be prepared for it: you cannot just use foreach for any collection. This support is provided through two main interfaces: IEnumerator and IEnumerable. These two interfaces are defined in the System.Collection namespace. Therefore, if you intend to use them, you can include this namespace in your source file.
The IEnumerator interface provides the means of identifying the class that holds a sample of the items that will be enumerated. This interface is equipped with one property and two methods. To use the functionalities provided by the IEnumerator interface, you must create a class that implements it. You can start the class as follows: public class Enumerator : IEnumerator { } If your collection is an array-based list, you can start by declaring the base array in the class: Here is an example: public class Enumerator : IEnumerator { private double[] numbers; } If the collection is not array-based, you can declare a variable for the class that would be enumerated. The role of the enumerator is to act on a collection. For this reason, the class should be prepared to receive an external collection. This can be done by passing it to a constructor of the enumerator. Here is an example: public class Enumerator : IEnumerator { private double[] numbers; public Enumerator(double[] list) { } } The internal collection would be used in the enumerator class. The external collection would be the source of the values of the list that would be enumerated. For these reasons, you can/should initialize the internal collection with the values of the external list. This can be done as follows: public class Enumerator : IEnumerator { private double[] numbers; public Enumerator(double[] list) { this.numbers = list; } }
In the previous lesson, when introducing some techniques of creating a list, we saw that you should have a type of tag, as a field, that allows you to monitor the item that is being currently accessed or used in the list. This is particularly valuable when visiting the members of the collection. The IEnumerator interface provides a property that is used to identify the current member of the list. This property is called Current. Because the current item is meant to be viewed only, the Current property is a read-only member. Based on the rules of abstract classes, remember that you must implement all members of an interface in the class that is based on it. To implement the Current property, you can define its get accessor to return the item at the current position. This can be done as follows: public class Enumerator : IEnumerator { private double[] numbers; private int cur; public Enumerator(double[] list) { this.numbers = list; } public object Current { get { return numbers[cur]; } } }
Although you should be able to identify the current item at any time, when the application starts, before the collection can be enumerated, the tag that is used to monitor the current item should be set to a value before the beginning of the count. This can be done by setting the tag to -1. Here is an example: public class Enumerator : IEnumerator { private double[] numbers; private int cur; public Enumerator(double[] list) { this.numbers = list; cur = -1; } public Object Current { get { return numbers[cur]; } } } While the collection is being used, at one moment you may want to reset the tag of the current item to its original position. To support this operation, the IEnumerator interface is equipped with a method named Reset. Its syntax is: void Reset(); When implementing this method, simply assign a non-existing value, which is usually -1, to the monitoring tag of the current item. This can be done as follows: public class Enumerator : IEnumerator { private double[] numbers; private int cur; public Enumerator(double[] list) { this.numbers = list; cur = -1; } public object Current { get { return numbers[cur]; } } public void Reset() { cur = -1; } } When using the implementer of the IEnumerator interface, if you try accessing an item beyond the maximum number of items, the compiler would throw an IndexOutOfRangeException exception. For this reason, when anticipating a bad behavior, you should catch this exception when implementing the Current property.
In the previous lesson, we saw that, when using the items of a collection, one way you could locate one item from another was to be able to jump from one item to the next. This operation is also very important when enumerating a collection. To support this operation, the IEnumerator interface is quipped with the MoveNext() method. Its syntax is: bool MoveNext(); When implementing this method, first increment the tag that monitors the current item of the collection. After incrementing the tag, check whether it is lower than the total number of items. If it is, return true. Otherwise, return false. This can be done as follows: public class Enumerator : IEnumerator { private double[] numbers; private int cur; public Enumerator(double[] list) { this.numbers = list; cur = -1; } public Object Current { get { return numbers[cur]; } } public void Reset() { cur = -1; } public bool MoveNext() { cur++; if (cur < numbers.Length) return true; else return false; } }
The IEnumerator interface is used to set up a collection for enumeration. The IEnumerator does not provide the functionality necessary to use foreach. The next step is to implement another interface called IEnumerable. While the IEnumerator interface is used to identify the class that holds each value that will be visited, the IEnumerable interface is used to communicate with the collection whose items will be enumerated. For this reason, when implementing this class, you should provide the means of accessing the external collection. This can be done by passing a collection of the class that holds the values, to a constructor of the IEnumerable implementer.
To implement the IEnumerable interface, start by deriving a class from it. While the class implemented by the IEnumerator interface represents an object, the class that implements the IEnumerable interface is a collection. Here is an example: public class Enumerable : IEnumerable { } The new class does not know what collection it will be asked to enumerate. For this reason, in the new class, you should declare a member variable of the class that holds the values that will be enumerated. If the collection is array-based, you can create the field as follows: public class Enumerable : IEnumerable { private double[] numbers; } Eventually, when instantiating the IEnumerable implementer, you will need to pass it a collection of values. To make this possible, you can create a method in the new class and pass that collection of objects. Here is an example: public class Enumerable : IEnumerable { private double[] numbers; public void Identify(double[] values) { } } In this method, you can assign the member variable to the argument. You should also assign each member of the argument to its equivalent of the member of the argument. This can be done with a for loop as follows: public class Enumerable : IEnumerable { private double[] numbers; public void Identify(double[] values) { numbers = values; for (int i = 0; i < values.Length; i++) numbers[i] = values[i]; } }
To support the use of the foreach loop, the IEnumerable interface is equipped with (only) a method named GetEnumerator that you must implement. The IEnumerable.GetEnumerator() method returns an IEnumerator object. When implementing this method, you can return an object of the class that implements the IEnumerator interface, passing it the collection that was declared in the IEnumerable implementer. This can be done as follows: public class Enumerable : IEnumerable { private double[] numbers; public void Identify(double[] values) { numbers = values; for (int i = 0; i < values.Length; i++) numbers[i] = values[i]; } public IEnumerator GetEnumerator() { return new Enumerator(numbers); } }
After implementing the IEnumerator and the IEnumerable interfaces, you can then use the foreach loop. To start, you must prepare the collection and its items for processing. Here is an example: public class Exercise { static int Main(string[] args) { double[] numbers = new double[5]; numbers[0] = 224.52; numbers[1] = 60.48; numbers[2] = 1250.64; numbers[3] = 8.86; numbers[4] = 1005.36; return 0; } } To enumerate the collection, declare a variable based on the implementer of the IEnumerable and pass the collection to its constructor. Once this is done, you can then use the foreach. Here is an example: using System; using System.Collections; public class Enumerator : IEnumerator { private double[] numbers; private int cur; public Enumerator(double[] list) { this.numbers = list; cur = -1; } public Object Current { get { return numbers[cur]; } } public void Reset() { cur = -1; } public bool MoveNext() { cur++; if (cur < numbers.Length) return true; else return false; } } public class Enumerable : IEnumerable { private double[] numbers; public void Identify(double[] values) { numbers = values; for (int i = 0; i < values.Length; i++) numbers[i] = values[i]; } public IEnumerator GetEnumerator() { return new Enumerator(numbers); } } public class Exercise { static int Main(string[] args) { double[] numbers = new double[5]; numbers[0] = 224.52; numbers[1] = 60.48; numbers[2] = 1250.64; numbers[3] = 8.86; numbers[4] = 1005.36; Enumerable coll = new Enumerable(); coll.Identify(numbers); foreach (double d in coll) Console.WriteLine("Item {0}", d); ; return 0; } }
The IEnumerable interface allows you to connect your collection to the class that handles its enumeration. Instead of implementing that interface, the C# language provides a shortcut through a contextual keyword named yield. The yield keyword allows a method to indicate that it performs an iteration for a collection class. To use yield, you have many options. In the previous sections, we saw that you can implement the IEnumerable interface and create a method that returns an IEnumerator object. As a shortcut, you can create a method name GetEnumerator that returns an IEnumerator object. In the method, you can use a while loop to visit each member of the collection. Every time you visit a member, return it to the compiler. To do this, combine the yield and the return keywords. Inside the loop, you must find a way to navigate from one member of the collection to the next. To do this, you can use a local variable that keeps a count and increment itself. Here is an example: using System;
using System.Collections;
public class ListNumbers
{
private double[] numbers;
private int size;
public ListNumbers()
{
size = 0;
numbers = new double[10];
}
public int Count
{
get
{
return size;
}
}
public void Add(double number)
{
numbers[size] = number;
size++;
}
public double this[int index]
{
get
{
return numbers[index];
}
set
{
numbers[index] = value;
size++;
}
}
public IEnumerator GetEnumerator()
{
int counter = 0;
while (counter < Count)
{
yield return numbers[counter];
counter++;
}
}
}
public class Exercise
{
public static int Main()
{
ListNumbers nbrs = new ListNumbers();
nbrs.Add(9.08);
nbrs.Add(79.13);
nbrs.Add(4.803);
nbrs.Add(725.45);
nbrs.Add(2937.468);
Console.WriteLine("--- Using \"for\"---");
for (int i = 0; i < nbrs.Count; i++)
Console.WriteLine("Number: {0}", nbrs[i]);
Console.WriteLine("--- Using \"foreach\"---");
foreach (double n in nbrs)
Console.WriteLine("Number: {0}", n);
Console.WriteLine();
return 0;
}
}
This would produce: --- Using "for"--- Number: 9.08 Number: 79.13 Number: 4.803 Number: 725.45 Number: 2937.468 --- Using "foreach"--- Number: 9.08 Number: 79.13 Number: 4.803 Number: 725.45 Number: 2937.468 Press any key to continue . . . Our collection class uses a simple type (a double type). Still, you can use the same approach if the items of the collection are composite (class type). Another technique to use yield consists of creating a method that returns an IEnumerable object. Inside the method, you can use a while loop to access each member of the collection and return it. Once again, you combine yield and return to indicate that the method is an iterator that returns a member of the collection. At the end of each loop, find a way to get to the next member unless you have gotten to the end of the collection. As done previously, you can use a local variable to take care of this. Here is an example: using System;
using System.Collections;
public class ListNumbers
{
private double[] numbers;
private int size;
public ListNumbers()
{
size = 0;
numbers = new double[5];
}
public int Count
{
get
{
return size;
}
}
public double this[int index]
{
get
{
return numbers[index];
}
set
{
numbers[index] = value;
size++;
}
}
public IEnumerable AllNumbers()
{
int counter = 0;
while (counter < Count)
{
yield return numbers[counter];
counter++;
}
}
}
In this example, we explicitly created a method named AllNumbers. If you do this, when using the collection, make sure you call this method. Here is an example: public class Exercise
{
public static int Main()
{
ListNumbers nbrs = new ListNumbers();
nbrs[0] = 9.08;
nbrs[1] = 79.13;
nbrs[2] = 4.803;
nbrs[3] = 725.45;
Console.WriteLine("--- Using \"for\"---");
for (int i = 0; i < nbrs.Count; i++)
Console.WriteLine("Number: {0}", nbrs[i]);
Console.WriteLine("--- Using \"foreach\"---");
foreach (double n in nbrs.AllNumbers())
Console.WriteLine("Number: {0}", n);
Console.WriteLine();
return 0;
}
}
This would produce: --- Using "for"--- Number: 9.08 Number: 79.13 Number: 4.803 Number: 725.45 --- Using "foreach"--- Number: 9.08 Number: 79.13 Number: 4.803 Number: 725.45 Press any key to continue . . . You can create the method as a static one. In this case you can pass the collection class as argument. When calling the method, use the name of the collection class to call the method and pass the variable as argument. Here is an example where the method is created in the body of the collection class: using System; using System.Collections; public class ListNumbers { private double[] numbers; private int size; public ListNumbers() { size = 0; numbers = new double[5]; } public int Count { get { return size; } } public double this[int index] { get { return numbers[index]; } set { numbers[index] = value; size++; } } public static IEnumerable AllNumbers(ListNumbers lst) { int counter = 0; while (counter < lst.Count) { yield return lst.numbers[counter]; counter++; } } } public class Exercise { public static int Main() { ListNumbers nbrs = new ListNumbers(); nbrs[0] = 9.08; nbrs[1] = 79.13; nbrs[2] = 4.803; nbrs[3] = 725.45; Console.WriteLine("--- Using \"for\"---"); for (int i = 0; i < nbrs.Count; i++) Console.WriteLine("Number: {0}", nbrs[i]); Console.WriteLine("--- Using \"foreach\"---"); foreach (double n in ListNumbers.AllNumbers(nbrs)) Console.WriteLine("Number: {0}", n); Console.WriteLine(); return 0; } } Here is an example where the method is created in the body of the class that calls it (the class that contains the Main() function): using System; using System.Collections; public class ListNumbers { private double[] numbers; private int size; public ListNumbers() { size = 0; numbers = new double[5]; } public int Count { get { return size; } } public double this[int index] { get { return numbers[index]; } set { numbers[index] = value; size++; } } } public class Exercise { public static IEnumerable AllNumbers(ListNumbers lst) { int counter = 0; while (counter < lst.Count) { yield return lst[counter]; counter++; } } public static int Main() { ListNumbers nbrs = new ListNumbers(); nbrs[0] = 9.08; nbrs[1] = 79.13; nbrs[2] = 4.803; nbrs[3] = 725.45; Console.WriteLine("--- Using \"for\"---"); for (int i = 0; i < nbrs.Count; i++) Console.WriteLine("Number: {0}", nbrs[i]); Console.WriteLine("--- Using \"foreach\"---"); foreach (double n in AllNumbers(nbrs)) Console.WriteLine("Number: {0}", n); Console.WriteLine(); return 0; } } Remember that, although we are using a simple type in our example, you can apply the same solution to a collection of objects.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||