Questions about refactoring switch statement to Strategy Pattern in C# - C# Discussion Boards

V.4-Dec-14 20:03

4-Dec-14 20:03

Many companies still use COBOL, or C to do much of the work, so although it looks like companies are only using the newest thing, this is not true at all. Some of the reasons include the price of software and "if it's not broken don't fix it" attitude.

If you're going C# I wouldn't start by taking directions. In fact, there is a learning curve and I would start with winforms then ASP.Net then WPF. Also learn WCF, but I found this to be a complete seperate track.

hope this helps.

V.

(MQOTD rules and previous solutions)

Re: general query related to C#

den2k885-Dec-14 0:35

den2k88

5-Dec-14 0:35

VB6 here. Have you heard the latest of Backstreet Boys? I was wondering if buying that cassette...

We switched from Win2k to XP a few years ago, to say... And we are in the top 5 companies worldwide in our sector.

Re: general query related to C#

Member 112872605-Dec-14 2:44

Member 11287260

5-Dec-14 2:44

I have completed my C# certification and have recently completed my SQL, HTML, CSS and Javascript.
That's why I am moving towards more advanced C#, now I am not sure where to start with, should I opt for win forms, aps or WPF.

and I don't want to go with ASP.NET, not right now, I already got my hands full.

So what say you.

Re: general query related to C#

V.5-Dec-14 2:46

5-Dec-14 2:46

1. aps does not exist.
2. please read my post more thouroughly as it holds the answer...

V.

(MQOTD rules and previous solutions)

Re: general query related to C#

Member 112872605-Dec-14 3:02

Member 11287260

5-Dec-14 3:02

I got it bro, thanks.
just read it again and got your point.
Sorry for not reading it thoroughly in the first go.

Re: general query related to C#

Member 112872605-Dec-14 3:39

Member 11287260

5-Dec-14 3:39

would you suggest some tutorial or books for win forms.
I did some research but couldn't find anything constructive.

Re: general query related to C#

V.5-Dec-14 4:36

5-Dec-14 4:36

O reilly´s have a good reputation. I would go through Amazon en check out the ratings.

V.

(MQOTD rules and previous solutions)

Questions about refactoring switch statement to Strategy Pattern in C#

Member 97402194-Dec-14 11:48

Member 9740219

4-Dec-14 11:48

There are many situations where an object has to be able to change its calculation based on a user input. Examples are applying discounts to an order, calculating shipping charges, and so many more. There are also many examples on the web that show if you implemented this logic using an enum, a switch statement and private methods, you should refactor the code using the strategy pattern. I don’t have an issue with understanding how to refactor the code. At the bottom of this post I have included sample code that shows the before and after refactored code.

However, I have found myself asking three questions after reading these posts and I am hoping that some one can explain it to me.

1. Suppose these calculations really will only apply to this object. Why do I want to make them more accessible by putting them in their own class that at a minimum is accessible to the entire assembly?

2. In the examples on the web; there is a class that calls the context object. The code below is typical of what you will find in posts. It creates the context object for each calculation method. I know it serves to show you the different results, however, I believe it hides an issue.

// Class that starts the calculation process - Sample Code seen on the web
        public class ClassThatCallsContextObject
        {
            public void StartCalculation ( )
            {
                decimal calculationResult = 0.0M;

                ContextObject contextObject;

                // Initilize the contextObject with the first calculation method
                contextObject = new ContextObject(new CalculationMethod1());
                calculationResult = contextObject.Calculation();
                // calculationResult would be set to 1.0

                // Initilize the contextObject with the second calculation method
                contextObject = new ContextObject(new CalculationMethod2());
                calculationResult = contextObject.Calculation();
                // calculationResult would be set to 2.0
            }
        }

In an actual application, only one of the choices will be selected. This selection can be made by the user, by a business rule, read in from the database, etc. That selection requires different code. I think you need to add back the enum and the switch statement to implement this code. Below is an example of the code.

// Define the possible calculation strategies
    public enum CalculationStrategies { Method1, Method2}

// Class that starts the calculation process - With User Selection
    public class ClassThatCallsContextObjectWithUserSelection
    {
        public void StartCalculation ( CalculationStrategies userInput )
        {
            decimal calculationResult = 0.0M;

            ContextObject contextObject;

            switch (userInput)
            {
                case CalculationStrategies.Method1:
                    // Initilize the contextObject with the first calculation method
                    contextObject = new ContextObject(new CalculationMethod1());
                    calculationResult = contextObject.Calculation();
                    // calculationResult would be set to 1.0
                    break;
                case CalculationStrategies.Method2:
                    // Initilize the contextObject with the second calculation method
                    contextObject = new ContextObject(new CalculationMethod2());
                    calculationResult = contextObject.Calculation();
                    // calculationResult would be set to 2.0
                    break;
            }
        }
    }

Is there a better way to implement this selection code?

3. Assuming item 2 is the best way to do this, when the time comes to add a new calculation (and we all know that it will be here eventually), the following work has to be done:

A. Add the new calculation to the enum.
B. Add a new switch statement to the object that calls the context object.
C. Create a new class that implements the new calculation.

If I didn't refactor the code, here is the work I would need to do.

A. Add the new calculation to the enum.
B. Add a new switch statement to the context object.
C. Create a new private method that implements the new calculation.

This looks like pretty much the same thing. So, what is the advantage of the refactored code?

Complete code examples for reference:

Switch Statement Code

// Class that starts the calculation process
    public class ClassThatCallsContextObject
    {
        public void StartCalculation ()
        {
            decimal calculationResult = 0.0M;

            ContextObject contextObject = new ContextObject();
            calculationResult = contextObject.Calculation(CalculationStrategies.Method1);
            // calculationResult would be set to 1.0
            calculationResult = contextObject.Calculation(CalculationStrategies.Method2);
            // calculationResult would be set to 2.0
        }
    }

    // Define the possible calculation strategies
    public enum CalculationStrategies { Method1, Method2}

    // Class that executes the calculations
    public class ContextObject
    {
        // Constructor
        public ContextObject ()
        {

        }

        // Method that has multiple ways to calculate the value depending on user input
        public decimal Calculation ( CalculationStrategies selectedStrategy )
        {
            decimal calcualtedValue = 0.0M;
            switch (selectedStrategy)
            {
                case CalculationStrategies.Method1:
                    calcualtedValue = CalculationMethod1();
                    break;
                case CalculationStrategies.Method2:
                    calcualtedValue = CalculationMethod2();
                    break;
            }
            return calcualtedValue;
        }

        // Calculation Method 1
        private decimal CalculationMethod1 (  )
        {
            return 1.0M;
        }

        // Calculation Method 2
        private decimal CalculationMethod2 ()
        {
            return 2.0M;
        }
    }

Strategy Pattern Code

// Class that starts the calculation process - Sample Code seen on the web
        public class ClassThatCallsContextObject
        {
            public void StartCalculation ( )
            {
                decimal calculationResult = 0.0M;

                ContextObject contextObject;

                // Initilize the contextObject with the first calculation method
                contextObject = new ContextObject(new CalculationMethod1());
                calculationResult = contextObject.Calculation();
                // calculationResult would be set to 1.0

                // Initilize the contextObject with the second calculation method
                contextObject = new ContextObject(new CalculationMethod2());
                calculationResult = contextObject.Calculation();
                // calculationResult would be set to 2.0
            }
        }

        // Define the possible calculation strategies
        public enum CalculationStrategies { Method1, Method2 }

        // Class that executes the calculations
        public class ContextObject
        {
            private ICalculationMethod _calculationMethod;

            // Constructor
            public ContextObject ( ICalculationMethod calculationMethod )
            {
                _calculationMethod = calculationMethod;
            }

            // Method that has multiple ways to calculate the value depending on user input
            public decimal Calculation ()
            {
                return _calculationMethod.CalculationMethod();
            }
        }

        interface ICalculationMethod
        {
            decimal CalculationMethod ();
        }

        public class CalculationMethod1 : ICalculationMethod
        {
            // Calculation Method 1
            private decimal CalculationMethod ()
            {
                return 1.0M;
            }
        }

        public class CalculationMethod2 : ICalculationMethod
        {
            // Calculation Method 2
            private decimal CalculationMethod ()
            {
                return 2.0M;
            }
        }

Re: Questions about refactoring switch statement to Strategy Pattern in C#

SledgeHammer014-Dec-14 12:24

SledgeHammer01

4-Dec-14 12:24

1) you don't new up the strategies on start, you new them up when they are requested

2) its not really necessary to cache the strategy objects unless they are expensive to create (new'ing up a class with an empty constructor is very cheap)

3) at the bottom of a strategy pattern is a switch

4) you are doing too much in the switch, all the switch should do is return the instance and the switch should be "internal". I.e., you'd have a static strategy factory class that has a static GetStrategy(someEnum) and returns the ICalculationMethod object. So your caller just needs to do StrategyFactory.GetStrategy(someEnum); The GetStrategy() method is going to have the switch.

5) your suggested use case is a trivial case and probably strategy is overkill for simple calculations... a more realstic scenario might be FedEx vs UPS shipping for example. Your strategy class will have methods for calculating the shipping, tracking, interacting with FedEx or UPS web service, etc. So it'll be much more then just a single method.

If your strategies are all 5 to 10 lines, I'd say strategy is overkill.

Re: Questions about refactoring switch statement to Strategy Pattern in C#

BillWoodruff4-Dec-14 18:36

BillWoodruff

4-Dec-14 18:36

Voted-up. Seems like a very good analysis to me. Why build a rocket-ship that can go to outer-space if your destination is the moon Smile | :)

«OOP to me means only messaging, local retention and protection and hiding of state-process, and extreme late-binding of all things. » Alan Kay's clarification on what he meant by the term "Object" in "Object-Oriented Programming."

Re: Questions about refactoring switch statement to Strategy Pattern in C#

Matt T Heffron5-Dec-14 6:51

Matt T Heffron

5-Dec-14 6:51

BillWoodruff wrote:
Why build a rocket-ship that can go to outer-space if your destination is the moon

I seem to recall that the moon is in outer space! Wink | ;)

A positive attitude may not solve every problem, but it will annoy enough people to be worth the effort.

postback issue with dynamically created datetimepicker

Dhyanga4-Dec-14 4:58

Dhyanga

4-Dec-14 4:58

Hi,

I am using dynamic datetimepicker that the user can add to any number. But I am having postback problem. Once I submit it, the return page should have all the submitted values including those added dynamic datetimepicker. Right now when i submit and when the page returns, all other static controls with data are there but no dynamically added datetimepicker controls.

Below is the code I used to add datatimepicker dynamically using javascript:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html>


<link rel="stylesheet" media="all" type="text/css" href="jquery-ui.css" />
<link rel="stylesheet" media="all" type="text/css" href="jquery-ui-timepicker-addon.css" />
<script type="text/javascript" src="jquery-1.11.1.min.js"></script>
<script type="text/javascript" src="jquery-ui.min.js"></script>
<script type="text/javascript" src="jquery-ui-timepicker-addon.js"></script>
<script type="text/javascript" src="jquery-ui-sliderAccess.js"></script>
<script type="text/javascript">
    $(function () {
      
         var counter = 1;
          jQuery("#<%= btnD1Add.ClientID %>").click(function (event) {
            event.preventDefault(); //this code is added to prevent the default submit functionality of the button

            jQuery("<li>").appendTo(".Date1More");
            jQuery("<label id='lblD1StartTime" + counter + "'>StartTime:</label><input type='text' id='txtD1StartTime"  +counter + "' name='txtStartTime1'/>").timepicker({
                hourGrid: 10,
                minuteGrid: 10,
                timeFormat: 'hh:mm tt'
            }).appendTo(".Date1More");
            
            counter++;
        });
    });
</script>
       <asp:Button ID="btnD1Add" runat="server" Text="Add more time slots" />
       

               <li>
                    <ul class="Date1More">
                                          
                </ul>
               </li>
</html>

Now how to use postback for these controls? I had tried to use following code in code but didn't work. I even tried it using !Ispostback() function under pageLoad but no effect.

protected string[] txtStartTime1;
   protected void Page_Load(object sender, EventArgs e)
        {
            
           
            txtStartTime1 = Request.Form.GetValues("txtStartTime1");
        }
protected void Page_Init(object sender, EventArgs e)
        {
            try
            {
                if (txtStartTime1.Length > 0)
                {
                    TextBox txtStTime1 = new TextBox();
                    txtStTime1.ID = "txtD1StartTime";
                    txtStTime1.CssClass = "timepicker";
                    this.Controls.Add(txtStTime1);

                }
              
            }
            catch(Exception ex){}

        }

Please help..

Dhyanga

modified 4-Dec-14 11:06am.

Re: postback issue with dynamically created datetimepicker using javascript

Richard MacCutchan4-Dec-14 5:01

Richard MacCutchan

4-Dec-14 5:01

Please use the correct forum: http://www.codeproject.com/Forums/12076/ASP-NET.aspx[^] or http://www.codeproject.com/Forums/1580226/JavaScript.aspx[^].

Re: postback issue with dynamically created datetimepicker using javascript

Dhyanga4-Dec-14 8:04

Dhyanga

4-Dec-14 8:04

Thanks. Yes I did.

Dhyanga

How to find a particular embedded resource in an assembly?

Sivaji15654-Dec-14 1:08

Sivaji1565

4-Dec-14 1:08

My function takes resource file name as input and returns Bitmap object.

GetManifestResourceNames returns fully qualified resource names. Sample Input is like "Fileopen.png".

How to find the required resource without doing any string operation ? Or any other way is also welcome.

public static Bitmap GetBitmap_FromResource(string ImageName)
        {
            Assembly assembly = Assembly.GetExecutingAssembly();
            string strResName = GetReqdResourceName(assembly, ImageName);
            if (string.IsNullOrEmpty(strResName))
                return null;

            Stream imgstream = assembly.GetManifestResourceStream(strResName);
            if (imgstream == null)
                return null;

            Bitmap Img = new Bitmap(imgstream);
            return Img;
        }

public static string GetReqdResourceName(Assembly assembly, string strResName)
        {
            string[] resNames = assembly.GetManifestResourceNames();

            // How to find the resource name without string operation
            return null;
        }

Re: How to find a particular embedded resource in an assembly?

Richard MacCutchan4-Dec-14 2:06

Richard MacCutchan

4-Dec-14 2:06

What exactly do you mean by "find the required resource without doing any string operation"? Please explain exactly what problem you are trying to solve.

C# Error check

Member 112847343-Dec-14 18:06

Member 11284734

3-Dec-14 18:06

Hi! I have a problem with how my program handles an invalid input.
It goes like this:

Enter string: Apple
Enter integer: 1

Enter string: Orange
Enter integer: 2

Search array: Apple
Search found with string name Apple with integer number 1

Search array: Banana
Search found with string name Apple with integer number 1

Object searchValue = null; 
string searchString; 
do 
{ 
// input 
searchString = GetCorrectInput.String(message); 

if (!searchString.Equals("Quit")) 
{ 
// array 
for (int i = 0; i < objectArray.Length; i++) 
{ 
if (searchString== objectArray[i].studentName) 
{ 
searchValue = objectArray[i]; 
} 

} 

// check input 
if (searchValue == null) 
{ 
Console.WriteLine("{0} was not found", searchString); 
} 
else 
{ 
Console.WriteLine("Search found string name {0} with integer number {1}", searchValue.stringName, searchValue.integerNumber); 
} 
} 
} while (!searchString.Equals("Quit")); 

Console.WriteLine("Thank you!");

Banana is supposed to be not found but its not Frown | :(

Re: C# Error check

Manfred Rudolf Bihy3-Dec-14 18:20

Manfred Rudolf Bihy

3-Dec-14 18:20

You forgot to set searchValue back to null. The proper place to do that would be at the top of the loop right after searchString is being set.

Regards,
Manfred

"I had the right to remain silent, but I didn't have the ability!"

Ron White, Comedian

Re: C# Error check

Member 112847343-Dec-14 18:34

Member 11284734

3-Dec-14 18:34

is there a way to avoid using null?

Re: C# Error check

Manfred Rudolf Bihy3-Dec-14 19:53

Manfred Rudolf Bihy

3-Dec-14 19:53

Member 11284734 wrote:
is there a way to avoid using null?

Why would you want to do that?
Please explain in detail!

Your code (if it even is your code which I'm now beginning to seriously doubt) uses the check for null explicitly so that would be the way to go.
You could of course use a special instance of whatever class ( it is not shown in your code sample) to signify that the searchString wasn't found, but that special instance also would have to be assigned to searchValue every time somewhere at the top of the loop before coig through the array comparing searchString against studentName.

Regards,

Manfred

"I had the right to remain silent, but I didn't have the ability!"

Ron White, Comedian

Re: C# Error check

Harsh Athalye5-Dec-14 0:14

Harsh Athalye

5-Dec-14 0:14

Apart from resetting searchValue variable to null, I think you also should break out if for loop as soon as you found the matching value in array.

How to copy a file from my computer to a shared folder of another computer with login username and password account

Derz3-Dec-14 14:58

Derz

3-Dec-14 14:58

Any answers are appreciated.

Thanks....

Re: How to copy a file from my computer to a shared folder of another computer with login username and password account

PIEBALDconsult3-Dec-14 16:23

PIEBALDconsult

3-Dec-14 16:23

What Operating System?

With Windows (7) I have been using NET USE to create a mapped drive, COPYing to it, and then using NET USE to /delete the mapped drive.

f:\>net use ?
The syntax of this command is:

NET USE
[devicename | *] [\\computername\sharename[\volume] [password | *]]
        [/USER:[domainname\]username]
        [/USER:[dotted domain name\]username]
        [/USER:[username@dotted domain name]
        [/SMARTCARD]
        [/SAVECRED]
        [[/DELETE] | [/PERSISTENT:{YES | NO}]]

invalid length for a base-64 char array or string

Jassim Rahma3-Dec-14 11:35

Jassim Rahma

3-Dec-14 11:35

Hi,

I would like to ask why I am getting the following error on the below code please?

The error is: invalid length for a base-64 char array or string

and this is the code:

string server = cipher_utility.Decrypt<RijndaelManaged>(xmlClass.read_registry("Softnames", "ServerHost"), "{1234567-1234-1234-1234-123456789012}", "{12345678-1234-1234-1234-123456789012}");

Thanks,
Jassim

Technology News @ www.JassimRahma.com

Re: invalid length for a base-64 char array or string

Richard MacCutchan3-Dec-14 20:56

Richard MacCutchan

3-Dec-14 20:56

Probably because the encrypted data has not been stored (or encrypted) correctly. And you would find debugging your code much easier if you did not write compound statements. It is just as efficient, and so much easier to debug, and to read, to write it as:

string encryptedData = xmlClass.read_registry("Softnames", "ServerHost");
string server = cipher_utility.Decrypt<RijndaelManaged>(encryptedData, "{1234567-1234-1234-1234-123456789012}", "{12345678-1234-1234-1234-123456789012}");

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.