Why to use C# and when to prefer other languages?!

Introduction and Background

Either it is Quora, or is it C# Corner or is it CodeProject, beginners and novice users are always asking questions like, “Which one to use, C++ or C#?”, and many more similar questions. It is harder to provide an answer to such questions on multiple platforms and to multiple questions. That is why, I thought, why not post a blog post to cover most of the aspects of the question. In this post I am going to cover a few of the concepts that you may want to understand to chose either language from the most widely used languages:

1. C or C++
2. Java
3. C# (or .NET framework itself)

Basically, my own preference in many cases is C#, but that depends on what I am going to build. I have been using C, C++, Java and C# in many applications depending on their needs, depending on how they allow me to write the programs and depending on whether it is the suitable programming language for this type of project and application architecture.

The way, I do that, is just a trick. Which, I am going to share with you, in this blog post. Among these a few of the very basic things to consider are:

1. How simple and easy it would be to use this language in the program?
2. What is the productivity of that language in your scenario!
3. Difficulty level and how many people know the language!

I will be talking about these things points in this post, so to make it clear where to use which programming language. But my major concern in this post would be to cover the aspects of C# programming language.

Productivity of language

First thing to always consider about the language is the productivity that a programming language can provide your team with. Productivity of the language depends on many factors, such as:

1. How programs are built in that language?
2. How fast and reliable is it to build the project regularly?
3. Is the program’s source code readable for other members on the team?
4. Is the language itself suitable for the case that we are going to use it in.

Productivity of the language, in my opinion, is the first thing to consider as the valid candidate to always discuss in the team and not just talk about in a black room. You should manage to get your teams up in a conference room, and then talk about a language. There may be many aspects where one language may fall short and other may get well. There would be many factors that most of the programmers don’t know but talking about them one of the geek would stand up and raise another important point that may guide your team to walk on the correct paths to build greater applications.

Let’s talk a bit graph.

Figure 1: Productivity graph of most widely used programming languages.

In this chart we can see that C is the most productive programming language, of which we are interested in; C, C++, Java and C#. C++ and C# are competing with each other, whereas Java is a bit behind than C# and so on.

In the above graph, it is clear that C# is an average programming language. The average doesn’t mean it is below the requirement. But it means that it can be used in any case, and its performance won’t fall as the data increases. The successful run graph shows that C# programs are much successful in many cases. That is why, in many cases, C# proves to be the valid candidate in many cases, since it is a general purpose programming language.

Then the question arises, “Productive in which sense?”

That is the most important part here. Now, this topic may take a bit more of the time to explain how productive a language is. For that, I have dedicated the topic, “Choosing the right language for the right task“. The right programming language for the right task would always help you to elevate the speed of the programming of your team! But among other aspects, the most important ones are:

1. What sort of project this is?
   • Of course you don’t want to use knife to unscrew a screw.
2. What are the IDE and compilation tools present?
   • Even a program written in C can go wrong if compile is buggy or inefficient.
3. How your team would like to manage the projects?
   • Does the language support good software engineering concepts?
   • Can it be used to generate good diagrams: Use-case, activity, class and other diagrams.

Thus, you should always consider thinking about the productivity of your development team while guessing which language to be used.

Choosing the right language for the right task

It is always stated to use the best tool for the best jobs! Then, why not use the best programming language for your projects, to get the best possible results? Before I continue any further, there is one thing I want to mentioned… “There is no best programming language ever built.” Every time a programming language is built for a purpose, another group of programmers jump in to create a new programming language, for the sake of fixing the errors in the previous language. What do you think motivated Bjarne to create C++ when there was C language and Scala already in the market?

In this section, there are many things to consider, from the performance and benefits to the clients, to the perk packages for the employees to the efficiencies of the project repositories and the tools provided for that programming languages.

C# was created by Microsoft and therefore, in my opinion, has, by far, the most efficient tools for programming. I mean, Visual Studio, alone is the beast in this game.

Figure 2: Visual Studio logo.

I am a huge fan of Visual Studio, and i would doubt someone who isn’t a fan of Visual Studio. C# has a better support and benefit of using the best IDE out there. Java has also a number of IDE, so does C++ and many of the C programs are written in minimal environments, like a small program to manage and compile the C programs; no offence geeks!

Figure 3: Graph of performance to safety ratio.

Now, if you look at this graph, it’s pretty much clear as to what it is trying to guide you with. Of course, as I had mentioned in the starting paragraph here, that there is no best programming language.

In many cases, C# and Java rule over C++ (and C) and in many cases, they rule over C# and Java. There are many factors, like the programming paradigms, performance of the code that is generated; Just-in-time compilation, memory-management delay and so on. While C# and Java may provide the best environment to build “managed” programs in, there are many cases where C# and java don’t work well, like writing a low-level program. Java developers wanted to build a Java OS, but they had to give up because something’s aren’t meant to be done in Java. 🙂

Always consider to search before making the final call. There are many companies working in the similar field that you are going to work in. There would be many packages and languages built for your own field that may help you to get started in no time!

Figure 4: Top 10 programming languages.

But, I think these are a bit backwards. I think, C is on the top because it causes a lot of trouble to beginners, so everyone is searching for “How to do {this} in C” on Google, raising the rankings. 😉

Selecting the best framework

I don’t totally agree with people when it comes to talk about frameworks like, Java, Qt (which I do like in many cases; like Ubuntu programming), and other programming frameworks available for programming applications to be run despite the architecture of the machine. In this case, my recommendation and personal views for .NET framework are very positive. As, already mentioned, I have programmed on Qt framework for Android, Ubuntu and Linux itself. It was a really very powerful framework to build applications on. But the downside was it was tough to learn, their compilers were modified, their C++ was tinkered.

While selecting the best framework for application development by choices are below:

1. How much flexible a framework is?
2. What language does it support?
   • Some frameworks support multiple languages, like .NET framework, it supports C#, VB.NET, Visual C++, JavaScript applications.
3. Is it cross-platform?
4. If not cross-platform, then does it support multiple architectures, at least?

Java framework is cross-platform, and entirely framework oriented. You simply have to target the framework despite the operating system or architecture being used. .NET framework on the other hand is a very beautiful framework to write applications on. It uses C#, VB.NET and C++ (no Java!) to write the applications, and then the compiled binaries can be executed on many machines that can support .NET framework. This provides an excellent cross-architecture support.

C# however, does not support Mac OS X, at the moment. Microsoft has started to roll out cross-platform binaries for C# programs. .NET Core has been a great success and once it gets released to a public version, I am sure most of the companies would start to target it. That is not going to happen, in a near future. In which case, Java and C++ are better than C#.

If you are interested in C# programming on multiple platforms, consider using Mono Project instead. You can read about that, on my blog here: Using C# for cross-platform development.

Figure 5: Top languages and their platforms of usage.

Java may be supported 100% in the rankings, but C# is also supporting multiple platforms and is rapidly growing making the language better than the rest. With the release of C# 6, Microsoft has proved that the language is much better than the rest in the race. There are many features that I like C# 6:

1. String interpolation
2. Getter-only auto-properties as lambdas
3. Improvements to lambdas

There are a few things Java still doesn’t have. For example, one statement code to write the data to the files and to extract the data. You have to get messy in the streams, or you have to write an entirely non-intuitive code with Files object and then to get the data from there… Yuk!

Performance of the compiled code

Writing the source code may be different in many ways:

1. Syntax of the programming language.
2. The way their objects and procedures are imported in the source code.
3. How clean the source code looks. Many programming languages are just nightmares.

But the major concern comes to mind when we are going to execute the programs. In many cases, or should I say in all the cases, the language which are bytecoded languages, lag behind than the languages that are compiled to native codes or Assembly codes. For example, C or C++ codes are faster, because they are not compiled to a bytecode instead they are generated as machine codes for a platform or architecture. C# or Java programs are compiled down to a bytecode which causes a JIT to occur when the programs are executed. This takes time.

However, you can see the following charts from https://attractivechaos.github.io/plb/ and see for yourself, the way bytecoded languages are similar and how much compiled languages are different, see for yourself.

Figure 6: Mathematical calculations.

Figure 7: Pattern matching and machine learning.

Which makes it pretty much clear, how much compiled language are faster than bytecoded, and then come the ones that are interpreted, like Ruby.

In many cases, there are other factors too, which cause bad performance:

1. Bad compiler.
2. A lot of resources being allocated.
3. Slow hardware resources; bad combination of CPU, RAM and other peripherals.
4. Bad programmer and bad code being used.
5. Bad practices of programming.
6. Keeping CPU halted for most of the times.

Much more are a valid candidates for the halting processes.

Finally… Final words

As I have already mentioned that there is no best programming language out there. One language is best in one case, other is best in another case and so on and so forth. In such cases, it is always better to communicate with your development team and then ask them questions and ask for their feedbacks. Choosing the good tools for your projects is always a good approach and good step in the process of programming.

If your software engineer or software architect are trying to find a good solution, ask them to work on the following questions:

1. What are the teams and developers qualified for?
   • Asking a team of C++ programmers to leave C++ and join Java or C# programming teams is not a good idea!
   • Always consider the best approach.
   • Time is money — Manage it with care.
   • Recruit more programmers with skills.
2. Is the programming language long lived, or a minor one?
3. Is programming language capable of making your application work?
   • I have used many programming languages, and thus I can decide which language to use. This is the job of your software architect to decide which languages to select.

If you follow these rules, you will save yourself from many of the future questions that cause a lot of problems. Many times developers ask questions like, “I have an application in Java, how to migrate it to C#?” Many web developers ask questions like, “I have a PHP application, how to convert PHP code to ASP.NET?” These questions have only answer. “Re-write the applications”.

There are many things that you should consider before designing the applications. Facebook is stuck with PHP, because it was written in PHP. Even if they wanted to migrate to ASP.NET, they won’t. Instead, they have found a work around the PHP bugs and downsides.

This is why, you should always consider using a conference or meeting to decide how to start the project, which programming language to use, what frameworks to target, who would lead the team and much more. These few days of discussion and designing would save you a lot of money and time in future.

Installing and setting up R environment on Windows

Introduction and Background

One day in the previous week, I heard my professor say something about “R language” and I said to myself, “What is this R language?”. Of course I had no idea about what that language was but I was a bit curious to know about it and to learn about it. Luckily, I was provided with a few concepts about R language on C# Corner’s R language category and later I Googled stuff around to learn more about it. In this post, I am going to walk you through different things that I have learnt during my learning time.

In this post, I will walk you through many steps to set up the environment for R programming. I will be covering the basic R environment set up and later I will talk about using Visual Studio as the default IDE for R programming, after all, “Who doesn’t like Visual Studio?!”.

What is R, anyways?

A good introduction to the R language and R environment is given by Mahesh on C# Corner, I would recommend that you read that one, What Is R Language.

R is a programming language, as well as an environment for computation. What sort of computation? Statistical computation is the major field for R. However, many fields are using R language for their purposes, for example, Data Scientists are using R language for many analysis that they have to perform on data sets. Data mining etc. come later in the picture.

I will point out a few of the major fields that R can be used in, later, but first a bit of the intro to R language would be useful.

Figure 1: R logo.

R language was developed in the early 1990s for statistical computation purposes. It started all with small modules, then it gathered and gained much reputation from online communities and it became a very popular environment to use in many fields of computer-related sciences. It was developed using C and FORTRAN as the basic language for interpretation.

R is distributed under GNU license and allows modifications to be made and shared online. R was developed in C and FORTRAN. That is why, you can modify the source code and recompile the binaries to make it suitable for your own needs. You can do that, too!

Why learn R?

As already mentioned that R is a programming language and environment for statistical programming and computation. That is why, R is a very prominent language to be used in data science projects. Specifically, the R language is used instead of C, Java or even C++ in these cases because the packages in R are already ready to be used and in no time, you get to use the best of the data science in this environment.

Figure 2: A chart for benefits of learning R language. 

Even if you are not an open source developer, you would still find R language very much helpful. The usage of R ranges from field to field. As already depicted in the diagram above, this project can be used in many ways, in many fields and for many purposes. A few of the basics where, this can be used can be categorized as following:

1. Working with data science projects.
2. Building reports for data analysis.
3. Creating presentations with graphical representations of results.
4. Adding your own extensions to the projects.
5. Educational use.

I will write some articles about using R in many ways, for many projects and tasks in computer science, but I would like to leave that all to a later time. Until now, just know that if you are going to analyze large amount of data sets, process data mining and similar projects. Then your best bid would be R language instead of C++ or Java.

Another beauty of R language is that it can be coupled with many other famous programming languages and environments and frameworks. For example, R functions can be called from C, C++, Java, Python and from the .NET environment, easily. You would be using the R language to:

1. Process the data from CSV or TXT files.
2. Generate summary reports in the form of graphics or tables.
3. Use a special type of Markdown format to build the results.
4. Use multiple formats to extract the data results.

There is much more to learn R for. Many giants such as Google and Facebook are already using R for their computation purposes, because it is open source and can be updated as per needs. It allows educational institutes to teach their students in a much simpler and easy way.

Setting up R environment

First step to perform would be to install the R environment packages on your machine. You can get the binaries for installation packages free of cost, from R website itself. Go to the CRAN network and select your own operating system and download the binaries to be used for installation purposes on your own machine.

I am using Windows operating system, so I used Windows instead. You can download the binaries for your own system. Start the setup and continue to follow the steps that are provided on the UI itself.

Figure 3: R installation setup for Windows.

Later you would be asked to select the location where you want to install the packages. Chose a location, but recommendation is to leave it as it is.

Figure 4: Location selection process.

Then the environment would be installed. You can run the environment programs by selecting either of the following icons.

Select the one that suits your architecture, mine was 64 bit so I executed the one for my architecture.

Figure 5: Running the R console.

Just to test that everything is running and set up fine, execute the following test on it.

> 3 + 3
[1] 6

FIgure 6: Running a test in R console.

This shows the R is set up and we can continue to use it in our projects for our own purposes. Now, you don’t want to work on your projects using this console only. For that, you would require a full-featured Integrated development environment; IDE. In the next steps, I will walk you through many steps to install the IDE for R.

Setting up R using Visual Studio

Visual Studio is, no doubt, the best IDE available for everything. It already supports C, C++, C#, VB.NET, Python and many more languages and frameworks such as web development languages of HTML, CSS and JavaScript.

R language support has also just arrived with the “R Tools for Visual Studio” extension that you can install to get started with using R language in Visual Studio. But for this to run, you need to have Visual Studio 2015 with Update 1, any edition would work from Community to Enterprise.

To set up, you can go to the URL provided above. Download the tools, you can also install the Microsoft R Open manually. Microsoft R Open is what we are going to use from now on. Go to this website for Revolution Analytics, and start setting up MRAN (Microsoft’s distribution of CRAN)

1. Download Microsoft R Open and install it.
2. Download MKL Math Library, which is going to enhance the performance of MRAN.

Figure 7: Microsoft R Open installer.

After that, install the MKL and you’re done!

Figure 8: MKL installer.

After that, just run the Visual Studio IDE to get started with your first R program. That, I am going to cover in an upcoming post.

Points of Interest

In this post, I have talked about a few of the stages that you can undergo to set up your R environment. Before that, I talked about why and how to use R. Apart from that, I also shared a few of the things required to set up R environment on Windows and to use Visual Studio as the default IDE for programming with R language.

In the later posts, I will write about using R language for your computation that require a lot of data processing. See you in the next posts. 🙂

Handling events in WPF in an easy and short hand way

I would talk about the WPF events and how to handle them easily, perhaps you might already know how to handle them; XAML gives you a lot of good handy functions, but there are a lot of other good ways of doing the same.

Events in WPF

Events in WPF are similar to what we had in Console and Windows Forms. They provide us with notifications and procedures to handle the business logic based on what state of application is. They allow us to manage what to happen when the application is starting, what to do when the application is closing and when the user interacts with the application.

Interaction includes the button clicks, input value change, window resize and many other similar processes that can be handled to ensure that our business logic is always applied. Validation and other checks can be easily applied to check the values.

Handling the events

Although handling the events is another function, thus we require a function to perform (or get triggered) when a certain event is triggered.

Note: Please note that I would be using Visual C# for this article’s purpose, it would be different as to what VB.NET would provide you as intuition and library resource, so try to follow Visual C# and not VB.NET at the moment.

In C#, you can handle the event, by attaching a function to it. It is similar to say, “When this happens, do that“. You tell your application to perform an action when something happens. The similar action is perform when a new email is received in your email clients, or when download is complete; if notification gets raised for that download complete. The attachment of the event to another function is pretty much simple, the only thing to worry is the required dependencies (the parameters or values or objects required to handle the event).

Look at the following code,

<Button Click="handleEvent">Click me</Button>

The above code is a XAML code (remember we are in WPF), the function to handle the event would be like this,

void handleEvent(object sender, RoutedEventArgs e) {
   // 1. sender is the Button object, cast it
   // 2. e contains the related properties for the Click
   // Code here... Any code!
}

Now, the function handleEvent is attached to the Click event of the Button object. Remember that event needs to be raised in order to trigger that function.

Pretty much easy it is to handle the events in WPF, just embed the code in the XAML markup, and Visual Studio would generate the back-end code. Write what you want to be done and you’re good to go!

Using back-end code

As, you know you can do anything with back-end code. You can also attach the event handler to the control’s event using the back-end code. In this section, I would explain how can you do that using back-end code.

Since we used Button for previous example, I would use the same for this one. Attaching the event is as easy as 1, 2, 3…

button.Click += handleEvent; // 1

void handleEvent(object sender, RoutedEventArgs e) { // 2
   // Handle the event
}

Step 3 is the click event that is performed by the user. 😉

Easy isn’t it? (IMO, it is not! Look in the next section to make it even better!)

Simple way…

This section is the actual section, which I wanted to talk about. The simple way to handle the events in WPF. It uses the lambda expression, to generate the delegate functions which in turn are executed when the event gets raised. If you are unclear as to what lambda expression or delegate functions are, I would suggest that before continuing, you learn them from the links I have provided. 🙂

A very simple way (IMO) to handle the event in WPF is like this,

myButton.Click += (sender, e) => 
{
   // Handle the event
}

Notice the above expressions. It is a simple lambda expression, being resolved as a delegate function with two paramters (sender and e). The parameters are required by the event handler itself, so that is why we need to pass them.

Points of Interest

The above is the simplest way anyone can handle the event in WPF and is a very much short hand, because you leave everything to the context of the object itself. You skip one more step for casting the object, you also don’t have to create different functions with different appended names such as, “myButton_Click”, “myButton_MouseEnter”. All you need to do is just create a lambda expression and use it as the event handler. Pretty much simple isn’t it?

That’s all for now. 🙂 I hope, I might have helped you out in making programming WPF applications easier. I would post more of such helpful tips for WPF developers soon. 🙂 Stay tuned.

Sending emails over .NET framework, and general problems – using C# code

Most of the times new developers stumble upon a very simple task, “sending emails” over .NET framework. Well, I am not going to talk about any language here (C# or VB.NET, Visual C++ you’re also included), I am going to talk generally about the .NET framework and the assemblies exposed by .NET framework for sending the emails, using your own SMTP server’s settings, such as username/password combination, port number and (most specially) the hostname for your SMTP server.

Background

Emails stands for electronic mail, and they are widely used in regular basis for communication, you can send email for sharing text data or you can send your albums over emails easily. Email has been a part of the internet entertainment since a great time, and people use many different email clients, some love the online clients eg. Gmail, Yahoo! etc, and some prefer an offline version of their email clients which use the internet connection to download the emails from a server, such as Thunderbird, Outlook etc.

But the fact is that all of them use the same protocol for transferring the emails over internet network. In this blog post, I will talk about sending the emails over the network, downloading the emails is a totally seperate topic and would have a seperate protocol working on the back end to download the emails from the server.

Sending the emails

Emails are sent using the SMTP protocol, over the internet. It is similar to the Hypertext protocol (not in way of communication, but in a way that is a protocol for communcation). For more on SMTP you can find yourself glad to read the Wikipedia page, I am not going in depth of the protocol here, instead I will just elaborate the methods to send the emails over the .NET framework.

What does .NET framework offer me?

.NET framework (who is oblivious to this one?) has a bunch of cool assemblies for us to work with, using our favourite languages, from C# to C++ and the assemblies in the .NET framework allow us to focus on the quality of the application and the logic, leaving the rest of the low-level coding to the framework itself, including and most specially the garbage collection like stuff and memory management.

.NET framework has a namespace, known as System.Net. This namespace is responsible for the network communication for the .NET applications. But we will be more concerned about the System.Net.Mail namespace, for working with the mail protocol, which exposes the SmtpClient, MailMessage classes for us to easily just pass our data to the objects and send the email using .NET framework.

Creating the module for sending email

Since .NET framework exposes a lot of frameworks to create your applications over, starting from as basic as Console application, to as much user-friendly as Windows Presentation Foundation. Interesting thing is, that in .NET framework same code can be used on the back-end of a Console app and the WPF application. So, the code that would be used to send the email in a Console application is just the same as you would be using for the WPF application. That is why I am not going to specify any framework, instead I am going to use a Console application for our project, for being simpler to be understood, and to focus more on the code instead. You can (in your own IDE) create any kind of application you want, from Windows Forms, to WPF to a web application (using ASP.NET).

Once your application has been created, you can create a simple module (function; not to be confused with the VB.NET’s Module). Inside that you can write the following code, don’t worry I will explain the code in the coming section of the blog post.

// You should use a using statement
using (SmtpClient client = new SmtpClient("<smtp-server-address>", 25))
{
   // Configure the client
   client.EnableSsl = true;
   client.Credentials = new NetworkCredential("<username>", "<password>");
   // client.UseDefaultCredentials = true;

   // A client has been created, now you need to create a MailMessage object
   MailMessage message = new MailMessage(
                            "justin17862@gmail.com", // From field
                            "justin17862@gmail.com", // Recipient field
                            "Hello", // Subject of the email message
                            "World!" // Email message body
                         );

   // Send the message
   client.Send(message);

   /* 
    * Since I was using Console app, that is why I am able to use the Console
    * object, your framework would have different ones. 
    * There is actually no need for these following lines, you can ignore them
    * if you want to. SMTP protocol would still send the email of yours. */
    
   // Print a notification message
   Console.WriteLine("Email has been sent.");
   // Just for the sake of pausing the application
   Console.Read();
}

Voila, (if you added correct details in the above code) you would have your emails sent to destination without any trouble; apart from Internet connection trouble. Now let us dissect the code into pieces and understand what happened, after that I will mention a few problems that arise in programming and cause a havoc for new developers in understanding the entire process, which includes the errors that are raised due to problems in connections and so on.

Explaination of the above code

First step in the code is the usage of the using statement. In .NET framework, you stumble upon different objects that use resources, which need to be disposed off properly, or atleast closed. For example, while a file is created it is required to call the Close() function, before any other process can use that file, similarly some processes requires that you can a Dispose() function on them, to release all the resources. But you can use a using statement, to let the .NET framework take care of all of the objects that need such functions to be called itself. For example, the following code,

using (SmtpClient client = new SmtpClient()) {
   // code here
}

is better than,

SmtpClient client = new SmtpClient();
// code here..
client.Dispose();

due to many particular factors, that I am not going to talk about here. That leaves the discussion about the using statement, you would find a few more deeper detail about using statements on MSDN documentations.

Next comes the SmtpClient object, the SmtpClient is used to create an object that would establish the connection between your machine and the SMTP server you’re using. SmtpClient requires you to set up a few things in it.

1. Host name, which is the name of your SMTP server’s address in string format.
2. Port that you will be using to connect, default is 25 (TCP port).
3. Most of the connections require that you set the SSL active. You can see that happening in our code too.
4. Credentials are required before you can use a service, most of the servers (Gmail, Outlook etc) require that you send a username/password combination in order to send email from your account using the SMTP protocol. That is why in most of the cases default credentials forward the developers into errors. We use NetworkCredential object (from System.Net namespace) to pass our username/password to the server.

Since SmtpClient is disposable, that is why we’re using it inside a using statement. We’re about to send an email, for that we create an object called MailMessage, and we pass our data to it. MailMessage object can set From, To, Subject and Body fields of an email message and then can be sent. You can see in our example, we’re using the constructor to create the MailMessage object that would hold the data for our From, To, Subject and Body fields.

Finally, we’re sending the email using the Send() function. Interesting thing is, in a GUI framework such as WPF or Win Forms, we should be using SendAsync for the sake of asynchrony in our application that would help us to create a fluid GUI for our application, otherwise the application woul stay stuck until the email has been sent and the control continues from this line of code. To learn more on asynchronous programming, please move to the MSDN link and learn more from there, they’ve got a great content for beginners like you.

A few errors in programming

Generally, there are always errors that developers miss and then tries to confuse himself for “Where did I miss it?”. Similarly, in sending the email and establishing a secure connection, there are usually a lot of problems, some are syntax, some are logically, but I would talk about the connection errors that might be raised. I tried to raise some exceptions myself to share them with you here, for you to understand when these exceptions might cause a problems for your environment.

Usually, the exceptions in the connection are raised only at the Send, or SendAsync method when the SmtpClient is not able to send your email successfully. It can be due to the connection problem, authentication problem or any other problem.

Problems with SMTP hostname

A general problem can be the hostname that you’re passing to the client to connect to, it must be correct and without the “http://“. You might stumble upon such a problem,

Hostname could not be resolved, because it is having “http://” in it. Just pass the smtp.gmail.com, if you’re using gmail as your SMTP server. Otherwise you should contact the SMTP developers for their SMTP hostname.

This would be resolved, by making sure that the hostname is correct. Every SMTP provider has its own setting for its server. Make sure you’re using the correct ones. This is the first problem you would stumble upon if you’re going to get any error. Failure sending mail can also be raised if the Firewall is blocking the network. 

Another problem with the SmtpClient is, if you’re not using the correct port number, then the connection might not establish and the worst thing is that there won’t be any exception raised. For example, use 295 port number. The command would continue to execute without any success message or exception. Make sure you’re using correct port number, otherwise use the default TCP port number; 25. For me, 25 port number works always.

Errors authenticating the user

While servers require the correct authentication, it is compulsary that you pass correct and required authentication details to the server. First stage is to enable the SSL over your connection. Mostly, servers close the connection if the connection isn’t over SSL. Recall the code in this blog post, and see the enable SSL command,

client.EnableSsl = true;

After this, you should make sure that you’re using the correct combination of your username and password. If they’re incorrect, server is free to close the connection. The following exception is raised if any of such (authentication) problem occurs in your application.

Server requires SSL connection, or correct username/password combination. Make sure you’re not wrong in both of these scenarios.

Once these problems are resolved, (and other problems don’t come out) your email would be sent, and you will see a success message in your application, mine showed me the following.

Email successfully sent! Success message in the Console application.

Points of Interest

In .NET framework you can use the System.Net and its namespace to work with network. For mailing, you use the System.Net.Mail namespace. System.Net.Mail exposes an SmtpClient object, that uses a hostname and a port to connect to the SMTP server for sending the emails. Some of the servers require SSL connection, and credentials (username/password combination).

MailMessage is the object you would use to send the email, you can fill this object with the From, To, Subject and Body field of your email message. SmtpClient would send this object, you can use Send or SendAsync methods to send email, depending on your framework and the methods that you would use to send the email from.

Exceptions are raised in SmtpClient, when the code reaches the Send (or SendAsync) function. That is because the connection problems occur at this stage, server tells the .NET framework for the errors in sending the email, and the exception is raised. Usually exceptions are raised due to following factors,

1. Username/Password is incorrect.
2. SSL is not enabled.
3. Hostname is not correct, so the SmtpClient was not able to establish the connection at all.
4. If the port number is incorrect, there is no error message at all. This is a tricky part for every developer. This can be minimized by using 25 (default TCP port).

SmtpClient exposes Dispose() function, that is why it is better to use the SmtpClient object in a using statement, not just as a simple (and ordinary) object to call dispose over later. Using a using statement lets you leave the releasing of the resources to the .NET framework itself.

For those, who are looking for a VB.NET code, you can use the Telerik converter to convert your C# code in VB.NET code (or vice versa).

.NET allows you to use the same code over different frameworks and platforms that run over .NET. Such as WPF, Console app, ASP.NET web applications. That is why, you can use this code above in almost all of your applications, no matter software apps, web apps or what-so-ever client application you’re creating until it runs over .NET framework. Because these assemblies are present in .NET, not in the language itself.