I wrapped up the Introduction to C# course today by reviewing methods and exception handling in C#. Like Java C# supports access controls such as private, public, protected, and static.
My certificate is “signed” by Satya Nadella, which motivates me.
I’ve never quite understood when the right time to use these types of access modifiers is. Conceptually it makes perfect sense, but I supposed I have not done enough OOP to come accross a case where I wouldn’t just want every single method to be public. From my understanding, it has to do with API design (in the strict sense of API not just “REST”). The whole purpose of OOP is to encapsulate pieces of code for further reuse. Public, private, static, and protected are meant to enforce the API contract that another developer using your library might be trying to implement.
You do not need to understand how the code in a method works. You may not even have access to the code, if it is in a class in an assembly for which you do not have the source, such as the .NET Framework class library.
This is something I hope to investigate and learn more about in the next few courses of this series.
I learned about ref and out, which are a bit strange to me. They allow you to return multiple values from a method call without having the method itself return anything at all. In python if we want to return multiple values we would usually return an array or some other list.
I also learned that you are able to used named parameters in C# methods. This is really cool. If you have a complex method that takes many optional arguments, you can specify the specific ones that you want by name. Or if you want to reduce ambiguity for which parameters are being passed in you can give them a name. I think this makes code a lot more readable.
C# Exception Handling
I also learned about C# Exception handling. This is actually something that I already knew a bit about since I spent a few yeasrs of my life in a previous job staring at C# call stacks trying to figure out what went wrong.
Overall, the course was fine. I think it was a bit short and I wish that it would have had more non trivial example problems. I am still going to keep going and start the next course in the series, Object Oriented Programming In C#, tomorrow.
I also spent a little bit of time playing around with Azure Notebooks. It is basically Jupyter as a service. I have used Jupyter quite a bit. Not so much for data science, but more for exploring various API’s in python along with notes. Azure notebooks is free to get started so if you have never played with Jupyter before I think its definitely worth checking out.