Why is KeyValueConfigurationCollection not sealed? [closed] - c#

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 6 years ago.
Improve this question
KeyValueConfigurationCollection is not sealed.
All other configuration collections are sealed for both netframework and corefx:
NameValueConfigurationCollection, ConnectionStringSettingsCollection, ProviderSettingsCollection, SchemeSettingElementCollection.
There are some subclasses of KeyValueConfigurationCollection in corefx repository, but only for test purposes.
Is KeyValueConfigurationCollection not sealed only for tests or there are other design reasons for that?

A class is unsealed so others can inherit from it. So presumably it was left unsealed because it was thought that people would want to subclass it. The fact that it is subclassed for test cases, indicates that it was the right thing to do. If subclassing it can help with testing it (or something else) then there are probably other scenarios where it is beneficial.

Related

why is Read7BitEncodedInt(); in BinaryReader protected? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 3 years ago.
Improve this question
I know you can easily create a wrapper around BinaryReader and expose Read7BitEncodedInt
But I'm just curious why the creators chose to not make it public
Is there a logic reason for it?
My guess is that this is internal implementation detail and isn't required to effectively use the BinaryReader. I wonder the opposite, why isn't it private? Presumably, there is a subclass out there that needs to use it or overwrite the implementation...

When to use abstract class over an interface with default implementations? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 3 years ago.
Improve this question
C# 8 adds the option for default method implementation in interfaces, meaning a method CAN be defined in the interface.
In a situation when I DON'T need a default constructor, nor inherited fields in the derived classes, what should I default to: An interface with a default implementation, or the old-school way of an abstract class?
Is there a convention for styling for this case?

Clean Code - naming related classes [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
I've read Clean Code by R.C. Martin and I'm trying to adopt his suggestions about clean code as broadly as possible.
But I'm not sure how to name related classes.
Let's say I have a class named TreeDirectoryList.
I want to cut implementation of this class into many smaller classes.
Let's say I'll create a class named ParentIndexStack.
ParentIndexStack will implement functionality very dependent on TreeDirectoryList, so it's very not probable that this implementation of ParentIndexStack will be useful with any other class in the future.
But the name of ParentIndexStack is very generic, it's possible, that I'll need another class with the same name, within the same project.
So I thought I'll name ParentIndexStack more precise, like TDLParentIndexStack (prefix TDL is from TreeDirectoryList).
Would it be correct ?
I'll end with many classes starting with TDLxxxxx.
One option is to put that set of classes in their own namespace. Then you can have simple, concise names that still communicate the full meaning of the class through the namespace context.

Define constants needed in more than one class [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 5 years ago.
Improve this question
In my smart client solution, I have a Project folder with:
IProjectView.cs
*ProjectView*
ProjectView.cs
ProjectView.Designer.cs
ProjectView.GeneratedCode.cs
ProjectView.resx
ProjectViewPresenter.cs
I want to define some constants for user by ProjectView.cs and ProjectViewPresenter.cs. Both of these classes implement IProjectView.cs, so were I back in Java, I'd put them there. If this were C++, I'd create a class ProjectConstants.cs and have the classes inherit it, but C# doesn't allow multiple inheritance.
How do I do this?
Can having a Read-Only Property in your interface solve your problem?
string MyReadOnlyProperty { get; }
I have no clue if this is very performance-wise compared to constant thought.

Is "massive" usage of #region considered a bad practice? [closed]

Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 9 years ago.
Improve this question
My MainViewModel has to deal with a lot of commands with complex actions inside, therefore its extension has grown inevitable. For keep the code organized I tend to use #region to group similar o related tasks.Does this considered an overuse of this feature or it's perfectly normal? You may say that it depends on me, if I feel right about it. I think that it helps a lot but I would like to know what do others. Here is a screenshot of how the code look like:
The bad practice is a massive single class. #region is simply hiding / coping with it.
If there are groups of methods, delegate them to another class.

Categories

Resources