(How)
C^# You Are - 17 December 2006

1. Which loop is more efficient?  Answer
   
   a) for (int nIdx = 0; nIdx < 1000; ++nIdx) {...};
   b) for (int nIdx = 0; nIdx < 1000; nIdx++) {...};

   a is the correct answer.  ++nIdx results in the following equivalent code.

   nIdx = nIdx + 1; return nIdx;

   Where as nIdx++ generate the following code.

   var tmp = nIdx; nIdx = nIdx + 1; return tmp;

   As a result the first code block generates less code and requires no temporary variables. You should prefer the first form whenever possible.  Note that the C# compiler is smart enough to optimize some common cases but it is always better to write efficient code when it can be done easily and cleanly.

   (Updated 28 Dec 2006 w/ comments received from James Curran)

2. What is an immutable type? Give an example?  Answer

   An immutable type is a type that stores a value that, once set, can not be changed. Immutable types are normally set when they are created. Value types should be immutable due to how the default implementation of some core methods work. An example of an immutable type would be String or DateTime.

3. What are the six visibility flags for type members in .NET?  Answer

   1. private
   2. public
   3. family
   4. assembly
   5. family and assembly
   6. family or assembly
4. Who can see an internal type member in C#?  Answer

   Access is limited to the owning assembly.

5. What are the valid visibility flags for a non-nested type in C#?  Answer

   public and internal.

6. What are the two uses for the keyword new in C#?  Answer

   The first usage is for creating new class instances. For example, MyClass cls = new MyClass();

   The second usage is to override a base class member that is not marked as virtual.

7. What is the character set used by .NET for strings?  Answer

   .NET uses Unicode exclusively. In order to use ANSI or another character set you must convert the string.

8. What is the difference between overloading and overriding?  Answer

   Overloading is the process of creating multiple methods with the same name but different signatures. Overloading is used to make a type easier to use by providing alternative versions of a method's parameters. Overloading requires no special keywords or other syntactical sugar. Note that the return type of a method is not part of the method's signature.

   Overriding is the process of replacing a base class member's implementation with a new implementation in a derived class. Overriding requires that the base member be marked as virtual or overriddable in the base class. Unlike overloading an overriddable member must be handled at runtime in order to take into account the polymorphic ability of types.

9. Given the following generic definition in C#. How many copies of the code are created at runtime?  Answer

   public class CollectionEx<T> : Collection<T>
   { ... }
   
   //Later in code
   CollectionEx<int> values = new CollectionEx<int>();
   CollectionEx<string> names = new CollectionEx<string>();
   CollectionEx<float> rates = new CollectionEx<float>();
   CollectionEx<Control> controls = new CollectionEx<Control>();
   

   For generic types one copy of the code is created for all reference types when the first reference type is used as the type parameter. A separate copy is created for each value type.

   Therefore there are 3 copies of the type in memory: int, float and reference.

10. How many generations are there in the .NET garbage collector?  Answer

    There are currently 3 as of v3.0 of .NET.