Easy mistakes in c#

ACCESS MODIFIERS

c# has some access modifiers as below:

public:class or member can be accessed by no restrictions 

private:members can only be accessed within the class itself

internal:class or member can only be accessed within the assembly,not others.

protected:members can only be accessed by derived class(or within the class of course), not others non-derived class.

protected internal: members can only be accessed by derived class, OR, in the same assembly.

NOTE:

1. not all access modifiers can be used on class,that's why there's underline on "class or member" and "members". it's nonsense to use private,protected,protected internal on a class. This may make mistake easily.
2. no access modifier allowed on namespace,but public is implicitly.
3. internal is default to a class
4. no access modifier allowed on a interface but public is implicitly.
5. private is default to members.

Modifiers and Static

Can not use override,virtual or abstract on a static member, cause these are for instance of inheriting

Size of basic types

type	size(bits)	Range
char	16(2 bytes of unicode)
byte(unsigned-byte)	8	0 ~ 255
sbyte(signed-byte)	8	-128 ~ 127
short	16	–32,768 ~ 32,767
ushort	16	0 ~ 65,535
int	32	–2,147,483,648 ~ 2,147,483,647
uint	32	0 ~ 4294967295
long	64	–9,223,372,036,854,775,808 ~ 9,223,372,036,854,775,807
ulong	64	0 ~ 18,446,744,073,709,551,615
decimal	128

dynamic

for most time, dynamic is just the same as object,(xx is dynamic) equals to (xx is object) so:

int a=1;
(a is dynamic) is true

but (null is object) is false, so:

string a=null;
(a is dynamic) is false,of course

dynamic i=1;
(i is object) is true

params

No additional prameters are permitted after "params" prameters, and only one params keyword is allowed in a method argument declaration.

A "params" parameter has to be an array. 

bitwise

&: computes the logical bitwise AND
|: computes the logical bitwise OR
^:if and only if one operand is true then true,or maybe i can say, if the operands is not the same then true otherwise false.
7^2=>111^010=>101=>5

class and struct

1. class is a reference type that stored in heap,and struct is value type that stored in stack. Struct often used to present light weight object which only contains data.
2. struct can have constructor which have parameter(s), default constructor(paramterless) is not allowed.
3. struct can have const,field,property,method,index,operator,event,but if you need these members,then a class may be better.
4. struct can implement interfaces,but can not inherit from a class,so members of struct can not be protected.
5. you can declare a struct object without "new" keyword: 
   MyStruct myStruct;
   
6. only class can have deconstructor

boxing & unboxing

boxing means value type data in stack "be boxed" to manage heap
unboxing mean reference type data in manage heap "be unboxed" to stack.

but why we call that "boxing" and "unboxing"? i think just because managed heap is a place where will be cleaned(by GC),so we will package up garbages,or  "box" garbages up and put them onto the heap.But sometime we don't want it to be collected by GC,so unbox it from heap and put them onto stack. Just kidding^^

Equals and ==

When we're talking about equal, actually we're talking about two kind of equal: logical equal(having the same value) or reference equal(having the same reference)

For reference objects, Equals method and the "==" operator have the same behavior, that is, comparing two objects to see if they have the same reference.But sometime we want a little bit change so we override the Equals method to let it compare the value rather than reference,String type is a classic example, when we use Equals method to compare two string,actually we are checking if they are the same sequence of chars.

Things is a little bit complicate for value type.The Equals method of value type is used to check the value equality, but value type don't have the "==" operator, unless you overload it.

Dictionary and Hashtable

1. Dictionary and Hashtable are based on hash,Dictionary is actually a hashtable. Dictionary supports generic type while Hashtable don't, so Dictionary is type-safe,and has a better performance, because we don't need to convert the value which is object type to what it actually is,so no risk of boxing/unboxing.
2. Both use the GetHashCode method to check the uniqueness of keys,but if there is already a key which have the same hashcode,then the Equals method will be called to check if they are ference equals.

Interface

1. interface can extend another interface
2. interface CAN have indexer! check it on MSDN
3. interface CAN have event! check it on MSDN
4. "Interfaces can contain events, indexers, methods, and properties." ——MSDN
5. You can implement multi interfaces which have method(s) with same name!! In this situation,you should implement them explicitly:
   

   public interface ITest {
       void Test();
   }
   public interface ITest2 {
       void Test();
   }
   public class Dual : ITest, ITest2
   {
       void ITest.Test() {
           Console.WriteLine("ITest.Test");
       }
       void ITest2.Test() {
           Console.WriteLine("ITest2.Test");
       }
   }

   And the methods can not be public.In order to access the method, you have to first convert the class to the interface type:

   var dual = new Dual();
   // Call the ITest.Test() function by first assigning to an explicitly typed variable
   ITest test = dual;
   test.Test();
   // Call the ITest2.Test() function by using a type cast.
   ((ITest2)dual).Test();

 What is serialization?

Serialization is a converting from data structure or object state in memory to byte stream so that it can be store,transfer,or turn into other format like xml,json under helping of formaters.

Foreground thread and background thread

1. A managed thread is either foreground thread or background thread.Foreground thread(s) keep application running,while background don't.
2. Process can not be stopped until all foreground threads are stopped.
3. Process won't be stopped by stopping Background thread.
4. Thread is foreground by default,you can set the IsBackground property to be ture if needed.

Mutable object and Immutable object

Mutable means "changeable" and Immutable means "unchangeable"

Immutable object means that once you create the object, you can never change its state.For example, string is immutable, when you declare a variable of string : string a="foo", then an object of string "foo" is created, when you do this:a="bar", you are actually creating a new instance of string ,the origin instance "foo" still exists and never ever be changed.

1. "obj is dynamic" is always true, nomatter what type is obj,unless obj is null。

2."params" parameter must be the last one,even to "out" parameter

3.Why can not a struct have default constructor??

Calling GetType on a Nullable type causes a boxing operation to be performed when the type is implicitly converted to Object. Therefore GetType always returns a Type object that represents the underlying type, not the Nullable type.

so basically your code is equal to:

    DateTime? dt = DateTime.Now;
    object box = (object)dt;
    Console.Write(box.GetType().ToString());

also, looking at "Boxing Nullable Types" on MSDN we read:

If the object is non-null -- if HasValue is true -- then boxing occurs, but only the underlying type that the nullable object is based on is boxed. Boxing a non-null nullable value type boxes the value type itself, not the System.Nullable(Of T) that wraps the value type.

this clearly explain the "strange" behavior of Nullable<T>.GetType()