Wednesday, September 21, 2011

Generic


C# generics and C++ templates are very similar but they work differently. Generic exist to write code that is reusable across different types.

Let’s imagine we need a stack of integers if we didn’t have generic types, one solution is to hard code a separate version of a class for every required type “IntStack” but when we need stack of string have do the same way “StrStack”. This would cause considerable code duplication.

However this wouldn’t work aAnother solution would be write a stack that is generalizes by using Object as element type.s well hard coded “IntStack” & “StrStack” an “ObjectStack” require boxing & down casting that could not be checked at compile time.

int x = (int)stack.Pop(); // down casting

What we need is both a general implementation of a stack that works for all element type, and a way to easily specialize that stack to a specific element type for increased type safety and reduced boxing and casing.

Generics provide this by allowing parameterize element type.  Stack<T>  has the benefits of both “ObjectStack” and “IntStack”.

Test this example for further understanding.

    class Part
    {
        private string _PartId;
        private string _PartName;
        private string _PartDescription;
        private double _Weight;

        public string PartId { get { return _PartId; } }

        public Part(string PartId, string PartName, string PartDescription, double Weight)
        {
            _PartId = PartId;
            _PartName = PartName;
            _PartDescription = PartDescription;
            _Weight = Weight;

        }

        public override string ToString()
        {
            return string.Format("Part Id: {0}, Part Name:{1}, Part Description:{2}, weight:{3}", _PartId, _PartName, _PartDescription, _Weight);
        }


    }

 

   //the actual indexer, which accepts the generic type
    //note the type must be a class (not a value type)

    class Indexer<T> where T:class
    {
        struct ItemStruct
        {
            public string Key;
            public T value;
            public ItemStruct(string key, T value)
            {
                this.Key = key;
                this.value = value;

            }

        }

        List<ItemStruct> _items = new List<ItemStruct>();

        // T must be a class so that can return null if not found
        public T Find(string key)
        {
            foreach (ItemStruct _ItemStruct in _items)
            {
                if (_ItemStruct.Key == key)
                {
                    return _ItemStruct.value;
                }
            }
            return null;
        }

        public void Add(string key, T value)
        {
            _items.Add(new ItemStruct(key, value));
        }

    }

 
// Programe to Test it

    class Program
    {
        static void Main(string[] args)
        {


            Indexer<Part> indexer = new Indexer<Part>();

            Part p1 = new Part("001", "Mother Board", "Intel Chip set Mother Board", 0.75);
            Part p2 = new Part("002", "Processor", "Intel Core i5", 0.2);

            indexer.Add(p1.PartId, p1);
            indexer.Add(p2.PartId, p2);

            Part p = indexer.Find("002");

            Console.WriteLine(p.ToString());

           
            Console.ReadKey();
        }
    }








Tuesday, September 20, 2011

Structs in C#


Struct is very similar to a class but strcut is value type and class is reference type, .Net supports the value type and reference type unlike Java, in Java you can define only reference type.  Strcuts may leads to good performance where it’s used properly.

Reference types instances allocating in the managed heap and are garbage collected when there are no further references. Value types are allocated in the stack and so allocated memory is reclaimed as soon as their scope ends.  Using a value type instead reference type may result in fewer objects on the managed heap, which results less load to garbage collector (GC) and better performance.

Since for a value type, a compiler by default doesn’t generate a default constructor, struct doesn’t have parameter less constructor. And no finalizer, no virtual members and it derives only from System.ValuType, This System.ValueType may again derives from System.Object .    But structs can implement interfaces, in C# all primitive data types are value type except System.String
However value types have their poor sides too, when passing around big struct costlier than a reference type.

An example of declaring constructors.
    struct Point
    {
       private int x;
       private int y;

        public Point(int x, int y)
        {
            this.x = x;
            this.y = y;
        }
    }

An example of calling struct constructors.
              Point point1 = new Point();
            Point point2 = new Point(1,3);

Wednesday, September 14, 2011

Crystal Report Problems solving with Visual Studio 2010


CRVS2010 Beta - file not found error crdb_adoplus.dll  

An error when running the product when have installed CR 2010.

"Could not load file or assembly 'file:///C:
Program Files
SAP BusinessObjects
SAP BusinessObjects Enterprise XI 4.0
win32_x86
dotnet1
crdb_adoplus.dll' or one of its dependencies.

Solution

Setting the useLegacyV2RuntimeActivationPolicy flag to True does "solve" the problem. However, that is a hack and not a true fix. In trying to create a purely .NET 4.0 application, the end user shouldn't have to add this flag to get an application to work. Hopefully someone at SAP will note this and fix it in the core Crystal Reports functionality before the final version for VS 2010 is released.

Add following section to your config file and it may resolve the issue.
<startup useLegacyV2RuntimeActivationPolicy="true">
  <supportedRuntime version="v4.0"/>
</startup>

Mapping data flows in Azure Data Factory

  Overview: Data flows are one of the features inside the Azure Data Factory which allows data engineers to develop data transformation logi...