How do I clone a generic list in C#?

Question

I have a generic list of objects in C#, and wish to clone the list. The items within the list are cloneable, but there doesn\'t seem to be an option to do list.Clone()

Answer 1

public class CloneableList<T> : List<T>, ICloneable where T : ICloneable
{
  public object Clone()
  {
    var clone = new List<T>();
    ForEach(item => clone.Add((T)item.Clone()));
    return clone;
  }
}

Answer 2

The following code should transfer onto a list with minimal changes.

Basically it works by inserting a new random number from a greater range with each successive loop. If there exist numbers already that are the same or higher than it, shift those random numbers up one so they transfer into the new larger range of random indexes.

// Example Usage
int[] indexes = getRandomUniqueIndexArray(selectFrom.Length, toSet.Length);

for(int i = 0; i < toSet.Length; i++)
    toSet[i] = selectFrom[indexes[i]];


private int[] getRandomUniqueIndexArray(int length, int count)
{
    if(count > length || count < 1 || length < 1)
        return new int[0];

    int[] toReturn = new int[count];
    if(count == length)
    {
        for(int i = 0; i < toReturn.Length; i++) toReturn[i] = i;
        return toReturn;
    }

    Random r = new Random();
    int startPos = count - 1;
    for(int i = startPos; i >= 0; i--)
    {
        int index = r.Next(length - i);
        for(int j = startPos; j > i; j--)
            if(toReturn[j] >= index)
                toReturn[j]++;
        toReturn[i] = index;
    }

    return toReturn;
}

Answer 3

public static Object CloneType(Object objtype)
{
    Object lstfinal = new Object();

    using (MemoryStream memStream = new MemoryStream())
    {
        BinaryFormatter binaryFormatter = new BinaryFormatter(null, new StreamingContext(StreamingContextStates.Clone));
        binaryFormatter.Serialize(memStream, objtype); memStream.Seek(0, SeekOrigin.Begin);
        lstfinal = binaryFormatter.Deserialize(memStream);
    }

    return lstfinal;
}

Answer 4

Another thing: you could use reflection. If you'll cache this properly, then it'll clone 1,000,000 objects in 5.6 seconds (sadly, 16.4 seconds with inner objects).

[ProtoContract(ImplicitFields = ImplicitFields.AllPublic)]
public class Person
{
       ...
      Job JobDescription
       ...
}

[ProtoContract(ImplicitFields = ImplicitFields.AllPublic)]
public class Job
{...
}

private static readonly Type stringType = typeof (string);

public static class CopyFactory
{
    static readonly Dictionary<Type, PropertyInfo[]> ProperyList = new Dictionary<Type, PropertyInfo[]>();

    private static readonly MethodInfo CreateCopyReflectionMethod;

    static CopyFactory()
    {
        CreateCopyReflectionMethod = typeof(CopyFactory).GetMethod("CreateCopyReflection", BindingFlags.Static | BindingFlags.Public);
    }

    public static T CreateCopyReflection<T>(T source) where T : new()
    {
        var copyInstance = new T();
        var sourceType = typeof(T);

        PropertyInfo[] propList;
        if (ProperyList.ContainsKey(sourceType))
            propList = ProperyList[sourceType];
        else
        {
            propList = sourceType.GetProperties(BindingFlags.Public | BindingFlags.Instance);
            ProperyList.Add(sourceType, propList);
        }

        foreach (var prop in propList)
        {
            var value = prop.GetValue(source, null);
            prop.SetValue(copyInstance,
                value != null && prop.PropertyType.IsClass && prop.PropertyType != stringType ? CreateCopyReflectionMethod.MakeGenericMethod(prop.PropertyType).Invoke(null, new object[] { value }) : value, null);
        }

        return copyInstance;
    }

I measured it in a simple way, by using the Watcher class.

 var person = new Person
 {
     ...
 };

 for (var i = 0; i < 1000000; i++)
 {
    personList.Add(person);
 }
 var watcher = new Stopwatch();
 watcher.Start();
 var copylist = personList.Select(CopyFactory.CreateCopyReflection).ToList();
 watcher.Stop();
 var elapsed = watcher.Elapsed;

RESULT: With inner object PersonInstance - 16.4, PersonInstance = null - 5.6

CopyFactory is just my test class where I have dozen of tests including usage of expression. You could implement this in another form in an extension or whatever. Don't forget about caching.

I didn't test serializing yet, but I doubt in an improvement with a million classes. I'll try something fast protobuf/newton.

P.S.: for the sake of reading simplicity, I only used auto-property here. I could update with FieldInfo, or you should easily implement this by your own.

I recently tested the Protocol Buffers serializer with the DeepClone function out of the box. It wins with 4.2 seconds on a million simple objects, but when it comes to inner objects, it wins with the result 7.4 seconds.

Serializer.DeepClone(personList);

SUMMARY: If you don't have access to the classes, then this will help. Otherwise it depends on the count of the objects. I think you could use reflection up to 10,000 objects (maybe a bit less), but for more than this the Protocol Buffers serializer will perform better.

Answer 5

After a slight modification you can also clone:

public static T DeepClone<T>(T obj)
{
    T objResult;
    using (MemoryStream ms = new MemoryStream())
    {
        BinaryFormatter bf = new BinaryFormatter();
        bf.Serialize(ms, obj);
        ms.Position = 0;
        objResult = (T)bf.Deserialize(ms);
    }
    return objResult;
}

Answer 6

public static object DeepClone(object obj) 
{
    object objResult = null;

    using (var ms = new MemoryStream())
    {
        var bf = new BinaryFormatter();
        bf.Serialize(ms, obj);

        ms.Position = 0;
        objResult = bf.Deserialize(ms);
     }

     return objResult;
}

This is one way to do it with C# and .NET 2.0. Your object requires to be [Serializable()]. The goal is to lose all references and build new ones.