使用Distinct和stringcompararer删除重复和忽略区分大小写。InvariantCultureIgnoreCase

string[] array = new string[] { "A", "a", "b", "B", "a", "C", "c", "C", "A", "1" };
var r = array.Distinct(StringComparer.InvariantCultureIgnoreCase).ToList();
Console.WriteLine(r.Count); // return 4 items

使用Distinct和stringcompararer删除重复和忽略区分大小写。InvariantCultureIgnoreCase

string[] array = new string[] { "A", "a", "b", "B", "a", "C", "c", "C", "A", "1" };
var r = array.Distinct(StringComparer.InvariantCultureIgnoreCase).ToList();
Console.WriteLine(r.Count); // return 4 items

可能是不存储重复元素的hashset，它会默默地忽略添加请求 重复。

static void Main()
{
    string textWithDuplicates = "aaabbcccggg";     

    Console.WriteLine(textWithDuplicates.Count());  
    var letters = new HashSet<char>(textWithDuplicates);
    Console.WriteLine(letters.Count());

    foreach (char c in letters) Console.Write(c);
    Console.WriteLine("");

    int[] array = new int[] { 12, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2 };

    Console.WriteLine(array.Count());
    var distinctArray = new HashSet<int>(array);
    Console.WriteLine(distinctArray.Count());

    foreach (int i in distinctArray) Console.Write(i + ",");
}

你可以使用一个LINQ查询来做到这一点:

int[] s = { 1, 2, 3, 3, 4};
int[] q = s.Distinct().ToArray();

int size = a.Length;
        for (int i = 0; i < size; i++)
        {
            for (int j = i + 1; j < size; j++)
            {
                if (a[i] == a[j])
                {
                    for (int k = j; k < size; k++)
                    {
                        if (k != size - 1)
                        {
                            int temp = a[k];
                            a[k] = a[k + 1];
                            a[k + 1] = temp;

                        }
                    }
                    j--;
                    size--;
                }
            }
        }

最好的方法是什么?很难说，HashSet方法看起来很快， 但是(取决于数据)使用排序算法(CountSort ?) 可以快得多。

using System;
using System.Collections.Generic;
using System.Linq;
class Program
{
    static void Main()
    {
        Random r = new Random(0); int[] a, b = new int[1000000];
        for (int i = b.Length - 1; i >= 0; i--) b[i] = r.Next(b.Length);
        a = new int[b.Length]; Array.Copy(b, a, b.Length);
        a = dedup0(a); Console.WriteLine(a.Length);
        a = new int[b.Length]; Array.Copy(b, a, b.Length);
        var w = System.Diagnostics.Stopwatch.StartNew();
        a = dedup0(a); Console.WriteLine(w.Elapsed); Console.Read();
    }

    static int[] dedup0(int[] a)  // 48 ms  
    {
        return new HashSet<int>(a).ToArray();
    }

    static int[] dedup1(int[] a)  // 68 ms
    {
        Array.Sort(a); int i = 0, j = 1, k = a.Length; if (k < 2) return a;
        while (j < k) if (a[i] == a[j]) j++; else a[++i] = a[j++];
        Array.Resize(ref a, i + 1); return a;
    }

    static int[] dedup2(int[] a)  //  8 ms
    {
        var b = new byte[a.Length]; int c = 0;
        for (int i = 0; i < a.Length; i++) 
            if (b[a[i]] == 0) { b[a[i]] = 1; c++; }
        a = new int[c];
        for (int j = 0, i = 0; i < b.Length; i++) if (b[i] > 0) a[j++] = i;
        return a;
    }
}

几乎没有分支。怎么做?调试模式，步进(F11)与一个小数组:{1,3,1,1,0}

    static int[] dedupf(int[] a)  //  4 ms
    {
        if (a.Length < 2) return a;
        var b = new byte[a.Length]; int c = 0, bi, ai, i, j;
        for (i = 0; i < a.Length; i++)
        { ai = a[i]; bi = 1 ^ b[ai]; b[ai] |= (byte)bi; c += bi; }
        a = new int[c]; i = 0; while (b[i] == 0) i++; a[0] = i++;
        for (j = 0; i < b.Length; i++) a[j += bi = b[i]] += bi * i; return a;
    }

有两个嵌套循环的解决方案可能需要一些时间， 特别是对于较大的数组。

    static int[] dedup(int[] a)
    {
        int i, j, k = a.Length - 1;
        for (i = 0; i < k; i++)
            for (j = i + 1; j <= k; j++) if (a[i] == a[j]) a[j--] = a[k--];
        Array.Resize(ref a, k + 1); return a;
    }