测试代码:
private void TestDistinct() { Task.Run(() => { //生成测试数据 DateTime dt = DateTime.Now; Random rnd = new Random(); List<MyData> list = new List<MyData>(); int total = 1000000; for (int i = 0; i < total; i++) { MyData info = new MyData(); info.id = rnd.Next(1, total * 10).ToString(); info.name = rnd.Next(1, total * 10).ToString(); list.Add(info); } double d = DateTime.Now.Subtract(dt).TotalMilliseconds; //方法一 DateTime dt1 = DateTime.Now; Dictionary<string, MyData> result1 = new Dictionary<string, MyData>(); foreach (MyData item in list) { MyData temp; if (!result1.TryGetValue(item.name, out temp)) { result1.Add(item.name, item); } } List<MyData> r1 = result1.Values.ToList(); double d1 = DateTime.Now.Subtract(dt1).TotalMilliseconds; //方法二 DateTime dt2 = DateTime.Now; List<MyData> result2 = list.ToLookup(item => item.name).ToDictionary(item => item.Key, item => item.First()).Values.ToList(); double d2 = DateTime.Now.Subtract(dt2).TotalMilliseconds; //方法三 DateTime dt3 = DateTime.Now; List<MyData> result3 = list.Distinct(new MyCompare()).ToList(); double d3 = DateTime.Now.Subtract(dt3).TotalMilliseconds; //方法四 DateTime dt4 = DateTime.Now; List<MyData> result4 = list.GroupBy(item => item.name).Select(item => item.First()).ToList(); double d4 = DateTime.Now.Subtract(dt4).TotalMilliseconds; this.BeginInvoke(new Action(() => { textBox1.Text = ""; textBox1.Text += "生成 " + list.Count.ToString("# ####") + " 条测试数据耗时:" + d + "毫秒\r\n\r\n"; textBox1.Text += "使用方法一去重耗时:" + d1 + "毫秒\r\n\r\n"; textBox1.Text += "使用ToLookup和ToDictionary去重耗时:" + d2 + "毫秒\r\n\r\n"; textBox1.Text += "使用Distinct去重耗时:" + d3 + "毫秒\r\n\r\n"; textBox1.Text += "使用GroupBy和Select去重耗时:" + d4 + "毫秒\r\n\r\n"; textBox1.Text += "去重后数量:" + r1.Count + "," + result2.Count + "," + result3.Count + "," + result4.Count + "" + "\r\n\r\n"; })); }); }
数据类:
public class MyData { public string id { get; set; } public string name { get; set; } } public class MyCompare : IEqualityComparer<MyData> { public bool Equals(MyData x, MyData y) { return x.name == y.name; } public int GetHashCode(MyData obj) { return obj.name.GetHashCode(); } }
测试结果:
结论:
方法一和方法三去重速度差不多,在一个数量级
方法二和方法四去重速度差不多,在一个数量级
方法二和方法四比方法一和方法三大约慢4、5倍左右
方法二和方法四比较方便,一行代码搞定,方法一和方法三代码行数相对较多,方法三要写个MyCompare
原文:https://www.cnblogs.com/zhang1f/p/11667029.html