一种反序列化字符串内容的方法

我已经解析这样的文本文件40年了。他是最好的方法

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.IO;
namespace ConsoleApplication1
{
    class Program
    {
        enum State
        {
            FIND_ID,
            FIND_VALUE
        }
        const string FILENAME = @"c:'temp'test.txt";
        static void Main(string[] args)
        {
            List<A> a_s = new List<A>();
            string inputLine = "";
            StreamReader reader = new StreamReader(FILENAME);
            State state = State.FIND_ID;
            A a = null;
            while ((inputLine = reader.ReadLine()) != null)
            {
                inputLine = inputLine.Trim();
                if (!inputLine.StartsWith("-") && inputLine.Length > 0)
                {
                    switch (state)
                    {
                        case State.FIND_ID :
                            if (inputLine.StartsWith("Id"))
                            {
                                string[] inputArray = inputLine.Split(new char[] { ':' });
                                a = new A();
                                a_s.Add(a);
                                a.Id = inputArray[1].Trim();
                                state = State.FIND_VALUE;
                            }
                            break;
                        case State.FIND_VALUE:
                            if (inputLine.StartsWith("Value"))
                            {
                                string[] inputArray = inputLine.Split(new char[] { ':' });
                                a.Value = inputArray[1].Trim();
                                state = State.FIND_ID;
                            }
                            break;
                    }
                }
            }
        }
    }
    class A
    {
        public string Id { get; set; }
        public string Value { get; set; }
    }
}
​

如果您可以修改您的A类，使其具有如下构造函数：

class A
{
    public string Id { get; set; }
    public string Value { get; set; }
    
    public A() { }
    public A(string s)
    {
        string[] vals = s.Split((new string[] { "'r'n" }), StringSplitOptions.RemoveEmptyEntries);
        this.Id = vals[0].Replace("Id : ", string.Empty).Trim();
        this.Value = vals[1].Replace("Value : ", string.Empty).Trim();
    }
    // only overridden here for printing
    public override string ToString()
    {
        return string.Format("Id : {0}'r'nValue : {1}'r'n", this.Id, this.Value);
    }
}

您可以实现以下内容：

public static List<A> GetValues(string file)
{
    List<string> vals = new List<string>(Regex.Split(System.IO.File.ReadAllText(file), "--------------------"));
    vals.RemoveAll(delegate(string s) { return string.IsNullOrEmpty(s.Trim()); });
    List<A> ret = new List<A>();
    vals.ForEach(delegate(string s) { ret.Add(new A(s)); });
    return ret;
}
public static void Main()
{
    foreach (A a in GetValues(@"C:'somefile.txt")) {
        Console.WriteLine(a);
    }
}

你最初提出的问题是为了避免循环；它没有显式的循环构造（for、foreach、do/while），但底层代码确实循环（例如，Regex.Split、vals.RemoveAll和vals.ForEach都是循环），因此正如注释所指出的，在这种情况下，您无法真正避免循环。

需要注意的是，在一些基准测试之后，如果要读取的文件的格式与您指定的格式完全相同，那么这种方法的速度会惊人地快。作为比较，我创建了一个文件，并复制/粘贴了你的示例模板（你发布的4个结果），总共有1032个结果，文件大小约为75k，XML文件的结果约为65k（由于---的文本较少），我编写了以下基准测试来运行：

public class A
{
    public string Id { get; set; }
    public string Value { get; set; }
    public A() { }
    public A(string s)
    {
        string[] vals = s.Split((new string[] { "'r'n" }), StringSplitOptions.RemoveEmptyEntries);
        this.Id = vals[0].Replace("Id : ", string.Empty).Trim();
        this.Value = vals[1].Replace("Value : ", string.Empty).Trim();
    }
    public A(string id, string val)
    {
        this.Id = id;
        this.Value = val;
    }
    // only overridden here for printing
    public override string ToString()
    {
        return string.Format("Id : {0}'r'nValue : {1}'r'n", this.Id, this.Value);
    }
}
public static List<A> GetValuesRegEx(string file)
{
    List<string> vals = new List<string>(Regex.Split(System.IO.File.ReadAllText(file), "--------------------"));
    vals.RemoveAll(delegate(string s) { return string.IsNullOrEmpty(s.Trim()); });
    List<A> ret = new List<A>();
    vals.ForEach(delegate(string s) { ret.Add(new A(s)); });
    return ret;
}
public static List<A> GetValuesXml(string file)
{
    List<A> ret = new List<A>();
    System.Xml.Serialization.XmlSerializer srl = new System.Xml.Serialization.XmlSerializer(ret.GetType());
    System.IO.FileStream f = new System.IO.FileStream(file,
                                                      System.IO.FileMode.OpenOrCreate,
                                                      System.IO.FileAccess.ReadWrite,
                                                      System.IO.FileShare.ReadWrite);
    ret = ((List<A>)srl.Deserialize(f));
    f.Close();
    return ret;
}
public static List<A> GetValues(string file)
{
    List<A> ret = new List<A>();
    List<string> vals = new List<string>(System.IO.File.ReadAllLines(file));
    for (int i = 0; i < vals.Count; ++i) {
        if (vals[i].StartsWith("---") && ((i + 3) < vals.Count) && (vals[i + 3].StartsWith("---"))) {
            ret.Add(new A(vals[i + 1].Replace("Id : ", string.Empty), vals[i + 2].Replace("Value : ", string.Empty)));
            i += 3;
        }
    }
    return ret;
}
public static List<A> GetValuesStream(string file)
{
    List<A> ret = new List<A>();
    string line = "";
    System.IO.StreamReader reader = new System.IO.StreamReader(file);
    int state = 0;
    A a = null;
    while ((line = reader.ReadLine()) != null) {
        line = line.Trim();
        if (!line.StartsWith("-") || line.Length > 0) {
            switch (state) {
                case 0:
                    if (line.StartsWith("Id")) {
                        string[] inputArray = line.Split(new char[] { ':' });
                        a = new A();
                        ret.Add(a);
                        a.Id = inputArray[1].Trim();
                        state = 1;
                    }
                    break;
                case 1:
                    if (line.StartsWith("Value")) {
                        string[] inputArray = line.Split(new char[] { ':' });
                        a.Value = inputArray[1].Trim();
                        state = 0;
                    }
                    break;
            }
        }
    }
    return ret;
}
public static void Main()
{
            System.Diagnostics.Stopwatch sw = new System.Diagnostics.Stopwatch();
    for (int x = 0; x < 5; ++x) {
        double avg = 0d;
        for (int i = 0; i < 100; ++i) {
            sw.Restart();
            List<A> txt = GetValuesRegEx(@"C:'somefile.txt");
            sw.Stop();
            avg += sw.Elapsed.TotalSeconds;
        }
        Console.WriteLine(string.Format("avg: {0} s", (avg / 100)));
        // best out of 5: 0.002380452 s
        avg = 0d;
        sw.Stop();
        for (int i = 0; i < 100; ++i) {
            sw.Restart();
            List<A> txt = GetValuesXml(@"C:'somefile.xml");
            sw.Stop();
            avg += sw.Elapsed.TotalSeconds;
        }
        Console.WriteLine(string.Format("avg: {0} s", (avg / 100)));
        // best out of 5: 0.002042312 s
        avg = 0d;
        sw.Stop();
        for (int i = 0; i < 100; ++i) {
            sw.Restart();
            List<A> xml = GetValues(@"C:'somefile.xml");
            sw.Stop();
            avg += sw.Elapsed.TotalSeconds;
        }
        Console.WriteLine(string.Format("avg: {0} s", (avg / 100)));
        // best out of 5: 0.001148025 s
        avg = 0d;
        sw.Stop();
        for (int i = 0; i < 100; ++i) {
            sw.Restart();
            List<A> txt = GetValuesStream(@"C:'somefile.txt");
            sw.Stop();
            avg += sw.Elapsed.TotalSeconds;
        }
        Console.WriteLine(string.Format("avg: {0} s", (avg / 100)));
        // best out of 5: 0.002459861 s
        avg = 0d;
        sw.Stop();
    }
    sw.Stop();
}

为了清晰起见，以下是在英特尔i7@2.2 GHz和5400 RPM HDD（碎片约0.1%）上运行时的结果：

GetValuesRegEx运行时间5次运行中的最佳平均值：0.002380452 s

GetValuesXml运行时间5次运行中的最佳平均值：0.002042312 s

GetValues（ReadAllLines/loop）运行时间5次运行中的最佳平均值：0.001148025s

GetValuesStream（StreamReader/loop）运行时间5次运行中的最佳平均值：0.002459861 s

您的结果可能会有所不同，并且这没有考虑任何错误处理，因此在使用代码时需要将其考虑在内。

希望这能有所帮助。