C# 将两个列表与重叠数据合并

Question

如果我有 2 个字符串列表

List<string> history = new List<string>(){ "AA", "BB", "CC", "AA" };
List<string> potentialNew = new List<string>(){ "CC", "AA", "DD", "EE", "FF", "AA"};

我需要一种方法来组合列表，同时防止“重叠”并保持相同的顺序。因此，在上面的示例中，将有一个组合列表：

AA, BB, CC, AA, DD, EE, FF, AA

换句话说，只有 DD、EE、FF 和 AA 被添加到 history 列表中。

我这几天一直在尝试解决这个问题，无数次搜索都没有找到解决方案。任何帮助将不胜感激！

Answer 1

这将为您在问题中提到的给定输入集提供预期的输出：

 List<string> history = new List<string>() { "AA", "BB", "CC", "AA" };
 List<string> potentialNew = new List<string>() { "CC", "AA", "DD", "EE", "FF" };
 var result = history.Concat(potentialNew.Where(x => !history.Contains(x)).ToList());

.Concat() 方法允许您连接两个列表。我们正在从potentialNew 中提取第一个列表中不存在的特定项目，并将它们与第一个列表连接。

更新：根据我们的讨论，我得出的结论是您正在寻找类似以下内容：

string lastItem = history.Last();
   int lastIndexToCheck=history.Count-2,i=0;
   for (; i < potentialNew.Count - 1; i++)
       {
          if (potentialNew[i] == lastItem && potentialNew[i - 1] == history[lastIndexToCheck])
              {
                 break;
              }
       }
       history.AddRange(potentialNew.Skip(i+1).ToList());  

现在历史将包含所需的元素集。

Answer 2

using System;
using System.Collections.Generic;
using System.Linq;

public class Program
{
    public static void Main()
    {
        List<string> history = new List<string>(){ "AA", "BB", "CC", "AA" };
        List<string> potentialNew = new List<string>(){ "CC", "AA", "DD", "EE", "FF" };
        // make lists equal length

        foreach(var x in history.ConcatOverlap(potentialNew)){
            Console.WriteLine(x);
        }
    }


}

public static class Ext{
    public static IEnumerable<string> ConcatOverlap(this List<string> history, List<string> potentialNew){
        var hleng = history.Count();
        var pleng = potentialNew.Count();
        if(pleng > hleng) history = history.Concat(Enumerable.Range(1, pleng - hleng).Select(x => string.Empty)).ToList();
        if(hleng > pleng) potentialNew = Enumerable.Range(1, hleng - pleng).Select(x => string.Empty).Concat(potentialNew).ToList();


        var zipped = history.Zip(potentialNew, (a,b)=> new {First=a,Next=b, Equal = (a.Equals(b) || string.IsNullOrEmpty(a) || string.IsNullOrEmpty(b))});
        var count = 0;
        var max = pleng > hleng ? pleng : hleng;
        Console.WriteLine("Max " + max);
        while(zipped.Any(x => !x.Equal) && count < max - 1){
            count++;
            potentialNew.Insert(0,string.Empty);
            history.Add(string.Empty);
            zipped = history.Zip(potentialNew, (a,b)=> new {First=a,Next=b, Equal = (a.Equals(b) || string.IsNullOrEmpty(a) || string.IsNullOrEmpty(b))});
        }
        return zipped.Select(x => string.IsNullOrEmpty(x.First) ? x.Next : x.First);
    }
}

再考虑一下：

public static IEnumerable<T> ConcatOverlap<T>(this IEnumerable<T> head, IEnumerable<T> tail){

    var skip = 0;
    var hLen = head.Count();
    while(head.Skip(skip).Zip(tail, (a,b) => a.Equals(b)).Any(x => !x) && skip < hLen){
        skip++;
    }

    return head.Take(skip).Concat(tail);
}

Answer 3

var history = new List<string>() { "AA", "BB", "CC", "AA" };
var potentialNew = new List<string>() { "CC", "AA", "DD", "EE", "FF" };

// Get the min. number of items to compare that 2 lists
for (int count = Math.Min(history.Count(), potentialNew.Count()); count >= 0; count--)
{
    // Get the items from the back of history list, and get the items from potential list
    // Compare them by SequenceEqual()
    if (history.Skip(history.Count() - count).Take(count).SequenceEqual(potentialNew.Take(count)))
    {
        // Add the items to the result if found. It must be the greatest common part
        return history.Concat(potentialNew.Skip(count));
    }
}

.Net Fiddle

Answer 4

这看起来很简单：

List<string> history = new List<string>(){ "AA", "BB", "CC", "AA" };
List<string> potentialNew = new List<string>(){ "CC", "AA", "DD", "EE", "FF", "AA"};

potentialNew.Aggregate(history, (h, p) =>
{
    if (!h.Skip(h.Count - 2).Contains(p))
    {
        h.Add(p);
    }
    return h;
});

结果是history 包含：

AA BB 抄送 AA DD EE 法郎 AA

Answer 5

不确定这在性能方面有多好，但我构建了一个逻辑来实现您想要的。任何人都可以进行调整以使其更清洁。

List<string> history = new List<string>() { "AA", "BB", "CC", "AA" };
List<string> potentialNew = new List<string>() { "CC", "AA", "DD", "EE", "FF", "AA" };

var result = ProcessChatLog(history,potentialNew);
//pass these two list to a function to process the chat log

核心逻辑在这里。

 public List<string> ProcessChatLog(List<string> history, List<string> potentialNew)
    {
        var lastChat = history.Last();  
        var lastChatIndex = history.Count - 1;
        var allIndexWithLastChat = potentialNew.Select((c, i) => new { chat = c, Index = i })
                                   .Where(x => x.chat == lastChat)
                                   .Select(x => x.Index).Reverse().ToList();       

        List<int> IndexToClear = new List<int>();
        bool overlapFound = false;

        foreach (var index in allIndexWithLastChat)
        {
            if (!overlapFound)
            {
                int hitoryChatIndex = lastChatIndex;
                IndexToClear.Clear();
                for (int i = index; i > -1; i--)
                {
                    if (potentialNew[i] == history[hitoryChatIndex])
                    {
                        IndexToClear.Add(i);
                        if (i == 0)
                        {
                            overlapFound = true;
                            break;
                        }
                        hitoryChatIndex--;
                    }
                    else
                    {
                        break;
                    }
                }
            }
            else
            {
                IndexToClear.Clear();
                break;
            }                             
        }

        if(IndexToClear.Count >0)
        {
            potentialNew.RemoveRange(IndexToClear.Min(), IndexToClear.Count);   
        }

        return history.Concat(potentialNew).ToList();
    }

这是一些结果

  history = { "AA", "BB", "CC", "AA" }
  potentialNew = { "CC", "AA", "DD", "EE", "FF", "AA"}

  Result = { "AA", "BB","CC", "AA", "DD", "EE", "FF", "AA"}

---

  history = { "AA", "BB","AA", "CC", "AA" }
  potentialNew = { "AA","CC", "AA", "DD", "EE", "FF", "AA"}

  Result = { "AA", "BB","AA","CC", "AA", "DD", "EE", "FF", "AA"}

---

  history = { "AA", "BB", "CC", "AA" }
  potentialNew = { "CC", "AA", "CC", "AA", "FF", "AA"}

  Result = { "AA", "BB","CC", "AA", "CC", "AA", "FF", "AA"}

---

  history = { "AA", "BB", "CC", "AA" }
  potentialNew = {"AA", "CC", "AA", "DD", "EE", "FF", "AA" }

  Result = { "AA", "BB","CC", "AA", "CC", "AA", "DD", "EE", "FF", "AA" }

如果这有帮助，请告诉我。

但我仍然说这不是您想要的理想输出。因为如果聊天包含相同的消息 20 次，并假设您在 2 个列表中分别包含 11 和 9 个项目，该怎么办。现在根据您想要的输出，您将省略所有 9 条消息新列表作为可能的重复和问题。所以我说不是这个修复，解决方案是跟踪聊天日志中传递的消息，并采取措施不在下一个日志中传递这些消息。这样可以保持逻辑和准确性