c# - 使用检查查找超集的 Linq 查询

Question

我想查询一个包含一些重复文本的列表，并且有/没有 id 值。

我有两个条件要匹配，

1. 选择任何 2 个具有 ID 的相同[text, type]组，否则选择唯一的一个。
2. 任意 2 个文本，其中一个包含另一个，则选择超集文本。[唱歌，跳舞唱歌=>跳舞唱歌]

Type      Text            Id

Name      John  
Name      John            22
Name      John Smith      2548
Hobby     Singing         
Hobby     Dancing Singing
School    XYZ             
School    XYZ             242

预期输出：

Type      Text            Id

Name      John Smith      2548
Hobby     Dancing Singing
School    XYZ             242

标签： c#linq

Answer 1

这很丑陋，但它有效：

class Program
{
    static void Main(string[] args)
    {
        
        List<Record> records = BuildTestData();
        List<Record> deduped = DeDupe(records);
        Console.Clear();

        foreach (Record r in deduped)
            Console.WriteLine($"Type:{r.Typ}, Text:{r.Txt}, ID:{r.ID} ");

        Console.ReadKey();
    }


    static List<Record> DeDupe(List<Record> dupes)
    {
        List<Record> excludes = new List<Record>();
        excludes.AddRange(dupes.GroupBy(x => new { x.Typ, x.Txt }).Where(y => y.Count() > 1).SelectMany(z => z.Where(a => string.IsNullOrEmpty(a.ID))));
        excludes.AddRange(dupes.Where(x => !excludes.Any(y => x == y) && dupes.Any(z => x != z && x.Txt != z.Txt && z.Txt.Contains(x.Txt))));
        return dupes.Where(x => !excludes.Any(y => x == y)).ToList();
    }


    static List<Record> BuildTestData()
    {
        return new List<Record>
        {
            new Record { Typ = "Name", Txt = "John", ID = null},
            new Record { Typ = "Name", Txt = "John", ID = "22"},
            new Record { Typ = "Name", Txt = "John Smith", ID = "2548"},
            new Record { Typ = "Hobby", Txt = "Singing", ID = null},
            new Record { Typ = "Hobby", Txt = "Dancing Singing", ID = null},
            new Record { Typ = "School", Txt = "XYZ", ID = null},
            new Record { Typ = "School", Txt = "XYZ", ID = "242"},
        };
    }
}

public class Record
{ 
    public string Typ { get; set; }
    public string Txt { get; set; }
    public string ID { get; set; }
}