如何对多列进行分组并计数？

Question

首先，我无法在标题中正确解释我的问题，但我认为它不是重复的。 我有一个 model 列表，它 id 喜欢按多个字段分组，然后将它们计入新字段，并将所有其他字段保留在结果中。 对于错误的解释，我很抱歉，我已经研究了几个小时如何做到这一点，但我无法弄清楚..我想从中得到：

销售编号	名称	产品	地址	电话	打印类型
112	乔	苹果	新街12	11223344	打印类型 1
112	乔	苹果	新街12	11223344	打印类型2
112	乔	苹果	新街12	11223344	打印类型2
112	乔	苹果	新街12	11223344	打印类型2
112	乔	苹果	新街12	11223344	打印类型3
113	乔	猕猴桃	新街12	11223344	打印类型3
114	简	橙子	新街19号	72754722	打印类型 1
115	约翰	橙子	新街11号	99236527	打印类型2
115	约翰	橙子	新街11号	99236527	打印类型2

按 SaleID 和 PrintType 分组，如下所示：

销售编号	名称	产品	地址	电话	NoOfPrintType1	NoOfPrintType2	NoOfPrintType3
112	乔	苹果	新街12	11223344	1个	3个	1个
113	乔	猕猴桃	新街12	11223344	0	0	1个
114	简	橙子	新街19号	72754722	1个	0	0
115	约翰	橙子	新街11号	99236527	0	2个	0

最好 id 喜欢使用 LINQ，但 id 也可以使用 SQL，id 只是想尽可能避免使用 for 循环。 编辑：有一定数量的打印类型，因此它不需要是动态的。

Answer 1

为了聚合一组 PrintTypes 的数据，您可以按 SaleId 分组，因为地址、电话和产品等附加数据对于 SaleId 是相同的（基于您的示例数据）。

var aggregated = from x in GenerateSales() 
    group x by x.SaleId into g // Assume that SaleId is the key that you want to group by
    select new Aggregate() 
    { 
        SaleId = g.Key, 
        // Additional data that are the same for all rows with the same SaleId
        Name = g.First().Name, 
        Product = g.First().Product, 
        Address = g.First().Address, 
        Phone = g.First().Phone, 
        // Calculate counts of Print Types
        NoOfPrintType1 = g.Where(x => x.PrintType == "PrintType1").Count(),
        NoOfPrintType2 = g.Where(x => x.PrintType == "PrintType2").Count(),
        NoOfPrintType3 = g.Where(x => x.PrintType == "PrintType3").Count(),
    };

Linq 语句首先按 SaleId 分组，然后为每个 SaleId 创建一个 object，它包括

• 销售编号
• 地址，电话等附加数据...
• 每个已知 PrintType 的计数（通过过滤组的项目然后计算行数来计算）

您可以在下面找到生成测试数据并输出结果的示例。

结果

112 | Joe | Apple | New street 12 | 11223344 | 1 | 3 | 1
113 | Joe | Kiwi | New street 12 | 11223344 | 0 | 0 | 1
114 | Jane | Orange | New street 19 | 72754722 | 1 | 0 | 0
115 | John | Orange | New street 11 | 99236527 | 0 | 2 | 0

示例代码

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

namespace LinqTest
{
    class Sale 
    {
        public string SaleId { get; set; }
        public string Name { get; set; }
        public string Product { get; set; }
        public string Address { get; set; }
        public string Phone { get; set; }
        public string PrintType { get; set; }
    }

    class Aggregate 
    {
        public string SaleId { get; set; }
        public string Name { get; set; }
        public string Product { get; set; }
        public string Address { get; set; }
        public string Phone { get; set; }
        public int NoOfPrintType1 { get; set; }
        public int NoOfPrintType2 { get; set; }
        public int NoOfPrintType3 { get; set; }

        public override string ToString() 
        {
            return $"{SaleId} | {Name} | {Product} | {Address} | {Phone} | {NoOfPrintType1} | {NoOfPrintType2} | {NoOfPrintType3}";
        }
    }

    class Program
    {
        static void Main(string[] args)
        {
            var aggregated = from x in GenerateSales() 
                group x by x.SaleId into g // Assume that SaleId is the key that you want to group by
                select new Aggregate() 
                { 
                    SaleId = g.Key, 
                    // Additional data that are the same for all rows with the same SaleId
                    Name = g.First().Name, 
                    Product = g.First().Product, 
                    Address = g.First().Address, 
                    Phone = g.First().Phone, 
                    // Calculate counts of Print Types
                    NoOfPrintType1 = g.Where(x => x.PrintType == "PrintType1").Count(),
                    NoOfPrintType2 = g.Where(x => x.PrintType == "PrintType2").Count(),
                    NoOfPrintType3 = g.Where(x => x.PrintType == "PrintType3").Count(),
                };
            foreach(var a in aggregated)
            {
                Console.WriteLine(a.ToString());
            }
        }

        static IEnumerable<Sale> GenerateSales()
        {
            yield return new Sale() { SaleId = "112", Name = "Joe", Product = "Apple", Address = "New street 12", Phone = "11223344", PrintType = "PrintType1" };
            yield return new Sale() { SaleId = "112", Name = "Joe", Product = "Apple", Address = "New street 12", Phone = "11223344", PrintType = "PrintType2" };
            yield return new Sale() { SaleId = "112", Name = "Joe", Product = "Apple", Address = "New street 12", Phone = "11223344", PrintType = "PrintType2" };
            yield return new Sale() { SaleId = "112", Name = "Joe", Product = "Apple", Address = "New street 12", Phone = "11223344", PrintType = "PrintType2" };
            yield return new Sale() { SaleId = "112", Name = "Joe", Product = "Apple", Address = "New street 12", Phone = "11223344", PrintType = "PrintType3" };
            yield return new Sale() { SaleId = "113", Name = "Joe", Product = "Kiwi", Address = "New street 12", Phone = "11223344", PrintType = "PrintType3" };
            yield return new Sale() { SaleId = "114", Name = "Jane", Product = "Orange", Address = "New street 19", Phone = "72754722", PrintType = "PrintType1" };
            yield return new Sale() { SaleId = "115", Name = "John", Product = "Orange", Address = "New street 11", Phone = "99236527", PrintType = "PrintType2" };
            yield return new Sale() { SaleId = "115", Name = "John", Product = "Orange", Address = "New street 11", Phone = "99236527", PrintType = "PrintType2" };
        }
    }
}

Answer 2

你可以使用这个 SQL 查询

with x as (
   SELECT  [SaleID],[Name],[Product],[Address],[Phone],[PrintType]
          ,case when [PrintType]='PrintType1' then count(*) else 0 end NoOfPrintType1
          ,case when [PrintType]='PrintType2' then count(*) else 0 end NoOfPrintType2
          ,case when [PrintType]='PrintType3' then count(*) else 0 end NoOfPrintType3
      FROM [Table_1]
      group by [SaleID] ,[Name],[Product],[Address],[Phone] ,[PrintType])
    
    select [SaleID],[Name],[Product],[Address],[Phone]
          ,sum(NoOfPrintType1)NoOfPrintType1
          ,sum(NoOfPrintType2)NoOfPrintType2
          ,sum(NoOfPrintType3)NoOfPrintType3 from x
   group by [SaleID]
          ,[Name],[Product],[Address],[Phone]