使用 LINQ 的 GROUP、SUM 和 DATE DIFFERENCE 实体框架

Question

我想将每个客户的所有消费金额相加，然后将结果除以客户在数据库中首次注册的日期；划分的结果将决定最赚钱和最不赚钱的客户。

Customer 表如下所示：

 public class Customer
{
    public int Id { get; set; }
    public string Name{ get; set; }
    public string Surname{ get; set; }
    public int DayOB{ get; set; }
    public int MonthOB{ get; set; }
    public int YearOB{ get; set; }
    public string Genre{ get; set; }
    public string Email { get; set; }
    public string Cellular{ get; set; }
    public bool Privacy { get; set; }
    public bool Profilation{ get; set; }
    public bool Marketing { get; set; }
    public decimal Points{ get; set; }
    public decimal ConversionRate{ get; set; }
    public DateTime RegisterDate{ get; set; }
    public string QRCode { get; set; }
}

而事务表像：

 public class Transaction
{
    public int Id { get; set; }
    public decimal SpentAmount{ get; set; }
    public decimal AddedCoupon{ get; set; }
    public decimal DeductCoupon{ get; set; }
    public DateTime TransactionDate{ get; set; }

    //FOREIGN KEY
    public int CustomerId { get; set; }
    //NAVIGATION
    public Customer Customer { get; set; }
}

目前我已经尝试过这个 LINQ 查询

  public async Task<IHttpActionResult> GetTransactions()
    {
        /*DateTime regDate= from c in db.Customers
                           select c.RegisterDate.Date;*/

        DateTime currentDay= DateTime.UtcNow;

        var query = from c in db.Customers
                    join t in db.Transactions on c.Id equals t.CustomerId
                    where t.Id != null
                    group t by t.CustomerId into ris
                    select new
                    {
                        CustomerId = ris.Key,
                        RegisterDate= ris.Select(c => c.Customer.RegisterDate.Date.Subtract(currentDay)),
                        Sum = ris.Sum(t => t.SpentAmount)
                    };

        return Ok(query.OrderByDescending(ris => ris.Sum));
    }

结果如下：

[
    {
        "CustomerId": 6,
        "RegisterDate": [
            "2020-03-01T00:00:00",
            "2020-03-01T00:00:00"
        ],
        "Sum": 46.00
    },
    {
        "CustomerId": 7,
        "RegisterDate": [
            "2019-11-01T00:00:00",
            "2019-11-01T00:00:00"
        ],
        "Sum": 45.50
    },
    {
        "CustomerId": 5,
        "RegisterDate": [
            "2020-02-01T00:00:00",
            "2020-02-01T00:00:00"
        ],
        "Sum": 31.00
    }
]

CustomerId 的分组部分和 Sum 是正确的结果，但是当我尝试在当天和注册日之间进行差异的部分时，然后将其除以花费的金额，这给了我错误。

Answer 1

抱歉，您不能在 ef.所以你有 2 个变体：

1) 在应用程序的 Web 部分中的 query.ToList() 之后进行计算。我想它会起作用的：

var query = from c in db.Customers
                    join t in db.Transactions on c.Id equals t.CustomerId
                    where t.Id != null
                    group t by t.CustomerId into ris;
                    select new
                    {
                        CustomerId = ris.Key,
                        RegisterDate = ris.Select(c => c.Customer.RegisterDate)
                        Sum = ris.Sum(t => t.SpentAmount)
                    };
var result = query.ToList().Select(x => new {x.CustomerId, PassedDay = RegisterDate.Date.Subtract(currentDay), Sx.Sum });
return Ok(result.OrderByDescending(ris => ris.Sum));

2) Ef 可以使用日期时间比较操作（'>'、'='...）和特殊的 dbfunctions（url 中的文档较低）。你可以用它重写代码。

dbfunctions - https://docs.microsoft.com/en-us/dotnet/api/system.data.entity.dbfunctions?view=entity-framework-6.2.0