LINQ 常规实践

Posted 2021-02-06 mi12205599

1.Linq 执行多列排序

OrderBy的意义是按照指定顺序排序，连续两次OrderBy，后面一个有可能会打乱前面一个的排序顺序，可能与预期不符。

要实现sql中的order by word,name类似效果; LINQ 有ThenBy可以紧接使用， ThenBy记住原本排序的值，然后再排其他值， 正因如此，ThenBy是针对IOrderEnumerable 进行调用的。

2. Linq主外键连接查询

group join操作符常用于返回‘主键对象-外键对象集合’的查询，例如‘产品类别-此类别下所有的产品’的模式。

  // 查询语法
 var query =  
        from c in db.Categories  
        join p in db.Products on c.CategoryID equals p.CategoryID into r  
        select  new  
        {  
            c.CategoryName,  
            Products = r  
        };  
 // 方法语法   
 var q =  
        db.Categories  
        .GroupJoin  
        (  
           db.Products,  
           c => c.CategoryID,  
           p => p.CategoryID,  
           (c, p) => new  
           {  
               c.CategoryName,  
               Products = p  
           }  
        );  
  这样就可以结合 DefaultIfEmpty 理解 left outer join的linq写法。

 group join生成的sql接近于：

select Categories.*, Products.* from Categories 
left join Products on  Categories.Id = Products.CateId 
order by Categories.Id 

 // group join 有点类似于 left join 数据膨胀的效果

https://docs.microsoft.com/en-us/dotnet/csharp/linq/perform-grouped-joins

3.linq2sql 指定形成sql左右连接

  linq2sql join语法默认得到的是inner join

 Model1Container model = new Model1Container();
            //内连接
            var query = from s in model.Student
                        join c in model.Course on s.CourseCno equals c.Cno
                        where c.Cno == 1
                        select new
                        {
                            ClassID = s.CourseCno,
                            ClassName = c.Cname,
                            Student = new
                            {
                                Name = s.Sname,
                                ID = s.Sno
                            }
                        };
    foreach (var item in query)
    {
        Response.Write("ClassID:" + item.ClassID + "ClassName:" + item.ClassName + "Name:" + item.Student.Name);
    }

 

 在sql profile里面监控到与上面Linq2sql 对应的sql是

SELECT [t0].[CourseCno] AS [ClassID], [t1].[Cname] AS [ClassName], [t0].[Sname] AS [Name], [t0].[Sno] AS [ID]
FROM [Student] AS [t0]
INNER JOIN [Course] AS [t1] ON [t0].[CourseCno] = [t1].[Cno]
WHERE [t1].[Cno] = @p0

  linq2sql 左连接

Model1Container model = new Model1Container();
            var query = from s in model.Student
                        join c in model.Course on s.CourseCno equals c.Cno into gc
                        from gci in gc.DefaultIfEmpty()
                        select new
                        {
                            ClassID = s.CourseCno,
                            ClassName = gci ==null ?String.Empty: gci.Cname,
                            Student = new
                            {
                                Name = s.Sname,
                                ID = s.Sno
                            }
                        };
    //Outer join时必须将join后的表into到一个新的变量gc中，然后要用gc.DefaultIfEmpty()表示外连接(没有匹配的记录字段设为NULL)
    
    foreach (var item in query)
    {
        Response.Write("ClassID:" + item.ClassID + "ClassName:" + item.ClassName + "Name:" + item.Student.Name);
    }
    
   // 上例中使用了DefaultIfEmpty操作符，它能够为Empty序列(注意是Empty序列而不是Null序列)返回一个默认元素序列，DefaultIfEmpty使用了泛型中的default关键字。
    
    gc.DefaultIfEmpty(new Course { Cname = "",Cperiod="" } )    //设置为空时的默认值
    

 

 与以上Linq2sql对应的sql 是

SELECT [t0].[CourseCno] AS [ClassID], [t1].[Cname] AS [ClassName], [t0].[Sname] AS [Name], [t0].[Sno] AS [ID]
FROM [Student] AS [t0]
LEFT OUTER JOIN [Course] AS [t1] ON [t0].[CourseCno] = [t1].[Cno]

 https://docs.microsoft.com/en-us/dotnet/csharp/linq/perform-left-outer-joins

4. 忽略Linq 检索某元素异常

在日常Linq实践中，按照某种预期写定的Linq， 在执行检索时某些元素会爆出异常，导致整个检索失败；如果我们能容忍某些元素的异常，继续完成整个Linq检索，可以写一个扩展方法，忽略报错元素。

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
 
namespace Enumerable {
    public static class EnumerableExtension {
        /// <summary>
        /// 在IEnumerable<T> 循环或者变为内存序列之前 忽略掉序列中存在的元素异常
        /// </summary>
        /// <typeparam name="T"></typeparam>
        /// <param name="values"></param>
        /// <returns></returns>
        public static IEnumerable<T> SkipException<T> (this IEnumerable<T> values) {
            using(var enumerator = values.GetEnumerator()) {
                var next = true;
                while(next) {
                    try{
                        // 如果枚举器成功推进到下一个元素，MoveNext为true；枚举数越过集合结尾，则为false，也就是说返回值只确定后续是否有值
                        next = enumerator.MoveNext();
                    } catch{
                        // catch到异常，忽略当前元素，继续循环
                        LogHelper.Write("cause exception", LogHelper.LogMessageType.Error);
                        continue;
                    }
                    // yield 返回枚举器指向的当前元素
                    if(next)
                        yield  return enumerator.Current;
                }
            }
        }
    }
}

 

这个扩展方法的思路是： 重写foreach检索语法的默认逻辑。

When you use the yield keyword in a statement, you indicate that the method, operator, or get accessor in which it appears is an iterator.
You use a yield return statement to return each element one at a time.
You consume an iterator method by using a foreach statement or LINQ query