SqlSugar ORM 入门篇2 【查询】 让我们实现零SQL

sqlsugar在查询的功能是非常强大的，多表查询、分页查询 、 一对一查询、二级缓存、一对多查、whencase等复杂函数、mapper功能、和拉姆达自定义扩展等，用好了是可以做到真正零sql的一款orm。

 

首先将sqlsugar更新到4.8版本，下面我就来一一讲解每种查询的写法

 

创建dbcontext

public class dbcontext
{
        public dbcontext()
        {
            db = new sqlsugarclient(new connectionconfig()
            {
                connectionstring = "server=.;uid=sa;pwd=sasa;database=sqlsugar4xtest",
                dbtype = dbtype.sqlserver,
                isautocloseconnection = true,//开启自动释放模式和ef原理一样我就不多解释了
                //initkey默认systemtable
            });
        }
        public sqlsugarclient db;//用来处理事务多表查询和复杂的操作
        public simpleclient<student> studentdb { get { return new simpleclient<student>(db); } }//用来处理student表的常用操作
        public simpleclient<school> schooldb { get { return new simpleclient<school>(db); } }//用来处理school表的常用操作
}

 

单表的简单查询

我们使用的simpleclient实现了简单的单表查询，如何扩展simpleclient可以看我的上一篇文章

            //调式代码 用来打印sql 
            db.aop.onlogexecuting = (sql, pars) =>
            {
                console.writeline(sql + "\r\n" + db.utilities.serializeobject(pars.todictionary(it => it.parametername, it => it.value)));
                console.writeline();
            };


            var data1 = studentdb.getbyid(1);//根据id查询
            var data2 = studentdb.getlist();//查询所有
            var data3 = studentdb.getlist(it => it.id == 1);  //根据条件查询      

            var p = new pagemodel() { pageindex=1,pagesize=2};// 分页查询
            var data4 = studentdb.getpagelist(it => it.name == "xx", p);
            console.write(p.pagecount);//返回总数


            // 分页查询加排序
            var data5 = studentdb.getpagelist(it => it.name == "xx", p,it=>it.name,orderbytype.asc);
            console.write(p.pagecount);//返回总数

 对于grid控件来说我一般用这个表单封装好了全部通用

  list<iconditionalmodel> conmodels = new list<iconditionalmodel>();
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.equal, fieldvalue = "1" });//id=1
            conmodels.add(new conditionalmodel() { fieldname = "student.id", conditionaltype = conditionaltype.equal, fieldvalue = "1" });//id=1
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.like, fieldvalue = "1" });// id like '%1%'
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.isnullorempty });
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.in, fieldvalue = "1,2,3" });
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.notin, fieldvalue = "1,2,3" });
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.noequal, fieldvalue = "1,2,3" });
            conmodels.add(new conditionalmodel() { fieldname = "id", conditionaltype = conditionaltype.isnot, fieldvalue = null });// id is not null
            var data6 = studentdb.getpagelist(conmodels,p,it=>it.name,orderbytype.asc); //组装条件当查询条件的 分页查询加排序

 

简单查询中拉姆达的使用技巧

基本上和ef差不太多

  var data3 = studentdb.getlist(it => it.name.contains("a"));  // like %a%  模糊查询

var p2 = new int[] { 1, 2, 3 };
var data31 = studentdb.getlist(it => p2.contains(it.id));  // id in (1,2,3)

我们还支持了sqlfunc.xxx一串方法来给我们使用，如下用法

var data311 = studentdb.getlist(it => sqlfunc.between(it.id,1,2));  // id between 1 and 2

 

动态拼表达式查询

            var exp = expressionable.create<student>()
                          .orif(1 == 1, it => it.id == 11)
                          .and(it => it.id == 1)
                          .andif(2 == 2, it => it.id == 1)
                          .or(it => it.name == "a1").toexpression();//拼接表达式

            var data311 = studentdb.getlist(exp);  // 动态表达式查询

 

 

扩展拉姆达方法

例如我们有自定义的sql函数或者sqlsugar不支持的我们可以自定扩展

具体看这个连接 http://www.codeisbug.com/doc/8/1162

 

复杂查询

我们上面看到的简单查询底层都是用复杂查询实现的

var data1 = studentdb.getbyid(1);
//等同于
var data2 = db.queryable<student>().single(it => it.id == 1);

 

多表查询

var list = db.queryable<student, school>((st, sc) => new object[] {
        jointype.left,st.schoolid==sc.id})
      .select((st,sc)=>new{id=st.id,name=st.name,schoolname=sc.name}).tolist();

生成的sql如下

select  [st].[id] as [id] , 
          [st].[name] as [name] , 
          [sc].[name] as [schoolname]  from [student] st 
          left join school sc on ( [st].[schoolid] =[sc].[id])

 

多表查询自支持自动填充到viewmodel

var s11 = db.queryable<student, school>((st, sc) => st.schoolid == sc.id)
                        .select<viewmodelstudent3>().tolist();

 public class viewmodelstudent3: student
{
         public string schoolname { get; set; }
 }

 

生成的sql如下

select
           sc.[name] as [schoolname],--这一列神奇的自动出现了
           sc.[id] as [scid],
           st.[id] as [id],
           st.[schoolid] as [schoolid],
           st.[name] as [name],
           st.[createtime] as [createtime]
           
           
            from [student] st  ,[school]  sc  where ( [st].[schoolid] = [sc].[id])

 

多表分页查询

 var list3 = db.queryable<student, school>((st, sc) => new object[] {
              jointype.left,st.schoolid==sc.id
            }).select<viewmodel>()
            .topagelist(pageindex,pagesize)

 

子查询

var getall = db.queryable<student, school>((st, sc) => new object[] {
jointype.left,st.id==sc.id})
.where(st => st.id == sqlfunc.subqueryable<school>().where(s => s.id == st.id).select(s => s.id))
.tolist();
      
//生成的mysql语句，如果是sqlserver就是top 1
select `st`.`id`,`st`.`schoolid`,`st`.`name`,`st`.`createtime` 
     from `student` st left join `school` sc on ( `st`.`id` = `sc`.`id` )  
      where ( `st`.`id` =(select `id` from `school` where ( `id` = `st`.`id` ) limit 0,1))

 

一对一的查询

var getall = db.queryable<student, school>((st, sc) => new object[] {
jointype.left,st.id==sc.id})
.select(st =>
       new{
              name = st.name,
              id = sqlfunc.subqueryable<school>().where(s => s.id == st.id).select(s => s.id)
       }).tolist();

 

 本文只讲重点，更多多表查询请看 api

 http://www.codeisbug.com/doc/8/1124

 

mapper功能

如果说 .select() 也可以实现一对一的查询或者一些sql函数但是毕竟是用来生成sql的所以有很多局限性，mapper是在查询出结果后进行处理所以任何c#方法都支持

也更强大

 var s12 = db.queryable<student, school>((st, sc) => st.schoolid == sc.id).select<viewmodelstudent3>()

                .mapper(it =>
                {

                    it.name = md5(it.name);
                    //有多少列要处理写多少列，能用mapper的就少用select兼容性更好些

                }).tolist();

 

高性能的一对多查询

我们也可以用mapper来实现一对多，弥补.select()不足

var s12 = db.queryable<student, school>((st, sc) => st.schoolid == sc.id).select<viewmodelstudent3>()

.mapper((it, cache) =>
{

    var allschools = cache.getlistbyprimarykeys<school>(vmodel => vmodel.schoolid);
    //in（viewmodelstudent3[0].schoolid , viewmodelstudent3[1].schoolid...）


    /*one to many*/
    it.schools = allschools.where(i => i.id == it.schoolid).tolist();


    /*c# syntax conversion*/
    it.name = it.name == null ? "null" : it.name;

}).tolist();

一对多查询的性能可以秒杀其它orm ，因为生成的sql只有2条，并且这2条不会多查询一条没用的记录，有幸趣的可以研究一下，其它的都内存处理

 

多queryable查询

union all查询将结果集合并

var getunionalllist2 = db.unionall(db.queryable<student>(), db.queryable<student>()).tolist();//union all

 

两个queryable联表查询（有人说我只支持12表join，那这样就可以支持24张表了）

var q1 = db.queryable<student, school>((st,sc)=>new object[] {
                jointype.left,st.schoolid==sc.id
            }).select((st, sc) => new viewmodelstudent4() { id=st.id, name=st.name,schoolname=sc.name });
 
var q2 = db.queryable<school>();
 
 
var innerjoinlist = db.queryable(q1, q2, (j1, j2) => j1.id == j2.id).select((j1, j2) => j1).tolist();//inner join
 
var leftjoinlist = db.queryable(q1, q2,jointype.left, (j1, j2) => j1.id == j2.id).select((j1, j2) => j1).tolist();/

 

二级缓存支持

二级缓存功能是对查询出来的数据进行缓存，在缓存不失效的情况下，下次同样的查询操作都会从缓存内读取

 

使用缓存查询

var list=db.queryable<student, school>((s1, s2) => s1.id == s2.id).select(s1 => s1).withcache().tolist();//可以设置过期时间withcache(60)

 

删除缓存

我们需要删除缓存也相当方便，只需要在对该表操作的时候加 removedatacache 就能把查询中引用该表的缓存全部清除

db.deleteable<student>().where(it => it.id == 1).removedatacache().executecommand();

//updateable和insertable一样用法

 

自动删除缓存

sqlsugarclient db = new sqlsugarclient(new connectionconfig() {
                connectionstring = config.connectionstring,
                moresettings =new  connmoresettings(){
                      isautoremovedatacache=true
                }

 

创建db对象

我们需要创建一个mycache类，你可以用我写好的也可以用你自已写的实现缓存

 

icacheservice mycache = new rediscache("10.1.249.196");//icacheservice
sqlsugarclient db = new sqlsugarclient(new connectionconfig()
{
connectionstring = config.connectionstring,
dbtype = dbtype.sqlserver,
isautocloseconnection = true,
configureexternalservices = new configureexternalservices()
{
datainfocacheservice = new rediscache() //rediscache是继承icacheservice自已实现的一个类
}
});

 

我写好的cache类可以作为参考

redis:

https://github.com/sunkaixuan/sqlsugar/blob/dev/src/asp.net/sqlsugar.extensions.datacache/rediscache.cs

.net自带cache:

https://github.com/sunkaixuan/sqlsugar/blob/dev/src/asp.net/sqlsugar.extensions.datacache/httpruntimecache.cs

 

永久开源，源码下：

https://github.com/sunkaixuan/sqlsugar

sqlsugar一直在默默进步，得到了一些大型企业的认可，但这只是开始，希望你喜欢。

 

上一篇