轻量型ORM框架Dapper的使用

在真实的项目开发中，可能有些人比较喜欢写SQL语句，但是对于EF这种ORM框架比较排斥，那么轻量型的Dapper就是一个不错的选择，即让你写sql语句了，有进行了关系对象映射。其实对于EF吧，我说下我个人的见解，EF是相对来说比较重的ORM框架，它是讲究通用性的框架，而非单纯的实现某种需求的框架，不得不否认的说，EF的开发效率要比用单纯的ADO.NET写sql语句的效率高的不是一点半点，可能很多人会说EF的性能低等等，针对这种问题有时候是需要折中的来讲，从另一个方面来讲，你用了ADO.NET写sql语句是性能好了，但是你得到的是弱类型的DataTable或者DataSet，常规在MVC的开发中，其实你难道不转换成集合或者单个对象吗？总不会拿着DataTable用吧，当你把弱类型的Datatable转换成List<T>的时候，一般来讲分两种转换，一种是利用反射，这其实也就成了关系对象映射了，跟ORM框架没什么区别了，第二种无非是循环赋值了，但是这种的开箱装箱的损耗也是很大的，最终你得到你想要的数据类型的时候其实折中算下来相比较EF直接得到，中间你又走了很多的弯路了，我在开发中一般EF打头阵，轻量的Dapper做辅助，因为有时候开发效率很重要，性能的问题我不会过多在SQL优化上下功夫，我一般会在中间缓存上做处理。下面是我简单的对Dapper的小包装，代码如下：

 

public class LiunianContext:IDisposable { private static readonly string _dbConnectionStr = ConfigurationManager.ConnectionStrings["XXXContext"].ConnectionString; private IDbConnection dbConnection; private IDbConnection DbConnection { get { if (_dbConnectionStr.IsNotNullOrEmpty()) dbConnection = new SqlConnection(_dbConnectionStr); else throw new ArgumentNullException("dbConnectionStr"); bool isClosed = dbConnection.State == ConnectionState.Closed; if (isClosed) dbConnection.Open(); return dbConnection; } } /// <summary> /// 执行增删改操作(包括批量操作) /// </summary> /// <param name="sql">sql语句(有参数参数化)</param> /// <param name="param">参数化值</param> /// <returns></returns> public bool Execute(string sql, object param) { bool isSuccess = false; if (sql.IsNotNullOrEmpty()) { try { int result = DbConnection.Execute(sql, param); isSuccess = result > 0 ? true : false; } catch { isSuccess = false; } } return isSuccess; } /// <summary> /// 执行存储过程操作 /// </summary> /// <param name="sql">存储过程名称</param> /// <param name="param">参数化值</param> /// <returns>返回存储过程是否执行成功</returns> public bool ExecuteStored(string storedName, object param) { bool isSuccess = false; if (storedName.IsNotNullOrEmpty()) { try { int result = DbConnection.Execute(storedName, param, commandType: CommandType.StoredProcedure); isSuccess = result > 0 ? true : false; } catch { isSuccess = false; } } return isSuccess; } /// <summary> /// 执行存储过程操作 /// </summary> /// <param name="storedName">存储过程名称</param> /// <param name="param">存储过程参数</param> /// <returns>返回存储过程要返回的值</returns> public DynamicParameters ExecuteStored(string storedName, DynamicParameters param) { if (storedName.IsNotNullOrEmpty()) { try { DbConnection.Execute(storedName, param, commandType: CommandType.StoredProcedure); } catch { } } return param; } /// <summary> /// 查询操作 /// </summary> /// <typeparam name="T">返回集合的类型</typeparam> /// <param name="sql">sql语句</param> /// <param name="param">参数化值</param> /// <returns></returns> public IEnumerable<T> Query<T>(string sql, object param) { IEnumerable<T> _list = default(IEnumerable<T>); if (!string.IsNullOrEmpty(sql)) { try { _list = DbConnection.Query<T>(sql, param); } catch { } } return _list; } /// <summary> /// 执行存储过程查询操作 /// </summary> /// <typeparam name="T">返回集合的类型</typeparam> /// <param name="storedName">存储过程</param> /// <param name="param">参数化值</param> /// <returns></returns> public IEnumerable<T> QueryStored<T>(string storedName, object param) { IEnumerable<T> _list = default(IEnumerable<T>); if (!string.IsNullOrEmpty(storedName)) { try { _list = DbConnection.Query<T>(storedName,commandType: CommandType.StoredProcedure); } catch { } } return _list; } /// <summary> /// 查询操作返回默认第一条数据(如返回null则创建默认类型) /// </summary> /// <typeparam name="T"></typeparam> /// <param name="sql"></param> /// <param name="param"></param> /// <returns></returns> public T FirstOrDefault<T>(string sql, object param) { var model = default(T); if (!string.IsNullOrEmpty(sql)) { try { model = DbConnection.Query<T>(sql, param).FirstOrDefault(); } catch { } } return model == null ? Activator.CreateInstance<T>() : model; } /// <summary> /// 查询一组SQL语句并返回值 /// </summary> /// <typeparam name="T1">第一条语句返回集合类型</typeparam> /// <typeparam name="T2">第二条语句返回集合类型</typeparam> /// <param name="sql">sql语句</param> /// <param name="param">参数化值</param> /// <returns></returns> public Tuple<IEnumerable<T1>, IEnumerable<T2>> Query<T1, T2>(string sql, object param) { IEnumerable<T1> _item1 = null; IEnumerable<T2> _item2 = null; if (!string.IsNullOrEmpty(sql)) { try { using (var multi = DbConnection.QueryMultiple(sql, param)) { _item1 = multi.Read<T1>(); _item2 = multi.Read<T2>(); } } catch { } } return Tuple.Create<IEnumerable<T1>, IEnumerable<T2>>(_item1, _item2); } /// <summary> /// 查询一组SQL语句并返回值 /// </summary> /// <typeparam name="T1">第一条语句返回集合类型</typeparam> /// <typeparam name="T2">第二条语句返回集合类型</typeparam> /// <typeparam name="T3">第三条语句返回集合类型</typeparam> /// <param name="sql">sql语句</param> /// <param name="param">参数化值</param> /// <returns></returns> public Tuple<IEnumerable<T1>, IEnumerable<T2>, IEnumerable<T3>> Query<T1, T2, T3>(string sql, object param) { IEnumerable<T1> _item1 = null; IEnumerable<T2> _item2 = null; IEnumerable<T3> _item3 = null; if (!string.IsNullOrEmpty(sql)) { try { using (var multi = DbConnection.QueryMultiple(sql, param)) { _item1 = multi.Read<T1>(); _item2 = multi.Read<T2>(); _item3 = multi.Read<T3>(); } } catch { } } return Tuple.Create<IEnumerable<T1>, IEnumerable<T2>, IEnumerable<T3>>(_item1, _item2, _item3); } public void Dispose() { if (dbConnection != null) { try { bool isClosed = dbConnection.State == ConnectionState.Closed; if (!isClosed) dbConnection.Close(); //dbConnection.Dispose();  } catch { } } } }

调用示例如下：

 

public abstract class BaseProvide { public Lazy<LiunianContext> lazyContext; protected LiunianContextDb { get { return lazyContext.Value; } } } public class EmployeeProvide : BaseProvide { /// <summary> /// 获取数据 /// </summary> /// <returns></returns> public List<Employee> ListEmployeeByParam(int[] param) { string sql = "select * from Employee where ID in @Ids"; return Db.Query<Employee>(sql, new { Ids = param }).ToList(); } /// <summary> /// 获取全部数据 /// </summary> /// <returns></returns> public List<Employee> ListEmployee() { string sql = "select * from Employee"; return Db.Query<Employee>(sql,null).ToList(); } /// <summary> /// 查询单个实体 /// </summary> /// <param name="Id"></param> /// <returns></returns> public Employee loadEmployee(int Id) { string sql = "select * from Employee where ID= @Id"; return Db.FirstOrDefault<Employee>(sql, new { Id = Id }); } /// <summary> /// 插入操作 /// </summary> /// <param name="model"></param> /// <returns></returns> public bool InsertEmployee(Employee model) { StringBuilder sql = new StringBuilder(); sql.Append("insert into Employee values (@UpdateTime,@CreateTime,@State,@LoginName,@UserName,"); sql.Append("@RoleId,@IsUse)"); return Db.Execute(sql.ToString(), model); } /// <summary> /// 更新操作 /// </summary> /// <param name="model"></param> /// <returns></returns> public bool UpdateEmployee(Employee model) { StringBuilder sql = new StringBuilder(); sql.Append("update Employee set UpdateTime = @UpdateTime where ID = @ID"); return Db.Execute(sql.ToString(), model); } /// <summary> /// 删除操作 /// </summary> /// <param name="Id"></param> /// <returns></returns> public bool DeleteEmployeeById(int Id) { string sql = "delete from Employee where ID = @Id"; return Db.Execute(sql.ToString(), new { Id = Id }); } /// <summary> /// 执行多个语句返回多个结果 /// </summary> /// <param name="Id"></param> /// <returns></returns> public Tuple<int,IEnumerable<Employee>> ListGroupEmployee() { string sql = "select Count(*) from Employee select * from Employee"; Tuple<IEnumerable<int>, IEnumerable<Employee>> queryGroup = Db.Query<int, Employee>(sql, null); return Tuple.Create<int, IEnumerable<Employee>>(queryGroup.Item1.FirstOrDefault(),queryGroup.Item2); } }

除了没把事务加上之外其他的增删改查都有了，同时在传对象的时候Dapper也解决了sql注入的问题，在性能上也是比较好的ORM框架，需要的拿去！