Passing Output parameters to stored procedure using dapper in c# code

I have a stored procedure in this format

CREATE PROCEDURE SP_MYTESTpROC
    @VAR1 VARCHAR(10),
    @VAR2 VARCHAR(20),
    @BASEID INT ,
    @NEWID INT OUTPUT
As Begin
   INSERT INTO TABLE_NAME(username, firstname)
      select @VAR1, @VAR2 
      WHERE ID = @BASEID

   SET @NEWID = SCOPE_IDENTITY() AS INT
END

I am calling this stored procedure from C# code using dapper. My question is: how do I pass in the output parameter to the stored procedure while using dapper?

Just searching the Test.cs file you could find this example

    public void TestProcSupport()
    {
        var p = new DynamicParameters();
        p.Add("a", 11);
        p.Add("b", dbType: DbType.Int32, direction: ParameterDirection.Output);
        p.Add("c", dbType: DbType.Int32, direction: ParameterDirection.ReturnValue);
        connection.Execute(@"create proc #TestProc 
                         @a int,
                             @b int output
                             as 
                             begin
                                 set @b = 999
                                 select 1111
                                 return @a
                             end");
        connection.Query<int>("#TestProc", p, commandType: CommandType.StoredProcedure).First().IsEqualTo(1111);
        p.Get<int>("c").IsEqualTo(11);
        p.Get<int>("b").IsEqualTo(999);
    }

So, I suppose that your C# code could be written as

    public void InsertData()
    {
        var p = new DynamicParameters();
        p.Add("VAR1", "John");
        p.Add("VAR2", "McEnroe");
        p.Add("BASEID", 1);
        p.Add("NEWID", dbType: DbType.Int32, direction: ParameterDirection.Output);
        connection.Query<int>("SP_MYTESTpROC", p, commandType: CommandType.StoredProcedure);
        int newID =  p.Get<int>("NEWID");
    }

As a side note, do not use SP as prefix for your stored procedure. It is reserved for system defined procedures and you could find yourself in troubles if Microsoft decides to use the same name. Albeit improbable it is a bad practice and why risk?

Further to "ath's" suggestion: To avoid reflection, DynamicParmers.AddDynamicParams() takes an anonymous object, after which you could add the return paramter like this...

var param = new { A="a", B="b" };
var dynamicParameters = new DynamicParameters();
dynamicParameters.AddDynamicParams(parameters);
dynamicParameters.Add("return", dbType: DbType.Int32, direction: ParameterDirection.ReturnValue);

now use the dynamicParameters object in your dapper call instead of the anonymous param object.

(You can also do this for an output parameter if preferred)

If you always have an OUTPUT parameter of INTEGER type named @id (@id = @id OUTPUT), you could make an extension method like this which would allow you to use the regular Dapper syntax passing the sql string and an anonymous object:

using Dapper;
using System.Data;
using System.Data.SqlClient;

public static int ExecuteOutputParam
            (this IDbConnection conn, string sql, object args)
        {
            // Stored procedures with output parameter require
            // dynamic params. This assumes the OUTPUT parameter in the
            // SQL is an INTEGER named @id.
            var p = new DynamicParameters();
            p.Add("id", dbType: DbType.Int32, direction: ParameterDirection.Output);

            var properties = args.GetType().GetProperties();
            foreach (var prop in properties)
            {
                var key = prop.Name;
                var value = prop.GetValue(args);

                p.Add(key, value);
            }

            conn.Execute(sql, p);

            int id = p.Get<int>("id");
            return id;
        }

This uses reflection to read all properties, but if you can take that penalty, you don't have to boilerplate the DynamicParameters for every call.

For transactions make an extension method on SqlTransaction passing it to Execute like so:

transaction.Connection.Execute(sql, p, transaction);