将 MS-SQL 存储过程转换为 MYSQL，但现在由于语法错误而无法创建

Question

I have a strong MS-SQL background but have never seen a line of mysql until yesterday. I need to convert my ms-sql stored proc to mysql and I thought I did it correctly BUT when I try to execute the sql to create the proc I just get incorrect syntax errors like this " #1064 - You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '//' at line 10". The stored proc is super simple so I feel dumb that I havent been able to figure this out. BTW, the recipe table has an identity on it called ID which is why I am using scope_identity.

MS-SQL

CREATE PROCEDURE ins_recipe
  @Name varchar(8000),
  @Directions varchar(8000);
AS
  IF NOT EXISTS(SELECT * FROM recipe WHERE Name = @Name)
    BEGIN
      INSERT INTO recipe ('name', 'directions')
      SELECT @Name, @Directions;

      SELECT SCOPE_IDENTITY;
    END
  ELSE
    BEGIN
      UPDATE recipe
        SET Directions = @Directions
      WHERE Name = @Name;

      SELECT ID FROM recipe WHERE Name = @Name;
    END
GO

I dont know how to do it all in mysql so this is just my starting attempt

DELIMITER //

CREATE PROCEDURE ins_recipe (IN
  p_Name varchar(8000),
  p_Directions varchar(8000))
BEGIN
  INSERT INTO `recipe` (Name, Directions) 
  SELECT * FROM (SELECT p_Name, p_Directions) AS tmp
  WHERE NOT EXISTS(SELECT Name FROM `recipe` WHERE Name = p_Name) LIMIT 1;

  SELECT LAST_INSERT_ID();
END //

Any help would be mucho appreciated. The error I get is that there is incorrect syntax near // and to refer to my correct mysql documentation. If anyone feels like helping me convert that whole procedure that would also be much appreciated but getting past the syntax errors at this point is my priority.

Answer 1

For MySQL, I would suggest the following. First, create a unique index/constraint, to prevent duplicate names:

create unique index unq_recipe_name on recipe(name);

Then, write the function using on duplicate key insert :

DELIMITER //

CREATE PROCEDURE ins_recipe (
    IN p_Name varchar(8000),
    IN p_Directions varchar(8000)
   )
BEGIN
    INSERT INTO `recipe` (Name, Directions) 
        VALUES (p_Name, p_Directions)
        ON DUPLICATE KEY UPDATE name = p_name;  -- this is a no op
    SELECT LAST_INSERT_ID();
END; //

The advantage to this approach is that the database maintains the data integrity. Without the index, two separate processes could insert two recipes with the same name into the database.

How does this work? If a duplicate is found, then the ON DUPLICATE KEY clause is invoked. The name = p_name is a no-op, because name already has that value.