Mysql - How to quit/exit from stored procedure

I have very simple question but i did't get any simple code to exit from SP using Mysql. Can anyone share with me how to do that?

CREATE PROCEDURE SP_Reporting(IN tablename VARCHAR(20))
BEGIN
     IF tablename IS NULL THEN
          #Exit this stored procedure here
     END IF;

     #proceed the code
END;

Solution 1:

CREATE PROCEDURE SP_Reporting(IN tablename VARCHAR(20))
proc_label:BEGIN
     IF tablename IS NULL THEN
          LEAVE proc_label;
     END IF;

     #proceed the code
END;

Solution 2:

If you want an "early exit" for a situation in which there was no error, then use the accepted answer posted by @piotrm. Most typically, however, you will be bailing due to an error condition (especially in a SQL procedure).

As of MySQL v5.5 you can throw an exception. Negating exception handlers, etc. that will achieve the same result, but in a cleaner, more precise manner.

Here's how:

DECLARE CUSTOM_EXCEPTION CONDITION FOR SQLSTATE '45000';

IF <Some Error Condition> THEN      
    SIGNAL CUSTOM_EXCEPTION
    SET MESSAGE_TEXT = 'Your Custom Error Message';
END IF;     

Note SQLSTATE '45000' equates to "Unhandled user-defined exception condition". By default, this will produce an error code of 1644 (which has that same meaning). Note that you can throw other condition codes or error codes if you want (plus additional details for exception handling).

For more on this subject, check out:

https://dev.mysql.com/doc/refman/5.5/en/signal.html

How to raise an error within a MySQL function

http://www.databasejournal.com/features/mysql/mysql-error-handling-using-the-signal-and-resignal-statements.html

Addendum

As I'm re-reading this post of mine, I realized I had something additional to add. Prior to MySQL v5.5, there was a way to emulate throwing an exception. It's not the same thing exactly, but this was the analogue: Create an error via calling a procedure which does not exist. Call the procedure by a name which is meaningful in order to get a useful means by which to determine what the problem was. When the error occurs, you'll get to see the line of failure (depending on your execution context).

For example:

CALL AttemptedToInsertSomethingInvalid;

Note that when you create a procedure, there is no validation performed on such things. So while in something like a compiled language, you could never call a function that wasn't there, in a script like this it will simply fail at runtime, which is exactly what is desired in this case!

Solution 3:

To handle this situation in a portable way (ie will work on all databases because it doesn’t use MySQL label Kung fu), break the procedure up into logic parts, like this:

CREATE PROCEDURE SP_Reporting(IN tablename VARCHAR(20))
BEGIN
     IF tablename IS NOT NULL THEN
         CALL SP_Reporting_2(tablename);
     END IF;
END;

CREATE PROCEDURE SP_Reporting_2(IN tablename VARCHAR(20))
BEGIN
     #proceed with code
END;