Mysql creates a stored procedure and uses Cursor to perform cyclic updates.

Use cursor)

1. Declare a cursor

Cursor for select_statement
This statement declares a cursor. You can also define multiple cursors in a subroutine, but each cursor in a block must have a unique name. A single operation is performed after the cursor is declared, but the SELECT statement cannot have an INTO clause.

2. cursor OPEN statement

This statement opens the previously declared cursor.

3. cursor FETCH statement

INTO var_name [, var_name]...
This statement reads the next row with the specified open cursor (if there is a next row) and advances the cursor pointer.

4. cursor CLOSE statement

This statement closes the previously opened cursor.

The following example shows how to use a cursor to update a stored procedure:

-- Procedure "useCursor" ddlcreate procedure 'usecursor '() BEGIN/* Definition of local variables declare */declare aid bigint default 0; declare mdsl bigint default 0; declare stop int default 0; declare cur cursor for (select count (area_info.id) as mdsl, area_info.id as aid from area_info right join subbranch_info on region = area_info.idwhere area_info. I _record_status = 0 and region _ Status = 0 group by area_info.id);/* mysql does not know why to add an exception to the judgment? * This captures a cursor exception * and sets the loop to jump out of the loop when the variable stop is null. */Declare continue handler for sqlstate '000000' SET stop = null;/* OPEN cursor */OPEN cur;/* move the cursor down, pay the two queried values to the two Defined variables */FETCH cur INTO mdsl and aid;/* loop body, which obviously adds and uses the names queried by the cursor; no. */WHILE (stop is not null) DOupdate zlgsydrsp. area_info set I _subbran_count = mdsl where id = aid;/* move the cursor one step down */FETCH cur INTO mdsl, aid; END WHILE; CLOSE cur; END;