Fetching from a Cursor Variable
The FETCH statement retrieve rows one at a time from the product set of a multi-row query. The syntax for the same is as shown:
FETCH {cursor_variable_name | :host_cursor_variable_name}
INTO {variable_name[, variable_name]... | record_name};
In the next example, you fetch rows from the cursor variable emp_cv into the
user-defined record emp_rec:
LOOP
/* Fetch from cursor variable. */
FETCH emp_cv INTO emp_rec;
EXIT WHEN emp_cv%NOTFOUND; -- exit when last row is fetched
-- process data record
END LOOP;
Any variables in the related query are evaluated only when the cursor variable is opened. To change the product set or the values of variables in the query, you should reopen the cursor variable with the variables set to their new values. Though, you can use a different INTO clause on separate fetches with similar cursor variable. Each fetch retrieve another row from the similar result set.
The PL/SQL ensures the return type of the cursor variable is compatible with the INTO clause of the FETCH statement. For each of the column value returned by the query related with the cursor variable, there should be a parallel, type-compatible field or variable in the INTO clause. The number of fields or variables should also equal the number of column values. Or else, you get an error.
The error occurs at the compile time, when the cursor variable is strongly typed or at run time, if it is weakly typed. At run time, the PL/SQL raises the predefined exception ROWTYPE_MISMATCH before the first fetch. Therefore, if you trap the error and execute the FETCH statement using a different INTO clause, then no rows are lost.