pl /sql procedure execution giving an error - plsql

Question

My stored proc is defined as
create or replace procedure TEST(
name IN table1.col_name%type,
price IN table1.col_price%type
)
is
begin
update table1 t set t.name =name where t.price = price;
commit;
end TEST;
I am trying to execute it as
exec TEST(name => 'John', price => 1000);
However, it gives invalid SQL error. What am i missing here?

Answer 1

Your input parameter %type statements claim the column names are col_name and col_price. But that is not how you refer to them in your stored procedure (name and price).
Bad things can happen when you name variables after column names. AskTom recommends a limited convention of variable naming conventions:
local variables start with L_
parameters start with P_
global package variables start with G_
That link has a good general discussion on PL/SQL naming conventions. I personally just use V_ for most variables (aside from indexes and other obvious things), but that's just me.
Lastly, the col_ in the column names seem redundant; simply use name and price as column names.
So, that said, I think this does what you want:
create table table1 (
name varchar2(30),
price number
);
create or replace procedure TEST(
p_name IN table1.name%type,
p_price IN table1.price%type
)
is
begin
update table1
set name = p_name
where price = p_price;
commit;
end TEST;
/
insert into table1 values ('John', 500);
commit;
select * from table1;
exec TEST(p_name => 'Bob', p_price => 500);
select * from table1;
-- Clean up test artifacts
drop procedure test;
drop table table1;
Giving the output:
table TABLE1 created.
PROCEDURE TEST compiled
1 rows inserted.
committed.
NAME PRICE
------------------------------ ----------
John 500
anonymous block completed
NAME PRICE
------------------------------ ----------
Bob 500
procedure TEST dropped.
table TABLE1 dropped.

Answer 2

I really don't understand the variable prefixing approach. Oracle don't do it with their own API's, and it would be extraordinarily irritating if they did. It always seems like a workaround, rather than a fix.
For me the fix is to namespace the variables with the procedure name. It keeps the argument names "clean" and makes your code 100% proof against capture:
create or replace procedure TEST(
name IN table1.col_name%type,
price IN table1.col_price%type)
is
begin
update table1 t
set name = test.name
where t.price = price;
commit;
end TEST;
Lots more info on capture here.