introduce increment/decrement operator (++/--) in SQL and PL/SQL for integer types

TPD-Opitz

I'd like to have increment/decrement operator (++/--) in SQL and PL/SQL.for integer types

in SQL this Operator should be available for updates like:

UPDATE some_table
SET integer_arrtibute ++,
,another_integer_Attribute--,
where ...

leading Operator or combination with assingments should not be allowed.

In PL/SQL the Operator should work as in Java, mening that the leading Operator is evalueated bevore an assingment and the trailing Operator is evaluated after an assingment:

declare
  integer v_test1 := 1;
  integer v_test2 := 5;
begin
v_test2 := v_test1++;
if(6=v_test2) then dbms_Output.put_line('success'); end if;
if(2=v_test1) then dbms_Output.put_line('success'); end if;
v_test2 := ++v_test1;
if(9=v_test2) then dbms_Output.put_line('success'); end if;
if(3=v_test1) then dbms_Output.put_line('success'); end if;
end;

bye

TPD

Pradeep Agrawal

Hi,

I think you posted this idea in the wrong place.

This community is for BigMachines.

Regards,

Pradeep

TPD-Opitz

Hi,

I think you posted this idea in the wrong place.

This community is for BigMachines.

Regards,

Pradeep

Thanks for the tipp,

I moved it to database.

bye

TPD

Pradeep Agrawal

Thanks for the tipp,

I moved it to database.

bye

TPD

welcome

abhinivesh.jain

Is this operation part of ANSI standards? Also what specific purpose will it solve?

TPD-Opitz

Is this operation part of ANSI standards? Also what specific purpose will it solve?

abhinivesh.jain wrote:

Is this operation part of ANSI standards?

is select * from tab_a, tab_b where a_attrib=b_attrib(+) ANSI standard?

Why am I restricted to ANSI standard with my enhancement suggestions?

Also what specific purpose will it solve? 

In SQL: easier handling of version numbering.

In PL/SQL: easier handling of counter variables in loops, especially when looping trough XML.

bye

TPD

TPD-Opitz

I just thought about the SQL operator and now I feel a bit uncomfortable with the trailing increment/decrement operator since this is kind of inconsistent:

It is some kind of assignment and the general convention for the trailing increment/decrement operator in other languages (like Java) is that the increment/decrement occurs after the assignment when the operator is trailing.

In conequence the SQL-proposal should be leading only instead of trailing only.

bye

TPD

Vadim Tropashko-Oracle

So, what the result of

select rownum++ from dual

connect by rownum++ < 10

is supposed to be?

William Robertson

I'm voting ++ for this as a PL/SQL feature. I can't see how it makes sense in SQL, though.

Christian Erlinger

Is it just me, but the last time I checked at least '--' is already part of the SQL and PL/SQL syntax - it indicates a single line comment...Of course one could argue that

i--;

clearly shows that you want to decrement i whereas

-- some comment

certainly is a comment.

however; what about this this

if --i = 5 then
  [...]

is this a comment?

or this

begin
  if --haha
    1 = 0 then
    dbms_output.put_line('blah');
  else --huhu;
    dbms_output.put_line('bleh');
  end if;
end;

? I haven't studied building compilers, but at least to me the introduction of a -- operator could possibly make existing, valid code invalid.

I have to deal with several languages from Java to PL/SQL in my day to day job, so I can sympathize as I frequently add '--' in my Java code to add comments and wonder about compiler errors

Maybe the related operators += and -= might prove more powerful and readable then ++ and --; e.g.

update some_table
set some_column +=5;

Or maybe

update some_table
set some_column -= another_column;

and of course the same in PL/SQL.

cheers

TPD-Opitz

So, what the result of

select rownum++ from dual

connect by rownum++ < 10

is supposed to be?

You may have noticed that i requested the SQL ++/-- operator for the set part in updates only.

bye

TPD