GoldenGate

As Frank said Certainly better to use documented function better than use undocumented.

And if only in the 12s, improved switching sql <-> pl / sql. So I think it's better in the pl/sql to use the documented function of sql (MOD (a,b))

----

Ramin Hashimzade

Not completely undocumented - it is buried away in an example here: 4 Using PL/SQL

PaulHorth wrote:

Not completely undocumented - it is buried away in an example here: 4 Using PL/SQL

Hmmm, that's a dubious example. 

It's got one mention in a bit of example code, but that example itself says "Note the use of the SQL numeric function MOD to check for no (zero) remainder. See "Using Numeric Functions" for information about SQL numeric functions." and if you go to look at the "Using Numeric Functions" section the example given there uses the "MOD(employee_id, 2)" style.

The actual syntax diagram for MOD clearly shows it in the latter style...

MOD

So that is the one I would say is the documented syntax.

I know the other style is used in other languages, and it could be that perhaps the person creating that example you found had created the example without thinking, and was just lucky that PL/SQL understood it that way.

PaulHorth wrote:

Not completely undocumented - it is buried away in an example here: 4 Using PL/SQL

Maybe. But i think better will be use MOD (a,b) with SQL , "a mod b" with pl/sql to exclude switching sql <-> pl / sql. It is only my opinion ))). I think Mr. Kyte could better answer to this question.

----

Ramin Hashimzade

Hi,

RaminHashimzadeh wrote:

... But i think better will be use MOD (a,b) with SQL , "a mod b" with pl/sql to exclude switching sql <-> pl / sql. It is only my opinion )))...


----
Ramin Hashimzade

Try some experiments to check your opinion.  You'll see there is no difference in performance.

Almost all of the SQL single-row functions, including MOD, are implemented in the PL/SQL package dbms_standard.  There is no context switching when you use a function like MOD in PL/SQL.

Hi Frank,

if it is really implemented then off course , there will be no difference. But really i don't know it is implemented or not.

----

Ramin Hashimzade

BluShadow wrote:


PaulHorth wrote:

Not completely undocumented - it is buried away in an example here: 4 Using PL/SQL


Hmmm, that's a dubious example.  

It's got one mention in a bit of example code, but that example itself says "Note the use of the SQL numeric function MOD to check for no (zero) remainder. See "Using Numeric Functions" for information about SQL numeric functions." and if you go to look at the "Using Numeric Functions" section the example given there uses the "MOD(employee_id, 2)" style.

The actual syntax diagram for MOD clearly shows it in the latter style...

MOD

So that is the one I would say is the documented syntax.

I know the other style is used in other languages, and it could be that perhaps the person creating that example you found had created the example without thinking, and was just lucky that PL/SQL understood it that way.

I'm not disagreeing with you that it's dubious. Strange that it is accepted by PL/SQL but not officially documented though

The fact that PL handles it but SQL doesn't is good enough reason not to use it, especially as mod is a common mathematical operator/function.  Better to use what is common between the both.

I wonder when  n1 MOD n2 was introduced. Could it be XQUERY related, to support:

SCOTT@orcl > select xmlquery('9 mod 7' returning content) from dual;

XMLQUERY('9MOD7'RETURNINGCONTENT)
--------------------------------------------------------------------------------
2

SCOTT@orcl >

SY.

BluShadow wrote:

The fact that PL handles it but SQL doesn't is good enough reason not to use it, especially as mod is a common mathematical operator/function.  Better to use what is common between the both.

Still not disagreeing with you

I was just intrigued as to why they added support for it and then didn't document it.

PaulHorth wrote:


BluShadow wrote:

The fact that PL handles it but SQL doesn't is good enough reason not to use it, especially as mod is a common mathematical operator/function.  Better to use what is common between the both.

Still not disagreeing with you 

I was just intrigued as to why they added support for it and then didn't document it.

I suspect it stems more from the Ada language upon which PL is based.  In Ada, the syntax is "X mod Y" style, but in Oracle they obviously like functions to be more "fn(arg1,arg2..)" style.

Thanks for replies.

I will use Method1 which is documented.

Why I found this 2 methods is I made the same logic VB2005 and PL/SQL.

First step is making VB2005 logic.

Second step is making PL/SQL logic(Of course bese code is copied from VB2005 code).