Comments on: Oracle 11g: New Pivot Table Feature

By: Dynamic subquery in Pivot Query - dBforums

Dynamic subquery in Pivot Query - dBforums — Fri, 26 Nov 2010 21:23:26 +0000

[...] We have someone who needs to do something like this. Right now the plan is to create some dynamic SQL. You may find this page useful. Oracle 11g: New Pivot Table Feature | The Pythian Blog [...]

By: Manan Sanghvi

Manan Sanghvi — Thu, 25 Jun 2009 21:54:21 +0000

Hey,
I am using the Oracle Warehouse Builder 11G. I am having an issue with the pivot operator. IN my source database I have a table with around 200 columns, each having say 1000 rows. Now I want to make each of the column into multiple rows by pivoting it. IN the pivot operator provided I have to type in the name of the columns manually which is not feasible for such big tables. Any one have any suggestions??

By: Chris

Chris — Wed, 17 Dec 2008 15:54:37 +0000

I’m currently writing some code that requires a pivot table. All I need to do is generate an output file with the pivoted data, and I’m doing it using SQL*Plus. In this case, I’m just going to generate the IN list using SQL-generating-SQL. I’ve got a variable list of values from 120-140 values going across the columns, so I can’t make it static. The SQL-generating-SQL technique works fine for this case. Even if I was writing a C/C++ program, I’d still use dynamic SQL for this and allocate my return buffers in a variable sized array. I personally don’t see this as a huge limitation of the pivot table feature in 11g. Relying on XML to make the columns variable was probably a good decision on Oracle’s part. XML gives you an overall structure to handle the returning data rather than using an array like I would do. My approach works, but it’s probably a little old-school. Using XML is more forward-thinking if you ask me.

By: MARTIN

MARTIN — Tue, 25 Nov 2008 11:21:57 +0000

Hello, Can anyone help me? Ive been trying a simple pivot byt just cant get it to work.

* i need the below select statement to have an output as below:
SELECT BGROUP, REFNO, SEQNO, PP02D, PP05P
FROM PRESERVED_PENSIONER_HISTORY
WHERE REFNO IN (’0084941′,’0079500′,
’0082833′,’0094388′,’0063814′,’0081368′)

*OUTPUT

REFNO DATE1(PP02D) DATE2(PP02D)
0084941 PP05P PP05P
0079500 PP05P PP05P
0082833 PP05P PP05P

There are many values for both values PP02D and PP05P.
Any help will be appreciated.

By: Mathew

Mathew — Wed, 23 Apr 2008 04:57:10 +0000

excelent functionality but the usage is very limited with the restriction of variable value in the pivot section.I dont know why the Orcale more concentrated with XML intsted of providing the functionality in the base noraml SQL.

By: Pythian Group Blog » Blog Archive » Log Buffer #66: a Carnival of the Vanities for DBAs

Fri, 12 Oct 2007 16:38:38 +0000

[...] difference between count(*) and count(1). Also here at home, Robert Hamel gives a bad rating to 11g’s new pivot table feature. “I was very pleased when I heard about Oracle adding pivot functionality in select [...]

By: ghassan

ghassan — Fri, 12 Oct 2007 08:55:02 +0000

Robert,
Doing what you want will not be easy, not only for Oracle, but for the clients: Supporting the sub-query and ANY without XML implies that CBO not only parses the SQL, but also executes the sub-query (or the query if you use ANY) in order to find out how many columns this query returns (it has to know in order to tell the ‘client’ programs how many columns there are so that they allocate the buffers, ….). A way out would be to augment the SQL with the MAX number of columns the user expects, but this puts much burden on the user, and the system will have to decide what to do in case of ANY surpassing this MAX.

By: Mr. Ed

Mr. Ed — Thu, 11 Oct 2007 07:50:17 +0000

2 out of 10. At this rate, it won’t work properly until Oracle 13g Release 1.