Skip to Main Content

SQL & PL/SQL

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

Interested in getting your voice heard by members of the Developer Marketing team at Oracle? Check out this post for AppDev or this post for AI focus group information.

Data Joins and Pivot Help

684479Jan 22 2010 — edited Jan 27 2010
Hi, I have one fact table and one huge history table that goes really deep with scores stored with month end dates.

I am looking for a way to manipulate the month and year of DateChanged in table A along with joins on the acct no to retrieve scores from previous months, resulting in a de-normalized data format as in the result table.

Any ideas?? Thanks!


Table A

Acct	DateChanged
-----------------------------------
123	1/12/2008
456	8/25/2008
 


Table B:
 
Acct           DateUpload        Score
-----------------------------------------
123           11/30/2007        620
123           12/31/2007        650
123           01/31/2008        712
123           02/29/2008        800
456           05/31/2008        520  
456           06/30/2008        630
456           07/31/2008        680
456           08/31/2008        710
456           09/30/2008        780
456           10/31/2008        786
456           11/30/2008        823



Result Table:
		        
Acct      Score2MonthPrior     Score1MonthPrior
-------------------------------------------------------
123            620                     650
456            630                     680
This post has been answered by Frank Kulash on Jan 22 2010
Jump to Answer

Comments

Frank Kulash

Hi,

Whenever you have a question, please post a little sample data (CREATE TABLE and INSERT statements, relevant columns only) for all the tables involved, and the exact results you want from that data, so that the people who want to help you can re-create the problem and test their ideas.

Explain, using specific examples, how you get those results from that data.

Always say what version of Oracle you're using (e.g. 11.2.0.2.0).

See the forum FAQ:

Use the analytic ROW_NUMBER function to assign numbers 1, 2, 3, ... to each row, with a separate series for each worker and date.  Each distinct combination of worker and assigned number will result in a row of output.

CarlosDLG

You will usually get quicker and more accurate answers if you post create table statements and inserts with some sample data, along with an explanation of the specific problem you are facing.

Here is one way to do it:

WITH test_data as

(

SELECT 'Project_A' project,'John' worker,date '2016-05-23' the_date FROM DUAL union all

SELECT 'Project_A','Mary',date '2016-05-23'  FROM DUAL union all

SELECT 'Project_A','Mary',    date '2016-05-24' FROM DUAL union all

SELECT 'Project_A','Steve',date '2016-05-24' FROM DUAL union all

SELECT 'Project_A','Mary',date '2016-05-25' FROM DUAL union all

SELECT 'Project_A','Mary',date '2016-05-26'  FROM DUAL union all

SELECT 'Project_B','John',date '2016-05-23' FROM DUAL union all

SELECT 'Project_B','Steve',date '2016-05-24' FROM DUAL

)

select worker,c23,c24,c25,c26 from

(

  SELECT t.*, dense_rank() over (PARTITION BY worker ORDER BY project) AS rn

  FROM test_data t

)

pivot

(

  MAX(project) FOR the_date in (date '2016-05-23' AS C23,date '2016-05-24' AS C24,date '2016-05-25' AS C25,date '2016-05-26' AS C26)

)

ORDER BY worker;

Results:

WORKERC23C24C25C26
JohnProject_A
JohnProject_B
MaryProject_AProject_AProject_AProject_A
Steve Project_A
Steve Project_B
Paulzip

You cannot dynamically generate your columns (e.g. generate your column names based on the data) without using special techniques (dynamic SQL), SQL has to have static columns at execution time.

Here's another way to do it.

select worker, d1, d2, d3, d4

from

(

  select project, worker,

         dense_rank() over (order by the_date) as date_rank,

         dense_rank() over (order by project) as project_rank

  from test_data t

)

pivot (

  max(project)

  for date_rank in (1 as D1, 2 as D2, 3 as d3, 4 as d4)

)

order by worker

WORKERD1D2D3D4
JohnProject_A
JohnProject_B
MaryProject_AProject_AProject_AProject_A
SteveProject_A
SteveProject_B
1 - 3
Locked Post
New comments cannot be posted to this locked post.

Post Details

Locked on Feb 24 2010
Added on Jan 22 2010
14 comments
2,190 views