Matrix report problem

Hi !!

I am using Oracle reports 6i to make a report (matrix report as far as i get it). I have following fields in the view which is coming from couple of tables and have some calculations to get these fields
(group by clause include section_code,group_code,machine_code,jobdate,organization_id)

Name Null? Type
----------------------------------------- -------- ---------------
SECTION_CODE NOT NULL VARCHAR2(5)
GROUP_CODE NOT NULL VARCHAR2(5)
MACHINE_CODE NOT NULL VARCHAR2(20)
JOBDATE NOT NULL DATE
ORGANIZATION_ID NOT NULL NUMBER(15)
MATPROC NUMBER
WASTAGE NUMBER
WASTPER NUMBER
REGRIND NUMBER
COSTKG NUMBER
COSTPC NUMBER
OUTMAN NUMBER
REJECTION NUMBER
UTILI NUMBER
EFFICIENCY NUMBER

Now they want the report in below mentioned format (group by section_code,group_code,machine_code)

------------01-Dec 02-Dec 03-Dec.......... 15-Dec-->(Jobdate field)
Matproc 150 220 170 200
Wastage 8 9 8 8
wastper .77 .99 .87 .98
regrind 1822 2288 2222 5544
costkg .033 .98 .76 .022
costpc .044 .77 .88 .83
outman 8.56 7.56 7.66 4.53
rejection 0 0 0 0
utili 12 12 15 14
effifciency 100 99 99 98


how i am suppose to do this in Oracle reports ..if anybody ahas any type of suggestion please send it as soon as possible..because its like a show stopper for me

regards

MNA

If jobdate contains million values , would you like to print across them?

Tamil

If there are an unknown/variable number of columns, then I would dump the data into Excel and use a Pivot Table. In this case you could "page" by Month.Year and have a reasonable 31 (30, 29, 28) column-wide report for each month.

Try this:
select SECTION_CODE, GROUP_CODE, MACHINE_CODE ,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'15', MATPROC , NULL )) DEC15,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'16', MATPROC , NULL )) DEC16,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'17', MATPROC , NULL )) DEC17,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'18', MATPROC , NULL )) DEC18,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'19', MATPROC , NULL )) DEC19,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'20', MATPROC , NULL )) DEC20,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'21', MATPROC , NULL )) DEC21,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'22', MATPROC , NULL )) DEC22,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'23', MATPROC , NULL )) DEC23,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'24', MATPROC , NULL )) DEC24,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'25', MATPROC , NULL )) DEC25,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'26', MATPROC , NULL )) DEC26,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'27', MATPROC , NULL )) DEC27,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'28', MATPROC , NULL )) DEC28,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'29', MATPROC , NULL )) DEC29,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'30', MATPROC , NULL )) DEC30,
SUM( decode( TO_CHAR(JOBDATE,'DD'),'31', MATPROC , NULL )) DEC31
from t10 WHERE TO_CHAR(JOBDATE, 'YYYYMM') = '200312'
group by SECTION_CODE, GROUP_CODE, MACHINE_CODE

You get the idea how cross tab report works.

Then modify it according to your need .

Tamil

Dear Tamilselvan !!

thanx for ur encouraging reply ... it really give me a move forward to think.. but stilll it didnt solve my problem.. the case is they want to select the date range at report runtime so u cant hardcode the months and dates and secondly .. they want to group by following feilds and date range selected at runtime ... but the thing is these are the column names and in a matrix report we need atleast Three columns to make matrix ...in my case there are only two columns that the date and the below mentioned fields

MATPROC NUMBER
WASTAGE NUMBER
WASTPER NUMBER
REGRIND NUMBER
COSTKG NUMBER
COSTPC NUMBER
OUTMAN NUMBER
REJECTION NUMBER
UTILI NUMBER
EFFICIENCY NUMBER

soo plz plz ... send any any comments ...which may help me
MNA

I gave a sample code for column MATPROC.

You can use union for other columns such as WASTAGE, WASTPER....

There are some restrictions when you use "cross tab report".

What will happen if the user gives a date range of 365 days?
Are you going to print across all the values in one line?

You can try CUBE / ROLLUP analytic functions.

Tamil