Cross tab Report Query

Hello All,
I have the below code.Iam  trying to put together a cross tab report with the below query for
columns NSN,SPEEDEX_ID and row will be each month total.
Please help me  with modifying the below query so that I can integrate this into the cross tab report.
Should I calculate each month's total one by one or is there a way I can modify the query to use pivot to pull the data like pivot.Please help with pivot if that could be an option

WITH DUAL as
(
	select 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -24, 0) ,101)  as sdate, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -23, 0) ,101)  as sdate2, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -22, 0) ,101)  as sdate3, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -21, 0) ,101)  as sdate4, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -20, 0) ,101)  as sdate5, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -19, 0) ,101)  as sdate6, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -18, 0) ,101)  as sdate7, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -17, 0) ,101)  as sdate8, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -16, 0) ,101)  as sdate9, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -15, 0) ,101)  as sdate10, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -14, 0) ,101)  as sdate11, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -13, 0) ,101)  as sdate12, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -12, 0) ,101)  as sdate13, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -11, 0) ,101)  as sdate14, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -10, 0) ,101)  as sdate15, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -9, 0) ,101)  as sdate16, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -8, 0) ,101)  as sdate17, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -7, 0) ,101)  as sdate18, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -6, 0) ,101)  as sdate19, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -5, 0) ,101)  as sdate20, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -4, 0) ,101)  as sdate21, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -3, 0) ,101)  as sdate22, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -2, 0) ,101)  as sdate23, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -1, 0) ,101)  as sdate24, 
	CONVERT(varchar(10),DATEADD(s,-1,DATEADD(mm, DATEDIFF(m,0,GETDATE()),0)),101) as edate
),
--select *  from DUAL

ALL_NSN_INFO AS (
select 
nsn,
speedex_id,
sum(qty_required) as total_qty
from d_item1, dual
where date_create >= sdate and date_create <= edate
and status between '01' and '79'
and program in ('770','772','773','774','775','776','778','779')
group by nsn, speedex_id
),
--select * from ALL_NSN_INFO order by nsn, speedex_id

MONTH_1 AS (
select 
nsn,
sum(qty_required) as month1_total
from d_item1, dual
where date_create >= sdate and date_create < sdate2
and status between '01' and '79'
and program in ('770','772','773','774','775','776','778','779')
group by nsn
),
--select * from MONTH_1 order by nsn

MONTH_2 AS (
select 
nsn,
sum(qty_required) as month2_total
from d_item1, dual
where date_create >= sdate2 and date_create < sdate3
and status between '01' and '79'
and program in ('770','772','773','774','775','776','778','779')
group by nsn
)
--select * from MONTH_2 order by nsn

select al.nsn
, al.speedex_id
, m1.month1_total
,m2.month2_total
, al.total_qty AS TOTAL
,(al.total_qty/24) AS DEMAND
from ALL_NSN_INFO al
left outer join MONTH_1 m1 on al.nsn = m1.nsn
left outer join MONTH_2 m2 on al.nsn = m2.nsn
order by al.nsn

Open in new window

Star79Asked:
Who is Participating?
Advertise Here
 
mlmccConnect With a Mentor Commented:
WHy don't you let Crystal do the summarizing and just provide the raw data.

Try this query

select
nsn,
speedex_id,
sum(qty_required) as total_qty
from d_item1 where date_create >=  DateAdd(24, 'M', CurrentDate) and status between '01' and '79'
and program in ('770','772','773','774','775','776','778','779')

You can then use the cross tab in Crystal to create the appropriate columns and rows.
For the date columns just set the group to be by MONTH

mlmcc
0
 
mlmccCommented:
Are you building a report in Crystal?

Does that query produce the data you desire?

If so you should be able to use it as a COMMAND datasource in Crystal to produce a cross tab.

There may be an easier way to do it in Crystal.  What data are you trying to show?

mlmcc
0
 
Star79Author Commented:
I've requested that this question be deleted for the following reason:

Re-submitting another question
0
Keep up with what's happening at Experts Exchange!

Sign up to receive Decoded, a new monthly digest with product updates, feature release info, continuing education opportunities, and more.

Sign Up
 
Star79Author Commented:
The screen shot attached is the data from sql.
Iam trying to display this in a crystal report.I guess a cross tab will help.
I have generated data for month1, month2, but it needs to go on till month24.
The report sample is atatched.I have to provide all the 24 months data along with the total.
Not sure if I need to modify the sql or can I play around in the cross tab.
Please advise.
Capture.PNG
reportSample.PNG
0
 
Star79Author Commented:
Sorry I would like to continue with the question
0
 
mlmccCommented:
Requestor wants to continue the discussion

mlmcc
0
 
James0628Connect With a Mentor Commented:
You're trying to create a column for each month, so each record contains all 24 months.  That won't really work for a cross-tab, because a cross-tab uses groups (one, or more, groups for the rows, and one, or more, groups for the columns), and each record in a report can only appear once in any given group.  You would need each record to be included in all 24 columns (ie. in each group), and CR just doesn't work like that.

 Then again, if you have a separate column for each month, it doesn't seem like you need a cross-tab.  You could just put each month field (month1_total, month2_total, etc.) on the report.

 However, a better way to handle this might be to put each month in a separate record (I guess this may essentially be the "pivot" option you asked about).  Your DUAL CTE has a separate column for each month (sdate, sdate2, etc.).  You could put each month in a separate row instead.  Something like:
WITH DUAL as
(
	SELECT 1 AS MONTH_COUNT,
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -24, 0) ,101)  as sdate,
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -23, 0) ,101)  as edate
        UNION
        SELECT 2 AS MONTH_COUNT, 
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -23, 0) ,101)  as sdate,
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -22, 0) ,101)  as edate
        UNION
 <and so on>
        SELECT 24 AS MONTH_COUNT,
    CONVERT(Varchar(10),DATEADD(MONTH, DATEDIFF(MONTH, 0, GETDATE()) -1, 0) ,101)  as sdate,
    CONVERT(varchar(10),DATEADD(s,-1,DATEADD(mm, DATEDIFF(m,0,GETDATE()),0)),101) as edate
),

Open in new window

I added the MONTH_COUNT column just for reference.

 Then you'd use date_create to Join d_item1 with DUAL, to get a row for each month.  You could include sdate in the query results, and then the report could have a cross-tab, using sdate for the columns (grouped by day).  Or you could include date_create in the query results, and use that for the cross-tab columns (grouped by month).

 One big advantage to the "separate row for each month" approach is that you'd just have one CTE for the months (DUAL), which would be Joined with your main data, and that's it.  With your approach, you'd presumably end up with 24 CTE's (one for each month, like MONTH_1 and MONTH_2), all of which would have to be Joined with your main data.


 However, I'm not sure how ALL_NSN_INFO and speedex_id figure into this.  ALL_NSN_INFO is grouped by nsn and speedex_id, but the month CTE's are only grouped by nsn, and don't reference speedex_id at all.  Are ALL_NSN_INFO and the month CTE's grouping on two different levels?  Or is the relationship between nsn and speedex_id unique, so grouping by nsn is the same as grouping by nsn and speedex_id ?

 If the grouping is really the same, I think you could just do the summary in the report, and don't really need ALL_NSN_INFO at all.  If you use a cross-tab, it will give you a total for each row.  Or, using your original approach, you could have a formula add the month columns (month1_total, month2_total, etc.) together.

 James
0
 
James0628Commented:
That's a very good point.  The DUAL CTE may not be necessary at all.

 James
0
 
mlmccCommented:
The only reason I see for that is to ensure there is data for all months in the time frame.

mlmcc
0
 
James0628Commented:
Yeah.  I got the idea to put the months in separate rows, but I was so focused on the old code, it didn't occur to me that they may not need the CTE at all.

 James
0
 
Star79Author Commented:
Still trying various options.thanks for all the options
0
 
Star79Author Commented:
We are using a different reporting tool called Pentaho which can handle dynamic cross tabs.I used that to acieve it.
But were it crytal the cross tab would have worked too.Closing the question
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.