I've this sp, which return a list of data, for each "month" (i.e. each row is a month). Somethings like that:
SELECT
*,
(CAST(t1.NumActivities AS DECIMAL) / t1.NumVisits) * 100 AS PercAccepted,
(CAST(t1.Accepted AS DECIMAL) / t1.Estimated) * 100 AS PercValue
FROM
(SELECT
MONTH(DateVisit) AS Month,
COUNT(*) AS NumVisits,
SUM(CASE WHEN DateActivity is not null THEN 1 ELSE 0 END) AS NumActivities,
SUM(Estimate) AS Estimated,
SUM(CASE WHEN DateActivity is not null THEN Estimate ELSE 0 END) AS Accepted
FROM [dbo].[Activities]
WHERE
DateVisit IS NOT NULL
AND (#year IS NULL OR YEAR(DateVisit) = #year)
AND (#clinicID IS NULL OR ClinicID = #clinicID)
GROUP BY MONTH(DateVisit)) t1
This is a result:
Month NumVisits NumActivities Estimated Accepted PercAccepted PercValue
1 5 1 13770.00 2520.00 20.00000000000 18.30065359477124
2 2 2 7900.00 7900.00 100.00000000000 100.00000000000000
3 1 0 2730.00 0.00 0.00000000000 0.00000000000000
8 1 1 3000.00 3000.00 100.00000000000 100.00000000000000
But as you can see, I could "miss" some Month (for example, here April "4" is missed).
Is it possible to insert, for the missing month/row, an empty (0) record? Such as:
Month NumVisits NumActivities Estimated Accepted PercAccepted PercValue
1 5 1 13770.00 2520.00 20.00000000000 18.30065359477124
2 2 2 7900.00 7900.00 100.00000000000 100.00000000000000
3 1 0 2730.00 0.00 0.00000000000 0.00000000000000
4 0 0 0 0 0 0
...
Here is a example with sample data:
CREATE TABLE #Report
(
Id INT,
Name nvarchar(max),
Percentage float
)
INSERT INTO #Report VALUES (1,'ONE',2.01)
INSERT INTO #Report VALUES (2,'TWO',3.01)
INSERT INTO #Report VALUES (5,'Five',5.01)
;WITH months(Month) AS
(
SELECT 1
UNION ALL
SELECT Month+1
FROM months
WHERE Month < 12
)
SELECT *
INTO #AllMonthsNumber
from months;
Your select query:
The left join will gives you the NULL for other months so just use ISNULL('ColumnName','String_to_replace')
\/\/\/\/
SELECT Month, ISNULL(Name,0), ISNULL(Percentage,0)
FROM AllMonthsNumber A
LEFT JOIN #Report B
ON A.Month = B.Id
EDIT:
Yes you can do it without creating AllMonthNumber Table:
You can use master..spt_values (found here) system table which contains the numbers so just with some where condition.
SELECT Number as Month, ISNULL(B.Name,0), ISNULL(Percentage,0)
FROM master..spt_values A
LEFT JOIN #Report B ON A.Number = B.Id
WHERE Type = 'P' AND number BETWEEN 1 AND 12
Related
We have a table with a parent child relationship, that represents a deep tree structure.
We are using a view with a CTE to query the data but the performance is poor (see code and execution plan below).
Is there any way we can improve the performance?
WITH cte (ParentJobTypeId, Id) AS
(
SELECT
Id, Id
FROM
dbo.JobTypes
UNION ALL
SELECT
e.Id, cte.Id
FROM
cte
INNER JOIN
dbo.JobTypes AS e ON e.ParentJobTypeId = cte.ParentJobTypeId
)
SELECT
ISNULL(Id, 0) AS ParentJobTypeId,
ISNULL(ParentJobTypeId, 0) AS Id
FROM
cte
A quick example of using the range keys. As I mentioned before, hierarchies were 127K points and some sections where 15 levels deep
The cte Builds, let's assume the hier results will be will be stored in a table (indexed as well)
Declare #Table table(ID int,ParentID int,[Status] varchar(50))
Insert #Table values
(1,101,'Pending'),
(2,101,'Complete'),
(3,101,'Complete'),
(4,102,'Complete'),
(101,null,null),
(102,null,null)
;With cteOH (ID,ParentID,Lvl,Seq)
as (
Select ID,ParentID,Lvl=1,cast(Format(ID,'000000') + '/' as varchar(500)) from #Table where ParentID is null
Union All
Select h.ID,h.ParentID,cteOH.Lvl+1,Seq=cast(cteOH.Seq + Format(h.ID,'000000') + '/' as varchar(500)) From #Table h INNER JOIN cteOH ON h.ParentID = cteOH.ID
),
cteR1 as (Select ID,Seq,R1=Row_Number() over (Order by Seq) From cteOH),
cteR2 as (Select A.ID,R2 = max(B.R1) From cteOH A Join cteR1 B on (B.Seq Like A.Seq+'%') Group By A.ID)
Select B.R1
,C.R2
,A.Lvl
,A.ID
,A.ParentID
Into #TempHier
From cteOH A
Join cteR1 B on (A.ID=B.ID)
Join cteR2 C on (A.ID=C.ID)
Select * from #TempHier
Select H.R1
,H.R2
,H.Lvl
,H.ID
,H.ParentID
,Total = count(*)
,Complete = sum(case when D.Status = 'Complete' then 1 else 0 end)
,Pending = sum(case when D.Status = 'Pending' then 1 else 0 end)
,PctCmpl = format(sum(case when D.Status = 'Complete' then 1.0 else 0.0 end)/count(*),'##0.00%')
From #TempHier H
Join (Select _R1=B.R1,A.* From #Table A Join #TempHier B on A.ID=B.ID) D on D._R1 between H.R1 and H.R2
Group By H.R1
,H.R2
,H.Lvl
,H.ID
,H.ParentID
Order By 1
Returns the hier in a #Temp table for now. Notice the R1 and R2, I call these the range keys. Data (without recursion) can be selected and aggregated via these keys
R1 R2 Lvl ID ParentID
1 4 1 101 NULL
2 2 2 1 101
3 3 2 2 101
4 4 2 3 101
5 6 1 102 NULL
6 6 2 4 102
VERY SIMPLE EXAMPLE: Illustrates the rolling the data up the hier.
R1 R2 Lvl ID ParentID Total Complete Pending PctCmpl
1 4 1 101 NULL 4 2 1 50.00%
2 2 2 1 101 1 0 1 0.00%
3 3 2 2 101 1 1 0 100.00%
4 4 2 3 101 1 1 0 100.00%
5 6 1 102 NULL 2 1 0 50.00%
6 6 2 4 102 1 1 0 100.00%
The real beauty of the the range keys, is if you know an ID, you know where it exists (all descendants and ancestors).
I have been given a query and trying to figure out a way to remove the cursor yet maintaining functionality, because the starting table can get into the millions of rows.
Example of data in table:
ID DollarValue Month RowNumber
1 $10 1/1/2014 1
1 $15 2/1/2014 2
1 -$40 3/1/2014 3
1 $50 4/1/2014 4
2 -$11 1/1/2014 1
2 $11 2/1/2014 2
2 $5 3/1/2014 3
Expected results:
ID DollarValue Month RowNumber TestVal
1 $10 1/1/2014 1 1
1 $15 2/1/2014 2 0
1 -$40 3/1/2014 3 -1
1 $50 4/1/2014 4 1
2 -$11 1/1/2014 1 -1
2 $11 2/1/2014 2 0
2 $5 3/1/2014 3 1
Here is the logic (pseudocode)that happens inside the cursor:
If a #ID <> #LastId AND #Month <> #LastMonth
Set #RunningTotal = #DollarValue
Set #LastMonth = '12/31/2099'
Set #LastID = #ID
Set #TestVal = Sign(#DollarValue)
Else
If Sign(#RunningTotal) = Sign(#RunningTotal + #DollarValue)
Set #TestVal = 0
Else
Set #TestVal = Sign(#DollarValue)
Set #RunningTotal = #RunningTotal + #DollarValue
Any idea how I can change this to set based?
You can use the windowed version of SUM to calculate running totals:
;WITH CTE AS (
SELECT ID, DollarValue, Month, RowNumber,
SUM ( DollarValue ) OVER (PARTITION BY ID ORDER BY RowNumber) as RunningTotal
FROM #mytable
)
SELECT C1.ID, C1.DollarValue, C1.Month, C1.RowNumber,
CASE WHEN C1.RowNumber = 1 THEN SIGN(C1.DollarValue)
WHEN SIGN(C1.RunningTotal) = SIGN(C2.RunningTotal) THEN 0
ELSE SIGN(C1.RunningTotal)
END AS TestVal
FROM CTE AS C1
LEFT JOIN CTE AS C2 ON C1.ID = C2.ID AND C1.RowNumber = C2.RowNumber + 1
Using LEFT JOIN on RowNumber you can get the previous record and compare the current running total with the previous one. Then use a simple CASE to apply rules pertinent to changes in SIGN of running total.
SQL FIDDLE Demo
P.S. It seems the above solution wont work in versions prior to SQL Server 2012. In this case the running total calculation inside the CTE has to be replaced by the "conventional" version.
This is 2008 solution
WITH CTE AS (
SELECT
AA.[ID]
,AA.[Month]
,AA.[RowNumber]
,AA.[DollarValue]
,SIGN(SUM(BB.[DollarValue])) AS RunTotalSign
FROM YourTable AS AA
LEFT JOIN YourTable AS BB
ON (AA.[ID] = BB.[ID] AND BB.[RowNumber] <= AA.[RowNumber])
GROUP BY AA.[ID],AA.[Month],AA.[DollarValue],AA.[RowNumber])
)
SELECT
AA.[ID]
,AA.[Month]
,AA.[RowNumber]
,AA.[DollarValue]
,CASE WHEN AA.RunTotalSign = CC.RunTotalSign Then 0
ELSE AA.RunTotalSign
END
AS TestVal
FROM CTE AS AA
LEFT JOIN CTE AS CC
ON (AA.[ID] = CC.[ID] AND AA.[RowNumber] = CC.[RowNumber]+1)
SELECT Column_A, SUM(CASE WHEN COLUMN_B = 'x' THEN 1 ELSE 'SET SUM TO 0' END) as total
FROM table
GROUP BY Column_A
ORDER BY Month
Is there a way to set the value of sum to 0 , I am trying to find the sum of last consecutive x value for cloumn_B if the consecution breaks I need to set the sum to 0
Table
ColumnA Column B Month
A x 1
A x 2
A x 3
A y 4
B x 1
B y 2
B x 3
B x 4
C x 1
C x 2
C y 3
C x 4
The expected result is:
A 0
B 2
C 1
EDIT:
DECLARE #TEMP TABLE (columnA varchar(50), maxMonth int )
INSERT INTO #TEMP (columnA, maxMonth)
SELECT columnA, MAX([Month]) FROM [table]
WHERE columnB != 'x'
GROUP BY columnA
SELECT original.columnA, SUM(CASE when original.[Month] > withMax.maxMonth THEN 1 ELSE 0 END) AS TOTAL
FROM #TEMP withMax JOIN [table] original ON withMax.columnA = original.columnA
WHERE original.columnB = 'x'
GROUP BY original.columnA
This query should work for you.
Just try this
SELECT Column_A, SUM(CASE WHEN COLUMN_B = 'x' THEN 1 ELSE 0 END) as total
FROM table
GROUP BY Column_A
ORDER BY some_date
SUM cannot be used for non-numeric data types
If you use, you will get error
Operand data type char is invalid for sum operator.
SELECT Column_A, ISNULL(CAST(SUM(CASE WHEN COLUMN_B = 'x' THEN 1 END) AS VARCHAR),'SET SUM TO 0') AS total
FROM table
GROUP BY Column_A
ORDER BY some_date
First of all my result looks like this:
KONTONR
Month
SELSKAPSKODE
BELOP
459611
1
BAGA
156000
459611
2
BAGA
73000
459611
4
BAGA
217000
459611
5
BAGA
136000
459611
1
CIVO
45896
459611
3
CIVO
32498
459611
4
CIVO
9841
330096
1
BAGA
42347
330096
3
BAGA
3695
I'm trying to show month 2 month bookings on several accounts, per account (KONTONR) there are several codes (SELSKAPSKODE) on which bookings are recorded (the sum of the bookings as BELOP). I would like to give an overview of the sum of the bookings (BELOP) per account (KONTONR) per month per code (SELSKAPSKODE). My problem is the codes don't show in a month if no bookings are made on that code. Is there a way to fix this? I understand why the codes don't show, since they're simply not in the table I'm querying. And I suspect that the solution is in making a 'fake' table which I then join (left outer join?) with 'another' table.
I just can't get it to work, I'm pretty new to SQL. Can someone please help?
My query looks like this (I only inserted the 'nested' query to make a set-up for a join, if this makes sense?!):
SELECT TOP (100) PERCENT KONTONR, Month, SELSKAPSKODE, BELOP
FROM (
SELECT SELSKAPSKODE, KONTONR, SKIPS_KODE, MONTH(POSTDATO) AS Month, SUM(BELOP) AS BELOP
FROM dbo.T99_DETALJ
WHERE (POSTDATO >= '2012-01-01') AND (BILAGSART = 0 OR BILAGSART = 2)
GROUP BY SELSKAPSKODE, KONTONR, SKIPS_KODE, MONTH(POSTDATO)
) AS T99_summary
GROUP BY KONTONR, SELSKAPSKODE, Month, BELOP
ORDER BY KONTONR, SELSKAPSKODE, Month
So concluding I would like to 'fill up' the missing months (see table at the start), for instance for account (KONTONR) 459611 month 3 is 'missing'. I would like to show month 3, with the sum of the bookings (BELOP) as '0'. Any help is greatly appreciated, thanks in advance!
You can query a table with the values 1-12 and left outer join your result.
Here is a sample using a table variable instead of your query and a CTE to build a table with numbers.
declare #T table
(
Month int
)
insert into #T values(1)
insert into #T values(1)
insert into #T values(1)
insert into #T values(3)
insert into #T values(3)
;with Months(Month) as
(
select 1
union all
select Month + 1
from Months
where Month < 12
)
select M.Month,
count(T.Month) Count,
isnull(sum(T.Month), 0) Sum
from Months as M
left outer join #T as T
on M.Month = T.Month
group by M.Month
Result:
Month Count Sum
----------- ----------- -----------
1 3 3
2 0 0
3 2 6
4 0 0
5 0 0
6 0 0
7 0 0
8 0 0
9 0 0
10 0 0
11 0 0
12 0 0
if you don't want to do all that you could also modify this: SUM(BELOP) with this:
Sum (case when BELOP is not null then 1 else 0 end)
You can also add in the year if you have a creation date for the interactions you are counting which may be helpful if your interactions span the course of many years.
with Months(Month) as
(
select 1
union all
select Month + 1
from Months
where Month < 12
)
select M.Month, year(CreatedOn) as Year,
count(amount) Count,
isnull(sum(amount), 0) Sum
from Months as M
left outer join Charge as C
on M.Month = (month(CreatedOn))
group by M.Month, year(CreatedOn) order by year(CreatedOn)
I want to create a report which aggregates the number of activities per customer per week.
If there has been no activites on that customer for a given week, 0 should be displayed (i.e week 3 and 4 in the sample below)
CUSTOMER | #ACTIVITIES | WEEKNUMBER
A | 4 | 1
A | 2 | 2
A | 0 | 3
A | 0 | 4
A | 1 | 5
B ...
C ...
The problem is that if there are no activities there is no data to report on and therefor week 3 and 4 in the sample below is not in the report.
What is the "best" way to solve this?
Try this:
DECLARE #YourTable table (CUSTOMER char(1), ACTIVITIES int, WEEKNUMBER int)
INSERT #YourTable VALUES ('A' , 4 , 1)
INSERT #YourTable VALUES ('A' , 2 , 2)
INSERT #YourTable VALUES ('A' , 0 , 3)
INSERT #YourTable VALUES ('A' , 0 , 4)
INSERT #YourTable VALUES ('A' , 1 , 5)
INSERT #YourTable VALUES ('B' , 5 , 3)
INSERT #YourTable VALUES ('C' , 2 , 4)
DECLARE #StartNumber int
,#EndNumber int
SELECT #StartNumber=1
,#EndNumber=5
;WITH AllNumbers AS
(
SELECT #StartNumber AS Number
UNION ALL
SELECT Number+1
FROM AllNumbers
WHERE Number<#EndNumber
)
, AllCustomers AS
(
SELECT DISTINCT CUSTOMER FROM #YourTable
)
SELECT
n.Number AS WEEKNUMBER, c.CUSTOMER, CASE WHEN y.Customer IS NULL THEN 0 ELSE y.ACTIVITIES END AS ACTIVITIES
FROM AllNumbers n
CROSS JOIN AllCustomers c
LEFT OUTER JOIN #YourTable y ON n.Number=y.WEEKNUMBER AND c.CUSTOMER=y.CUSTOMER
--OPTION (MAXRECURSION 500)
OUTPUT:
WEEKNUMBER CUSTOMER ACTIVITIES
----------- -------- -----------
1 A 4
1 B 0
1 C 0
2 A 2
2 B 0
2 C 0
3 A 0
3 B 5
3 C 0
4 A 0
4 B 0
4 C 2
5 A 1
5 B 0
5 C 0
(15 row(s) affected)
I use a CTE to build a Numbers table, but you could build a permanent one look at this question: What is the best way to create and populate a numbers table?. You could Write the Query without a CTE (same results as above):
SELECT
n.Number AS WEEKNUMBER, c.CUSTOMER, CASE WHEN y.Customer IS NULL THEN 0 ELSE y.ACTIVITIES END AS ACTIVITIES
FROM Numbers n
CROSS JOIN (SELECT DISTINCT
CUSTOMER
FROM #YourTable
) c
LEFT OUTER JOIN #YourTable y ON n.Number=y.WEEKNUMBER AND c.CUSTOMER=y.CUSTOMER
WHERE n.Number>=1 AND n.Number<=5
ORDER BY n.Number,c.CUSTOMER
Keep a table of time periods separately, and then outer left join the activities to it.
Like:
select *
from ReportingPeriod as p
left join Activities as a on a.ReportingPeriodId = p.ReportingPeriodId;