Sql : Based on column value do sum/substract - sql

I am having data in sqltable like below :
+----+-------+-------------+--------+
| Id | PayId | DeductionId | Amount |
+----+-------+-------------+--------+
| 1 | 1 | 0 | 100 |
| 2 | 2 | 0 | 250 |
| 1 | 0 | 3 | 50 |
| 2 | 0 | 4 | 75 |
+----+-------+-------------+--------+
So in output, need group by on Id and when PayId is non-zero then do sum of amount and when DedctionId is non-zero then do sum of amount and substract of those two values. So need Output like below :
+----+--------+
| Id | Amount |
+----+--------+
| 1 | 50 |
| 2 | 175 |
+----+--------+
How to do that?

Hopefully, both are never non-zero. You just want conditional aggregation:
select id,
sum(case when payid > 0 then amount
when deductionid > 0 then - amount
else 0
end) as amount
from t
group by id;

Use a CASE statement for the SUM:
select
Id,
sum(
case when payid <> 0 then amount else 0 end -
case when deductionid <> 0 then amount else 0 end
) Amount
from tablename
group by id
See the demo.
Results:
> Id | Amount
> -: | -----:
> 1 | 50
> 2 | 175

A simple case expression should work here for you.
select Id
, Amount = sum(case when PayID > 0 then Amount else Amount * -1 end)
from YourTable
group by Id

Related

how to get balance sheet (debit , credit , balance) from transactions table in SQL?

if I have transactions table like that:
+----+--------+------------+-------------+--------+
| id | userID | debitAccID | creditAccID | amount |
+----+--------+------------+-------------+--------+
| 1 | 1 | 1 | 2 | 500 |
| 2 | 1 | 1 | 3 | 600 |
| 3 | 1 | 3 | 1 | 200 |
+----+--------+------------+-------------+--------+
how what query to use to get a table for account with id 1 like that:
+----+--------+------------+-------------+--------+
| debit | credit |balance |
+----+--------+------------+-------------+--------+
| | 500 | | 500 |
| | 600 | | 1100 |
| | | 200| 900 |
+----+--------+------------+-------------+--------+
900
Assuming the id column shows the correct order of transactions, you can use case and window with the default of rows between unlimited preceding and current row to get your output:
select id, user_id,
case when user_id = debit_acc_id then amount else 0 end as debit,
case when user_id = credit_acc_id then amount else 0 end as credit,
sum(case when user_id = debit_acc_id then amount else 0 end) over w
- sum(case when user_id = credit_acc_id then amount else 0 end) over w as balance
from transactions
where user_id = 1
window w as (partition by user_id order by id)
order by user_id, id;
db<>fiddle here

Partition By - Sum all values Excluding Maximum Value

I have data as follows
+----+------+--------+
| ID | Code | Weight |
+----+------+--------+
| 1 | M | 200 |
| 1 | 2A | 50 |
| 1 | 2B | 50 |
| 2 | | 350 |
| 2 | M | 350 |
| 2 | 3A | 120 |
| 2 | 3B | 120 |
| 3 | 5A | 100 |
| 4 | | 200 |
| 4 | | 100 |
+----+------+--------+
For ID 1 the max weight is 200, I want to subtract sum of all weights from ID 1 except the max value that is 200.
There might be a case when there are 2 rows containing max values for same id. Example for ID 2 we have 2 rows containing max value i.e. 350 . In such scenario I want to sum all values except the max value. But I would mark weight 0 for 1 of the 2 rows containing max value. That row would be the one where Code is NULL/Blank.
Case where there is only 1 row for an ID the row would be kept as is.
Another scenario could be one where there is only row containing max weight but Code is NULL/Blank in such case we would simply do what we did for ID 1. Sum all values except max value and subtract from row containing max value.
Desired Output
+----+------+--------+---------------+
| ID | Code | Weight | Actual Weight |
+----+------+--------+---------------+
| 1 | M | 200 | 100 |
| 1 | 2A | 50 | 50 |
| 1 | 2B | 50 | 50 |
| 2 | | 350 | 0 |
| 2 | M | 350 | 110 |
| 2 | 3A | 120 | 120 |
| 2 | 3B | 120 | 120 |
| 3 | 5A | 100 | 100 |
| 4 | | 200 | 100 |
| 4 | | 100 | 100 |
+----+------+--------+---------------+
I want to create column Actual Weight as shown above. I can't find a way to apply partition by excluding max value and create column Actual Weight.
dense_rank() to identify the row with max weight, dr = 1 is rows with max weight
row_number() to differentiate the max weight row for Code = blank from others
with cte as
(
select *,
dr = dense_rank() over (partition by ID order by [Weight] desc),
rn = row_number() over (partition by ID order by [Weight] desc, Code desc)
from tbl
)
select *,
ActWeight = case when dr = 1 and rn <> 1
then 0
when dr = 1 and rn = 1
then [Weight]
- sum(case when dr <> 1 then [Weight] else 0 end) over (partition by ID)
else [Weight]
end
from cte
dbfiddle demo
Hmmm . . . I think you just want window functions and conditional logic:
select t.*,
(case when 1 = row_number() over (partition by id order by weight desc, (case when code <> '' then 2 else 1 end))
then weight - sum(case when weight <> max_weight then weight else 0 end) over (partition by id)
else weight
end) as actual_weight
from (select t.*,
max(weight) over (partition by id, code) as max_weight
from t
) t

Select maximum value from all previous rows (oracle sql)

I have a table that looks as follows:
personid opendate fraud
1 20190105 0
1 20190715 1
1 20191011 0
5 20181205 0
5 20190105 0
5 20190705 1
I want to get the maximum value of fraud up until each date grouped by personid. That is, I want to get the following table:
personid opendate fraud ever_fraud
1 20190105 0 0 <- no preceding frauds yet
1 20190715 1 1 <- fraud committed
1 20191011 0 1 <- fraud committed in previous credit
5 20181205 0 0
5 20190105 0 0
5 20190705 1 1
You can use window functions:
select
t.*,
max(fraud) over(partition by personid order by opendate) ever_fraud
from mytable t
Demo on DB Fiddle:
PERSONID | OPENDATE | FRAUD | EVER_FRAUD
-------: | -------: | ----: | ---------:
1 | 20190105 | 0 | 0
1 | 20190715 | 1 | 1
1 | 20191011 | 0 | 1
5 | 20181205 | 0 | 0
5 | 20190105 | 0 | 0
5 | 20190705 | 1 | 1
It seems like suggesting lag() analytic function :
with t2 as
(
select t.*,
lag(fraud,1,0) over (partition by personid order by opendate) as ever_fraud0
from t
)
select personid, opendate, fraud,
case when fraud = 1 then fraud else ever_fraud0 end as ever_fraud
from t2
Demo

SQL "Group" and "Count" categories

Edit. This is a follow up from another question. To simplify the question. Assume a table
date | id | type
01/01 | 1 | F
02/01 | 1 | F
02/01 | 1 | F
03/01 | 1 | S
03/01 | 1 | S
04/01 | 1 | F
04/01 | 1 | S
05/01 | 1 | S
I am looking for a way to summarise the above table by combination of transaction types per day. If a person (id) has only one transaction per day it counts as a Single type. If they have more than one it counts as a Multiple one. I've done that with my original query and it works. The output from the above table would be:
date | Single | Multiple
01/01 | 1 | 0
02/01 | 0 | 1
03/01 | 0 | 1
04/01 | 0 | 1
05/01 | 1 | 0
I got that far and it works. What's I'm struggling with (ie. don't have a clue of how to start) is how set up a query to show all possible combinations of Type (SS, FF, FS) instead of just counting the multiple transactions. The desired output would be like:
date | Single | # FF | # FS | # SS
01/01 | 1 | 0 | 0 | 0
02/01 | 0 | 1 | 0 | 0
03/01 | 0 | 0 | 0 | 1
04/01 | 0 | 0 | 1 | 0
05/01 | 1 | 0 | 0 | 0
Any constructive hints or ideas will be much appreciated.
this is assuming that you have max 2 types per date.
You can use the CASE WHEN statement with MIN() and MAX() to check for combination of FF, FS or SS
select [date],
case when count(*) = 1 then 1 else 0 end as Single,
case when count(*) >= 2
and min([type]) = 'F'
and max([type]) = 'F'
then 1
else 0
end as [# FF],
case when count(*) >= 2
and min([type]) = 'F'
and max([type]) = 'S'
then 1
else 0
end as [# FS],
case when count(*) >= 2
and min([type]) = 'S'
and max([type]) = 'S'
then 1
else 0
end as [# SS]
from yourtable
group by [date]
EDIT :
for more then 3 types, just change the count(*) = 2 to count(*) >= 2 as long as the type are either F or S

SQL column sum and difference

my table, I want to create three columns in single-column, reprinted now.
id | date | type | total
------ | ------ | ------ | -----
1 | 01.10.2016| Paypal | 50
2 | 03.10.2016| credit | 40
3 | 05.10.2016| Cash | 50
4 | 06.10.2016| payment| 100
5 | 07.10.2016| Cash | 20
6 | 15.10.2016| Skrill | 10
7 | 18.10.2016| payment| 20
8 | 19.10.2016| Paypal | 10
9 | 19.10.2016| payment| 20
10 | 22.10.2016| Cash | 40
11 | 23.10.2016| Skrill | 10
my table, I want to create three columns in single-column, reprinted now.
SELECT id,date,type,total
(select (
sum(case when type="Paypal" then total else 0 end)+
sum(case when type="credit" then total else 0 end))+
sum(case when type="Cash" then total else 0 end) ) as receiv,
(Select(
sum(case when type="payment" then total else 0 end)) AS payment,
(Select sum(receiv -payment) FROM totals t2
WHERE (t2.date <= t1.date) and (t2.id <= t1.id) order by t1.date) AS remainder
FROM totals t1
group by date, type
order by id,date
--
The following query for the sql code?
Type = "Paypal, credit, Cash" sums "receiv" sums and Type = "payment" sums will be added to the "remainder" column.
id | date | type | receiv| payment| remainder
------ | ------ | ------ | ------| ------ | ------
1 | 01.10.2016| Paypal | 50 | 0 | 50
2 | 03.10.2016| credit | 40 | 0 | 90
3 | 05.10.2016| Cash | 50 | 0 | 140
4 | 06.10.2016| payment| 0 | 100 | 40
5 | 07.10.2016| Cash | 20 | 0 | 60
6 | 15.10.2016| Skrill | 10 | 0 | 70
7 | 18.10.2016| payment| 0 | 20 | 50
8 | 19.10.2016| Paypal | 10 | 0 | 60
9 | 19.10.2016| payment| 0 | 20 | 40
10 | 22.10.2016| Cash | 40 | 0 | 80
11 | 23.10.2016| Skrill | 10 | 0 | 90
Running total is easier in other databases which have analytical functions. In MySQL, you can do this with a correlated sub-query.
select id,dt,type,
case when type <> 'payment' then total else 0 end receiv,
case when type = 'payment' then total else 0 end payment,
case when type <> 'payment' then total else 0 end
- case when type = 'payment' then total else 0 end
+ coalesce((select sum(case when type <> 'payment' then total else 0 end)
- sum(case when type = 'payment' then total else 0 end)
from yourtable where id < y.id),0)
from yourtable y
Sample Demo