I have a table with records belonging to same person but the person was assigned with 2 different id's.
I need to combine the sales and then hold on to the id having highest sales.
For Example:
ID Name Sales
1 ABC 10
4 ABC 60
5 xyz 100
6 xyz 10
I need result as
ID Name Sales
4 ABC 70
5 XYZ 110
Please help me with a sql query for the above.
try this:
create table #mytable
(id int,
name nvarchar(20),
Sales int)
insert into #mytable
values
(1,'ABC',10),
(4,'ABC',60),
(5,'xyz',100),
(6,'xyz',10)
select (select top(1) ID
from #mytable r2
where r2.name=r1.name
and r2.Sales=MAX(r1.Sales))as ID,
name,
sum(Sales)
from #mytable r1
group by name
drop table #mytable
Related
SELECT max(salary),
(SELECT MAX(SALARY) FROM EMPLOYEE
WHERE SALARY NOT IN(SELECT MAX(SALARY) FROM EMPLOYEE)) as 2ND_MAX_SALARY;
This is giving me the error: FROM keyword not found where expected
You want the top 2 of your table ordered by one of the columns (the FETCH NEXT clause is available from Oracle 12c R1)
SELECT Salary FROM Employee ORDER BY Salary DESC LIMIT 2
FETCH NEXT 2 ROWS ONLY;
Use
SELECT Salary FROM Employee ORDER BY Salary DESC LIMIT 2
FETCH NEXT 2 ROWS WITH TIES;
if you want to return all employees that have the 1st or 2nd highest salary: There might only be one highest salary amount in the company, but more than one employee who gets that amount. Those rows are the ties.
If you're on Oracle database version lower than 12c, rank analytic function might help.
For sample rows:
SQL> select * from employee order by salary desc;
ENAME SALARY
---------- ----------
KING 5000 --> highest salary
FORD 3000 --> Ford and Scott "share" the 2nd
SCOTT 3000 --> highest salary
JONES 2975
BLAKE 2850
CLARK 2450
ALLEN 1600
TURNER 1500
MILLER 1300
WARD 1250
MARTIN 1250
ADAMS 1100
JAMES 950
SMITH 800
14 rows selected.
In a subquery (or a CTE, as I did), calculate rank for each salary and then, in the main query, select rows that rank as to top salaries:
SQL> with temp as
2 (select ename,
3 salary,
4 rank() over (order by salary desc) rnk
5 from employee
6 )
7 select ename, salary
8 from temp
9 where rnk <= 2
10 order by rnk desc;
ENAME SALARY
---------- ----------
SCOTT 3000
FORD 3000
KING 5000
SQL>
SELECT MAX(salary) AS max_salary,
(SELECT MAX(salary)
FROM employee
WHERE salary NOT IN (SELECT MAX(salary)
FROM employee
)
) AS 2nD_max_salary
FROM employee;
I have an 'order' column in a table in a postgres database that has a lot of missing numbers in the sequence. I am having a problem figuring out how to replace the numbers currently in the column, with new ones that are incremental (see examples).
What I have:
id order name
---------------
1 50 Anna
2 13 John
3 2 Bruce
4 5 David
What I want:
id order name
---------------
1 4 Anna
2 3 John
3 1 Bruce
4 2 David
The row containing the lowest order number in the old version of the column should get the new order number '1', the next after that should get '2' etc.
You can use the window function row_number() to calculate the new numbers. The result of that can be used in an update statement:
update the_table
set "order" = t.rn
from (
select id, row_number() over (order by "order") as rn
from the_table
) t
where t.id = the_table.id;
This assumes that id is the primary key of that table.
I have a table called _Invoice in SQL Server 2016 - like this:
Company InvoiceNo
-----------------
10 1
10 2
10 3
20 1
20 2
20 3
20 4
I want to get the highest value from all companies.
Like this:
Company InvoiceNo
-----------------
10 3
20 3
I want this data to then update another table that is called InvoiceSeries
where the InvoiceNo is higher than the NextNo in InvoiceSeries table
I am stuck with getting the highest data from InvoiceNo:
UPDATE InvoiceSeries
SET NextNo = -- Highest number from each company--
FROM InvoiceSeries ise
JOIN _Invoice i ON ise.InvoiceSeries = i.InvoiceSeries
WHERE i.InvoiceNo > ise.NextNo
Some example data:
Columns in InvoiceSeries Columns in _Invoices
Company NextNo Company InvoiceNo
10 9007 10 9008
20 1001 10 9009
10 9010
10 9011
10 9012
20 1002
20 1003
20 1004
If I understand correctly, you are looking for the HIGHEST common invoice number
Example
Select A.*
From YourTable A
Join (
Select Top 1 with ties
InvoiceNo
From YourTable
Group By InvoiceNo
Having count(Distinct Company) = (Select count(Distinct Company) From YourTable)
Order By InvoiceNo Desc
) B on A.InvoiceNo=B.InvoiceNo
Returns
Company InvoiceNo
10 3
20 3
EDIT - Updated for comment
Select company
,Invoice=max(invoiceno)
From YourTable
Group By company
This answer assumes there will be a record in the Invoice Series table.
--Insert Sample Data
CREATE TABLE #_Invoice (Company INT, InvoiceNo INT)
INSERT INTO #_Invoice(Company, InvoiceNo)
VALUES
(10 , 1),
(10 , 2),
(10 , 3),
(20 , 1),
(20 , 2),
(20 , 3),
(20 , 4)
CREATE TABLE #InvoiceSeries(Company INT, NextNo INT)
INSERT INTO #InvoiceSeries(Company, NextNo)
VALUES
(10, 1),
(20 ,1)
UPDATE s
SET NextNo = MaxInvoiceNo
FROM #InvoiceSeries s
INNER JOIN (
--Get the Max invoice number per company
SELECT Company, MAX(InvoiceNo) as MaxInvoiceNo
FROM #_Invoice
GROUP BY Company
) i on i.Company = s.Company
AND s.NextNo < i.MaxInvoiceNo --Only join to records where the 'nextno' is less than the max
--Confirm results
SELECT * FROM #InvoiceSeries
DROP TABLE #InvoiceSeries
DROP TABLE #_Invoice
I have one table
id employee leave_days leave_type type
1 ABC 10 sick remove
2 ABC 20 sick add
3 ABC 15 Annual remove
4 ABC 50 Annual add
5 XYZ 10 sick remove
6 XYZ 20 sick add
7 XYZ 15 Annual remove
8 XYZ 50 Annual add
From the above table I will group by the column name called leave_type and then I will merge rows and the output should be as follows.
I have to group by column name leave_type and add new column called leave_allocated . In the leave_allocated column, the column type with value add only will come.
id employee leave_days leave_type leave_allocated
1 ABC 10 sick 20
2 ABC 15 Annual 50
3 XYZ 10 sick 20
4 XYZ 15 Annual 50
I tried with sub query I could not match the inner query with outer query .
This should help
SELECT id,
employee,
leave_dates,
leave type,
(SELECT leave_days
FROM TABLE t2
WHERE t2.id = t1.id
AND t2.type = 'add'
) leave_allocated
FROM TABLE t1
WHERE t1.type = 'remove'
I'm using SQL Server 2012 and C#.
Imagine have something similar to a shopping cart and now need to create an order with the following items:
productA - 4 (qty)
productB - 1 (qty)
productC - 9 (qty)
In my C# code I have a list that looks like this:
id : "productA" , qty : "4"
id : "productB" , qty : "1"
id : "productV" , qty : "9"
Questions:
How can I pass the list of 2 values to the stored procedure?
How can I have the stored procedure run 3 while loops each one running 4 times, then once then 9 times in order to physically create one record x request?
Note: In my case I don't have a QTY column in the table, I need to specifically create one record x item on the order.
You can done this by Table Value Parameter in SQL.
Sql Authority
MSDN
You can done this by passing TVP as #table format
declare #table table(product varchar(10), qty int)
insert into #table
select 'product1', 4 union
select 'product2', 2
;WITH cte AS (
SELECT product, qty FROM #table
UNION ALL
SELECT product, qty-1 FROM cte WHERE qty > 1
)
SELECT t.product, t.qty
FROM cte c
JOIN #table t ON c.product = t.product
ORDER BY 1
Reference for the CTE : Creating duplicate records for a given table row
To pass a table into the stored procedure use table-valued parameter.
At first create a type:
CREATE TYPE [dbo].[ProductsTableType] AS TABLE(
[ID] [varchar](50) NOT NULL,
[qty] [int] NOT NULL
)
Then use this type in the stored procedure. The #ParamProducts is a table and can be used in all queries where a table can be used.
CREATE PROCEDURE [dbo].[AddProducts]
#ParamProducts ProductsTableType READONLY
AS
BEGIN
...
END
To actually insert required number of rows I would use a table of numbers , http://web.archive.org/web/20150411042510/http://sqlserver2000.databases.aspfaq.com/why-should-i-consider-using-an-auxiliary-numbers-table.html
In my database I have a table called Numbers with a column Number that contains numbers from 1 to 100,000. Once you have such table it is trivial to get the set that you need.
DECLARE #T TABLE (ID varchar(50), qty int);
INSERT INTO #T (ID, qty) VALUES ('productA', 4);
INSERT INTO #T (ID, qty) VALUES ('productB', 1);
INSERT INTO #T (ID, qty) VALUES ('productV', 9);
SELECT *
FROM
#T AS Products
INNER JOIN dbo.Numbers ON Products.qty >= dbo.Numbers.Number
;
Result set
ID qty Number
productA 4 1
productA 4 2
productA 4 3
productA 4 4
productB 1 1
productV 9 1
productV 9 2
productV 9 3
productV 9 4
productV 9 5
productV 9 6
productV 9 7
productV 9 8
productV 9 9
This is an example. In your case you would have this SELECT inside INSERT INTO YourFinalTable.