99

I need to declare 12 decimal variables, corresponding to each month's year, with a cursor I sum values to this variables, then later I Update some sales information.

I don't know if sql server has this syntax

 Declare MonthsSale(1 to 12) as decimal(18,2)

This code works Ok. !

CREATE PROCEDURE [dbo].[proc_test]
AS
BEGIN

--SET NOCOUNT ON;

DECLARE @monthsales TABLE ( monthnr int,    amount decimal(18,2)    )


-- PUT YOUR OWN CODE HERE


-- THIS IS TEST CODE
-- 1 REPRESENTS JANUARY, ...
INSERT @monthsales (monthnr, amount) VALUES (1, 100)
INSERT @monthsales (monthnr, amount) VALUES (1, 100)

INSERT @monthsales (monthnr, amount) VALUES (2, 200)
INSERT @monthsales (monthnr, amount) VALUES (3, 300)
INSERT @monthsales (monthnr, amount) VALUES (4, 400)
INSERT @monthsales (monthnr, amount) VALUES (5, 500)
INSERT @monthsales (monthnr, amount) VALUES (6, 600)
INSERT @monthsales (monthnr, amount) VALUES (7, 700)
INSERT @monthsales (monthnr, amount) VALUES (8, 800)
INSERT @monthsales (monthnr, amount) VALUES (9, 900)
INSERT @monthsales (monthnr, amount) VALUES (10, 1000)
INSERT @monthsales (monthnr, amount) VALUES (11, 1100)
INSERT @monthsales (monthnr, amount) VALUES (12, 1200)


SELECT monthnr, SUM(amount) AS SUM_MONTH_1 FROM @monthsales WHERE monthnr = 1 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_2 FROM @monthsales WHERE monthnr = 2 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_3 FROM @monthsales WHERE monthnr = 3 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_4 FROM @monthsales WHERE monthnr = 4 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_5 FROM @monthsales WHERE monthnr = 5 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_6 FROM @monthsales WHERE monthnr = 6 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_7 FROM @monthsales WHERE monthnr = 7 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_8 FROM @monthsales WHERE monthnr = 8 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_9 FROM @monthsales WHERE monthnr = 9 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_10 FROM @monthsales WHERE monthnr = 10 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_11 FROM @monthsales WHERE monthnr = 11 GROUP BY monthnr
SELECT monthnr, SUM(amount) AS SUM_MONTH_12 FROM @monthsales WHERE monthnr = 12 GROUP BY monthnr

-- END TEST CODE
END
0

5 Answers 5

161

You could declare a table variable (Declaring a variable of type table):

declare @MonthsSale table(monthnr int)
insert into @MonthsSale (monthnr) values (1)
insert into @MonthsSale (monthnr) values (2)
....

You can add extra columns as you like:

declare @MonthsSale table(monthnr int, totalsales tinyint)

You can update the table variable like any other table:

update m
set m.TotalSales = sum(s.SalesValue)
from @MonthsSale m
left join Sales s on month(s.SalesDt) = m.MonthNr
1
  • I would like to use in-memory solution because I want to protect my SSD from excessive write access. Is it possible? Data amount in the array will be small, but it will be created very frequently.
    – Paul
    Commented Jan 18, 2023 at 22:44
28

Is there a reason why you aren't using a table variable and the aggregate SUM operator, instead of a cursor? SQL excels at set-oriented operations. 99.87% of the time that you find yourself using a cursor, there's a set-oriented alternative that's more efficient:

declare @MonthsSale table
(
MonthNumber int,
MonthName varchar(9),
MonthSale decimal(18,2)
)

insert into @MonthsSale
select
    1, 'January', 100.00
union select    
    2, 'February', 200.00
union select    
    3, 'March', 300.00
union select    
    4, 'April', 400.00
union select    
    5, 'May', 500.00
union select    
    6, 'June', 600.00
union select    
    7, 'July', 700.00
union select    
    8, 'August', 800.00
union select    
    9, 'September', 900.00
union select    
    10, 'October', 1000.00
union select    
    11, 'November', 1100.00
union select    
    12, 'December', 1200.00

select * from @MonthsSale   
select SUM(MonthSale) as [TotalSales] from @MonthsSale
2
  • 14
    Apparently in MSSQL2012 you can now insert in this format: VALUES(1, 'January', 100.00),(2, 'February', 200.00) - source: blog.sqlauthority.com/2012/10/27/…
    – andrewb
    Commented Sep 13, 2013 at 1:15
  • 3
    This feature totally escaped my notice; apparently that works in SQL 2008 as well.
    – Paul Smith
    Commented Sep 13, 2013 at 14:44
8

T-SQL doesn't support arrays that I'm aware of.

What's your table structure? You could probably design a query that does this instead:

select
month,
sum(sales)
from sales_table
group by month
order by month
1
  • Only as a side-comment, I would note the syntax T[n].v is a bit more concise than (select v from T where T.i = n). Actually, it's a lot more concise. I'd quite like to see T-SQL add it.
    – debater
    Commented Aug 7, 2015 at 13:21
3

Great question and great idea, but in SQL you'll need to do this:

For data type datetime, something like this-

declare @BeginDate    datetime = '1/1/2016',
        @EndDate      datetime = '12/1/2016'
create table #months (dates datetime)
declare @var datetime = @BeginDate
   while @var < dateadd(MONTH, +1, @EndDate)
   Begin
          insert into #months Values(@var)
          set @var = Dateadd(MONTH, +1, @var)
   end

If all you really want is numbers, do this-

create table #numbas (digit int)
declare @var int = 1        --your starting digit
    while @var <= 12        --your ending digit
    begin
        insert into #numbas Values(@var)
        set @var = @var +1
    end
0

You can declare an array using the VALUES keyword. Your example can be expressed succintly in the following form:

SELECT * FROM (VALUES 
    (1,100), 
    (2,200), 
    (3,300), 
    (4,400), 
    (5,500), 
    (6,600), 
    (7,700), 
    (8,800), 
    (9,900), 
    (10,1000), 
    (11,1100), 
    (12,1200)
) MonthSale (monthnr, Amount)

Feel free to INSERT it into a table or place a WHERE or GROUP condition for your purposes.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.