I am trying to get the accumulated value of the column MontanteRecuperacao into montanterecuperacao_acumulada, by date and contract.
But the decimal datatype is becoming a problem and I don't know why, this works when the variables are of INT type.
This is the error message
Msg 4187, Level 16, State 1, Procedure spCalculaRecuperacao, Line 123
Data type decimal of receiving variable cannot store all values of the data type decimal of column 'montanterecuperacao_acumulada' without data loss.
Code:
DECLARE #MontanteRecuperacao_running DECIMAL,
#montanterecuperacao_acumulada_running DECIMAL,
#Contrato VARCHAR(10) = '' ;
WITH CTE AS
(
SELECT TOP 100 PERCENT
MontanteRecuperacao, montanterecuperacao_acumulada, ContratoId
FROM
MovimentoRecuperacao
ORDER BY
[ContratoId], [DtReferencia] DESC
)
UPDATE CTE
SET #montanterecuperacao_acumulada_running = montanterecuperacao_acumulada =
CASE
WHEN #Contrato <> ContratoId
THEN montanterecuperacao_acumulada
ELSE #montanterecuperacao_acumulada_running - #MontanteRecuperacao_running
END,
#MontanteRecuperacao_running = MontanteRecuperacao,
#Contrato = ContratoId
You've declared your DECIMAL parameters without specifying precision or scale:
#MontanteRecuperacao_running DECIMAL , #montanterecuperacao_acumulada_running DECIMAL
The default values are DECIMAL(18,0), which is essentially an integer. In order to get the behavior you desire, you'll need to assess your data for proper values to assign.
This post has the following code:
DECLARE #A DECIMAL(3, 0), #B DECIMAL(18, 0), #F FLOAT
SET #A = 3
SET #B = 3
SET #F = 3
SELECT 1 / #A * 3.0, 1 / #B * 3.0, 1 / #F * 3.0
SELECT 1 / #A * 3 , 1 / #B * 3 , 1 / #F * 3
Using float, the expression evaluates to 1. Using Decimal, the expression evaluates to some collection of 9s after the decimal point. Why does float yield the more accurate answer in this case? I thought that Decimal is more accurate / exact per Difference between numeric, float and decimal in SQL Server and Use Float or Decimal for Accounting Application Dollar Amount?
The decimal values that you have declared are fixed width, and there are no points after the decimal place. This affects the calculations.
SQL Server has a rather complex formula for how to calculate the precision of arithmetical expressions containing decimal numbers. The details are in the documentation. You also need to take into account that numeric constants are in decimal format, rather than numeric.
Also, in the end, you need to convert back to a decimal format with the precision that you want. In that case, you might discover that float and decimal are equivalent.
I'm using this query to round off the numbers and this round off the next value. Now I need to round off to the before value means if the value is 45.67 then the value should be 45. I tried these two queries and still I need to tweak the values.
Method1:
parsename('$' + convert(varchar,convert(money,round(sum(Column1 * Column2),0)),1),2)
Method2:
parsename('$' + convert(varchar,convert(money,floor(Column1 * Column2),0),1),2)
Really appreciate any suggestions.
The CEILING function returns the smallest integer greater than or equal to the specified numeric expression. The FLOOR function returns the largest integer less than or equal to the specified numeric expression. For example, in considering a numeric expression of 12.9273, CEILING returns 13 and FLOOR returns 12. The return value of both FLOOR and CEILING has the same data type as the input numeric expression.
SELECT CEILING(12.9273);
Here is the result set.
13
SELECT FLOOR(12.9273);
Here is the result set.
12
http://technet.microsoft.com/en-us/library/ms190927%28v=sql.105%29.aspx
To round down you can use FLOOR()
E.G.
DECLARE #number numeric(5,2)
SET #number = 45.67
SELECT FLOOR(#number)
You'd get the result 45
With your example, it looks like it's already working?
declare #number1 numeric(5,2)
declare #number2 numeric(5,2)
set #number1 = 1.23
set #number2 = 21.69
select parsename('$' + convert(varchar,convert(money,floor(#number1 * #number2),0),1),2)
select #number1 * #number2
Results
$26
26.6787
In SQL Server I declare a decimal variable, then set it from an equation. The decimal variable loses precision. However, if I just select the equation, the precision is intact. How do I set the decimal variable without losing precision?
SQL:
DECLARE #_oDiscount decimal(10,2)
SET #_oDiscount = CAST(9.99 AS decimal(10,2)) * CAST(.5 AS float)
SELECT #_oDiscount AS DecimalVariable, CAST(9.99 AS decimal(10,2)) * CAST(.5 AS float) AS Equation
OUTPUT:
DecimalVariable | Equation
-------------------------------
5.00 | 4.995
Well, YOU'VE defined the decimal to contain only 2 digits after the comma!
Therefore, the result of the calculation being 4.995, it will be rounded to 5.00.
If you would be using DECIMAL(12,4) instead, then there is no loss of precision!
DECLARE #_oDiscount decimal(10,4)
SET #_oDiscount = CAST(9.99 AS decimal(10,4)) * CAST(0.5 AS DECIMAL(10,4))
SELECT #_oDiscount
--> return 4.9950
Also: I would recommend against using FLOAT whenever possible!
And furthermore: there's really no need for all those casts..... just use
SET #_oDiscount = 9.99 * 0.5
and you'll get just the same results.
The DECIMAL(p, s) defines how precise the decimal value will be: p (precision) stands for the total number of digits, while s (scale) stands for the number of digits after the decimal point.
I'm trying to determine the best way to truncate or drop extra decimal places in SQL without rounding. For example:
declare #value decimal(18,2)
set #value = 123.456
This will automatically round #value to be 123.46, which is good in most cases. However, for this project, I don't need that. Is there a simple way to truncate the decimals I don't need? I know I can use the left() function and convert back to a decimal. Are there any other ways?
ROUND ( 123.456 , 2 , 1 )
When the third parameter != 0 it truncates rather than rounds.
Syntax
ROUND ( numeric_expression , length [ ,function ] )
Arguments
numeric_expression
Is an expression of the exact numeric or approximate numeric data
type category, except for the bit data type.
length
Is the precision to which numeric_expression is to be rounded. length must be an expression of type tinyint, smallint, or int. When length is a positive number, numeric_expression is rounded to the number of decimal positions specified by length. When length is a negative number, numeric_expression is rounded on the left side of the decimal point, as specified by length.
function
Is the type of operation to perform. function must be tinyint, smallint, or int. When function is omitted or has a value of 0 (default), numeric_expression is rounded. When a value other than 0 is specified, numeric_expression is truncated.
select round(123.456, 2, 1)
SELECT Cast(Round(123.456,2,1) as decimal(18,2))
Here's the way I was able to truncate and not round:
select 100.0019-(100.0019%.001)
returns 100.0010
And your example:
select 123.456-(123.456%.001)
returns 123.450
Now if you want to get rid of the ending zero, simply cast it:
select cast((123.456-(123.456%.001)) as decimal (18,2))
returns 123.45
Actually whatever the third parameter is, 0 or 1 or 2, it will not round your value.
CAST(ROUND(10.0055,2,0) AS NUMERIC(10,2))
Do you want the decimal or not?
If not, use
select ceiling(#value),floor(#value)
If you do it with 0 then do a round:
select round(#value,2)
Another truncate with no rounding solution and example.
Convert 71.950005666 to a single decimal place number (71.9)
1) 71.950005666 * 10.0 = 719.50005666
2) Floor(719.50005666) = 719.0
3) 719.0 / 10.0 = 71.9
select Floor(71.950005666 * 10.0) / 10.0
Round has an optional parameter
Select round(123.456, 2, 1) will = 123.45
Select round(123.456, 2, 0) will = 123.46
ROUND(number, decimals, operation)
number => Required. The number to be rounded
decimals => Required. The number of decimal places to round number to
operation => Optional. If 0, it rounds the result to the number of decimal. If another value than 0, it truncates the result to the number of decimals. Default value is 0
SELECT ROUND(235.415, 2, 1)
will give you 235.410
SELECT ROUND(235.415, 0, 1)
will give you 235.000
But now trimming0 you can use cast
SELECT CAST(ROUND(235.415, 0, 1) AS INT)
will give you 235
This will remove the decimal part of any number
SELECT ROUND(#val,0,1)
SELECT CAST(Value as Decimal(10,2)) FROM TABLE_NAME;
Would give you 2 values after the decimal point. (MS SQL SERVER)
Another way is ODBC TRUNCATE function:
DECLARE #value DECIMAL(18,3) =123.456;
SELECT #value AS val, {fn TRUNCATE(#value, 2)} AS result
LiveDemo
Output:
╔═════════╦═════════╗
║ val ║ result ║
╠═════════╬═════════╣
║ 123,456 ║ 123,450 ║
╚═════════╩═════════╝
Remark:
I recommend using built-in ROUND function with 3rd parameter set to 1.
I know this is pretty late but I don't see it as an answer and have been using this trick for years.
Simply subtract .005 from your value and use Round(#num,2).
Your example:
declare #num decimal(9,5) = 123.456
select round(#num-.005,2)
returns 123.45
It will automatically adjust the rounding to the correct value you are looking for.
By the way, are you recreating the program from the movie Office Space?
Try like this:
SELECT cast(round(123.456,2,1) as decimal(18,2))
If you desire to take some number like 89.0904987 and turn it into 89.09 by simply omitting the undesired decimal places, simply use the following:
select cast(yourColumnName as decimal(18,2))
The following screenshot is from W3Schools SQL Data Types section, which describes what decimal(18,2) is doing:
Therefore,
select cast(89.0904987 as decimal(18,2))
gives you: 89.09
Please try to use this code for converting 3 decimal values after a point into 2 decimal places:
declare #val decimal (8, 2)
select #val = 123.456
select #val = #val
select #val
The output is 123.46
I think you want only the decimal value,
in this case you can use the following:
declare #val decimal (8, 3)
SET #val = 123.456
SELECT #val - ROUND(#val,0,1)
I know this question is really old but nobody used sub-strings to round. This as advantage the ability to round really long numbers (limit of your string in SQL server which is usually 8000 characters):
SUBSTRING('123.456', 1, CHARINDEX('.', '123.456') + 2)
I think we can go much easier with simpler example solution found in Hackerrank:
Problem statement: Query the greatest value of the Northern Latitudes
(LAT_N) from STATION that is less than 137.2345. Truncate your answer
to 4 decimal places.
SELECT TRUNCATE(MAX(LAT_N),4)
FROM STATION
WHERE LAT_N < 137.23453;
Solution Above gives you idea how to simply make value limited to 4 decimal points. If you want to lower or upper the numbers after decimal, just change 4 to whatever you want.
Mod(x,1) is the easiest way I think.
select convert(int,#value)