Decimals not being picked up in SQL case expression - tsql

I have a case expression and whole numbers are working fine but when it gets to the decimal it is returning null, so when #inputUnits = '1' and #unitInCase = 2 #setUnit = 'Two' but when #unitInCase = .1 #setUnit = NULL not sure what is happening.
DECLARE
#inputUnits varchar(50),
#unitInCase numeric,
#setUnit varchar(50)
SET #inputUnits = '1'
SET #unitInCase = .1
SET #setUnit =
CASE
WHEN #inputUnits = '1' THEN
CASE
WHEN #unitInCase= .1 THEN 'One'
WHEN #unitInCase= 2 THEN 'Two'
WHEN #unitInCase= 3 THEN 'Three'
WHEN #unitInCase = 4 THEN 'Four'
END
WHEN #inputUnits = '2' THEN
CASE
WHEN #unitInCase= .1 THEN 'One'
WHEN #unitInCase = .2 THEN 'Two'
WHEN #unitInCase= .3 THEN 'Three'
WHEN #unitInCase= .4 THEN 'Four'
END
END

You need to specify what type of numeric you need e.g., numeric(10,2) so your code should start with (say)
DECLARE
#inputUnits varchar(50),
#unitInCase numeric(10,2),
See https://learn.microsoft.com/en-us/sql/t-sql/data-types/decimal-and-numeric-transact-sql?view=sql-server-ver15 for info about numeric and decimal data types (which are synonymous). Broadly speaking, the first number represents the number of digits you wish to track (on both sides of the decimal point), whereas the second number is the number of digits after the decimal point.
See db<>fiddle with the code before and after.
In the example below, it shows what happens if you don't set the scale and precision.
DECLARE #x numeric;
SET #x = 1.234567890;
SELECT #x;
-- Result is 1

Related

Slow running query comparing dates

Could someone please offer some advice. I have the following query that is using roughly 200,000 records. I need to evaluate a 'DateTime' field to evaluate if the revenue occurs during the correct time slot. I am currently using CASE statements to evaluate the DateTime field and it is an absolute pig, it runs over 5 minutes. Is there a faster more efficient way to do this? Note the variables #cur_date, #end_date, #prev_yr_qtr_start, #cur_date_yr_prev etc are all strings and r.pw_ship_date is of type DATETIME. So in essence I'm comparing r.pw_ship_date to strings ie:'2017-01-01 00:00'
Note: it took 4:00 minutes to run this query when I added 'SELECT TOP(500)' for 200,000 records it would take forever.
Thanks in advance
DECLARE #total TABLE
(
acct_number VARCHAR(50),
pro_nbr VARCHAR(50),
sales_rep VARCHAR(50),
bill_to_name VARCHAR(50),
billing_addr1 VARCHAR(50),
billing_addr2 VARCHAR(50),
billing_city CHAR(50),
billing_state CHAR(2),
billing_zip CHAR(10),
cur_month_bills INT,
cur_month_rev DECIMAL(30, 6),
cur_qtr_bills INT,
cur_qtr_rev DECIMAL(30, 6),
prev_yr_qtr_bills INT,
prev_yr_qtr_rev DECIMAL(30, 6),
cur_ytd_bills INT,
cur_ytd_rev DECIMAL(30, 6),
prev_ytd_bills INT
)
INSERT INTO #total
SELECT TOP(50000) f.acct_number ,
r.pro_nbr ,
r.sales_rep ,
r.bill_to_name ,
r.billing_addr1 ,
r.billing_addr2 ,
r.billing_city ,
r.billing_state ,
r.billing_zip ,
'cur_month_bills' = MAX(( CASE WHEN r.pw_ship_date BETWEEN #cur_date AND #end_date THEN 1 ELSE 0 END )) ,
'cur_month_rev' = MAX(ROUND(( CASE WHEN r.pw_ship_date BETWEEN #cur_date AND #end_date THEN f.tot_revenue ELSE 0 END ), 2)) ,
'cur_qtr_bills' = MAX((CASE WHEN r.pw_ship_date BETWEEN #cur_date AND #end_date THEN 1 ELSE 0 END )) ,
'cur_qtr_rev' = MAX(ROUND(CASE WHEN r.pw_ship_date BETWEEN #cur_date AND #end_date THEN f.tot_revenue ELSE 0 END, 2)) ,
'prev_yr_qtr_bills' = MAX(CASE WHEN r.pw_ship_date BETWEEN #prev_yr_qtr_start AND #cur_date_yr_prev THEN 1 ELSE 0 END ) ,
'prev_yr_qtr_rev' = MAX(ROUND(CASE WHEN r.pw_ship_date BETWEEN #prev_yr_qtr_start AND #cur_date_yr_prev THEN f.tot_revenue ELSE 0 END , 2)) ,
'cur_ytd_bills' = MAX(CASE WHEN r.pw_ship_date BETWEEN #first_day_cur_yr AND #end_date THEN 1 ELSE 0 END ),
'cur_ytd_rev' = MAX(ROUND(CASE WHEN r.pw_ship_date BETWEEN #first_day_cur_yr AND #end_date THEN f.tot_revenue ELSE 0 END , 2)) ,
'prev_ytd_bills' = MAX(CASE WHEN r.pw_ship_date BETWEEN #first_day_prev_yr AND #end_date THEN 1 ELSE 0 END )
FROM #summed f
INNER JOIN #raw r ON f.acct_number = r.acct_number AND f.pro_nbr = r.pro_nbr
GROUP BY f.acct_number ,
r.pro_nbr ,
r.sales_rep ,
r.bill_to_name ,
r.billing_addr1 ,
r.billing_addr2 ,
r.billing_city ,
r.billing_state ,
r.billing_zip;
Change your table variables #raw and #summed to temporary tables. Table variables have no statistics and are extremely limited with regard to indexing (you can only have one). Because of this, SQL Server assumes that your table variables have only one row (2012 and older) or 100 rows (2014+). This means that you almost certainly are getting a bad execution plan for your query, and that's going to ruin you.
Once you've changed #raw and #summed into #raw and #summed, put an index on them - at a minimum, index your foreign keys (the fields you're joining on), acct_number and pro-nbr. It may be worth creating a clustered index and/or a primary key as well, but that's something you'll need to experiment with to find the performance you require.
The other thing that is killing your performance is comparing datetimes to strings. This is causing a type conversion and that can drag you down significantly. If you're working with a date/time, use the appropriate data type - not a string that looks like a date.
If this is still not running quickly enough, move your CASE statements out of your aggregate functions.
MAX(( CASE WHEN r.pw_ship_date BETWEEN #cur_date AND #end_date THEN 1 ELSE 0 END ))
Move the CASE statement into the query that populates #raw.pw_ship_date so that when you're performing the aggregate, you're just looking at integers all the way down.

How to extract letters from a string using Firebird SQL

I want to implement a stored procedure that extract letters from a varchar in firebird.
Example :
v_accountno' is of type varchar(50) and has the following values
accountno 1 - 000023208821
accountno 2 - 390026826850868140H
accountno 3 - 0700765001003267KAH
I want to extract the letters from v_accountno and output it in o_letter.
In my example: o_letter will store H for accountno 2 and KAH for accountno 3.
I tried the following stored procedure, which obviously won't work for accountno 3. (Please help).
CREATE OR ALTER PROCEDURE SP_EXTRACT_LETTER
returns (
o_letter varchar(50))
as
declare variable v_accountno varchar(50);
begin
v_accountno = '390026826850868140H';
if (not (:v_accountno similar to '[[:DIGIT:]]*')) then
begin
-- My SP won't work in for accountno 3 '0700765001003267KAH'
v_accountno = longsubstr(v_accountno, strlen(v_accountno), strlen(v_accountno));
o_letter = v_accountno;
end
suspend;
end
One solution would be to replace every digits with empty string like:
o_letter = REPLACE(v_accountno, '0', '')
o_letter = REPLACE(o_letter, '1', '')
o_letter = REPLACE(o_letter, '2', '')
...
Since Firebird 3, you can use substring for this, using its regex facility (using the similar clause):
substring(v_accountno similar '[[:digit:]]*#"[[:alpha:]]*#"' escape '#')
See also this dbfiddle.

Removing decimal places

I have a set of decimal values, but I need to remove the values after the decimal if they are zero.
17.00
23.50
100.00
512.79
become
17
23.50
100
512.79
Currently, I convert to a string and replace out the trailing .00 - Is there a better method?
REPLACE(CAST(amount as varchar(15)), '.00', '')
This sounds like it is purely a data presentation problem. As such you should let the receiving application or reporting software take care of the formatting.
You could try converting the .00s to datatype int. That would truncate the decimals. However, as all the values appear in one column they will have to have the same type. Making everything an int would ruin your rows with actual decimal places.
As a SQL solution to a presentation problem, I think what you have is OK.
I would advice you to compare the raw decimal value with itself floored. Example code:
declare
#Val1 decimal(9,2) = 17.00,
#Val2 decimal(9,2) = 23.50;
select
case when FLOOR ( #Val1 ) = #Val1
then cast( cast(#Val1 as int) as varchar)
else cast(#Val1 as varchar) end,
case when FLOOR ( #Val2 ) = #Val2
then cast( cast(#Val2 as int) as varchar)
else cast(#Val2 as varchar) end
-------------
17 | 23.50

tsql how to validate a number's scale

I need to validate the number of digits to the right of the decimal (the scale)
0, is a valid number in any of the places (tenths, hundredths, thousandths, etc.).
Any tips or tricks?... w/o an extensive regex library, and no built in function, I would prefer a function that accepts the number, the number of places the scale should equal, and then return a bit.
Following up with Maess's suggestion I came up with this:
CREATE FUNCTION [dbo].[GetScale]
(
#tsValue varchar(250)
, #tiScale int
)
RETURNS int
AS
BEGIN
DECLARE
#tiResult int
, #tiValueScale int
SET #tiResult = 0
SELECT #tiValueScale = LEN( SUBSTRING ( #tsValue, PATINDEX('%.%', #tsValue) + 1, LEN(#tsValue) ) )
IF (#tiValueScale = #tiScale)
SET #tiResult = 1
RETURN #tiResult
END
GO
Seems to work as desired. Thanks for the help.
Just as a followup... i ran into an issue where a number didnt have a decimal (which returns the patindex to 0) and the number was the same size as the scale, it would return a false positive... so i add an additional select from the patindex to determine if it does exist or not... it now looks like this:
- =============================================
ALTER FUNCTION [dbo].[GetScale]
(
#tsValue varchar(250)
, #tiScale int
)
RETURNS int
AS
BEGIN
DECLARE
#tiResult int
, #tiValueScale int
, #tiDecimalExists int
SET #tiResult = 0
SET #tiDecimalExists = 0
SELECT #tiDecimalExists = PATINDEX('%.%', #tsValue)
IF (#tiDecimalExists != 0)
BEGIN
SELECT #tiValueScale = LEN( SUBSTRING ( #tsValue, #tiDecimalExists + 1, LEN(#tsValue) ) )
IF (#tiValueScale = #tiScale)
SET #tiResult = 1
END
RETURN #tiResult
END
I tried Anthony's solution with some success, but there is some undesirable side effects when the first whole number is 9.
For example...
select 0.11 as fraction, Math.NumberOfDecimalPlaces(0.11) dp union
select 9.1, Math.NumberOfDecimalPlaces(9.1) union
select 9.01, Math.NumberOfDecimalPlaces(9.01) union
select 9.0, Math.NumberOfDecimalPlaces(9.0) union
select 99.0, Math.NumberOfDecimalPlaces(99.0) union
select 10999.0, Math.NumberOfDecimalPlaces(10999.0) union
select 8.0, Math.NumberOfDecimalPlaces(8.0) union
select 0, Math.NumberOfDecimalPlaces(0)
Produces...
0.00 0
0.11 2
8.00 0
9.00 -1
9.01 2
9.10 1
99.00 -2
10999.00 -3
Which shows some incorrect calculations when 9 is the first whole number.
I've made a small improvement to Anthony's original function.
CREATE FUNCTION [Math].[NumberOfDecimalPlaces]
(
#fraction decimal(38,19)
)
RETURNS INT
AS
BEGIN
RETURN FLOOR(LOG10(REVERSE(ABS(#fraction % 1) +1))) +1
END
This simply strips of the whole number part of the fraction. Which when implemented produces...
0.00 0
0.11 2
8.00 0
9.00 0
9.01 2
9.10 1
99.00 0
10999.00 0
The correct result
CREATE FUNCTION dbo.DecimalPlaces(#n decimal(38,19))
RETURNS int
AS
BEGIN
RETURN FLOOR(LOG10(REVERSE(ABS(#n % 1) + 1))) + 1
END
Edit (Feb 5 '15):
Thanks sqlconsumer. I have included your fix for nines before the decimal point.

How do I use T-SQL's Case/When?

I have a huge query which uses case/when often. Now I have this SQL here, which does not work.
(select case when xyz.something = 1
then
'SOMETEXT'
else
(select case when xyz.somethingelse = 1)
then
'SOMEOTHERTEXT'
end)
(select case when xyz.somethingelseagain = 2)
then
'SOMEOTHERTEXTGOESHERE'
end)
end) [ColumnName],
Whats causing trouble is xyz.somethingelseagain = 2, it says it could not bind that expression. xyz is some alias for a table which is joined further down in the query. Whats wrong here? Removing one of the 2 case/whens corrects that, but I need both of them, probably even more cases.
SELECT
CASE
WHEN xyz.something = 1 THEN 'SOMETEXT'
WHEN xyz.somethingelse = 1 THEN 'SOMEOTHERTEXT'
WHEN xyz.somethingelseagain = 2 THEN 'SOMEOTHERTEXTGOESHERE'
ELSE 'SOMETHING UNKNOWN'
END AS ColumnName;
As soon as a WHEN statement is true the break is implicit.
You will have to concider which WHEN Expression is the most likely to happen. If you put that WHEN at the end of a long list of WHEN statements, your sql is likely to be slower. So put it up front as the first.
More information here: break in case statement in T-SQL
declare #n int = 7,
#m int = 3;
select
case
when #n = 1 then
'SOMETEXT'
else
case
when #m = 1 then
'SOMEOTHERTEXT'
when #m = 2 then
'SOMEOTHERTEXTGOESHERE'
end
end as col1
-- n=1 => returns SOMETEXT regardless of #m
-- n=2 and m=1 => returns SOMEOTHERTEXT
-- n=2 and m=2 => returns SOMEOTHERTEXTGOESHERE
-- n=2 and m>2 => returns null (no else defined for inner case)
If logical test is against a single column then you could use something like
USE AdventureWorks2012;
GO
SELECT ProductNumber, Category =
CASE ProductLine
WHEN 'R' THEN 'Road'
WHEN 'M' THEN 'Mountain'
WHEN 'T' THEN 'Touring'
WHEN 'S' THEN 'Other sale items'
ELSE 'Not for sale'
END,
Name
FROM Production.Product
ORDER BY ProductNumber;
GO
More information - https://learn.microsoft.com/en-us/sql/t-sql/language-elements/case-transact-sql?view=sql-server-2017