To Replace Number Values
Apr 6, 2006'000000' is a string with 6 charaters
Requirement:
I want to Replace 0 with 1.
Replace 0 to 1 at position 3 output= '001000'
Replace 0 to 1 at position 5 output= '000010'
ADVERTISEMENT
SQL Server 2012 :: Replace All Values In String With Values From Look Up Table
Mar 19, 2014I have a table that lists math Calculations with "User Friendly Names" that look like the following:
([Sales Units]*[AUR])
([Comp Sales Units]*[Comp AUR])
I need to replace all the "User Friendly Names" with "System Names" in the calculations, i.e., I need "Sales Units" to be replaced with "cSalesUnits", "AUR" replaced with "cAUR", "Comp Sales Units" with "cCompSalesUnits", and "Comp AUR" with "cCompAUR". (It isn't always as easy as removing spaces and added 'c' to the beginning of the string...)
The new formulas need to look like the following:
([cSalesUnits]*[cAUR])
([cCompSalesUnits]*[cCompAUR])
I have created a CTE of all the "Look-up" values, and have tried all kinds of joins, and other functions to achieve this, but so far nothing has quite worked.
How can I accomplish this?
Here is some SQL for set up. There are over 500 formulas that need updating with over 400 different "look up" possibilities, so hard coding something isn't really an option.
DECLARE @Synonyms TABLE
(
UserFriendlyName VARCHAR(128)
, SystemNames VARCHAR(128)
)
INSERT INTO @Synonyms
( UserFriendlyName, SystemNames )
[Code] .....
T-SQL (SS2K8) :: Count Number Of Values That Exist In A Row Based On Values From Array Of Numbers
Apr 16, 2014How to count the number of values that exist in a row based on the values from an array of numbers. Basically the the array of numbers I want to look for are in row 1 of table [test 1] and I want to search for them and count the "out of" in table [test 2]. Excuse me for not using the easiest way to convey my question below. I guess in short I have 10 numbers and like to find how many of those numbers exist in each row. short example:
Table Name: test1
Columns: m1 (int), m2 (int), m3 (int) >>> etc
Array/Row1: 1 2 3 4 5 6 7 8 9 10
------
Table Name: test2
Columns: n1 (int), n2 (int), n3 (int), n4 (int), n5 (int)
Row 1: 3, 8, 18, 77, 12
Row 2: 1, 4, 5, 7,18, 21
Row 3: 2, 4, 6, 8, 10
Answer: 2 out of 5
Answer: 4 out of 5
Answer: 5 out of 5
Replace Negative Values
Sep 13, 2004hi,
In the result of a function in my query, there are negative numbers.
How do I replace them with a 0 or is there a function like ISNULL that replaces the values that are negative?
thanks,
maarten
How To Replace NULL Values With 0
Apr 1, 2013I need to separate the data from one column in to two columns based on the transaction type TRAN_TYPE (C is credit, D is Debit)
TRAN-TYPE| AMOUNT
C 20.00
D 30.00
C 50.00
To do that I have this code:
select
CASE WHEN TRAN_TYPE = 'D'THEN CAST (ISNULL(amount, 0) as varchar (30)) end as DEBIT,
CASE WHEN TRAN_TYPE = 'c'THEN CAST (ISNULL(amount, 0) as varchar (30)) end as CREDIT
FROM HISTORY
And my output is:
DEBIT | CREDIT
----------------
NULL | 20.00
------|-------
30.00 | NULL
------|-------
NULL | 50.00
------|-------
how to replace the null values with 0
Find And Replace Values
Dec 7, 2007I was wondering how to do a find and replace with SQL? Would I use a SET statement?
I need to find a specific value and replace all of the results with a different value.
Thanks for all responses.
Replace The NULL Values
Dec 18, 2007Thank u Chirag....
but i should display all the records of that table not only price column.
consider the column price in titles table in pubs database. If the price of any record is not defined it should be retrieved as -9999.
Replace Values Of Null
Mar 4, 2008select distinct
case when item is null then replace(item,null,'-')
else itemend
end as item
from itemstable
i want to replace all null values to '-'...but still i m getting null values..
syntax is correct still not getting results..
can anyone help?
thanks.
How To Replace Values In Different Columns At The Same Time
May 14, 2004HI,
I AM HAVING A TABLE WHICH HAS INCREMENTAL COLUMNS,WHERE COLUMNS GETS ADDED EVERY MONTH TO THE TABLE AND THE TABLE THEN CONTAINS PREVIOUS MONTH AND PRESENT MONTH DATA ABOUT CUSTOMERS ,DETAILS AND TRANSACTIONS.
THE PROBLEM WITH THIS DATA IS ,IF THE CUSTOMER IS NEW ,THEN IN PREVIOUS MONTHS HIS INFORMATION IS NULL,WHICH HAVE TO BE CODED HAS "NOT PRESENT".
NOW,
HOW DO WE CONVERT ALL THE PREVIOUS COLUMNS FOR A PARTICULAR CUSTOMER HAS NULL AT THE SAME TIME ?.
HERE IS HOW THE PROC WRITTENED FOR IT GOES :-
DROP PROCEDURE DE_NAT
CREATE PROCEDURE DE_NAT
AS
BEGIN
DECLARE @MONMIN1 NVARCHAR(100),MON NVARCHAR(100),@YEAR NVARCHAR(100) , @MONYEAR NVARCHAR(100)
SET @MONMIN1 = DATENAME((MONTH),DATEADD(MONTH,-1,GETDATE()))
SET @MON = MONTH(GETDATE())
SET @YEAR = YEAR(GETDATE())
SET @MONYEAR = @MON + @YEAR
EXEC('select A.CUSTOMERS,B.*,CAST(A.RFM_40D AS FLOAT) AS R40
INTO TSD_' + @MONYEAR
+ ' from TSD_20 A
LEFT OUTER JOIN SD20 ' + @MONMIN1 + ' B
ON A.CUSTOMERS = B.CUSTOMER')
END
THIS PROC JUST ADDS THE PRESENT MONTHS DATA TILL LAST MONTHS DATA.
BUT IF A CUSTOMER IS NEW, THEN HOW DO I REPLACE THE NULL VALUES FOR THE PREVIOUS DATA TO 'NOT PRESENT'
FOR EG :- IF THERE IS A NEW CUSTOMER ,HOW DO WE CHANGE :-
CUSTOMERS ERTYYTRE RTYUUYTR TYUIIUYT QWERREWQ DFGHHGFD
----------- ---------- ---------- ---------- ----------- ----------
101023 <NULL> <NULL> <NULL> <NULL> 1.0
102022 1.0 1.62.3 3.4 4.5
NOW, AS YOU CAN SEE, THAT FOR CUSTOMERS = '101023'.
THE COLUMN DFGHHGFD IS, THIS MONTHS DATA , I WANT TO CHANGE ALL NULL VALUES PRESIDING IT AS "INACTIVE"
CAN I CHANGE , ALL COLUMNS FROM NULL TO "INACTIVE" , AT THE SAME TIME. ?
AS NEXT MONTH, AGAIN THE COLUMNS IS GONNA INCREASE WHICH WILL
AGAIN CAUSE A PROBLEM .
PLS TELL ME A METHOD , SO THAT I CAN DO THE NEEDFUL.
Transact SQL :: Replace Multiple Values Without Looping
Jul 13, 2015I need one query:
create table #task(TaskId bigint unique, Name varchar(2000))
insert into #task values(1, 'Text Text Text Text Text Text Text <<Name>> Â Text Text Text <<Salary>>')
insert into #task values(2, 'Text Text Text <<Name>> Text Text Text Text <<Company>> Â Text Text Text <<Salary>> Â Text Text Text')
[Code] ....
Now I need to create an inline function who resolve the task name with appropriate values and return me the resolved task name
select * from fn_TaskResolver(1, 'Text Text Text Text Text Text Text <<Name>> Â Text Text Text <<Salary>>')
I try this function but its return multiple rows as i just want to return one row. as I have big data set so i don't want to use scaler or Multi Line function.
create function fn_TaskResolver(@TaskId bigint, @name varchar(2000)
Return table
as
return
[Code] ....
Single String Replace Logic - 0 Needs To Be Ignored For Decimal Values
May 10, 2012The string column value looks like as below. Each value has a size of 15 withing a string
'2.2020 30 4.0000'
The column value should match with user input as below. The result should show equal when it is compared. Currently, it results not equal since it is a string comparision. The last digit '0' needs to be ignored for decimal values.
'2.202 30 4.0'
I need to handle the decimal values in such a way, if staring value with '.' and last digit is 0 then replace with space ''. So, it should look like
'2 2 2 30 4 ' = '2 2 2 30 4 '
When this string is compared, it results in EQUAL.
I tried the below logic, which even replaces the integer value like 30 to 3 and 3000 to 3 and results in equal which is incorrect.
RTRIM(REPLACE(REPLACE(RT1.rate,'''+@DOT+''','''+@S PACE+'''), '''+@ZERO+''', '''+@SPACE+''')) = '''+REPLACE(REPLACE(@Rate,'.',' '), '0', ' ')+''' '
Ex:'2.2020 300 4.00' = '2.20200 30 4.0'
After replace, string looks like
Ex:'2 2 2 3 4 ' = '2 2 2 3 4 '
It results as EQUAL which is incorrect. I need only decimal value to be replaced not integer.
I am looking for a single string replace logic.
SQL Server 2012 :: Replace Multiple Values Without Looping
Jul 13, 2015I need one query...
create table #task(TaskId bigint unique, Name varchar(2000))
insert into #task values(1, 'Text Text Text Text Text Text Text <<Name>> Text Text Text <<Salary>>')
insert into #task values(2, 'Text Text Text <<Name>> Text Text Text Text <<Company>> Text Text Text <<Salary>> Text Text Text')
-- select * from #task
[Code] ....
Now I need to create an inline function who resolve the task name with appropriate values and return me the resolved task name
select * from fn_TaskResolver(1, 'Text Text Text Text Text Text Text <<Name>> Text Text Text <<Salary>>')
I try this function but its return multiple rows as i just want to return one row. as I have big data set so i don't want to use scaler or Multi Line function.
create function fn_TaskResolver(@TaskId bigint, @name varchar(2000)
Return table
as
return
(
with data as
[Code] ....
REPLACE Funtion For Multiple Values In SP/Reporting Services
Mar 4, 2008Running into a problem with trying to pull multiple values in a stored procedure.
Ran Profiler for trace on what variable is coming back as when running in Reporting Services.
Profiler shows this:
exec apGetCompanysByRep @Product = N'Absorbent Pads,Airedale Terrier'
Trying to run Replace function but not quite getting the syntaxt correct. The stored procedure will work but when running report will often get unclosed quotation marks.
What I have:
Set @Product=
Replace(@Product, '''' + ',' + @product + '''', (@Product + '''' + ',' + ' ' + '''' + @product + ''''))
This is probably a little too much coding but I've tried several ways.
In the SP the code is as follows:
CREATE PROCEDURE [dbo].[apGetCompanysByRep]
@Magazine varchar(6) = null
, @DirectsMagazine varchar(6) = null
, @Category varchar(50) = null
, @SubCategory varchar(50) = null
--, @FirstName varchar(30)
, @LastName varchar(30)= null
, @Product varchar(1100)
AS
declare @SQL varchar(2000)
Set @Product=Replace(@Product, '''' + ',' + @product + '''', (@Product + '''' + ',' + ' ' + '''' + @product + ''''))
set @SQL = '
Thanks for any assistance anyone can give.
T-SQL (SS2K8) :: Replace Multiple Characters Based On Table Values
Dec 12, 2014There is a table [Formula_Calc] with formula calculations that need to be replaced with relevant values based on another table [Totals]
[Totals]
RowNo|Total
F1|240
F2|160
F3|180
F11|1000
F12|1500
F13|2000
For example we've got a row from [Formula_Calc] table 'F1+F3' as a string that needs to be transformed as 240+160=400
The below code works for the above example but if I pick 'F11+F3' instead , returns 2561 which comes from 2401+16.
Probably replaces F1 value instead of F11 and adds 1st digit (1) if I got it right ...
DECLARE @formula NVARCHAR(100);
DECLARE @Total NVARCHAR(100);
SET @formula = 'F11+F3';
SELECT @formula = REPLACE(@formula,RowNo,Total)
FROM [Totals]
SET @Total='select '+@formula
EXECUTE sp_executesql @Total;
PRINT @Total;
Updating Trigger Problem : Trying To Replace The Values From The Upadate Statement
Apr 25, 2007I am having an issue when trying to do something that looks simple on the face of it.. I want to replace the value during update in the same table.
Here's my situation
tst_update
ID int
Col1 varchar(10
I want to ensure that everytime Col1 is updated to "A" I want to set it to "X". Otherwise leave it as is.
Can someone help.
Thanks,
Ashish
SQL Server 2012 :: How To Replace Multiple Values In Single Select Statement
Aug 18, 2015how we can replace the multiple values in a single select statement? I have to build the output based on values stored in a table. Please see below the sample input and expected output.
DECLARE @V1 NVARCHAR(100)
SELECT @V1 = 'FirstName: @FN, LastName: @LN, Add1: @A1, Add2: @A2 '
DECLARE @T1 TABLE
(FN VARCHAR(100), LN VARCHAR(100), A1 VARCHAR(100), A2 VARCHAR(100))
[code]....
Add Symbol To Column Values And Convert Column Values To Western Number System
Feb 12, 2014I want to add $ symbol to column values and convert the column values to western number system
Column values
Dollar
4255
25454
467834
Expected Output:
$ 4,255
$ 25,454
$ 467,834
My Query:
select ID, MAX(Date) Date, SUM(Cost) Dollars, MAX(Funded) Funding from Application
COST is the int datatype and needs to be changed.
Integration Services :: Replace Blank Strings Values To NULL And Convert To Integer Data Type
Oct 5, 2015I need to convert a a string column to integer. Before converting, I need to check if it has blank values then convert it to NULL. Someone told me that its easier to convert it to NULL before converting to integer.
View 5 Replies View RelatedGetting The Number Of Records With Like Values
Jun 14, 2006I have a resultset that looks something like this:
Anzahl users_statdata_hobbies
---------------------
499 Andere
266 Essen
60 Essen,Andere
127 Essen,Musik
10 Essen,Musik,Party,Andere
30 Essen,Party
4 Essen,Party,Andere
51 Kunst
4 Kunst,Andere
13 Kunst,Essen
4 Kunst,Essen,Andere
I get this with this query which might be altered somehow:
SELECT COUNT(*) AS Anzahl, users_statdata_hobbies
FROM vgetAuswertung2
GROUP BY users_statdata_hobbies
ORDER BY users_statdata_hobbies
Of course this is not normalized but I can't change this.
Nevertheless I need to get the full number of each Hobby and not only the combination of them.
So instead or in addition to the existing recordset I need e.g
357 Essen which ist the sum of all records containing 'Essen' in the above example
The list of individual hobbies is defined therefor I could loop through the list manually and search for 'WHERE Hobbies LIKE '%ESSEN%' and count but since it's quiet a big resultset and there are several other similar tasks already I'm looking for a more performant way and I'm sure it could be done in SQL directly.
Any ideas someone?
Not A Number Values In The Database
Jul 28, 2007Our application has a table, which is populated via ADO.Net from C# with data originating from a C++ COM call. Today I encountered an entry that is C++ code for an undefined value: -1.#IND stored in the database. However, I could only discover what was stored in the table by Casting the value to a varchar -- simply selecting returned an error.
Is this expected behavior or a bug? It does not seem correct that SQL Server should store a value that cannot be displayed. In essence, either the value should not be allowed in the table because it violated the domain or SQL Server ought to have a way to display it with a Select *.
As fas as our application is concerned, we will be masking these values -- initially by ignoring them in the queries and eventually the loading program will convert to null.
Unknown Number Of Values
Mar 19, 2008I'm on SQL Server 2005 SP2.
I have the old age question of how to process a string parameter that is passed to a Stored Procedure that has an unknown number values. The example below has 5 values but it could be anywhere between 1 and 20.
I basically need to extract each value to Insert these values into the appropriate tables.
In the SQL 2000 days I use to do this with some T-SQL code that determines where the comma is and then I get the value and so on.....
I have read somewherethat this can be achieved using the XML Data Type.
Can someone show me that or atleast get me started on how to achiev this?
DECLARE @Range VARCHAR(200)
SET @Range = '10, 4, 8, 6, 22'
How Can I Have A Variable Number Of Parameter Values In A Dataset?
Apr 20, 2007I have a strongly typed dataset, and I need to be able to do a search on multiple values of a parameter. The problem is I don't know how many. I have a textbox that the user can enter search words in. The select string is built from the string of words that are entered, like this:For iCount = 0 To UBound(sArray) strSQL = strSQL & "Description LIKE '%" & sArray(iCount) & "%' OR "Next Can I do this is a dataset method? How? If I can't, what are my options?Diane
View 6 Replies View RelatedHow To Re Number Integer Field Values In A Table
Sep 22, 2014I am facing problem with re numbering of field records in a table.I am having one table with below records.
sid(identity/primarykey) stickyId(Integer) UserId(integer)
102 0 171
103 1 171
104 2 171
105 3 171
here how to renumbering stickyId values when deleted particular stickyId from UI. Here stickyId field is Integer type only. not primarykey/identity field.
Get ColorID Based On Number Of Values Passed
Dec 20, 2014I have a requirement regarding a color combination data. I have a lookup table that holds a colorid, p1, p2, p3, p4 to p8 which will be having colors Red, Green and Amber. P1 to P8 columns holds these three colors based on their combinations.
I have attached the look up table data for reference.
I need to pass the color values to p1 to p8 and need to retrieve the color id based on the passed color.
If we pass values for all p1 to p8 then it is easy to get the color code, however it will not happen. The passed values may be dynamic. ie we will not have all 8 values all the times. sometimes we will have 2 colors passed, sometimes 5 colors will be passed.
If i pass only two colors say red and red, i need the color id of only the row that has red and red for p1 and p2 alone. i dont want want all the colorid's that has red and red in p1 and p2 and some other colors in p3 to p4.
The exact colorid of the combination must be returned on passing the values to p1 and p2.
To elaborate; I am passing Red and Red as values to P1 and P2. In the look up table we can have 10 rows that has red and red i p1 and p2 like
colorid p1p2p3p4p5p6p7p8
1 redred
10 redredred
20 redredred
30 redredredred
40 redredredredred
50 redredredredredred
60 redredredredredredred
70 redredredredredredredred
So the result must have only the colorid 1 and not all the colorid's listed above. when I pass 3 red as values for p1, p2, p3 then the result must be 10. Colorid 1, 20, 30, 40, 50, 60 and 70 must not come in the result.
I need a function or procedure that will accept the arguments and provide me the result based on the values.
[URL]
T-SQL (SS2K8) :: Separating Values - Number And Alphabets
Jul 28, 2015Below are the same data
with Sample(Size) as (
select '16.3 Oz.' union all
'1' as union all
'2 Tablespoons' union all
'46. Oz. Each' )
Iwant to separate number and alphabets. But i wanted to keep the dot to have decimal values. Expected output
select '16.3' as val1 'Oz' as val2 union all
'1' val1 as union all
'2' as val1 'Tablespoons' as val2 union all
'46' as val1 'Oz Each' as val2
Please note that i need to remove the extra dots at the end of the the val1 and no dots in val2
I rid some of the functions like dbo.fn_StripCharacters and dbo.fn_GetAlphabetsOnly ffound in the internet. evey with my own logic. i couldn't remove the dot which appear at the end of val1
Count The Number Of Null Values In A Column
Feb 3, 2008Hi,
I have a table employee with 4 columns,
empno fname lname deptno
1 abc def 10
2 fff hhh 20
3 abc def NULL
4 abc def NULL
5 abc def 50
suppose i want to know the total number of null values in a particular column say deptno how shuld i write a query?
select count(deptno) from employee
where deptno IS NULL..
When i query this i get the result as 0..
Show Number Of Values As % Of Total Records..?
Feb 15, 2007HiI'm migrating from Access til MySQL.Works fine so far - but one thing is nearly killing me:I got the count of total records in a variabel - (antalRecords)I got the count for the Field Q1 where the value value is = 'nej'Now I just need to calculate how many % of my records have the value 'nej'I access this worked very fine - but with MySQL ( and ASP) I just cant getit right!!! I go crazy ....My code looks like this :strSQL="SELECT COUNT(Q1) AS Q1_nej FROM Tbl_evaluering " &_"WHERE Q1 = 'NEJ' "set RS = connection.Execute(strSQL)antal_nej = RS("Q1_nej")procent_nej = formatNumber((antal_nej),2)/antalrecords * 100Hope ...praying for help ...Please ;-)best wishes -Otto - Copenhagen
View 3 Replies View RelatedInsert Error: Column Name Or Number Of Supplied Values
Oct 12, 2007In SQL 2005 I created table as
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[logMsg](
[logMsgID] [int] IDENTITY(1,1) NOT FOR REPLICATION NOT NULL,
[msg] [nvarchar](256) COLLATE Latin1_General_CI_AS NOT NULL,
[AppId] [int] NULL,
CONSTRAINT [PK_logMsg] PRIMARY KEY CLUSTERED
(
[logMsgID] ASC
)WITH (IGNORE_DUP_KEY = OFF) ON [PRIMARY]
) ON [PRIMARY]
and trying to insert values with
INSERT INTO [ProxyDB].[dbo].[logMsg]
([msg]
,[AppId])
VALUES
('Text Test',1)
Getting error message:
Msg 213, Level 16, State 1, Procedure TrgInslogMsg, Line 14
Insert Error: Column name or number of supplied values does not match table definition.
Urgent help is required
Count Number Of Positive Dollar Values In Query?
Feb 3, 2012How do I count the number of positive dollar values in a query? I have various negative and positive dollar values and want to count how many positive I have then I will know how many are negative. I cannot put >0 or >0.00 in the where because it still returns everything cuz the programmer set the datatype to be money
View 4 Replies View RelatedTime Series: Number Of Previous Values Used In Model
Nov 8, 2006As the Microsoft Time Series algorithm implementation is based upon the Autoregressive Tree approach described in:
C. Meek, D. M. Chickering, D. Heckerman. Autoregressive Tree Models for Time-Series Analysis. In Proc. 2nd Intl. SIAM Conf. on Data Mining, 2002 (SDM-02). SIAM, pp. 229 €“ 244. http://www.siam.org/meetings/sdm02/proceedings/sdm02-14.pdf.
The model estimated is refererred to as an instance of "... autoregressive tree models of length p, denoted ART(p). An ART(p) model is an ART model in which each leaf node of the decision tree contains an AR(p) model, and the split variables for the decision tree are chosen from among the previous p variables in the time series..." (see the last paragraph of p. 2 of the paper).
What is the value of "p" used in the Microsoft Time Series implementation -- specifically, how many previous time series variables are used in estimating the model? It doesn't appear that this value can be specified in the algorithm parameters -- is that correct?
Thanks,
- Paul
NULL Values To Sequential Number (The Field Is Nvarchar Datatype)
Jun 16, 2012Ok I have upgraded my works database from a poorly designed Access database to a SQL database. The previous system allowed NULL values and duplicates to be inserted into a field that should NOT ALLOW NULL Values or duplicates. Therefore, this issue has now been moved across to my new system as I cannot set these constraints on the field that has multiple NULL values.
My solution would be to use a sequential operator, so whatever = NULL would be changed to a sequential number that us as administrators would know was a bogus number starting at something like = 999999900 counting up from that. There are only 250 records that would require updating.
To make things more interesting this field is not a integer type, its a Nvarchar type as its a Hardware ID. Both numerical and characters are require.
Insert Rows Based On Number Of Distinct Values In Another Table?
May 21, 2014I have a table with PO#,Days_to_travel, and Days_warehouse fields. I take the distinct Days_in_warehouse values in the table and insert them into a temp table. I want a script that will insert all of the values in the Days_in_warehouse field from the temp table into the Days_in_warehouse_batch row in table 1 by PO# duplicating the PO records until all of the POs have a record per distinct value.
Example:
Temp table: (Contains only one field with all distinct values in table 1)
Days_in_warehouse
20
30
40
Table 1 :
PO# Days_to_travel Days_in_warehouse Days_in_warehouse_batch
1 10 20
2 5 30
3 7 40
Updated Table 1:
PO# Days_to_travel Days_in_warehouse Days_in_warehouse_batch
1 10 20 20
1 10 20 30
1 10 20 40
2 5 30 20
2 5 30 30
2 5 30 40
3 7 40 20
3 7 40 30
3 7 40 40
how can I update Table 1 to get desired results?
SQL 2012 :: CheckSum Agg Function Returns 0 For Even Number Of Repeated Values?
Aug 14, 2014From what I've seen, the CheckSum_Agg function appears to returns 0 for even number of repeated values. If so, then what is the practical use of this function for implementing an aggregate checksum across a set of values?
For example, the following work as expected; it returns a non-zero checksum across (1) value or across (2) unequal values.
declare @t table ( ID int );
insert into @t ( ID ) values (-7077);
select checksum_agg( ID ) from @t;
-----------
-7077
declare @t table ( ID int );
insert into @t ( ID ) values (-7077), (-8112);
select checksum_agg( ID ) from @t;
-----------
1035
However, the function appears to returns 0 for an even number of repeated values.
declare @t table ( ID int );
insert into @t ( ID ) values (-7077), (-7077);
select checksum_agg( ID ) from @t;
-----------
0
It's not specific to -7077, for example:
declare @t table ( ID int );
insert into @t ( ID ) values (-997777), (-997777);
select checksum_agg( ID ) from @t;
-----------
0
What's curious is that (3) repeated equal values will return a checksum > 0.
declare @t table ( ID int );
insert into @t ( ID ) values (-997777), (-997777), (-997777);
select checksum_agg( ID ) from @t;
-----------
-997777
But a set of (4) repeated equal values will return 0 again.
declare @t table ( ID int );
insert into @t ( ID ) values (-997777), (-997777), (-997777), (-997777);
select checksum_agg( ID ) from @t;
-----------
0
Finally, a set of (2) uneuqal values repeated twice will return 0 again.
declare @t table ( ID int );
insert into @t ( ID ) values (-997777), (8112), (-997777), (8112);
select checksum_agg( ID ) from @t;
-----------
0