I currently have some SQL code that is used to build a string that is a concatination of string values across multiple rows. The subqueries in the script sometimes return NULL values so I use the following statement to change the default behavior of the concatination operator which prevents my query from returning NULL:
SET CONCAT_NULL_YIELDS_NULL ON
Here's the code snippet:
select DISTINCT
(SELECT CASE WHEN (t1.MaskValue & HDR.TranTypeID)=1 THEN ' ' + t1.description ELSE '' END FROM transactiontypes t1 WHERE (t1.MaskValue & HDR.TranTypeID)=1) +
(SELECT CASE WHEN (t2.MaskValue & HDR.TranTypeID)=2 THEN ' ' + t2.description ELSE '' END FROM transactiontypes t2 WHERE (t2.MaskValue & HDR.TranTypeID)=2) +
(SELECT CASE WHEN (t3.MaskValue & HDR.TranTypeID)=4 THEN ' ' + t3.description ELSE '' END FROM transactiontypes t3 WHERE (t3.MaskValue & HDR.TranTypeID)=4) +
(SELECT CASE WHEN (t4.MaskValue & HDR.TranTypeID)=8 THEN ' ' + t4.description ELSE '' END FROM transactiontypes t4 WHERE (t4.MaskValue & HDR.TranTypeID)=8) +
(SELECT CASE WHEN (t5.MaskValue & HDR.TranTypeID)=16 THEN ' ' + t5.description ELSE '' END FROM transactiontypes t5 WHERE (t5.MaskValue & HDR.TranTypeID)=16)) as 'Transaction Type'
FROM HDResponse HDR
Here's the underlying table structure:
CREATE TABLE [dbo].[TransactionTypes](
[ID] [int] IDENTITY(1,1) NOT NULL,
[Description] [varchar](50) COLLATE SQL_Latin1_General_CP1_CI_AS NULL, [MaskValue] [int] NOT NULL) ON [PRIMARY]
CREATE TABLE [dbo].[HDResponse](
[ResponseID] [int] IDENTITY(1,1) NOT NULL,
[LoggedDateTime] [datetime] NULL,
[ResponseTypeTripID] [int] NULL,
[ResponseTypeID] [int] NULL,
[ResponseTypeObjectID] [int] NULL,
[ObjectID] [int] NULL,
[IDHolderID] [int] NULL,
[TransportCode] [int] NULL,
[CardID] [int] NULL,
[IssueCode] [smallint] NULL,
[EventDateTime] [datetime] NULL,
[Response] [bit] NULL,
[TranTypeID] [int] NULL)
ON [PRIMARY]
The problem I am having is I need to be able to use the query above in a view used for reporting. Unfortunately, you cannot use SET CONCAT_NULL_YIELDS_NULL ON in a view. This causes my query to return NULL if any of the subqueries return NULL. I could create a function to do something similar and reference the function in the query but I can't help but think there must be a way to get this done in a single query.
Any thoughts or ideas would be greatly appreciated.
Thanks!!!!!
What version of SQL Server are you using? SET CONCAT_NULL_YEILDS_NULL is ON by default unless you have changed the settings after connecting to SQL Server. This is true for any connections made via ODBC/OLEDB. You can use the PIVOT operator in SQL Server 2005 to solve this problem. You can also do this using a standard SQL technique like below:
select max(case p.trantype when 1 then p.description else '' end) +
max(case p.trantype when 2 then ' ' + p.description else '' end) +
max(case p.trantype when 4 then ' ' + p.description else '' end) +
max(case p.trantype when 8 then ' ' + p.description else '' end) +
max(case p.trantype when 16 then ' ' + p.description else '' end) as "Transaction Type"
from (
select t.description, (t.MaskValue & h.TranTypeID) as trantype
from HDResponse as h
join transactiontypes as t
on (t.MaskValue & h.TranTypeID) in (1, 2, 4, 8, 16)
) as p
No comments:
Post a Comment