Best practice to avoid code replication?

Question

currently I have to change a view, a big part of the view code looks like this:

CAST('My_Tag_' + CASE 
        WHEN FieldX isnull
            THEN ''
        WHEN FieldX = '123'
            THEN 'some_number'
        WHEN FieldX = 'abc'
            THEN 'some_text'
        ELSE 'strange' ) AS VARCHAR(128) AS myField
)

Just a chunk of code, that puts together a string (the code itself doesn't even matter right now, I have like 50 other examples, where I have a lot of code replication). Now I have exact the same code for 30 more fields in the view, just the 'My_Tag_' and FieldX is changing. If this would be C#, I would just write a little helper function. Of course I could write a function here, too. But as this is a bigger project with a lot of tables, views, etc, I would have hundreds of functions soon.

Now I am pretty new to SQL and normally my home is the OOP-world. But there has to be a solution to avoid code replication and to avoid having hundreds of helper functions in the database?

What's best practice in this case?

Answer 1

The best practice may be to create a user defined function.

The arguments would be the fields that change and it would return the intended value.

Answer 2

You can use a CTE to add a field to a table:

; with  TableWithExtraField as
        (
        select  case ... end as NewField
        ,       table1
        )
select   NewField
from     TableWithExtraField

Or a subquery also works:

select  NewField
from    (
        select  case ... end as NewField
        ,       table1
        ) as TableWithExtraField

Answer 3

CREATE FUNCTION dbo.MyTag(@myfield VARCHAR(MAX))
RETURNS VARCHAR(128)
AS
BEGIN
    RETURN CAST('My_Tag_' + CASE 
        WHEN @myfield IS NULL 
            THEN ''
        WHEN @myfield = '123'
            THEN 'some_number'
        WHEN @myfield = 'abc'
            THEN 'some_text'
        ELSE 'strange' END AS VARCHAR(128)) 
)
END