Question

The following two statements both have negligible execution times over a relatively large table. Subsequently I was wondering which would be deemed better practice? Personally I think that the ‘case’ (option 2) statement reads a lot easier to somebody who make look at this a year from now however option 1 seems a lot cleaner.

Option #1:

select *
from imaginarytable
where   1=1
    and (@section_name = 'ALL' or upper(section_name) = upper(@section_name))

Option #2:

select *
from imaginarytable
where   1=1
    and upper(section_name)=
        case when @section_name <> 'ALL' 
             then upper(@section_name
             else upper(section_name)
        end

Answer 1

Personally, I cannot imagine anyone finding the CASE version more clear. The first option is basic boolean logic, which every CS student learns and with which we all should find very comfortable.

Plus, I’m not even sure your 2 expressions are equivalent, in part because the CASE expression is basically incomprehensible.