Question

Why can’t I do this?

EXEC mySproc NCHAR(0xA5)

I get

Incorrect syntax near ‘0xa5’.

yet, I can do this

DECLARE @foo NCHAR
SET @foo = NCHAR(0xA5)
EXEC mySproc @foo

and even this

SELECT NCHAR(0xA5)

It is interesting how SQL server chooses to evaluate expressions. Any thoughts?

Answer 1

Because it violates an T-SQL stored procedure call syntax which states:

Execute a stored procedure or function
[ { EXEC | EXECUTE } ]
    { 
      [ @return_status = ]
      { module_name [ ;number ] | @module_name_var } 
        [ [ @parameter = ] { value 
                           | @variable [ OUTPUT ] 
                           | [ DEFAULT ] 
                           }
        ]
      [ ,...n ]
      [ WITH RECOMPILE ]
    }
[;]

where value

Is the value of the parameter to pass to the module or pass-through
command. If parameter names are not specified, parameter values must
be supplied in the order defined in the module.

When executing pass-through commands against linked servers, the order
of the parameter values depends on the OLE DB provider of the linked
server. Most OLE DB providers bind values to parameters from left to
right.

If the value of a parameter is an object name, character string, or
qualified by a database name or schema name, the whole name must be
enclosed in single quotation marks. If the value of a parameter is a
keyword, the keyword must be enclosed in double quotation marks.

If a default is defined in the module, a user can execute the module
without specifying a parameter.

The default can also be NULL. Generally, the module definition
specifies the action that should be taken if a parameter value is
NULL.

thus

you should first perform all the calculations, place the results into variables and then pass the variables into SP call