RE: Subroutines vs Subprocedures was RE: Indicators

Hi All,
I know that having unique host variable names within the program is
frustrating.  Especially when the SQL starts acting funny and the error
isn't obvious.  Before v5r3, the type of a duplicate name was not
dependable and may even affect data structures.  That is why the manual
states the names must be unique.   In v5r3, the precompiler changed the way
duplicate names are handled.  As some of you have seen, you are now getting
SQL0314.  Having QUALIFIED helps a little.  I also did make a small
exception, that if there is a stand alone defined exactly like another
stand alone, I won't error with name not unique.  If you look in the XREF,
you will see that it is only in there once.  Note that I said small
exception...it currently doesn't even support duplicate names defined with
the same LIKE.  I'm currently working on a v5r3 PTF that will expand the
exception to parameter names which are defined exactly the same and stand
alone/parameters that are defined with LIKE.

Gina


>date: Wed, 25 Aug 2004 10:36:48 -0400
>from: CWilt@xxxxxxxxxxxx
>subject: RE: Subroutines vs Subprocedures was RE: Indicators
>
>Well, like I said I end up relearning this all the time.
>
>But try having the global zip have different attributes.
>
>Also, try putting it into a DS.
>
>>From the manual:
>v5r2 - "The names of host variables must be unique within the program,
even
>if the host variables are in different procedures."
>
>v5r3 - "The names of host variables must be unique within the program,
even
>if the host variables are in different procedures. However, if a data
>structure has the QUALIFIED keyword, then the subfields of that data
>structure can have the same name as a subfield in a different data
structure
>or as a stand-alone variable. The subfield of a data structure with the
>QUALIFIED keyword must be referenced using the data structure name to
>qualify the subfield name."
>
>
>Charles