Re: QCLSRC, QRPGSRC, etc... best practice?

[Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>]

Booth,

Hard wiring a file name flies in the face of everything IBM taught 
you?!  Huh?!?!

It's ALWAYS been standard practice to hard-code a filename on an 
F-spec.  (or DCL-F).

Its also standard to hard-code a table name (which is the same as a file 
name on IBM i) in SQL statements.  Every single example on planet earth 
does this.

I think you are confusing file names with library names, because I've 
never seen an application of any size or in any language on IBM i that 
doesn't hard-code file names throughout the application.   Library 
names, however, are another matter entirely.

So use /COPY CPYSRC,MEMBER

and *not* /COPY MYLIB/CPYSRC,MEMBER



On 10/29/2018 4:29 PM, Booth Martin wrote:
> Exactly my point.  Therefore why waste my time messing with the many 
> various schemes?
>
> I'll have just one development library and let the 3rd party use 
> whatever solution they prefer.   Hard-wiring /COPY's file name flies 
> in the face of every rule IBM has taught me. Until now.  My obligation 
> is to be sure I don't create a solution that requires one specific 
> hard-wired naming scheme.
>
> (hmmmm....  I wonder if **free changes the rule?)