RE: Free-Form vs Fixed Form.

Scott,

I find there is a subtle syntax checking in fixed-format EVAL and CALLP with
respect to grammar errors (parms, semicolons, etc.) but that's about it.
Which seems to be a little bit more than nothing in /free. 

-Bob Cozzi
www.RPGxTools.com
RPG xTools - Enjoy programming again.


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Scott Klement
Sent: Tuesday, January 03, 2006 12:32 PM
To: RPG programming on the AS400 / iSeries
Subject: Re: Free-Form vs Fixed Form.


> /free syntax checking in SEU is essentially non-existent in v5r2.  I'm not
> sure the syntax checker has *ever* flagged anything within a /free -
> /end-free block.  And, believe me, I had all kinds of newbie syntax errors
> early on, but had to rely on a compile listing *ERR scan to find them.

Have you found it to be better in WDSC?  I haven't.

In my experience, /FREE statements are pretty much always EVAL or CALLP 
statements.  Fixed-format statements don't validity check the "extended 
factor 2" portion of those statements, either.  The only real advantage as 
far as validity checking goes is when you're using other opcodes besides 
EVAL and CALLP... that's why it's "slightly better" for fixed format stuff 
vs. free format stuff...  because it's better for those opcodes besides 
EVAL and CALLP, which make only a small percentage of my programs.

> Not intending to beat *this* dead horse again, but, dang, if we could just
> get rid of the statement-ending semicolons!

We're not talking about how the language shoudl've been designed (in this 
thread)  This discussion is about the pros and cons of fixed vs. free. 
Can we please not go off on the tangent of telling IBM how to design a 
language that's already been designed and is already in production where 
it can't be changed?  Thanks.