Free-format SQL and mod-marks

PTF SI25533 fixes that problem.   The PTF for SEU syntax check
(qsqchks) will be PTFed soon.

Thanks,
Gina


date: Tue, 30 Jan 2007 11:55:53 -0500
from: rob@xxxxxxxxx
subject: Re: Free-format SQL and mod-marks

Had the same issue.  Don't know why I didn't open a pmr then.  Probably
because I was not a big fan of that mod markers and twas a coworker who
had the problem.  I just removed the mod markers.  I think WDSC made the
issue slightly easier to spot than SEU and tried to use that as a selling
aid to the stegosaurus.
Try opening a pmr.  At one time the gal from the precompiler team was on
the list but a pmr is your best bet.

Rob Berendt
--
Group Dekko Services, LLC
Dept 01.073
PO Box 2000
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





"Bruce Guetzkow" <bruceg@xxxxxxxxxxxxxxxxxxxxxxx>
Sent by: rpg400-l-bounces@xxxxxxxxxxxx
01/30/2007 11:41 AM
Please respond to
RPG programming on the AS400 / iSeries <rpg400-l@xxxxxxxxxxxx>


All:

For those of you using free-format SQL embedded in RPGLE, have you tried
putting mod-marks in columns 1-5 on the SQL lines?  I'm getting a
pre-compiler error on the following code:

V102S    exec sql
V102S      insert
V102S        into cltdebitpf
V102S      values (
V102S             :alphaclt,
V102S             :name,
V102S             :alphaaba,
V102S             :acctnbr,
V102S             :trancode,
V102S             :alphaamt
V102S             );

Here is the error message:
MSG ID  SEV  RECORD  TEXT
SQL0104  30    1004  Position 17 Token V102S was not valid. Valid tokens:
(
                     END GET SET CALL DROP FREE HOLD LOCK OPEN WITH ALTER
                     BEGIN.

Removing the mod-marks from all but the first line eliminates the problem,
as the following compiles cleanly:

V102S    exec sql
           insert
             into cltdebitpf
           values (
                  :alphaclt,
                  :name,
                  :alphaaba,
                  :acctnbr,
                  :trancode,
                  :alphaamt
                  );

SEU and WDSc are consistent with the pre-compiler...all indicate that the
first example is an error.  Does free-form SQL in RPGLE use columns 1-5?
No
other free-form statements do.  Is this documented somewhere?

I am on V5R4 and should be pretty current with PTFs.  Anyone have any
information on this?

Thanks,
--Bruce Guetzkow