Re: Supported Syntax in SQLRPGLE in V6R1. Changed? - Solved

Pete

Verrrrrrrrrrrry eenterrrrrrrresting!

I see that this parameter has been around since V5R3


RPG preprocessor options (RPGPPOPT)

Specifies if the ILE RPG compiler will be called to preprocess the source member before the SQL precompile is run. Preprocessing the SQL source member will allow some compiler directives to be handled before the SQL precompile. The preprocessed source will be placed in file QSQLPRE in QTEMP. This source will be used for the SQL precompile.

*_*NONE_*
The compiler is not called for preprocessing. **LVL1*
The compiler is called for preprocessing to expand /COPY and handle
the conditional compilation directives except the /INCLUDE directive. **LVL2*
The compiler will be called for preprocessing to expand /COPY and
/INCLUDE and handle the conditional compilation directives.


Pete Helgren wrote:

For now.

The solution was using the preprocessor option *LVL2 (RPGPPOPT(*LVL2)) on the CRTSQLRPGI command

Not sure why that option is needed. I was contacted by an IBM developer who asked me to try it and it fixed the problem. I am waiting for a reply on the "why" part. If necessary I'll change the command default but for now I'll just have to remember to change the option each time I compile.

Pete


Kurt Anderson wrote:

Sorry if it's already been said, but I'm pretty sure the issue you're
running into is actually a result of v5r4 not v6r1. In v5r4 the
top/down order of SQL within the program now matters. Also, at my old
job we got bit by using a procedure called SQLOpen which is now a
reserved word at v5r4 - note: bad idea to start a name with SQL*.
However there are SQL changes in 6.1 as well I believe, including the
use of local host variables.

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Pete Helgren
Sent: Friday, December 05, 2008 10:59 AM
To: RPG programming on the IBM i / System i
Subject: Re: Supported Syntax in SQLRPGLE in V6R1. Changed?

I change the command defaults so I don't even have to remember anything (except to change the defaults on a new release!)

I just got this from Common:

COMMON Webcast - What's New in RPG for V6R1

So I guess I'll check it out to find out why my SQLRPGLE programs from V5R3M0 no longer compile in V6R1M0 (Barbara Morris should know if anyone

does!)

Pete


Tommy.Holden@xxxxxxxxxxxxxxxxxxxxx wrote:

the SET OPTION is not mandatory...but is much better than trying to remember and manually key them in on the compile options 8^)

Thanks,
Tommy Holden



From:
Pete Helgren <Pete@xxxxxxxxxx>
To:
RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>
Date:
12/05/2008 10:10 AM
Subject:
Re: Supported Syntax in SQLRPGLE in V6R1. Changed?
Sent by:
rpg400-l-bounces@xxxxxxxxxxxx



Even if none were ever used before? That is, is the Set Option now a

required entry in an SQLRPGLE program? I usually set the options at compile time.

Pete


Helge Bichel wrote:

The SQL Set Option has to appear first in the source when on V5R4.

exec Sql Set Option Commit = *NONE, CloSQLCsr = *ENDMOD;
Brgds
Helge


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [

mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Vern Hamberg

Sent: Friday, December 05, 2008 5:38 AM
To: RPG programming on the IBM i / System i
Subject: Re: Supported Syntax in SQLRPGLE in V6R1. Changed?

Pete

There is something about stuff that has to appear first in source now

-

might have started in V5R4 - I hope this tickles someone's memory.

But

it is like DECLARE CURSOR and such need to be at the top of the

source.

I dunno!

Vern

Pete Helgren wrote:

Well, I don't get it. Seems like there is something different about

the

way host variables are defined and used since V5R3M0. I went back

to

using the 270 I have at V5R3M0 so I could get this fix compiled and

out

to the customer but it is awfully strange that an SQLRPGLE source

member

from a V5R3M0 box can't be pulled over and compiled in V6R1M0.

Once I get this out to the customer, I'll call IBM. Seems like it

would

be something simple but after searching all afternoon on the 'Net, I

haven't found anything promising.

Pete


Brian Johnson wrote:

On Thu, Dec 4, 2008 at 3:17 PM, Pete Helgren <Pete@xxxxxxxxxx>

wrote:

Thanks Brian.

Do you know if host variable use changed in 6.1?

Sorry, I'm not there yet, still working at V5R3. The work-around

for

the

problem I'm thinking about was to use CRTSQLRPGI GENLVL(30) to

accept

the

'error' and allow the compile to continue. Perhaps your command

default was

changed at V5R3 but hasn't been propagated to the V6 box.