RE: Strange CL problem with SUBR

Well, it would be *less* unfortunate. As I said, some people may actually have code using this quirk -not stopping at a SUBR statement- intentionally, or eve unintentionally, so if IBM fixes is, IBM may actually break the code of that people.

I know it's an unlikely scenario, but I don't like seeing fixes that break existing code. That's one of the success secrets of the iSeries.

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Crispin Bates
Sent: Tuesday, January 29, 2008 12:40 PM
To: Midrange Systems Technical Discussion
Subject: Re: Strange CL problem with SUBR

Why is it unfortunate if they fix it? It is clearly a bug, and is not
working as documented.

Still, as Adam pointed out, there's already a PTF for it...


----- Original Message -----
From: "Luis Colorado" <LuisC@xxxxxxxxxxxxxxxx>
To: "Midrange Systems Technical Discussion" <midrange-l@xxxxxxxxxxxx>
Sent: Tuesday, January 29, 2008 12:34 PM
Subject: RE: Strange CL problem with SUBR

This is unfortunate. It's very easy to forget putting that GOTO or RETURN
before the subroutine. In many cases the developer may not even note the
problem right away. This is going to become a classic trap for
inexperienced developers.

This is unfortunate because if IBM fixes this bug, it may break existing
programs that depended on this quirk. And we know that IBM tries not to
break existing code.

So, unlucky if they fix it, and unlucky if they don't.