RE: File Exception/Error Subroutine - INFSR

[Dan <dbcerpg@xxxxxxxxx>]

Buck, the MONITOR op group sounds like a good idea, once we upgrade
from v4r5.  I'm not sure this group is ready for file I/O isolated in a
subroutine.  One step at a time.

What happens when a File Exception/Error Subroutine is defined for a
file and you check for %error right after a chain op?  Will the "local"
%error test override the INFSR?

TIA, Dan

--- Buck Calabro <Buck.Calabro@commsoft.net> wrote:
> >*every*, I/O operation involving a data
> >file is checked with %error.
> -snip-
> >I just would desire to "hide" the code in a
> >faraway routine
> -snip-
> >Does anyone have practical experience using INFSRs?
>
> Yes.  They are great for catching unexpected errors (like object
> damage),
> doing a post-mortem DUMP and exiting with an error code set so the
> caller
> can handle the fact that this program failed.  Great for commitment
> control.
>
> As you note, you can't resume at the point of failure.  For more
> information
> on that, check the MONITOR operation.  This very list had a thread on
> this
> topic not too long ago.
>
> Whatever you decide to do, I might argue that you should keep the I/O
> and
> it's error handling together.  To me, that means wrapping the I/O in
> either
> a subr call or a procedure call.  I prefer a subr because it is very
> clear
> to me that subrs deal with global variables (like file fields.)
>   --buck

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus – Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com