Re: CEE* FC handling (was RE: date math in CL)

[Barbara Morris <bmorris@xxxxxxxxxx>]

G Armour wrote:
> ...
> Don't have a lot of time to get into it now, but for those interested, it
> appears that the IBM docs on the FC are in the ILE Concepts manual,
> chapter "Exception and Condition Management", section "Condition
> Handling".  The short of it seems to be is that, if the first 4 bytes of
> the FC are hex zeros, then the CEE proc call was successful.  Other than
> that, I haven't had time to dig in further, so if others are so inclined,
> please report back here.  If someone has already invented this wheel,
> please let us know.
> ...

The feedback structure is also called a "condition token".  You can use
the CEEDCOD (Decompose a Condition Token) API to get the individual bits
from it.

Me, I think one of the greatest things about the iSeries is the joblog. 
I like the way you can see so much information about what happened if
you get an error.  With the feedback for the CEE APIs and the error-code
parameter for the other APIs, your error messages don't go to the joblog
unless you write code to put them there using the info in the
feedback/errcode.  If you don't write code to do that, any information
about the error is lost once your program ends.  At least with the
error-code parameter, you do get full information about the error; with
the feedback parameter, I don't think it's possible to get the
replacement text of the message, so even if you wanted to log your own
messages based on the feedback, you wouldn't necessarily be able to do
it fully.

Besides that, in general, I'd rather monitor for exceptions than check
if an error occurred.  If you use exception-based code, you don't have
to worry about forgetting an opportunity to check for errors.  If you
forget to check the feedback, you can have silent errors; forgetting to
code an exception handler gives you a nice noisy error.

So most of the time I code *OMIT for the feedback and bytesProvided = 0
for the errcode.

Just my own opinion.