RE: Standards question: What is your feeling about %found & %eofvs. %found(file) & %eof(file) ?

I actually came across a program that didn't specify the file in the BIF
and by all accounts, it should have worked because there wasn't any other
reads, except for chains... the loop got short-circuited somehow, all I do
know is that when I specified the file, the program worked as expected.


Michael Schutte
Admin Professional
Bob Evans Bob-B-Q: You haven't had barbeque until you've had Bob Evans
Bob-B-Q. Try our six Bob-B-Q dishes, starting at $5.99. For more
information, visit www.BobEvans.com.


rpg400-l-bounces@xxxxxxxxxxxx wrote on 07/02/2008 02:25:46 PM:

Greetings all... The topic is quite timely. I was just in a

conversation

this morning with a fellow programmer who insists that lack of use of the
file name can cause "unintended results".....even if you check

immediately

after the read/chain/setll.

Under normal practice I do not qualify the bifs for the simple reason of
cut-n-paste. Tests for valid reads ALWAYS occur immediately after a
read/chain/setll/etc. (Why continue processing if you don't have the
record.....beside the program has to do something because of the failure,
why not do it now?). Qualifying the bif means that when code is

copied,

not only does the read statement change, but the test using the bif has

to

change also.

I've only had one issue with the bif and that was chaining to an update
file. The program had a one time failure because either the "%found"

was

not qualified, or the "and not %error" was not specified. The program
tried to update a record that was never read apparently due to a record
lock. Both "fixes" were implemented so I don't have a definitive answer
as to which was the real problem. I tend to believe the real issue was
the lack of the "and not %error" portion of the test. I never could
reproduce the error any other way.

Has anyone had any real issues in not qualifying the bif with the file
name......other than cases where the I/O Opcode and the test were

separated

by more than one statement number?


--
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing

list

To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.