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

Joe Pluta wrote:

I hate to argue this point, but the qualified BIF is absolutely clearer because it doesn't require me to read another line of code to figure out what it does. It's sort of a slam dunk. Any other argument is based on making life easier for the original programmer.

I have a customer file that's hierarchical in nature. There's a parent account and many sub-accounts, like Sears National (parent) and Sears Sheboygan, Sears Troy, Sears Saginaw... (sub-accounts) Each sub-account points up to the parent account.

That means I have two key lists:
chain parent_account custhist
chain sub_account custhist
or worse, one key list with an eval:
if parent<>0
acct#=parent
endif
chain acct# custhist

If I do a CHAIN at line 100, then at line 500 do a IF %FOUND(CUSTHIST), it's going to be difficult to immediately understand which key list was used to set that %FOUND. It's not possible to read that one line of code and know whether I'm working with parent accounts or sub-accounts. The file name alone isn't enough information.

Setting a boolean with a decent name is much clearer here than the bald file name on the BIF:
parent_account chain custhist
if %found()
parent_found=*on
else...
...
exsr processa

processa begsr (400 lines later)
if parent_found
...do some stuff based on the fact that this is a parent account