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

[Buck <kc2hiz@xxxxxxxxx>]

Alan Shore wrote:

> and within each process there are other chains on other files
>
> QUESTION: for those people who do NOT qualify with file names, how would
> you handle this piece of logic?

Qualifying the %found seems reasonable here, but then so does matching 
record.  I don't think the word standard should be construed to imply an 
immutable constraint in our industry.  If I were forced to work for a 
place where we had an immutable Thou Shalt Never Qualify %Found() 
standard, I'd use %found() to set appropriately named booleans.

That might seem like extra work (not the typing, but the maintenance of 
more variables both in the code and in my brain as I work with this code 
next year) but reasonable names help a lot.  What I mean is that I would 
NOT name the boolean after the file:
chain key1 file1:
if %found
  file1_found=*on;
else
  file1_found=*off
endif

Rather, I would name the boolean after the nominal function I'm trying 
to perform.  I could be going off to CUSTHIST to see if the customer has 
any prior purchases at all:
if %found
  custHasPriorPurchases=*on

or I might be interested in the past 90 days:
if %found
  custHasPurchases90Day=*on

When I'm in the bowels of PROCESSA and I see
if %found(custhist)
  ...

I'm forced to go looking to see what key list was used to set %found. 
Was it the 90 day purchases, the customer umbrella group (i.e. any 
K-Mart store), a specific store (i.e. Sheboygan) or some other reason I 
went to CUSTHIST?  One hopes the context makes it clear, but sometimes a 
well named boolean can clarify that.
  --buck