Re: Sub procedure end specification

ditto. IMO it's just a good programming practice to do those things.


Thanks,
Tommy Holden



From:
Bryce Martin <BMartin@xxxxxxxxxxxx>
To:
RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>
Date:
05/28/2009 07:32 AM
Subject:
Re: Sub procedure end specification
Sent by:
rpg400-l-bounces@xxxxxxxxxxxx



David,
It seems that you have allowed your coding principals to be compromised.
Sure, you forget from time to time to change the name in the ending spec
when you copy from another procedure. I do this as well. But what you
should remember is that its still better coding practice to specify the
name of the procedure you are ending than to leave it blank. This also
goes for putting the names of your parameters in your procedure interface.

They don't mean a thing, except for someone trying to maintain the code.
I put my prototypes in copybooks, so having the names of the parameters in

the PI is a big help. Don't give up on the good fight. Keep your coding
principals, guard them. Once you start to make concessions, its all
downhill.




Thanks
Bryce Martin
Programmer/Analyst I
Ext. 4777



David FOXWELL <David.FOXWELL@xxxxxxxxx>
Sent by: rpg400-l-bounces@xxxxxxxxxxxx
05/28/2009 03:13 AM
Please respond to
RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>


To
RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>
cc

Subject
Sub procedure end specification






I used to suggest to my colleagues that they systematically put the name
of a sub procedure in the end specification instead of :

D E

Our prototypes our declared in the source of the module. When searching
for the name of a sub procedure in the source, you get a hit in a comment
line before the prototype, the PR declaration, the comment before the sub
procedure, the B declaration, the PI declaration, the call to the
subprocedure and maybe other comments, but : never in the end of the
subprocedure even though the subprocedure might be pages and pages of
code.

Well, this still hasn't caught on. So I've decided, if you can't beat'em,
join'em and I no longer put the name of the subprocedure in the end spec.
It makes for quicker compiling, as I always copy another sub procedure and

forget to change the name at the end spec.

My reasoning that the end spec name isn't necessary : if I can't see the
start spec on the same page (ok 2 pages for greenscreeners), then my sub
procedure is probably too dang long!!!!