Re: Blank Suppress

[boldt@xxxxxxxxxx]

John Carr wrote:
>> Hmmm, let me try to explain how the trim BIFs work.  %TRIM
>> performs two tasks:  1) sloughing leading blanks; and 2)
>> sloughing trailing blanks.  %TRIML and %TRIMR each perform
>> only one of these two tasks.  It is obvious to me that if
>> task 1 takes n units of time and task 2 takes m units of
>> time, doing both would require n+m units of time.
>
>Hans
>
>What's your opinion of someone who worries about this type of thing,
>and at the same time does 1000 I/O's between enter keys??

Hmmm, I'd say that person was a computer programmer!
Heck, I'll bet most of us at some time worry too much
about the little details and miss the overall picture.

Bob's question did prompt us to do a bit of profiling,
and one of the results was a bit surprising.  I wrote
a program to compare performance of %TRIM(CH) vs
%TRIML(%TRIMR(CH)) vs %TRIMR(%TRIML(CH)).  The 2nd was
a bit slower than the 1st, but surprisingly, the
%TRIMR(%TRIML(CH)) performed very slightly faster than
%TRIM(CH), in spite of the fact that the W-Code was
almost identical!  In practice, however, the difference
is too small to worry about, and actually disappears
with full optimization.  In my test program, full
optimization made the trims run about 2.5 times faster.
(But as they say, your milage may vary!)

Anyways, perhaps that's todays lesson:  Try using
OPTIMIZE(*FULL).

Cheers!  Hans

Hans Boldt, ILE RPG Development, IBM Toronto Lab, boldt@ca.ibm.com


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* This is the RPG/400 Discussion Mailing List!  To submit a new         *
* message, send your mail to "RPG400-L@midrange.com".  To unsubscribe   *
* from this list send email to MAJORDOMO@midrange.com and specify       *
* 'unsubscribe RPG400-L' in the body of your message.  Questions should *
* be directed to the list owner / operator: david@midrange.com          *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *