> Le 22 avr. 2017 à 13:16, Bruno LEGAY via 4D_Tech <[email protected]> a
> écrit :
>
> I haven't measured if EXECUTE METHOD has a significant overhead to be
> honest...
it depends ;-)
+++
ARRAY LONGINT($ms_al;2)
C_LONGINT($count_l;$duration_l;$tk_l)
$duration_l:=60*3
//**loop 1**
$tk_l:=Tickcount+$duration_l
$count_l:=0
Repeat
$count_l:=$count_l+1
$result:=TestMethod (...)
Until (Tickcount>=$tk_l)
$ms_al{1}:=$count_l
//**loop 2**
$tk_l:=Tickcount+$duration_l
$count_l:=0
Repeat
$count_l:=$count_l+1
EXECUTE METHOD("TestMethod";$result;...)
Until (Tickcount>=$tk_l)
$ms_al{2}:=$count_l
C_REAL($ratio_r)
$ratio_r:=$ms_al{1}/$ms_al{2}
ALERT(String($ms_al{1})+"\r"+String($ms_al{2})+"\r"+String($ratio_r))
+++
I tried with 3 different 'TestMethod'
a - a simple wrapper of 'Random' function
b - a home made random based on generate UUID + hexadecimal convert
c - explode a delimited text in a text array
Just to mention that from a to c, the called TestMethod is more and more time
consuming (in other words, regarding $count_l, a > b > c).
• interpreted
no difference, loop 1 & 2 are executed the same number of times.
• compiled
a - loop 1 executed 88 times more than loop 2
b - loop 1 executed 3 times more than loop 2
c - loop 1 ~ loop 2
In conclusion, the more time consuming the called method, the most negligible
EXECUTE METHOD.
--
Arnaud de Montard
**********************************************************************
4D Internet Users Group (4D iNUG)
FAQ: http://lists.4d.com/faqnug.html
Archive: http://lists.4d.com/archives.html
Options: http://lists.4d.com/mailman/options/4d_tech
Unsub: mailto:[email protected]
**********************************************************************
