@P.T.B
The average time needed is (testing gave me a break ...)
[c] LC 8.0.0-dp7
repeat-PMB 3.794
repeat-GC 1.475 <- fastest
repeat-hh 1165.0 <- :-( No difference to LC 7.1.0!
sort-GC 7.240
I have indeed data strings of the tested size (10-15 MByte).
Advanced statistical analysis without arrays is close to useless.
And LC > 6 has the useful option to split and combine by
delimiters of length > 1.
Is the library 'com.livecode.foreign' a target of hope?
p.s. Your HTML5-link from 'This Week in Livecode' forces me to
improve that test. What a trick! ;-)
@G.C.
Of course is a large smaller-than-set the worst case for the PMBrepeat.
But meanwhile I think, fully convinced by the tests with LC 7 and 8
(see above) that your 'direct' function greatestLessThanRepeat is the
way to go.
It works also great for smallestGreaterThan (without equality) or both:
-- Here v is the critical value and v1,v2 are starting values
-- for min and max, coming from a pre-knowledge about the data
-- or for example v1=-10^16, v2=10^16
function miniMaxGC L,v,v1,v2
put v1 into vMax; put v2 into vMin
repeat for each item i in pList
if i < v and i > vMax then put i into vMax
else if i > v and i < vMin then put i into vMin
end repeat
return ( vMax, v, vMin )
end miniMaxGC
[Perhaps a switch will be better? Or eliminate at first v by "replace"?]
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
