Can this be done faster ?

[Jim Ault]

I would use the following type of  loop

put the ticks into timer
set the itemdel to tab  --or what applies
filter databaseList without empty

repeat for each line txtLine in databaseList
    put doaction(txtLine) & cr after newListing
end repeat

filter newListing without empty
put the ticks - timer
sort newListing by item 4 of each




Jim Ault
Las Vegas

On 7/20/06 7:27 PM, "John Miller" <johnmiller1950 at sbcglobal.net> wrote:

> JB,
> 
> I once wrote an inventory control program for a company using
> Revolution.  They had many thousands of items that I read out of
> their database and then sorted and processed.
> 
> I found that using a repeat for that many items was extremely slow.
> The solution that I found which made things blazingly fast was to
> only process about 50 items at a time.
> 
> For instance:
> 
> 
> put 50 into XX
> put the ticks into timer
> repeat until listXX is empty
>    put line 1 to XX of listXX into listYY
>    delete line 1 to XX of listXX
>    repeat with x = 1 to (number of lines in listYY)
>       doaction to line x of listYY
>    end repeat
> end repeat
> put the ticks - timer
> 
> 
> Using the format above, you can experiment to see which value for XX
> is the most advantageous. I was able to whittle my time down from 45
> seconds to less than 1.
> 
> Good Luck
> John Miller
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution