Matchtext script results

[J. Landman Gay]

Robert Brenstein wrote:
>> And that is what surprised me -- that no tinkering with arrays, or 
>> matchtext, or anything else is faster than the most straightforward 
>> Revolution syntax. I was thinking this would take a long time, but in 
>> fact it is the fastest way to do it (that I've seen so far, anyway.) 
>> We've mentioned this on the list before, but I guess I need to be hit 
>> on the head with the facts occasionally, just to remind me how good 
>> we've got it.
>>
>> Surprise. Rev wins out again.
>>
>> -- 
>> Jacqueline Landman Gay         |     jacque at hyperactivesw.com
>> HyperActive Software           |     http://www.hyperactivesw.com
> 
> 
> I think that the time needed to prepare the text for searching should be 
> included since it is a required step and for large files can use enough 
> ticks to alter the results. Best would be to see both the total time and 
> search-only time.

If the processing had to be re-done for each file, then I included it in 
the timing. But there were a few scripts where a one-time variable had 
to be set up before any files were processed. I didn't think this was 
important; it was almost always only one line of script and it only 
happened once. I suppose it added a millisecond or two to the total.

I also didn't consider text preparation in my tests, because the text 
I'll be reading in won't need any cleanup, and the words I'll be looking 
for are stripped of punctuation before being passed to the scan. For 
general use,  you're probably right that we should include that.

I don't have time to redo all the tests (it took a while) but anyone who 
wants to could grab the scripts off the list and see what they get. It's 
always good to get more info, and I'm sure my results aren't definitive.

-- 
Jacqueline Landman Gay         |     jacque at hyperactivesw.com
HyperActive Software           |     http://www.hyperactivesw.com