On Apr 14, 2004, at 9:50 PM, Chris Yavelow wrote:
I'm conducting an informal speed test comparing identical projects in SuperCard (4.1.2), Revolution (2.2), and RealBasic (5.5.1). The projects match 3,000 short phrases against an 81,920 word text. There has already been some feedback and these simple projects are now running even faster. Each one is only a dozen lines of code.
The way I do massive matching in massive data sets is to grab small overlapping chunks of the word text and do matching within that chunk, keeping track of the start char offsets so that I can calculate where in the word text the phrase resides. Then I grab the next chunk and repeat. The overlap is big enough to make sure that if a phrase crosses a boundary, in one of the chunks it will be the whole phrase. I find that breaking the word text into smaller chunks and doing all the searches on each chunk speeds things up. Sorry I don't have time to play with the example stack to run time tests to compare my technique with those currently proposed.
Kee Nethery
_______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
