Re: High Speed Text Searching Algorythms...

Paradox wrote:
>
> As an interesting CS project, I'm looking into different text searching
> algorythms, everything from simple strncmp to things like
> Knuth-Morris-Pratt and Boyer-Moore algorythms. But I was wondering, I
> know company's like Google (and proably other high speed internet search
> engines) have their own implimentations of search engines. But it seems
> as though for text searching, those are extreamly fast, proably
> searching records in the billions in under 1 second. What kind of
> algorythms do systems like that use? Are there any places online that I
> can find descriptions of these algorythms?

Full-text search is a different matter than simply searching for a string in another string. Sites like Google and Northern Lights maintain a database of the words in the documents they have on file.

You can find sample programs on freshmeat.net by entering the terms "full text search database".

Newsgroups trimmed.

-- 
Steve Furlong, Computer Condottiere     Have GNU, will travel
   518-374-4720     sfurlong_at_acmenet.net