At 20:21 01/07/2000 -0400, [EMAIL PROTECTED] wrote:
> And while I'm composing this email, I might as well also ask for
>recommendations on any good search engines that can a) index local text files
>b) do incremental updates c) possibly work with oracle if they are sql based
>d) a script will be processing the output, so no fancy HTML output needed
>with a solid perl interface. I prefer them to be free for commercial use (I
>am in somewhat of a hurry and might not be able to get the funds that quickly
I've recently come accross the DBIx::FullTextSearch module which I hadn't
seen before (maybe it wasn't on CPAN) and from my point of view it would be
more appropriate than what I had bumped into before for that kind of
operation: http://search.cpan.org/search?dist=DBIx-FullTextSearch. I've
only toyed around with it a little, but I've managed to get a search up and
running pretty fast. You'd still need to create a frontend indexer for your
HTML, but using HTML::Parser for that is a matter of ten lines of code
(you're only interested in the text parts). It's written to work with
MySQL, but if you stay away from some parts of it (eg the blobs stuff)
chances are that it might be quite easy to port, if it needs any porting.
-- robin b.
I'm not a complete idiot - some parts are missing!
