Marc G. Fournier wrote:
> okay, first off ... what URL? I've been working on, and testing, archives
> all week, and altho getting things down below a 2min search appears to be
> near impossible (anyone know something *better* then mnogosearch that will
> handle >250k URLs and growing ... ? we tried ht/Dig before, and the
> resources it suck'd were outrageous) ...
(I assume we're talking about the email archives).
How about a quick and dirty script (perl, maybe) that grabs everything,
"intelligently" chucks the pseudo contents (stripping message quotes for
instance) into a pgsql table, and then see if tsearchv2 can be used to
search it via another quick and dirty CGI? Have to find some place to
host it though, the db might be huge.
Eating our own dog food and all that.
--
Linux homer 2.4.18-14 #1 Wed Sep 4 13:35:50 EDT 2002 i686 i686 i386
GNU/Linux
6:30pm up 253 days, 9:51, 5 users, load average: 5.00, 5.00, 5.00