Re: [ams@oryx.com: [ANNOUNCE] Archiveopteryx 1.10 released]

> -----Original Message-----
> From: pgsql-www-owner(at)postgresql(dot)org
> [mailto:pgsql-www-owner(at)postgresql(dot)org] On Behalf Of Joshua D. Drake
> Sent: 19 September 2006 23:12
> To: Jim C. Nasby
> Cc: pgsql-www(at)postgresql(dot)org
> Subject: Re: [pgsql-www] [ams(at)oryx(dot)com: [ANNOUNCE]
> Archiveopteryx 1.10 released]
>
> Jim C. Nasby wrote:
> > So... since we all *love* the current archive engine (and
> it's search)
> > so much...
> >
> > What about switching over to this? It appears it should
> work fine with
> > majordomo, and I'd bet we could get someone from
> Archiveopteryx to help
> > with this.
>
> It also has a Tsearch2 interface :)

Meaning what exactly? I'm sure any one of us is capable of writing a
small script that could catch the mail messages from each list, and
store them in a tsearch enabled database. The difficult bit is adding a
good *web* interface to it with all the appropriate thread links and
index pages, and without breaking all those pesky legacy URLs. There are
also tens of thousands of documents in the search index that aren't
emails that would also need to be handled.

If you want to move to a new tsearch search engine (which I'm not
against btw) then using our existing crawler script as a basis for a
generic indexer would bypass all those problems, leaving just a web
interface to tsearch to build.

Regards, Dave.