[Xapian-discuss] Unicode and complex queries from Python

[Olly Betts]

On Sun, Sep 25, 2005 at 05:15:00PM +0100, James Aylett wrote:
> I've asked about this on the SWIG list, but without a helpful reply.

Are you giving up on this for now?  I guess I might try and have another
fiddle if you have.  I suspect you understand it better, but I probably
have the advantage of being able to spend more time on it...

> > Any ETA on the unicode aware QueryParser?
> 
> It won't happen before the 1.0 release of Xapian, for the reasons Olly
> has given.

That's true, but perhaps a bit misleading.

It true because we wouldn't want to switch in a unicode queryparser
until 1.0 (probably not even as an option, since we'd need to upgrade
the snowball stemmers to get the utf=8 versions and it wouldn't really
work to have incompatible stemmers for iso-8859-1 and utf-8).

But it's misleading because there's no particular reason to hold off
releasing 1.0 when the unicode queryparser is ready to release.
If the magic "1.0" badge is an issue, we can just release 0.10.0...

I've got a patch which makes the queryparser unicode - it's in
production use on search.gmane.org (the xapian search there is now
fully live incidentally).  Stemming isn't handled yet though, and
it's not the prettiest patch in the world.  I'd imagine we're talking
weeks rather than months to clean it up and finish it off.

Cheers,
    Olly