Xapian discuss - Feb 2006 - Matching second index?

Afternoon All,

I have set up a second index, using scriptindex, with the prefix of XDESC.

I can see the words when I use the bit of code Olly provided (cheers!) to see
the term frequencys, and they start with:

XDESC:R (I lowercased them all)

Now if I run a search on omega with the following:

XDESCtastefully I get zero results..when tastefully has 1000+ matches.

Or XDESC:tastefully..etc.

What am I doing wrong?

I have also tried it through my basic PHP bindings code just passing it through:

$database = new_Database("/home2/omega/boris2");
$enquire = new_Enquire($database);
$stemmer = new_stem ("english");
$query_parser = new_queryparser();
queryparser_set_stemmer ($query_parser, $stemmer);
queryparser_set_stemming_strategy ($query_parser, 1);
queryparser_set_database ($query_parser, $database);

$query = queryparser_parse_query ($query_parser, "XDESCtastefully");

I am missing something I know...someone point me in the right direction.

Oh I am using flint now, halfed the size of my normally database, and my new
larger DB with description as an index is only 1.6 gig in size which is nice.

Cheers
John

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

On Sun, Feb 05, 2006 at 02:54:09PM +0000, John Wards
wrote:
> I am missing something I know...someone point me in the right direction.
The prefix that the QueryParser understands isn't the same as the prefix
which is actually on the term in the database.  This document explains:

http://svn.xapian.org/*checkout*/trunk/xapian-applications/omega/docs/termprefixes.txt

See the "Probabilistic Fields" section in particular.
> Oh I am using flint now, halfed the size of my normally database, and my
new
> larger DB with description as an index is only 1.6 gig in size which is
nice.
There should be further size reductions in the future too.

Cheers,
    Olly