Re: Tsearch2 / Create rule on select

On Fri, 9 Mar 2007, Jean-Michel Pour? wrote:

> Le vendredi 09 mars 2007 ЪЪ 10:58 +0100, Magnus Hagander a ЪЪcrit :
>> No idea. Assuming you want to do it beforehand. otherwise, just create
>> the index and see how large it got?
>
> Thank you for your comments. I will add TSeach2 support to phpBB 3.x
> soon.
>
> I had incredible response time on simple queries on more than plain-text
> 200.000 row. About 1 millisecond! Incredible but true.
>
> explain analyse verbose SELECT * FROM phpbb_posts_text WHERE idxfti @@
> 'jmp'::tsquery limit 100;
>
> Limit (cost=0.00..444.47 rows=100 width=934) (actual time=0.046..0.824
> rows=100 loops=1)
> -> Index Scan using idxfti_idx on phpbb_posts_text
> (cost=0.00..1053.38 rows=237 width=934) (actual time=0.040..0.418
> rows=100 loops=1)
> Index Cond: (idxfti @@ '''jmp'''::tsquery)
> [color=red]Total runtime: 1.068 ms
>
> TSearch2 will be used by wikimedia shortly (it is supported in beta
> version).

I've already seen tsearch2 in 1.9.X release.

>
> IMHO, it would be better if PostgreSQL parser was able to find Tseach
> indexes alone, rewritting the query automatically. Maybe it is on the
> radar list of hackers.

btw, we're working on new FTS feature of PostgreSQL, it's basically
tsearch integrated into the pg core. But it has some new features and
new SQL commands for configuring of FTS, so if the matter is not pressing I'd recommend
to check http://mira.sai.msu.su/~megera/pgsql/ftsdoc

>
> Kind regards,
> Jean-Michel
>
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 2: Don't 'kill -9' the postmaster
>

Regards,
Oleg
_____________________________________________________________
Oleg Bartunov, Research Scientist, Head of AstroNet (www.astronet.ru),
Sternberg Astronomical Institute, Moscow University, Russia
Internet: oleg(at)sai(dot)msu(dot)su, http://www.sai.msu.su/~megera/
phone: +007(495)939-16-83, +007(495)939-23-83