Re: 7.2 items - Mailing list pgsql-hackers

From mlw
Subject Re: 7.2 items
Date
Msg-id 3AFFD2EF.C1024B0C@mohawksoft.com
Whole thread Raw
In response to Re: 7.2 items  (Lincoln Yeoh <lyeoh@pop.jaring.my>)
List pgsql-hackers
Lincoln Yeoh wrote:
> 
> At 01:20 PM 10-05-2001 -0400, Bruce Momjian wrote:
> >Here is a small list of big TODO items.  I was wondering which ones
> >people were thinking about for 7.2?
> >
> >---------------------------------------------------------------------------
> 
> Well since you asked, here's my wish list for Postgresql 7.2.
> 
> 1) Full text index to be used by LIKE queries.
> e.g.
> create index myfti_idx on mytable ( mysoundex(story,'british english')
> fti_ops);
> Usage:
> select * from mytable where mysoundex(story,'british english') like
> '%tomato%';
> select * from mytable where mysoundex(story,'us english') like '%either%';
> select * from mytable where mysynonym(story) like '%excellent%';
> 
> First select indexed. Other selects not indexed.

This is not as easy as it looks. Full text search requires one of two
approaches, either a trigger function which updates a full text index on insert
or update, or a system which periodically scans a database and builds a full
text index. The fulltextindex method that is in contrib and my FTSS system are
examples of both respectively.

Either way it is a bit of overhead, and typically outside normal SQL. Most
people would not want the amount of overhead required to maintain a full text
index on each insert or update.

Also, I have been trying to talk the guys into doing some things with indexes,
but my understanding is that indexes are one of the last bastions of black
magic in Postgres.


-- 
42 was the answer, 49 was too soon.
------------------------
http://www.mohawksoft.com


pgsql-hackers by date:

Previous
From: Franck Martin
Date:
Subject: Re: Re: 7.2 items
Next
From: Jan Wieck
Date:
Subject: Re: Coping with huge deferred-trigger lists