Re: wildcard search support for pg_trgm

Tom Lane <tgl@sss.pgh.pa.us>

From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alexander Korotkov <aekorotkov@gmail.com>
Cc: Jan Urbański <wulczer@wulczer.org>, Postgres - Hackers <pgsql-hackers@postgresql.org>
Date: 2011-02-01T02:40:32Z
Lists: pgsql-hackers
=?UTF-8?B?SmFuIFVyYmHFhHNraQ==?= <wulczer@wulczer.org> writes:
> OK, now it works flawlessly as far as I can tell. Will mark it as Ready
> for Committer.

Applied with mostly-stylistic corrections, plus addition of
documentation and a minimal regression test.

I did *not* apply this bit:

>> 2) I found gist index not very useful with default SIGLENINT = 3. I've
>> changed this value to 15 and I found gist index performs very good on
>> dictionary. But on longer strings greater values of SIGLENINT may be
>> required (probably even SIGLENINT > 122 will give benefit in some cases in
>> spite of TOAST).

AFAICT that would break on-disk compatibility of pg_trgm GIST indexes.
I don't believe we have adequate evidence to justify doing that, and
in any case it ought to be a separate patch rather than buried inside a
mostly unrelated feature patch.

			regards, tom lane