On 2/19/07, Oleg Bartunov oleg@sai.msu.su wrote:
pg_trgm was developed for spelling corrrection and there is a threshold of
similarity, which is 0.3 by default. Readme explains what does it means.
Yes, I read it.
Similarity could be very low, since you didn't make separate column and length
On Mon, 19 Feb 2007, Guillaume Smet wrote:
On 2/19/07, Oleg Bartunov oleg@sai.msu.su wrote:
pg_trgm was developed for spelling corrrection and there is a threshold of
similarity, which is 0.3 by default. Readme explains what does it means.
Yes, I read it.
Similarity could be very low, since
On 2/19/07, Oleg Bartunov oleg@sai.msu.su wrote:
You need to wait GiN support.
OK. Thanks.
If you need testers for this one, feel free to contact me. I'm very
interested in testing pg_trgm in conjunction with tsearch2.
--
Guillaume
---(end of
Hi Oleg,
On 2/17/07, Oleg Bartunov oleg@sai.msu.su wrote:
3-gram is implemented as a contrib/pg_trgm. It currently uses GiST index,
but may be enhanced with the GiN.
As I'm facing the same problem, I've taken a look to pg_trgm. At the
moment, my opinion is quite mixed but perhaps I did
On Sun, 18 Feb 2007, Guillaume Smet wrote:
Hi Oleg,
On 2/17/07, Oleg Bartunov oleg@sai.msu.su wrote:
3-gram is implemented as a contrib/pg_trgm. It currently uses GiST index,
but may be enhanced with the GiN.
As I'm facing the same problem, I've taken a look to pg_trgm. At the
moment, my
Thanks. I'll look into this.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
3-gram is implemented as a contrib/pg_trgm. It currently uses GiST index,
but may be enhanced with the GiN.
Oleg
On Sat, 17 Feb 2007, Tatsuo Ishii wrote:
Hi,
Is anybody working on implementing n-gram search
Thanks. I'll look into this.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
3-gram is implemented as a contrib/pg_trgm. It currently uses GiST index,
but may be enhanced with the GiN.
Oleg
On Sat, 17 Feb 2007, Tatsuo Ishii wrote:
Hi,
Is anybody working on implementing n-gram search
Hi,
Is anybody working on implementing n-gram search functionality for
text type data? tsearch2 is great for long text but it's not
appropreate for short (10-100 bytes) text data. What I want to achieve
is a fast partial match search using indexes, i.e. foo ~ 'bar' or foo
LIKE '%bar%' type
3-gram is implemented as a contrib/pg_trgm. It currently uses GiST index,
but may be enhanced with the GiN.
Oleg
On Sat, 17 Feb 2007, Tatsuo Ishii wrote:
Hi,
Is anybody working on implementing n-gram search functionality for
text type data? tsearch2 is great for long text but it's not
