Postgres Pro
Downloads
Home   >   mailing lists

Thread: Phrase search distance syntax

Phrase search distance syntax

From
Bruce Momjian
Date:
Sorry to be asking another phrase search syntax question, and so close
to final release, but ...

Why does the phrase distance operator assume <1> means adjacent words,
and not <0>.  (FYI, <-> is the same as <1>.)

For example:
 select to_tsvector('park a a house') @@ to_tsquery('park <3> house');

seems like it would be more logical as <2>, meaning two lexems between
the words.

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com

+ As you are, so once was I. As I am, so you will be. +
+                     Ancient Roman grave inscription +

Re: Phrase search distance syntax

From
Teodor Sigaev
Date:
> Sorry to be asking another phrase search syntax question, and so close
> to final release, but ...
Really close...
>
> Why does the phrase distance operator assume <1> means adjacent words,
> and not <0>.  (FYI, <-> is the same as <1>.)
Because
1 it is a result of subtruction of word's positions
2 <0> could be used as special case like a word with two infinitives:
# create text search dictionary xx (template = 'ispell', 
DictFile='ispell_sample', AffFile='ispell_sample');
# alter text search configuration english ALTER MAPPING FOR asciiword WITH xx, 
english_stem;

# select to_tsvector('english', 'bookings');     to_tsvector
---------------------- 'book':1 'booking':1

# select to_tsvector('english', 'bookings') @@ 'book <0> booking'; ?column?
---------- t
(1 row)



-- 
Teodor Sigaev                                   E-mail: teodor@sigaev.ru
  WWW: http://www.sigaev.ru/

Re: Phrase search distance syntax

From
Bruce Momjian
Date:
On Fri, Sep 23, 2016 at 05:07:26PM +0300, Teodor Sigaev wrote:
> >Sorry to be asking another phrase search syntax question, and so close
> >to final release, but ...
> Really close...
> >
> >Why does the phrase distance operator assume <1> means adjacent words,
> >and not <0>.  (FYI, <-> is the same as <1>.)
> Because
> 1 it is a result of subtruction of word's positions
> 2 <0> could be used as special case like a word with two infinitives:
> # create text search dictionary xx (template = 'ispell',
> DictFile='ispell_sample', AffFile='ispell_sample');
> # alter text search configuration english ALTER MAPPING FOR asciiword WITH
> xx, english_stem;
> 
> # select to_tsvector('english', 'bookings');
>      to_tsvector
> ----------------------
>  'book':1 'booking':1
> 
> # select to_tsvector('english', 'bookings') @@ 'book <0> booking';
>  ?column?
> ----------
>  t
> (1 row)

OK, thanks.  I just found it as unusual.

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com

+ As you are, so once was I. As I am, so you will be. +
+                     Ancient Roman grave inscription +

Re: Phrase search distance syntax

From
Tom Lane
Date:
Teodor Sigaev <teodor@sigaev.ru> writes:
>> Why does the phrase distance operator assume <1> means adjacent words,
>> and not <0>.  (FYI, <-> is the same as <1>.)

> Because
> 1 it is a result of subtruction of word's positions
> 2 <0> could be used as special case like a word with two infinitives:

This is actually documented, in 12.1.2:
A special case that's sometimes useful is that <0> can be used torequire that two patterns match the same word.
        regards, tom lane