PostgreSQL Row-Level Full-Text Search
By Digoal.
In some applications, you may need to search all the fields in a table, and other fields may still require a precise query, while others in contrast may require a fuzzy query or a full-text search.
Now consider this scenario: the selection of drop-down boxes on some front-end pages. The selection of these boxes can actually be a headache for application developers because writing SQL statements can turn out to be quite a big hassle sometimes.
Now consider the code below:
postgres=# create table t(phonenum text, info text, c1 int, c2 text, c3 text, c4 timestamp);
CREATE TABLE
postgres=# insert into t values ('13888888888','i am digoal, a postgresqler',123,'china','PRC, Alibaba, ah',now());
INSERT 0 1
postgres=# select * from t;
phonenum | info | c1 | c2 | c3 | c4
-------------+-----------------------------+-----+-------+------------------------------+----------------------------
13888888888 | i am digoal, a postgresqler | 123 | china | PRC, Alibaba, Ah | 2016-04-19 11:15:55.208658
(1 row)Now, based on this code, the following is a query the t table, shown above. If any field matches digoal, this following record is returned:
select * from t where phonenum='digoal' or info ~ 'digoal' or c1='digoal' or ......;Of course, though, each field requires a condition, such as the precise matching condition, and the full-text search condition. However, with that said, Row-level full-text search can greatly simplify this query.
Below, we will go over how you can create a row-level text index to make things a bit easier.
Create a Row-Level Text Index
Consider Jieba as an example. The source code can be found here, and the pg_scws based on SCWS.
All of the above support custom dictionaries. The Installation process for these dictionaries is omitted here. The usage of these dictionaries is as follows:
postgres=# select t::text from t;
t
-----------------------------------------------------------------------------------------------------------------
(13888888888,"i am digoal, a postgresqler",123,china,PRC, Alibaba, Ah,"2016-04-19 11:15:55.208658")
(1 row)
postgres=# select to_tsvector('jiebacfg',t::text) from t;
to_tsvector
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
' ':6,8,11,13,33 '04':30 '11':34 '123':17 '13888888888':2 '15':36 '19':32 '2016':28 '55.208658':38 'china':19 'digoal':9 'postgresqler':14 'PRC':21 'Alibaba':23
(1 row)Use t::text to convert rows into a large text.
postgres=# select to_tsvector('jiebacfg',t::text) @@ to_tsquery('digoal & china') from t;
?column?
----------
t
(1 row)
postgres=# select to_tsvector('jiebacfg',t::text) @@ to_tsquery('digoal & post') from t;
?column?
----------
f
(1 row)To create a row-level text index, an immutable function index is required.
postgres=# create or replace function f1(regconfig,text) returns tsvector as
$$
select to_tsvector($1,$2);
$$
language sql immutable strict;
CREATE FUNCTION
postgres=# create or replace function f1(text) returns tsvector as
$$
select to_tsvector($1); $$
language sql immutable strict;
CREATE FUNCTION
postgres=# alter function record_out(record) immutable;
ALTER FUNCTION
postgres=# alter function textin(cstring) immutable;
ALTER FUNCTION
postgres=# create index idx_t_1 on t using gin (f1('jiebacfg'::regconfig,t::text)) ;
CREATE INDEX
Verification:
postgres=# select * from t where f1('jiebacfg'::regconfig,t::text) @@ to_tsquery('digoal & post') ;
phonenum | info | c1 | c2 | c3 | c4
----------+------+----+----+----+----
(0 rows)
postgres=# select * from t where f1('jiebacfg'::regconfig,t::text) @@ to_tsquery('digoal & china') ;
phonenum | info | c1 | c2 | c3 | c4
-------------+-----------------------------+-----+-------+------------------------------+----------------------------
13888888888 | i am digoal, a postgresqler | 123 | china | PRC, Alibaba | 2016-04-19 11:15:55.208658
(1 row)
postgres=# select * from t where f1('jiebacfg'::regconfig,t::text) @@ to_tsquery('digoal & Alibaba') ;
phonenum | info | c1 | c2 | c3 | c4
-------------+-----------------------------+-----+-------+------------------------------+----------------------------
13888888888 | i am digoal, a postgresqler | 123 | china | PRC, Alibaba, Ah | 2016-04-19 11:15:55.208658
(1 row)
postgres=# explain select * from t where f1('jiebacfg'::regconfig,t::text) @@ to_tsquery('digoal & Alibaba') ;
QUERY PLAN
------------------------------------------------------------------------------------------------------
Seq Scan on t (cost=0.00..1.52 rows=1 width=140)
Filter: (to_tsvector('jiebacfg'::regconfig, (t.*)::text) @@ to_tsquery('digoal & Alibaba'::text))
(2 rows)If the number of records is large, the indexes will be used. When the number of records is small, we can use Hint or Switch to force the index:
postgres=# set enable_seqscan=off;
SET
postgres=# explain select * from t where f1('jiebacfg'::regconfig,t::text) @@ to_tsquery('digoal & Alibaba') ;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------
Bitmap Heap Scan on t (cost=12.25..16.77 rows=1 width=140)
Recheck Cond: (to_tsvector('jiebacfg'::regconfig, (t.*)::text) @@ to_tsquery('digoal & Alibaba'::text))
-> Bitmap Index Scan on idx_t_1 (cost=0.00..12.25 rows=1 width=0)
Index Cond: (to_tsvector('jiebacfg'::regconfig, (t.*)::text) @@ to_tsquery('digoal & Alibaba'::text))
(4 rows)
