Postgres Pro
Downloads
Home   >   mailing lists

Thread: array variables

array variables

From
"Marcin Krawczyk"
Date:
Hi all. Can anyone help me solve this problem ? Say I have something like this:<br /><br />CREATE TEMP TABLE idx (id
integer,p integer, idd integer);<br />INSERT INTO idx (id, p, idd) VALUES (3048, 1, 12043);<br />INSERT INTO idx (id,
p,idd) VALUES (3048, 2, 2321);<br /> INSERT INTO idx (id, p, idd) VALUES (3048, 5, 12002);<br />INSERT INTO idx (id, p,
idd)VALUES (3048, 6, 2387);<br />INSERT INTO idx (id, p, idd) VALUES (3048, 13, 4301);<br />INSERT INTO idx (id, p,
idd)VALUES (3048, 16, 1022);<br /> INSERT INTO idx (id, p, idd) VALUES (3048, 19, 321);<br />INSERT INTO idx (id, p,
idd)VALUES (3049, 2, 7543);<br />INSERT INTO idx (id, p, idd) VALUES (3050, 1, 56324);<br />INSERT INTO idx (id, p,
idd)VALUES (3050, 2, 8746);<br /><br />How do I put (SELECT * FROM idx) into multidimensional array variable ?
(plpgsql)Is is possible at all ?<br /><br /><br />regards<br />mk<br />

Re: array variables

From
"Marcin Krawczyk"
Date:
I know I can do one column like this : <br /><br />a := ARRAY(SELECT id FROM idx);<br /><br />but how about more than
one?<br />Because if I try this :<br /><br />a := ARRAY(SELECT id, p FROM idx);<br /><br />I get<br />ERROR: subquery
mustreturn only one column<br /> SQL state: 42601<br /><br />regards<br />mk<br />

Re: array variables

From
"Pavel Stehule"
Date:
Hello

2008/11/13 Marcin Krawczyk <jankes.mk@gmail.com>:
> I know I can do one column like this :
>
> a := ARRAY(SELECT id FROM idx);
>
> but how about more than one ?
> Because if I try this :
>
> a := ARRAY(SELECT id, p FROM idx);
>
> I get
> ERROR: subquery must return only one column
> SQL state: 42601
>

you can't do it directly :( now.

postgres=# create or replace function accum() returns int[] as
$$declare s int[] = '{}'; r record; begin for r in select * from f2
loop s := s || array[[r.a, r.b]]; end loop; return s; end;$$ language
plpgsql;
\CREATE FUNCTION
postgres=# select accum();      accum
-------------------{{10,10},{20,20}}
(1 row)

postgres=# select * from f2;a  | b
----+----10 | 1020 | 20
(2 rows)

regards
Pavel Stehule


> regards
> mk
>

Re: array variables

From
"Marcin Krawczyk"
Date:
Thank you guys, I appreciate your help.<br /><br /><br />regards<br />mk<br /><br />

Re: array variables

From
Achilleas Mantzios
Date:
Στις Thursday 13 November 2008 11:13:30 ο/η Marcin Krawczyk έγραψε:
> I know I can do one column like this :
>
> a := ARRAY(SELECT id FROM idx);
>
> but how about more than one ?
> Because if I try this :
>
> a := ARRAY(SELECT id, p FROM idx);
>
> I get
> ERROR: subquery must return only one column
> SQL state: 42601

how about smth like:
SELECT ARRAY[ARRAY(SELECT id FROM idx),ARRAY(SELECT p FROM idx)];

you will get 2 rows with "select count(*) from idx" columns each.
you can access it like
SELECT (ARRAY[ARRAY(SELECT id FROM idx),ARRAY(SELECT p FROM idx)])[i][j];
1<=i<=2
1<=j<=select count(*) from idx
>
> regards
> mk
>



--
Achilleas Mantzios

Re: array variables

From
Tom Lane
Date:
"Pavel Stehule" <pavel.stehule@gmail.com> writes:
> 2008/11/13 Marcin Krawczyk <jankes.mk@gmail.com>:
>> Because if I try this :
>> a := ARRAY(SELECT id, p FROM idx);
>> I get
>> ERROR: subquery must return only one column

> you can't do it directly :( now.

Sure you can, if you're using a version new enough to have arrays of
composite types.

regression=# create table t1 (f1 int, f2 text);
CREATE TABLE
regression=# insert into t1 values (1,'one');
INSERT 0 1
regression=# insert into t1 values (2,'two');
INSERT 0 1
regression=# select array(select row(t1.*)::t1 from t1);      ?column?        
-----------------------{"(1,one)","(2,two)"}
(1 row)

Whether this is a good idea for a large table is a different question ;-)
        regards, tom lane

Re: array variables

From
"Pavel Stehule"
Date:
2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
> "Pavel Stehule" <pavel.stehule@gmail.com> writes:
>> 2008/11/13 Marcin Krawczyk <jankes.mk@gmail.com>:
>>> Because if I try this :
>>> a := ARRAY(SELECT id, p FROM idx);
>>> I get
>>> ERROR: subquery must return only one column
>
>> you can't do it directly :( now.
>
> Sure you can, if you're using a version new enough to have arrays of
> composite types.
>
> regression=# create table t1 (f1 int, f2 text);
> CREATE TABLE
> regression=# insert into t1 values (1,'one');
> INSERT 0 1
> regression=# insert into t1 values (2,'two');
> INSERT 0 1
> regression=# select array(select row(t1.*)::t1 from t1);
>       ?column?
> -----------------------
>  {"(1,one)","(2,two)"}
> (1 row)
>
> Whether this is a good idea for a large table is a different question ;-)

I don't expect so user use devel version ;) - and result is array of
some composite type, not two dimensional array (but arrays of records
is nice feature too).

regards
Pavel

>
>                        regards, tom lane
>

Re: array variables

From
Tom Lane
Date:
"Pavel Stehule" <pavel.stehule@gmail.com> writes:
> 2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
>> Sure you can, if you're using a version new enough to have arrays of
>> composite types.

> I don't expect so user use devel version ;)

My example was done in 8.3.

> - and result is array of
> some composite type, not two dimensional array

Well, if the columns are of different types then you'll never be able to
represent them as a 2-D array, so I thought this was a more general answer.
        regards, tom lane

Re: array variables

From
"Pavel Stehule"
Date:
2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
> "Pavel Stehule" <pavel.stehule@gmail.com> writes:
>> 2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
>>> Sure you can, if you're using a version new enough to have arrays of
>>> composite types.
>
>> I don't expect so user use devel version ;)
>
> My example was done in 8.3.
>
>> - and result is array of
>> some composite type, not two dimensional array
>

I tested it with error:

postgres=# create table f(a int, b int);
CREATE TABLE
postgres=# insert into f values(10,20);
INSERT 0 1
postgres=# select array(select row(a,b) from f);
ERROR:  could not find array type for datatype record
postgres=# select version();                                             version
----------------------------------------------------------------------------------------------------PostgreSQL 8.3.0 on
i686-pc-linux-gnu,compiled by GCC gcc (GCC)
 
4.1.2 20070925 (Red Hat 4.1.2-33)
(1 row)

I forgot on casting, so I was confused.

Regards
Pavel Stehule


> Well, if the columns are of different types then you'll never be able to
> represent them as a 2-D array, so I thought this was a more general answer.

ok

>
>                        regards, tom lane
>

Re: array variables

From
Tom Lane
Date:
"Pavel Stehule" <pavel.stehule@gmail.com> writes:
> 2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
>> My example was done in 8.3.

> I tested it with error:

> postgres=# select array(select row(a,b) from f);
> ERROR:  could not find array type for datatype record

You left out the cast to a named rowtype --- that's not optional,
unfortunately.
        regards, tom lane

Re: array variables

From
"Pavel Stehule"
Date:
2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
> "Pavel Stehule" <pavel.stehule@gmail.com> writes:
>> 2008/11/13 Tom Lane <tgl@sss.pgh.pa.us>:
>>> My example was done in 8.3.
>
>> I tested it with error:
>
>> postgres=# select array(select row(a,b) from f);
>> ERROR:  could not find array type for datatype record
>
> You left out the cast to a named rowtype --- that's not optional,
> unfortunately.
>

it works on devel

[pavel@localhost pgsql]$ psql postgres
\psql (8.4devel)
Type "help" for help.

postgres=# create table foo(a int, b varchar);
CREATE TABLE
postgres=# insert into foo values(10, 'kuku'),(20,'aba');
INSERT 0 2
postgres=# select array(select row(a,b) from foo);        ?column?
--------------------------{"(10,kuku)","(20,aba)"}
(1 row)

regards
Pavel Stehule

>                        regards, tom lane
>