Re: plpgsql arrays - Mailing list pgsql-performance

From Matthew Wakeling
Subject Re: plpgsql arrays
Date
Msg-id alpine.DEB.2.00.0904071719320.791@aragorn.flymine.org
Whole thread Raw
In response to Re: plpgsql arrays  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-performance
On Tue, 7 Apr 2009, Tom Lane wrote:
> Matthew Wakeling <matthew@flymine.org> writes:
>> CREATE OR REPLACE FUNCTION overlap_gene_primer() RETURNS SETOF RECORD AS $$
>> DECLARE
>>      left location;
>>      retval RECORD;
>> BEGIN
>>      DECLARE left_cursor NO SCROLL CURSOR FOR SELECT location FROM location, gene WHERE location.id = gene.id ORDER
BYobjectid, start, end; 
>>      left = FETCH left_cursor;
>> END;
>> $$ LANGUAGE plpgsql;
>
> Well, the DECLARE for the cursor should go in the DECLARE section,
> and the syntax for the FETCH should be
>     FETCH cursorname INTO recordvariablename;
> and I'm too lazy to check right now but I think you might be missing
> an OPEN for the cursor.

Yeah, thanks to Justin I found the plpgsql docs for cursors. The main
cursors docs should really link there.

This seems to do what I want:

CREATE OR REPLACE FUNCTION overlap_gene_primer() RETURNS SETOF RECORD AS
$$
DECLARE
     left_cursor NO SCROLL CURSOR FOR SELECT location.* FROM location, gene WHERE location.subjectid = gene.id ORDER BY
objectid,start, end; 
     left location;
BEGIN
     OPEN left_cursor;
     FETCH left_cursor INTO left;
END;
$$ LANGUAGE plpgsql;

Matthew

--
 Lord grant me patience, and I want it NOW!

pgsql-performance by date:

Previous
From: Andrew Sullivan
Date:
Subject: Re: Best replication solution?
Next
From: Andrew Sullivan
Date:
Subject: Re: Best replication solution?