Hi Viktor,
I'm not sure what your requirements are in terms of performance and
stability of the your result set. See Pavel's response. A cursor issues
a single query and renders a single result set. The result set is
static, the cursor just gives you finer control/performance when
retrieving rows
Hello
using a "window" implemented via LIMIT OFFSET is not good - it is
solution on some systems where cursors are not available, but it is
bad solution on PostgreSQL. Use a cursor instead - it is significantly
more efective with less memory requests.
Regards
Pavel Stehule
2011/7/8 Viktor Bojov
Thanx Wayne,
at the end i did it that way and it works.
The code is below.
CREATE FUNCTION pattern_counter1("patLength" integer) RETURNS character
varying
LANGUAGE plperl
AS $_X$
my $rvCnt = spi_exec_query("select count(1) as cnt from entry");
#my $rowCountAll = $rvCnt->{processed};
my $row = $rvC
I'm have the same situation with large tables. Take a look at using a
cursor to fetch several thousand rows at a time. I presume what's
happening is that perl is attempting to create a massive list/array in
memory. If you use a cursor the list should only contain X number of
rows where X in the
