Horton

Re: why does plperl cache functions using just a bool for is_trigger

Alex Hunsaker <badalex@gmail.com>

View thread
From: Alex Hunsaker <badalex@gmail.com>
To: Hannu Krosing <hannu@2ndquadrant.com>
Cc: Jan Urbański <wulczer@wulczer.org>, Tom Lane <tgl@sss.pgh.pa.us>, Andrew Dunstan <andrew@dunslane.net>, Postgres - Hackers <pgsql-hackers@postgresql.org>
Date: 2010-11-04T20:29:49Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Remove unnecessary use of trigger flag to hash plperl functions

On Thu, Nov 4, 2010 at 13:43, Hannu Krosing <hannu@2ndquadrant.com> wrote:
> So your plan was to have some savepoint before each execute ?
>
> How would one rollback the latest transaction ?

It is always rolled back.  Its how plperl works today:
create or replace function foo() returns int as $$
eval {
    spi_exec_query('create table uniq (num int primary key');
    spi_exec_query('insert into uniq (num) values (1), (1);', 1);
};

if($@) {
 # do something ... $@ == "duplicate key value violates unique
constraint "uniq_pkey" at line 2."
  warn $@;
}

# oh well do something else
# note the transaction is _not_ aborted
spi_exec_query('select 1;', 1);
return 1;
$$ language plperl;

=# begin;
=# select foo();
=# select 1;
=# commit;

It does not matter if you use eval or not, its always in a sub transaction.

> I see. "exception when unique violation" in plpgsql  does automatic
> rollback to block start (matching BEGIN) so I assumed that your
> try/except sample is designed to do something similar

Basically, minus the rollback to start.  Its per query.