Hi,
On 2019-09-22 20:16:15 -0400, Tom Lane wrote:
> I think you're considering a much smaller slice of the system than
> what seems to me to be at risk here. As an example, an ExprState
> tree would also include any fn_extra-linked state that individual
> functions might've set up.
> > Hm. I int
Hi,
On 2019-09-22 20:16:15 -0400, Tom Lane wrote:
> Andres Freund writes:
> > On 2019-09-22 18:43:23 -0400, Tom Lane wrote:
> >> I'm not convinced that it'd be safe to re-use an ExprState after a
> >> previous execution failed (though perhaps Andres has a different
> >> opinion?)
>
> > I don't i
Andres Freund writes:
> On 2019-09-22 18:43:23 -0400, Tom Lane wrote:
>> I'm not convinced that it'd be safe to re-use an ExprState after a
>> previous execution failed (though perhaps Andres has a different
>> opinion?)
> I don't immediately see why it'd be problematic to reuse at a later
> time
Hi,
On 2019-09-22 18:43:23 -0400, Tom Lane wrote:
> I'm not convinced that it'd be safe to re-use an ExprState after a
> previous execution failed (though perhaps Andres has a different
> opinion?)
I don't immediately see why it'd be problematic to reuse at a later
time, as long as it's guarantee
Pavel Stehule writes:
> When I tested some hypothesis I wrote buggy code. It was surprise how fast
> I lost all free memory
> do $$
> begin
> for i in 1..300
> loop
> begin
> -- do some error
> if i then end if;
> exception when others then
> -- do nothing
>
Hi
When I tested some hypothesis I wrote buggy code. It was surprise how fast
I lost all free memory
do $$
begin
for i in 1..300
loop
begin
-- do some error
if i then end if;
exception when others then
-- do nothing
end;
end loop;
end;
$$;
problem is somew
