On 10/24/2010 06:44 PM, Tom Lane wrote:
> =?UTF-8?B?SmFuIFVyYmHFhHNraQ==?=<wulczer(at)wulczer(dot)org> writes:
>> I see that plperl uses a triple of (function oid, is_trigger flag, user
>> id) as a hash key for caching compiled functions. OTOH pltcl and plpgsql
>> both use (oid, trigger relation oid, user id). Is there any reason why
>> just using a bool as plperl does would be wrong?
> plpgsql needs to consider the trigger relation OID because datatypes of
> the trigger relation's columns will make their way into cached plans
> for the function. The same function, if applied as a trigger on two
> different rels, could therefore have different cached plans so it needs
> two separate cache entries.
> In PLs where this kind of dependency isn't possible, there's no need for
> separate function cache entries.
> I'm not certain that plperl is actually correct to do it that way,
> but that's the basic idea.
Why do we need the is_trigger flag at all for the plperl hash key? At
first glance it strikes me as unnecessary.
In response to
pgsql-hackers by date
|Next:||From: Greg Stark||Date: 2010-10-24 23:11:51|
|Subject: Re: ask for review of MERGE|
|Previous:||From: Tom Lane||Date: 2010-10-24 22:59:34|
|Subject: Re: Range Types, discrete and/or continuous |