Re: [PATCH 07/15] core: Free specs after last element is sent
- From: "Juan A. Suarez Romero" <jasuarez igalia com>
- To: grilo-list gnome org
- Subject: Re: [PATCH 07/15] core: Free specs after last element is sent
- Date: Mon, 09 Aug 2010 16:25:21 +0200
On Mon, 2010-08-09 at 11:38 +0200, Iago Toral Quiroga wrote:
> About the idea of freeing the stuff in the idle loop so devs can use
> the
> callback data after invoking the result callback with remaining=0, my
> preference is to avoid it if possible. My reasons:
>
>
So, if I understand correctly, you're right with the ref/unref feature,
but keeping the automatic unreffing after invoking the callback with
remaining==0 as it is: that is, when it is invoking callback with
remaining==0, the spec is automatically freed, unless the user has
invoked ref() previously.
In this case, I'm fine with it. But this must be stressed in
documentation, as it leaded to some mistakes in the past.
J.A.
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]