Libxfce4ui spawn api break

Nick Schermer nickschermer at gmail.com
Sun Nov 1 10:10:53 CET 2009


2009/11/1 Brian J. Tarricone <brian at tarricone.org>:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 10/31/2009 04:21 PM, Nick Schermer wrote:
>> 2009/10/31 Jannis Pohlmann <jannis at xfce.org>:
>>> If we can agree on that, go ahead. I personally think
>>> xfce_spawn_on_screen() should be at least as powerful as
>>> gdk_spawn_on_screen(). And if it misses the PID, then we should add it
>>> as long as we can.
>>
>> In that case we should also add (GSpawnChildSetupFunc child_setup,
>> gpointer user_data), does anybody think we needs this?
>
> Nah, from what I'm told the child setup function is generally a bad
> idea.  I can't think of any use cases offhand, aside from calling
> setsid(), which is rarely necessary (and should be done by default
> anyway, unless G_SPAWN_DO_NOT_REAP_CHILD is passed).

Ok, that's what I thought. We always add reap-child to watch the
process for startup notification, so no setsid for us.

Nick



More information about the Xfce4-dev mailing list