Re: [PATCH] Gtk2.xs (gtk_init): do gtk_init_check in PL_minus_c mode

[Kevin Ryde <user42 zip com au>]

Torsten Schoenfeld <kaffeetisch gmx de> writes:
> I agree that it would be better to handle this in Gtk2::import, but I
> was unable to find a way to access PL_minus_c from Perl.

Actually, I now think that to change -init would be incompatible for
anyone relying on it to Gtk2->init at BEGIN time.

I've been striking various widget things in my test scripts lately that
demand an init where I hadn't thought it would be needed.  If you've
written '-init' it may well be because you do in fact need init at BEGIN
time, and to suppress that gives a chance of bombing badly.

What about better to advise programs to have their own separate

    use Gtk2;
    Gtk2->init;

if the code can get to CHECK stage without init, and if you care about
that (after all, there's more than one way to do it :).