Re: Re: ext/libglade complete?

Only if you have at least half a pint of strong coffee first :) ----- Original Message ----- From: "Andrei Zmievski" <andrei@gravitonic.com> To: "Steph Fox" <steph@zend.com> Cc: "PHP-GTK dev" <php-gtk-dev@lists.php.net> Sent: Sunday, October 30, 2005 9:37 PM Subject: Re: [PHP-GTK-DEV] Re: [PHP-GTK] ext/libglade complete? > Go have a beer. :) > > -Andrei > > > On Oct 30, 2005, at 1:33 PM, Steph Fox wrote: > > >>> Erm, wait, I just realized what you meant by that. > >>> > >>> No, it's not cleaner, because of the way Glade works. In PHP-GTK > >>> 1, if you > >>> set up the object to be passed in Glade it never worked, because > >>> signal_autoconnect() didn't recognise it. > >>> > >> > >> PHP-GTK 1 only cared about functions, really. > >> > >> > >>> This way, it does work, and > >>> signal_autoconnect() caters for four separate kinds of signals. > >>> Its only > >>> problems are a) passing data and b) connecting to a callback > >>> defined within > >>> a class, non-statically. > >>> > >> > >> I'm still not sure what you're talking about here, "four separate > >> kinds of signals". I think it may be best if I just re-implement > >> signal_autoconnect() and take a vote on the implementation. > >> > > > > connect(), connect_after(), connect_object() and > > connect_object_after() are > > the types of signals that can be created from within Glade. Please > > don't > > mess with that, it's the first time it's worked properly! > > > > > >>> I think it would be sensible to have a separate function that > >>> passed an > >>> object as first parameter if you wanted to be able to access > >>> callbacks > >>> within that class, and data mapping as a second parameter. I don't > >>> think it > >>> would be a good idea to call it signal_autoconnect_object() though, > >>> because > >>> that means something other. > >>> > >> > >> Ok, I definitely need to reimplement it now. Build faster, Glade2! > >> > > > > Huh? > > > > > >> > >> - Andrei > > >