pHilipp Zabel píše v Út 09. 02. 2010 v 22:44 +0100:
> 2010/2/9 Jiří Zárevúcky <[email protected]>:
> > pHilipp Zabel píše v Út 09. 02. 2010 v 22:05 +0100:
> >> 2010/2/9 Jiří Zárevúcky <[email protected]>:
> >> > pHilipp Zabel píše v Út 09. 02. 2010 v 16:37 +0100:
> >> >> On Tue, Feb 9, 2010 at 3:53 PM, Levi Bard
> >> >> <[email protected]> wrote:
> >> >> >> While this is indeed correct workaround, the Hildon binding should
> >> >> >> probably be updated (so you get type-checking from the compiler).
> >> >> >
> >> >> > In that case, can that be a different vapi so that we can still use
> >> >> > hildon-1 with diablo (maemo 4.x)?
> >> >> > Or at least keep the legacy version as hildon-1-diablo.vapi or 
> >> >> > something?
> >> >>
> >> >> These are Maemo specific changes to GTK+ (properties of Gtk.Entry, not
> >> >> of Hildon.Entry). This lead to the slightly awkward Hildon.gtk_...
> >> >> helper functions. I forgot to wrap all of them - here is a patch that
> >> >> adds the missing Hildon.gtk_entry_... functions to
> >> >> hildon-1-custom.vapi.
> >> >>
> >> >> The helper function in question is Hildon.gtk_entry_set_input_mode
> >> >> (Gtk.Entry entry, Hildon.GtkInputMode mode). I guess we could also
> >> >> manually add the hildon-input-mode property without accessor methods
> >> >> to Hildon.Entry, but this is the official API.
> >> >>
> >> >
> >> > You could manually specify the names of the accessors. Is there any
> >> > valid reason for binding those as ugly static methods except being
> >> > "politically correct"?
> >>
> >> Vala doesn't let me add methods to Gtk classes from hildon-1.vapi.
> >> Changing upstream gtk+-2.0.vapi for a vendor specific modification
> >> doesn't seem right. What would you suggest?
> >>
> >
> > I would suggest that adding vendor specific modifications to existing
> > classes is way too ugly
> 
> Agreed.
> 
> > to be accepted and you should bind it to hildon
> > classes. How can they even add more properties to an already defined
> > class?
> 
> They ship a modified GTK+ library. Hildon depends on that.

That's evil (for binding creators at least). Well, what about shipping a
modified Gtk bindings for Maemo? It would be much more natural than
trying to fit a separate extension package on top of the original ones.

Just out of curiosity, will that changes get pulled to mainstream, or is
it just mission-specific stuff?

> I guess we could get away with just binding the hildon_gtk_entry_*
> methods to Hildon.Entry instead of Gtk.Entry. Is there a way of doing
> this via metadata before the vapigen step?
> But there are a few others like hildon_gtk_widget_set_theme_size or
> hildon_gtk_window_set_progress_indicator that can't be mapped to
> Hildon classes, as they are needed to work on Gtk.Button and
> Gtk.Dialog, for example.
> 

Ah, then it's not so simple. Thanks for enlightening. :)


Attachment: signature.asc
Description: Toto je digitálně podepsaná část zprávy

_______________________________________________
Vala-list mailing list
[email protected]
http://mail.gnome.org/mailman/listinfo/vala-list

Reply via email to