Re: [gtk-list] Re: key bind class
- From: "Oskar Liljeblad" <osk hem passagen se>
- To: <gtk-list redhat com>
- Cc: <otaylor redhat com>
- Subject: Re: [gtk-list] Re: key bind class
- Date: Tue, 28 Jul 1998 12:41:04 +0200
From: Owen Taylor <otaylor@redhat.com>
>You might want to look at:
>
> http://archive.redhat.com/gtk-list/1998-July/0525.html
>
>That briefly summerizes the "binding" system for GTK+.
>The binding system is intended for doing things like
>setting up keystroke bindings for Entry and Text widgets.
What if an application wants to be notified by a key stroke, which has no
relation whatsoever to an existing widget? Should it listen to
key_press_event's or it is possible to use the new GTK binding system or the
existing accelerator system?
>Neither of the systems allow multiple key sequences, of course,
>and there is no builtin abstraction that would allow modifying
>toolbars to activate menu items, so it isn't really exactly
>what you are talking about with your commandapp.
In the command API, configurability is archived through separating menus,
toolbars and key bindings as much as possible. A toolbar button or key
binding cannot rely on the existance of a certain menu item. In the future,
toolbars and menubar will be the same - it's up to the user to decide what
his toolbar/menubar should look like. The key binding shown in a menu item
is looked up in the list of key bindings, but otherwise there is not a
specific key binding for a menu item.
>However,
>there is enough overlap there that more comprehensive systems
>should possibly be built on top of these facilities instead
>of being put into GTK+ along side.
Again, what facility should I use? Right now the key parser class I created
listens to
key_press_events of widgets (usually the main window). Do you suggest I
change to a different system?
Regards,
Oskar Liljeblad (osk@hem.passagen.se)
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]