Re: GtkBuilder status

2006-05-15 Thread Damon Chaplin
On Thu, 2006-05-11 at 12:00 -0300, Johan Dahlin wrote: > I understand your point of view, you'd like to avoid updating glade again > to support the new builder. > > Until glade is updated, the users will be able use a script to convert from > the old format to the new format. I'm afraid it is un

Re: GtkBuilder status

2006-05-13 Thread Tristan Van Berkom
Tristan Van Berkom wrote: I'm afraid I'll have to take at least one day to better my knowlage about exactly how GtkAction works, but from the little I already know and my first glances... I dont see why the menu items cant be prefabricated with actions attached, and the action groups made ava

Re: GtkBuilder status

2006-05-12 Thread Johan Dahlin
> > Yes, these simple cases would be possible to handle somewhat, it would > prevent you from using the delimiter character in object names, which > wouldn't be a too big deal. > However, specifying model data for GtkListStore and GtkTreeStore is not > really possible using properties, unless you

Re: GtkBuilder status

2006-05-12 Thread Tristan Van Berkom
Hi all, /me calls mail_thread_join() on this thread... Nothing wrong with allowing buildable menus or toolbars, but how do you propose that they should achieve what uimanager does, mainly separation of ui and actions, and merging of uis ? Well, the merging of UI's is not very complex, I wo

Re: GtkBuilder status

2006-05-11 Thread Owen Taylor
On Thu, 2006-05-11 at 15:55 -0500, Yevgen Muntyan wrote: > Owen Taylor wrote: > > >>It's, err, wrong to claim that another piece of soft is perfect, and > >>make this piece of soft "perfect assuming that's perfect", > >>and say "is that's not perfect, complain about it, but this piece of > >>sof

Re: GtkBuilder status

2006-05-11 Thread Yevgen Muntyan
Owen Taylor wrote: It's, err, wrong to claim that another piece of soft is perfect, and make this piece of soft "perfect assuming that's perfect", and say "is that's not perfect, complain about it, but this piece of soft will still assume that piece of soft is perfect". Are we living in ideal

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Dan Winship wrote: > Johan Dahlin wrote: >>> Over the last couple of weeks Henrique Romano and I have been working on >>> GtkBuilder, a UI constructor intended for inclusion in GTK+. >>> I'd like to discuss the API and some of the decisions before making the >>> code available for public review. >>

Re: GtkBuilder status

2006-05-11 Thread Owen Taylor
On Thu, 2006-05-11 at 09:47 -0500, Yevgen Muntyan wrote: > Johan Dahlin wrote: > > >> o My real concern is about supporting menus and toolbars built by > >>UIManager. > >>- Is the motivation here only a "time-to-market" thing ? > >>- If so, do you have any plan or stratagy to take baby-ste

Re: GtkBuilder status

2006-05-11 Thread Dan Winship
Johan Dahlin wrote: >> Over the last couple of weeks Henrique Romano and I have been working on >> GtkBuilder, a UI constructor intended for inclusion in GTK+. >> I'd like to discuss the API and some of the decisions before making the >> code available for public review. >> > I went ahead and at

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Damon Chaplin wrote: > On Wed, 2006-05-10 at 11:51 -0300, Johan Dahlin wrote: > >> We've made a couple of important decisions: >> * GMarkup based parser which parses and creates the object tree in one step >> go instead of saprving a whole tree in memory. >> * breaking xml format compatibility w

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
>> It'll be fairly easy to add support for this in the signal tag, >> so you can specify the data argument. It would require two attributes >> one for the data and one for the type, then you'd do: >> >> >data-type="GObject"/> >> >> void quit_cb (GtkAction *action, GtkWindow *window); >> >

Re: GtkBuilder status

2006-05-11 Thread Yevgen Muntyan
Johan Dahlin wrote: [snip] There should be only one obvious way of doing a specific task. GtkUIManager is the currently the obvious way of creating menus and toolbars Here you go, python programmer in action :) Seriously speaking, GtkUIManager is not and may not be perfect. The single

Re: GtkBuilder status

2006-05-11 Thread Nickolay V. Shmyrev
> >I never claimed GtkUIManager was perfect. I just said it was the obvious way >of creating menus and toolbars. It might be overkill in smaller >applications, but if they use gtkbuilder they won't even notice. > >Johan > Probably I've missed something but can someone give a link where a decision

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Damon Chaplin wrote: > On Wed, 2006-05-10 at 11:51 -0300, Johan Dahlin wrote: > >> We've made a couple of important decisions: >> * GMarkup based parser which parses and creates the object tree in one step >> go instead of saprving a whole tree in memory. >> * breaking xml format compatibility w

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
>> [snip] >> There should be only one obvious way of doing a specific task. >> GtkUIManager is the currently the obvious way of creating menus and >> toolbars >> >> > Here you go, python programmer in action :) > > Seriously speaking, GtkUIManager is not and may not be perfect. The > single fact

Re: GtkBuilder status

2006-05-11 Thread Yevgen Muntyan
Yevgen Muntyan wrote: Johan Dahlin wrote: o My real concern is about supporting menus and toolbars built by UIManager. - Is the motivation here only a "time-to-market" thing ? - If so, do you have any plan or stratagy to take baby-steps and eventually get all the ui building code in

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Morten Welinder wrote: > When Glade is used with a recent gtk+, it will write lots of > properties that old gtk+ will not understand. Now if those > properties were non-default that is unavoidable, but it is > irritating that one cannot readily edit a glade file on a > new system without hand-patc

Re: GtkBuilder status

2006-05-11 Thread Yevgen Muntyan
Johan Dahlin wrote: o My real concern is about supporting menus and toolbars built by UIManager. - Is the motivation here only a "time-to-market" thing ? - If so, do you have any plan or stratagy to take baby-steps and eventually get all the ui building code into the IBuildable ? F

Re: GtkBuilder status

2006-05-11 Thread Morten Welinder
When Glade is used with a recent gtk+, it will write lots of properties that old gtk+ will not understand. Now if those properties were non-default that is unavoidable, but it is irritating that one cannot readily edit a glade file on a new system without hand-patching using Emacs afterwards. Ca

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Yevgen Muntyan wrote: > Johan Dahlin wrote: > >> >>> Over the last couple of weeks Henrique Romano and I have been working on >>> GtkBuilder, a UI constructor intended for inclusion in GTK+. >>> I'd like to discuss the API and some of the decisions before making the >>> code available for public r

Re: GtkBuilder status

2006-05-11 Thread Yevgen Muntyan
Kalle Vahlman wrote: On 5/11/06, Yevgen Muntyan <[EMAIL PROTECTED]> wrote: Why remove custom properties? They are still needed by applications, Out of curiosity, for what kind of functionality do _applications_ use custom properties for? I can understand if they are used in creation of the

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
Behdad Esfahbod wrote: > On Thu, 11 May 2006, Yevgen Muntyan wrote: > >> Johan Dahlin wrote: >> Over the last couple of weeks Henrique Romano and I have been working on GtkBuilder, a UI constructor intended for inclusion in GTK+. I'd like to discuss the API and some of the decisions

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
>>> Over the last couple of weeks Henrique Romano and I have been working on >>> GtkBuilder, a UI constructor intended for inclusion in GTK+. >>> I'd like to discuss the API and some of the decisions before making the >>> code available for public review. >>> >> >> I went ahead and attached the

Re: GtkBuilder status

2006-05-11 Thread Johan Dahlin
> o first of all, GtkBuilder & GtkBuildable is a nice api and seemingly > covers >the api that in libglade allowed you to provide a constructor > function and >a "build_children" function... this will be important for people to > port any >custom libglade extentions that were using the

Re: GtkBuilder status

2006-05-11 Thread Damon Chaplin
On Wed, 2006-05-10 at 11:51 -0300, Johan Dahlin wrote: > We've made a couple of important decisions: > * GMarkup based parser which parses and creates the object tree in one step > go instead of saprving a whole tree in memory. > * breaking xml format compatibility with libglade > * not supporti

Re: GtkBuilder status

2006-05-10 Thread Behdad Esfahbod
On Thu, 11 May 2006, Yevgen Muntyan wrote: > Johan Dahlin wrote: > > > > >> Over the last couple of weeks Henrique Romano and I have been working on > >> GtkBuilder, a UI constructor intended for inclusion in GTK+. > >> I'd like to discuss the API and some of the decisions before making the > >> c

Re: GtkBuilder status

2006-05-10 Thread Kalle Vahlman
On 5/11/06, Yevgen Muntyan <[EMAIL PROTECTED]> wrote: Why remove custom properties? They are still needed by applications, Out of curiosity, for what kind of functionality do _applications_ use custom properties for? I can understand if they are used in creation of the widgets by the loading sy

Re: GtkBuilder status

2006-05-10 Thread Yevgen Muntyan
Johan Dahlin wrote: Over the last couple of weeks Henrique Romano and I have been working on GtkBuilder, a UI constructor intended for inclusion in GTK+. I'd like to discuss the API and some of the decisions before making the code available for public review. I went ahead and attached the

Re: GtkBuilder status

2006-05-10 Thread Yevgen Muntyan
Johan Dahlin wrote: Over the last couple of weeks Henrique Romano and I have been working on GtkBuilder, a UI constructor intended for inclusion in GTK+. I'd like to discuss the API and some of the decisions before making the code available for public review. I went ahead and attached the

Re: GtkBuilder status

2006-05-10 Thread Matthias Clasen
o One concern I have is about the format, the way it seems... oddball cases are being treated by introducing new tags in the format specificly for those cases. Is there a reason they cant be handled as properties ? example: STRING, INT, INT entry1, entry2 My feeling

Re: GtkBuilder status

2006-05-10 Thread Tristan Van Berkom
On 5/10/06, Johan Dahlin <[EMAIL PROTECTED]> wrote: HiOver the last couple of weeks Henrique Romano and I have been working onGtkBuilder, a UI constructor intended for inclusion in GTK+.I'd like to discuss the API and some of the decisions before making the code available for public review.This wor

Re: GtkBuilder status

2006-05-10 Thread Johan Dahlin
Over the last couple of weeks Henrique Romano and I have been working on GtkBuilder, a UI constructor intended for inclusion in GTK+. I'd like to discuss the API and some of the decisions before making the code available for public review. I went ahead and attached the patch to http://bugzil

GtkBuilder status

2006-05-10 Thread Johan Dahlin
Hi Over the last couple of weeks Henrique Romano and I have been working on GtkBuilder, a UI constructor intended for inclusion in GTK+. I'd like to discuss the API and some of the decisions before making the code available for public review. This work has been sponsored by Nokia. We've made a c