Nabble - Gtk+

Re: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T20:40:38Z

On Nov 10, 2009, at 8:32 PM, Jon Cruz wrote:

>
> On Nov 10, 2009, at 5:44 PM, John Ralls wrote:
>
>> For those who prefer a web forum, we have one of those, too, at http://sourceforge.net/apps.sourceforge.net/phpbb/gtk-osx/
>> You need to sign up for a sourceforge userid to post to it.
>
> 404 on that forum link

Sorry, that's an old link. I'll have to update the Support page, too.
The correct link is:

http://sourceforge.net/apps/phpbb/gtk-osx/

Thanks for pointing it out.
Regards,
John Ralls
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T20:32:16Z

On Nov 10, 2009, at 5:44 PM, John Ralls wrote:

> For those who prefer a web forum, we have one of those, too, at http://sourceforge.net/apps.sourceforge.net/phpbb/gtk-osx/
> You need to sign up for a sourceforge userid to post to it.

404 on that forum link
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: RFC: Adding zlib dependency to libgio

2009-11-10T18:54:53Z

On Tue, Nov 10, 2009 at 1:41 PM, Matthias Clasen
<matthias.clasen@...> wrote:

> On Tue, Nov 10, 2009 at 5:49 AM, nf2 <nf2.email@...> wrote:
>> On Tue, Nov 10, 2009 at 9:44 AM, Alexander Larsson <alexl@...> wrote:
>>> On Mon, 2009-11-09 at 23:03 -0500, Paul Davis wrote:
>>>> On Mon, Nov 9, 2009 at 6:23 AM, Alexander Larsson <alexl@...> wrote:
>
>>
>> As I'm reading the word Gtk+ here more often: I still believe that a
>> VFS API shoudn't be tied to a certain UI toolkit. That would be
>> repeating the mistake KIO did. What about Mozilla, OpenOffice, VLC and
>> many many others. They should all link GIO (as a VFS API). If libgio
>> dupulicates too many things they already have, they might be put off.
>
> Whether GIO contains DBus support or not has probably minimal
> influence on whether or not Mozilla or OpenOffice use it. They make
> their own decisions about what platform to build on and don't care if
> you think they 'should' use anything in particular.
>

The modularity of the GLib stack is a nice feature. A lot of things
which are built with it, might be very useful far outside Gtk+ and
Gnome.

Let's take for instance libsoup: It already links GIO. But for what
sake would a HTTP client library always need to carry D-Bus around?

On my system, Gtk+ links 44 libraries. I guess one less or more won't
make any difference. Or, for instance a "gvfs-ls /" , which probably
has to load about 15 libraries, takes 0.03 seconds. Therfore - I
reckon - gathering unrelated APIs into a single *.so won't have any
significant performance benefit. But maybe i'm wrong.

Norbert
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T17:44:01Z

On Nov 10, 2009, at 4:57 PM, Shawn Bakhtiar wrote:

> Re-build using jhbuild today:
>

Let's not clutter up this list with support requests for Gtk-OSX. Gtk-
OSX has its own support mailing list at gtk-osx-users@...
; you can subscribe at http://lists.sourceforge.net/lists/listinfo/gtk-osx-users/subscribe

For those who prefer a web forum, we have one of those, too, at http://sourceforge.net/apps.sourceforge.net/phpbb/gtk-osx/
You need to sign up for a sourceforge userid to post to it.

Regards,
John Ralls

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: dynamic mixing of primitives and background picture - semi-gui cropping utility

2009-11-10T17:41:57Z

On Nov 8, 2009, at 10:33 AM, <postdoc38@...> <postdoc38@...>
wrote:

> Dear Fellow Gtk-Perlers,
>
> See my attempt below at a "semi-gui" jpeg lossless cropping utility.
> The idea is to cycle through a few jpeg files, set the crop
> parameters by moving sliders with cursor keys, and crop the jpegs
> losslessly.
>
> y program works, but is quite "Mc Gyver"-like. I go back and forth
> between perlmagick and gtk2 to draw the cropping rectangle and the
> background picture. And I think it should be much faster to only
> redraw the part of the background were the previous crop rectangle
> was instead of redrawing the whole thing.
>
> A little background: I have a bunch of jpegs that are produced by a
> camera shooting old slides (a cheap and dirty slide scanner). They
> all have black borders due to the copying setup. What I aim is to
> check that the crop parameters are correct on each picture. Lossless
> means cropping along multiple of 8 pixels using appropriate outside
> utilities (namely jpegtran).
>
> Thanks for any input!
>
> Guillaume
>
>
> <auto_crop_new>

You forgot "use strict"!

Instead of hard-coded numeric key values, you should use the keysyms
hash.

use Gtk2::Gdk::Keysyms;

...

if ($key_val == $Gtk2::Gdk::Keysyms{GDK_Up}) {

etc

(Or mess around with an accelerator table. For what you have, the key
value processing is probably nice and user-friendly, as it requires no
modifier keys.)

Instead of connecting to delete-event to quit the main loop, do that
in the destroy signal on the window. Then you can add "q" as a
hotkey, and just call $window->destroy and the right thing will happen.

In auto_crop(), should check $color for an empty value; i loaded up a
jpeg that didn't have a border, and imagemagic returned no border for
it, causing a bunch of uninitialized value warnings.

Oh, actually my imagemagick is too old to support the special info:
name. You might want to do a version check on convert before doing
that stuff.

If you want to learn about gdk, you can try to do the drawing all in
the app instead of bouncing back and forth with imagemagick. You
could also take the slow route and draw on the pixels in perl, but
that's a bit icky.

--
It's all very complicated and would take a scientist to explain it.
-- MST3K

_______________________________________________
gtk-perl-list mailing list
gtk-perl-list@...
http://mail.gnome.org/mailman/listinfo/gtk-perl-list

From forum: Gtk+ - Perl

RE: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T16:57:54Z

Re-build using jhbuild today:

1) Take new OS X Snow Leopard out of box (Nice!)
2) Apply all updates and installed xCode
3) Download and install latest version of git from http://code.google.com/p/git-osx-installer
4) open terminal:
5) type "sh gtk-osx-build-setup.sh" (let run)
6) when done; type "export PATH=$PATH:/Users/shawn/.local/bin"
6.1) obviously use your own name instead of shawn (me :) )

7) type "jhbuild bootstrap" (12/12 completed no problem)
8) type "jhbuild build meta-gtk-osx-bootstrap" (13/13 completed with no problem)

9) type: "jhbuild build meta-gtk-osx-core"
9.1) ERROR: no libiconv but header used
9.2) Open a new shell
9.3) download the latest version of libiconv from http://www.gnu.org/software/libiconv
9.4) Unpack the files into source tree (for me : /Users/shawn/gtk/source );
9.5) in new shell type "jhbuild shell"; cd into the source
9.6) type "./configure --prefix=/Users/shawn/gtk/inst", which will configure it to go into the same place you GTK builds will go
9.7) type "make; make install"
9.8) return to 9.1 and simply re-run the configuration stage (I believe choice # 7)

10) ige-mac-integration warns about nested and implicit decelerations ....
10.1) start shell by choosing 4
10.2) Edit the src/Makefile.am and remove $(WARN_CFLAGS)
10.3) Re-run stage configuration (Option # 7)
10.4) 11/11 completed no problems

11) type "jhbuild build libglade" (completed no problems)

Optional (I need MySQL):
12) download source form mysql site and unpack in source directory
13) type "configure --prefix=/User/shawn/gtk/inst"
14) type "make; make install" (have a cig)

I am able to fully compile my application with no problems. However I am running into a serious issue, and I don't know if it is my crappy programming or if this is a real issue on the 64bit, all of a sudden there are tons of execution errors, for example,

<--in isidisplay.c -->

GtkWidget *
isi_display_get_widget(IsiDisplay *self, const gchar* object)
{

     /* Sanity Check */
    g_return_val_if_fail(self != NULL, FALSE);
    g_return_val_if_fail(self->priv != NULL, FALSE);
    g_return_val_if_fail(self->priv->dispose_has_run != TRUE, FALSE);

    /* Get teh XML File and load XML Object */
    return glade_xml_get_widget (self->priv->xml, object);

return NULL;}

<-- in isiapp.c-->

gboolean
isi_app_initialize_default_interface(IsiApp *self){

    /* Local vars */
    GdkWindow *g_window = NULL;


    GtkWindow *main_window        = NULL; /* Temp widget pointer for search otpions setup */

........

    main_window = isi_display_get_widget(self->display,"orderdesk");
    g_object_get((GObject*)main_window,"window",&g_window,NULL);

.......

}

crud out with the following error in gdb:

Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x0000000002019230
0x0000000100c343b0 in g_object_get (_object=0x2019230, first_property_name=0x1000973c5 "window") at gobject.c:1630
1630    g_return_if_fail (G_IS_OBJECT (object));
(gdb) bt
#0 0x0000000100c343b0 in g_object_get (_object=0x2019230, first_property_name=0x1000973c5 "window") at gobject.c:1630
#1 0x000000010001b7ff in isi_app_initialize_default_interface (self=0x10201a000) at isiapp.c:6209
#2 0x000000010000ae0a in main (argc=1, argv=0x7fff5fbfec08) at isimain.c:53
(gdb) bt
#0 0x0000000100c343b0 in g_object_get (_object=0x2019230, first_property_name=0x1000973c5 "window") at gobject.c:1630
#1 0x000000010001b7ff in isi_app_initialize_default_interface (self=0x10201a000) at isiapp.c:6209
#2 0x000000010000ae0a in main (argc=1, argv=0x7fff5fbfec08) at isimain.c:53

{scratching head dumbfounded} I comment out the above code and tried again, getting an error same thing but in a completely different area:

gboolean
isi_app_setup_menu(IsiApp *self)
{

    /* Local Variables */
    GtkWidget *menu = NULL;
    GtkWidget **widget = NULL;
    IsiCategory* c;

    guint num_rows;
    guint *p_index;             /* Index of top level cats */
    guint *c_index;            /* Index of child level cats */
    guint i,j,k;            /* Loop variables */

    /* Sanity Check */
    g_return_val_if_fail(self != NULL, FALSE);
    g_return_val_if_fail(self->priv != NULL, FALSE);
    g_return_val_if_fail(self->priv->dispose_has_run != TRUE, FALSE);

    /* If valid rows */
    if(self->priv->user_perms != NULL && self->priv->user_perms->rows != NULL){

        /* Get number of rows */
        guint num_rows = g_list_length(self->priv->user_perms->rows);

......

     }

/* By default return FALSE to permission checks */
return FALSE;
}

gdb output:

Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x00000000018639f8
0x000000010001a3c6 in isi_app_setup_menu (self=0x10201b000) at isiapp.c:2583
2583        if(self->priv->user_perms != NULL && self->priv->user_perms->rows != NULL){
(gdb) bt
#0 0x000000010001a3c6 in isi_app_setup_menu (self=0x10201b000) at isiapp.c:2583
#1 0x000000010001b816 in isi_app_initialize_default_interface (self=0x10201b000) at isiapp.c:6204
#2 0x000000010000ae0a in main (argc=1, argv=0x7fff5fbfec08) at isimain.c:53

I am going to try an do what Jralls suggest and perhaps re-build with both -g and see if I can not trace this better, also try to build the library as 32bit and see what happens than. But this application has been running great using this technique on the 32 bit interface with no problems what so ever! I can't figure out why I have all kinds of bad memory references.

I just noticed the there seems to be 32bit memory addresses intermingled with 64bit? IE in the above output self is 32bit but the address lookup is 64bit ?? Is that right? Same with the code??

Do I need to post this to gtk-app or is this something wrong in the backend (perhaps I am compiling with wrong options ??)

HELP :S :'S :"S :''''S

EMAILING FOR THE GREATER GOOD
Join me

Subject: Re: Gtk-OSX Frameworks (was: Why are developers...)
From: jacob@...
Date: Tue, 10 Nov 2009 18:20:26 +0100
CC: gtk-devel-list@...
To: shashaness@...

On Nov 10, 2009, at 4:46 PM, Shawn Bakhtiar wrote:

For building an application... I couldn't agree more, about the framework vs. jhbuild and autotools. You definitely want the latter. I like XCode's editor. when looking at source code (the colors man the colors). It also has a lot of nice features such as collapsible sections, an intuitive way of knowing if you {} are correct, as well as a jump to function feature that list all functions in the current file in a drop down menu. However, you can use the editor, and build in shell (jhbuild shell). In any case, gdb is a much better debugger to boot.

But yeah.. just try to build mysql with it, or even use it in a build. Good luck!!

Also using the ige-mac-bundler, users now simple drag and drop the latest package (application) to their application folder, and they are done, especially if you adhere to the XDG file system.

I don't know what all the complaint is about... I have been using the jhbuild scripts with little to no problems. I have had a few dependency issues but nothing that can not be figured out with a little reading of the script itself and attention to what I am doing. In any case, anything that is missing, simple download to source directory, and build inside the jhbuild shell, your done!

Like I said, I'm not too good with the back-end stuff, but it looks like I will be getting my own Snow Leopard today, I can re-run the jhbuild stuff from scratch, and see if I can't get a framework out. Would this help?

That would be great!

I've been trying to build it on Snow Leopard, butI i'm stuck now with

jhbuild meta-gtk-osx-core failing to build ige-mac-integration:

*** Building ige-mac-integration *** [10/11]

make

make all-recursive

Making all in src

if /bin/sh ../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -Wall -Wunused -Wchar-subscripts -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wcast-align -std=c99 -Wno-sign-compare -Wno-pointer-sign -Werror -I/Users/dacobi/gtk/inst/include -I/Users/dacobi/gtk/inst/include/gtk-2.0 -I/Users/dacobi/gtk/inst/lib/gtk-2.0/include -I/Users/dacobi/gtk/inst/include/atk-1.0 -I/Users/dacobi/gtk/inst/include/cairo -I/Users/dacobi/gtk/inst/include/pango-1.0 -I/Users/dacobi/gtk/inst/include/glib-2.0 -I/Users/dacobi/gtk/inst/lib/glib-2.0/include -I/Users/dacobi/gtk/inst/include/pixman-1 -I/Users/dacobi/gtk/inst/include/freetype2 -I/Users/dacobi/gtk/inst/include/libpng12 -xobjective-c -g -O2 -MT libigemacintegration_la-ige-mac-menu.lo -MD -MP -MF ".deps/libigemacintegration_la-ige-mac-menu.Tpo" -c -o libigemacintegration_la-ige-mac-menu.lo `test -f 'ige-mac-menu.c' || echo './'`ige-mac-menu.c; \

then mv -f ".deps/libigemacintegration_la-ige-mac-menu.Tpo" ".deps/libigemacintegration_la-ige-mac-menu.Plo"; else rm -f ".deps/libigemacintegration_la-ige-mac-menu.Tpo"; exit 1; fi

libtool: compile: gcc -DHAVE_CONFIG_H -I. -I. -I.. -I.. -Wall -Wunused -Wchar-subscripts -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wcast-align -std=c99 -Wno-sign-compare -Wno-pointer-sign -Werror -I/Users/dacobi/gtk/inst/include -I/Users/dacobi/gtk/inst/include/gtk-2.0 -I/Users/dacobi/gtk/inst/lib/gtk-2.0/include -I/Users/dacobi/gtk/inst/include/atk-1.0 -I/Users/dacobi/gtk/inst/include/cairo -I/Users/dacobi/gtk/inst/include/pango-1.0 -I/Users/dacobi/gtk/inst/include/glib-2.0 -I/Users/dacobi/gtk/inst/lib/glib-2.0/include -I/Users/dacobi/gtk/inst/include/pixman-1 -I/Users/dacobi/gtk/inst/include/freetype2 -I/Users/dacobi/gtk/inst/include/libpng12 -xobjective-c -g -O2 -MT libigemacintegration_la-ige-mac-menu.lo -MD -MP -MF .deps/libigemacintegration_la-ige-mac-menu.Tpo -c ige-mac-menu.c -fno-common -DPIC -o .libs/libigemacintegration_la-ige-mac-menu.o

cc1obj: warnings being treated as errors

ige-mac-menu.c: In function ‘menu_flash_off_cb’:

ige-mac-menu.c:77: warning: implicit declaration of function ‘FlashMenuBar’

ige-mac-menu.c:77: warning: nested extern declaration of ‘FlashMenuBar’

ige-mac-menu.c: In function ‘carbon_menu_free’:

ige-mac-menu.c:139: warning: implicit declaration of function ‘DisposeMenu’

ige-mac-menu.c:139: warning: nested extern declaration of ‘DisposeMenu’

ige-mac-menu.c: In function ‘carbon_menu_item_free’:

ige-mac-menu.c:182: warning: implicit declaration of function ‘DeleteMenuItem’

ige-mac-menu.c:182: warning: nested extern declaration of ‘DeleteMenuItem’

ige-mac-menu.c: In function ‘carbon_menu_item_get_checked’:

ige-mac-menu.c:294: warning: implicit declaration of function ‘GetMenuItemProperty’

ige-mac-menu.c:294: warning: nested extern declaration of ‘GetMenuItemProperty’

ige-mac-menu.c: In function ‘carbon_menu_item_update_state’:

ige-mac-menu.c:337: warning: implicit declaration of function ‘ChangeMenuItemAttributes’

ige-mac-menu.c:337: warning: nested extern declaration of ‘ChangeMenuItemAttributes’

ige-mac-menu.c: In function ‘carbon_menu_item_update_active’:

ige-mac-menu.c:347: warning: implicit declaration of function ‘CheckMenuItem’

ige-mac-menu.c:347: warning: nested extern declaration of ‘CheckMenuItem’

ige-mac-menu.c: In function ‘carbon_menu_item_update_submenu’:

ige-mac-menu.c:361: warning: implicit declaration of function ‘SetMenuItemHierarchicalMenu’

ige-mac-menu.c:361: warning: nested extern declaration of ‘SetMenuItemHierarchicalMenu’

ige-mac-menu.c:367: warning: implicit declaration of function ‘CreateNewMenu’

ige-mac-menu.c:367: warning: nested extern declaration of ‘CreateNewMenu’

ige-mac-menu.c:373: warning: implicit declaration of function ‘SetMenuTitleWithCFString’

ige-mac-menu.c:373: warning: nested extern declaration of ‘SetMenuTitleWithCFString’

ige-mac-menu.c: In function ‘carbon_menu_item_update_label’:

ige-mac-menu.c:394: warning: implicit declaration of function ‘SetMenuItemTextWithCFString’

ige-mac-menu.c:394: warning: nested extern declaration of ‘SetMenuItemTextWithCFString’

ige-mac-menu.c: In function ‘carbon_menu_item_update_accelerator’:

ige-mac-menu.c:417: warning: implicit declaration of function ‘SetMenuItemModifiers’

ige-mac-menu.c:417: warning: nested extern declaration of ‘SetMenuItemModifiers’

ige-mac-menu.c:423: warning: implicit declaration of function ‘SetMenuItemCommandKey’

ige-mac-menu.c:423: warning: nested extern declaration of ‘SetMenuItemCommandKey’

ige-mac-menu.c: In function ‘carbon_menu_item_create’:

ige-mac-menu.c:588: warning: implicit declaration of function ‘InsertMenuItemTextWithCFString’

ige-mac-menu.c:588: warning: nested extern declaration of ‘InsertMenuItemTextWithCFString’

ige-mac-menu.c:592: warning: implicit declaration of function ‘SetMenuItemProperty’

ige-mac-menu.c:592: warning: nested extern declaration of ‘SetMenuItemProperty’

ige-mac-menu.c: In function ‘nsevent_handle_menu_key’:

ige-mac-menu.c:724: warning: implicit declaration of function ‘IsMenuKeyEvent’

ige-mac-menu.c:724: warning: nested extern declaration of ‘IsMenuKeyEvent’

ige-mac-menu.c:727: warning: implicit declaration of function ‘GetMenuItemCommandID’

ige-mac-menu.c:727: warning: nested extern declaration of ‘GetMenuItemCommandID’

ige-mac-menu.c:740: warning: implicit declaration of function ‘GetMenuID’

ige-mac-menu.c:740: warning: nested extern declaration of ‘GetMenuID’

ige-mac-menu.c:742: warning: implicit declaration of function ‘GetMenuEventTarget’

ige-mac-menu.c:742: warning: nested extern declaration of ‘GetMenuEventTarget’

ige-mac-menu.c:742: warning: passing argument 2 of ‘SendEventToEventTarget’ makes pointer from integer without a cast

ige-mac-menu.c: In function ‘sync_menu_shell’:

ige-mac-menu.c:921: warning: implicit declaration of function ‘GetMenuAttributes’

ige-mac-menu.c:921: warning: nested extern declaration of ‘GetMenuAttributes’

ige-mac-menu.c:927: warning: implicit declaration of function ‘ChangeMenuAttributes’

ige-mac-menu.c:927: warning: nested extern declaration of ‘ChangeMenuAttributes’

ige-mac-menu.c: In function ‘parent_set_emission_hook_remove’:

ige-mac-menu.c:988: warning: implicit declaration of function ‘ClearMenuBar’

ige-mac-menu.c:988: warning: nested extern declaration of ‘ClearMenuBar’

ige-mac-menu.c:989: warning: implicit declaration of function ‘DeleteMenu’

ige-mac-menu.c:989: warning: nested extern declaration of ‘DeleteMenu’

ige-mac-menu.c: In function ‘window_focus’:

ige-mac-menu.c:1001: warning: implicit declaration of function ‘SetRootMenu’

ige-mac-menu.c:1001: warning: nested extern declaration of ‘SetRootMenu’

ige-mac-menu.c: In function ‘ige_mac_menu_set_quit_menu_item’:

ige-mac-menu.c:1068: warning: implicit declaration of function ‘GetIndMenuItemWithCommandID’

ige-mac-menu.c:1068: warning: nested extern declaration of ‘GetIndMenuItemWithCommandID’

ige-mac-menu.c:1071: warning: implicit declaration of function ‘SetMenuItemCommandID’

ige-mac-menu.c:1071: warning: nested extern declaration of ‘SetMenuItemCommandID’

make[2]: *** [libigemacintegration_la-ige-mac-menu.lo] Error 1

make[1]: *** [all-recursive] Error 1

make: *** [all] Error 2

*** Error during phase build of ige-mac-integration: ########## Error running make *** [10/11]

Anyone?

/Jacob

EMAILING FOR THE GREATER GOOD
Join me

> From: jralls@...
> Subject: Re: Gtk-OSX Frameworks (was: Why are developers...)
> Date: Tue, 10 Nov 2009 07:10:09 -0800
> To: gtk-devel-list@...
>
>
> On Nov 10, 2009, at 4:16 AM, Paul Davis wrote:
>
> > On Mon, Nov 9, 2009 at 1:10 PM, Jack Skellington <dacobi@...>
> > wrote:
> >
> >> Also if a native Gtk+ OS X framework were available people who are
> >> maintaining Gtk+ apps would have the option to extend their user base
> >> to OS X quite quickly.
> >
> > All it requires to use it is to build the GTK stack yourself using the
> > instructions provided (i wish the instructions had not been moved away
> > from gnome.org, but they are still easy to find). I should put "all"
> > in quotes because if you choose to follow bleeding edge GTK then you
> > will find that maintaining your built version can be quite a lot of
> > work in the face of breakages and changes in many different parts of
> > the stack. This is not true (or at least, it is MUCH less true) if you
> > use a recent release version (there are instructions on how to do that
> > included in the gtk osx build info).
> >
> > I do not believe that using a pre-built GTK OS X framework is
> > desirable for users or developers. Include GTK as part of your app
> > bundle. Its not hard to do and gives you complete control over which
> > version of GTK is used by your app. We do this for Ardour (and now
> > Mixbus) (screenshots at http://ardour.org/ and
> > http://mixbus.harrisonconsoles.com/). Users download a single app, and
> > run it. No framework installation or maintainance.
>
> I haven't built and made available updated frameworks because the
> approach Richard used to create the first one (still hanging around on gtk-osx.org
> , as previously noted elsewhere) didn't produce a result that I think
> I can support. I have in mind a modification of ige-mac-bundler which
> I think will provide better results, but other tasks have higher
> priority at the moment.
>
> Some others, including me, have done some work on the gtk-osx-
> frameworks, and the network graph at github shows that my tree (http://github.com/jralls/gtk-osx-framework
> ) is current with all of them . Do be aware that there are 3 branches,
> of which master is probably the only one which will get you close
> enough to use.
>
> I also agree with Paul here: The Apple Framework idiom doesn't make
> sense for cross-platform development. It uses special #include syntax
> and special linking. It doesn't play well with pkg-config. Yes, those
> are solvable problems, but why? The regular gnu autotools work very
> well indeed on OSX, and one needs to use it anyway for building on
> Linux. Why deal with another build chain just for the dubious benefit
> of using XCode instead of emacs or vim?
>
> Add to that that it's hard to build a non-trivial application using
> only gtk+. You're going to need a bunch of other libraries, either
> from gnome, gnu, or freedesktop. They're not going to be in
> Frameworks, so you're going to have to integrate them the autotools
> way, so what do you gain from having gtk+ in a set of frameworks?
>
> There is one exception to which I'm quite sympathetic: PyGtk. At
> present building a downloadable PyGtk app bundle is a royal pain, and
> a PyGtk framework *might* be part of the solution.
>
> Regards,
> John Ralls
>
> _______________________________________________
> gtk-devel-list mailing list
> gtk-devel-list@...
> http://mail.gnome.org/mailman/listinfo/gtk-devel-list
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

gdk_pixbuf_composite leaves artifacts in new image

2009-11-10T16:54:03Z

Hello list,

I am trying to create a thumbnail from a digital image, which should not
exceed 128 px width or height. After loading the image, I apply the EXIF
orientation information, if available. In order to prettify the
resulting display, I would like to place the pixbuf horizontally
centered onto another transparent pixbuf with dimensions 128x128 px.
Here is (part of) the code I use (removed the error checking for
simplification):

gchar *imgSrcName;
GdkPixbuf *src, *interm, *result;
guint width, height, xOffset;
int i = 0;
...
src = gdk_pixbuf_load_from_file_at_scale( imgSrcName, 128, 128, TRUE,
NULL );
result = gdk_pixbuf_new( GDK_COLORSPACE_RGB, TRUE,
gdk_pixbuf_get_bits_per_sample( src ), 128, 128 );

interm = gdk_pixbuf_apply_embedded_orientation( src );
gdk_pixbuf_unref( src );
width = gdk_pixbuf_get_width( interm );
height = gdk_pixbuf_get_height( interm );
while ( i < 2 )
{
gchar *save;

gdk_pixbuf_fill( result, 0xffffff00 );
xOffset = i++ * ( 128 - width ) / 2;
gdk_pixbuf_composite( interm, result, xOffset, 0, width,
height, 0, 0, 1.0, 1.0, GDK_INTERP_NEAREST, 255 );

save = g_strdup_printf( "test%d.jpg", i );
gdk_pixbuf_save( result, save, "jpeg", NULL, "quality", "100",
NULL );
g_free( save );
}
gdk_pixbuf_unref( interm );
gdk_pixbuf_unref( result );

The resulting images can be seen at:
[1] http://www.inselpix.com/img/833790624900.jpg
[2] http://www.inselpix.com/img/478403985121.jpg

As you can see, the result is OK only if the xOffset on translation is
0. The centered version shows artifacts on the right border of the
image. And it does not make a speck of difference, if the interpolation
type is _BILINEAR or _HYPER ...

Could anyone please help in solving this ?

Regards

Todor
_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: GPtrArray has no insert?

2009-11-10T16:50:45Z

Il giorno Mon, 09 Nov 2009 18:10:49 +0100
Martin Kalbfuß <ma.kalbfuss@...> ha scritto:

> So a GPtrArray would be the right choice. But I can't see any
> insertion function like the GArray has. Is such a function not
> present or is it simply not documented?
>
> And what happens when I try to access a non existent element with an
> out of range index? The documentation says nothing about this?

Hi,

there is no g_array_insert_val() equivalent for GPtrArray, you must
implement it by yourself. If you just want to write data, access directy
the pdata field or use g_ptr_array_index().

If you access out of bounds items you get the same results you encounter
with any array: plain garbage if you are reading and a segmentation
fault if you are writing (and lucky).

Ciao.
--
Nicola
_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: gdk_pixbuf_composite leaves artifacts in the dest pixbuf, if, dest_x is > 0. Help please?

2009-11-10T16:45:46Z

That did not work either, so I uploaded the 2 images on the Net and here
are the links for the interested:

[no artifacts] http://www.inselpix.com/img/833790624900.jpg

[artifacts] http://www.inselpix.com/img/478403985121.jpg

I hope someone can tell me, what I missed/did wrong when composing the
target image.

Kind regards,

Todor

On 11/10/2009 06:49 PM, Todor Todorov wrote:

> OK, something happened to the attached images, so let's see about
> inlining them ...
>
> offset is zero, no artifacts offset is zero, no artifacts
>
> offset is 16, artifacts at the right side of the imageoffset is 16,
> artifacts at the right side of the image
>
> On 11/10/2009 06:39 PM, Todor Todorov wrote:
>> Hello list,
>>
>> I am trying to create a thumbnail from a digital image, which should
>> not exceed 128 px width or height. After loading the image, I apply
>> the EXIF orientation information, if available. In order to prettify
>> the resulting display, I would like to place the pixbuf horizontally
>> centered onto another transparent pixbuf with dimensions 128x128 px.
>> Here is (part of) the code I use (removed the error checking for
>> simplification):
>>
>> gchar *imgSrcName;
>> GdkPixbuf *src, *interm, *result;
>> guint width, height, xOffset;
>> int i = 0;
>> ...
>> src = gdk_pixbuf_load_from_file_at_scale( imgSrcName, 128, 128, TRUE,
>> NULL );
>> result = gdk_pixbuf_new( GDK_COLORSPACE_RGB, TRUE,
>> gdk_pixbuf_get_bits_per_sample( src ), 128, 128 );
>>
>> interm = gdk_pixbuf_apply_embedded_orientation( src );
>> gdk_pixbuf_unref( src );
>> width = gdk_pixbuf_get_width( interm );
>> height = gdk_pixbuf_get_height( interm );
>> while ( i < 2 )
>> {
>> gchar *save;
>>
>> gdk_pixbuf_fill( result, 0xffffff00 );
>> xOffset = i++ * ( 128 - width ) / 2;
>> gdk_pixbuf_composite( interm, result, xOffset, 0, width,
>> height, 0, 0, 1.0, 1.0, GDK_INTERP_NEAREST, 255 );
>>
>> save = g_strdup_printf( "test%d.jpg", i );
>> gdk_pixbuf_save( result, save, "jpeg", NULL, "quality",
>> "100", NULL );
>> g_free( save );
>> }
>> gdk_pixbuf_unref( interm );
>> gdk_pixbuf_unref( result );
>>
>> As you can see in the attached images, the result is OK only if the
>> xOffset on translation is 0. The centered version shows artifacts on
>> the right border of the image. And it does not make a speck of
>> difference, if the interpolation type is _BILINEAR or _HYPER ...
>>
>> Could anyone please help in solving this ?
>>
>> Regards
>>
>> Todor
>>
>>
>> _______________________________________________
>> gtk-app-devel-list mailing list
>> gtk-app-devel-list@...
>> http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list
> _______________________________________________
> gtk-app-devel-list mailing list
> gtk-app-devel-list@...
> http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list
>

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

GDBus in gio? [was Re: Minutes of the GTK+ Team Meeting - 2009-11-10]

2009-11-10T16:22:24Z

Emmanuele Bassi wrote:
> 1. bug 600141 (desrt)
> - dbus-1 brings in pthread
> - gio-2.0 doesn't
> - dlopen()-ing a shared object linking to pthread from a library
> that doesn't is a big no-no which kind of works in Linux but
> breaks things like gdb
> - big hammer: make gio-2.0 depend on gthread-2.0
> - small hammer: make gio-2.0 pkg-config file include -pthread
> ACTION: desrt should involve gtk-devel-list and gather feedback

Is there a reason the dbus stuff won't become its own glib module along
side gio, gobject, gthread etc...?

Please forgive me if I'm asking something that's been asked before. I'm
searching through my archives of this mailing list...

Cheers,

Stef

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: Converting clib FILE stream in GFileInputStream

2009-11-10T16:11:33Z

On 11/10/2009 06:32 PM, Martin Kalbfuß wrote:
>
> Wouldn't that be platform specific? I'm new to the GTK+ world so I'm
> still confused about which file to use. Currently that's even not that
> important, because the other library I use has it's own stream type like
> GIO has and it hasn't the ability to convert it to a standard file
> stream. So mixing the two APIs isn't possible. What a shame.

Sure you can. Just create a GMemoryInputStream or GMemoryOutputStream
depending on the direction of the data flow between the two APIs.

If using GMemoryInputStream, in the write callback from your other API, you
call g_memory_input_stream_add_data() (perhaps on a copy of the data).

If using GMemoryOutputStream, humm, seems like the API doesn't provide a
callback for this one.

Regardless, you can definitely subclass GInputStream / GOutputStream yourself
to achieve what you want.

behdad
_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: gdk_pixbuf_composite leaves artifacts in the dest pixbuf, if, dest_x is > 0. Help please?

2009-11-10T16:10:11Z

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: gdk_pixbuf_composite leaves artifacts in the dest pixbuf, if, dest_x is > 0. Help please?

2009-11-10T15:49:15Z

OK, something happened to the attached images, so let's see about
inlining them ...

offset is zero, no artifacts offset is zero, no artifacts

offset is 16, artifacts at the right side of the imageoffset is 16,
artifacts at the right side of the image

On 11/10/2009 06:39 PM, Todor Todorov wrote:

> Hello list,
>
> I am trying to create a thumbnail from a digital image, which should
> not exceed 128 px width or height. After loading the image, I apply
> the EXIF orientation information, if available. In order to prettify
> the resulting display, I would like to place the pixbuf horizontally
> centered onto another transparent pixbuf with dimensions 128x128 px.
> Here is (part of) the code I use (removed the error checking for
> simplification):
>
> gchar *imgSrcName;
> GdkPixbuf *src, *interm, *result;
> guint width, height, xOffset;
> int i = 0;
> ...
> src = gdk_pixbuf_load_from_file_at_scale( imgSrcName, 128, 128, TRUE,
> NULL );
> result = gdk_pixbuf_new( GDK_COLORSPACE_RGB, TRUE,
> gdk_pixbuf_get_bits_per_sample( src ), 128, 128 );
>
> interm = gdk_pixbuf_apply_embedded_orientation( src );
> gdk_pixbuf_unref( src );
> width = gdk_pixbuf_get_width( interm );
> height = gdk_pixbuf_get_height( interm );
> while ( i < 2 )
> {
> gchar *save;
>
> gdk_pixbuf_fill( result, 0xffffff00 );
> xOffset = i++ * ( 128 - width ) / 2;
> gdk_pixbuf_composite( interm, result, xOffset, 0, width,
> height, 0, 0, 1.0, 1.0, GDK_INTERP_NEAREST, 255 );
>
> save = g_strdup_printf( "test%d.jpg", i );
> gdk_pixbuf_save( result, save, "jpeg", NULL, "quality", "100",
> NULL );
> g_free( save );
> }
> gdk_pixbuf_unref( interm );
> gdk_pixbuf_unref( result );
>
> As you can see in the attached images, the result is OK only if the
> xOffset on translation is 0. The centered version shows artifacts on
> the right border of the image. And it does not make a speck of
> difference, if the interpolation type is _BILINEAR or _HYPER ...
>
> Could anyone please help in solving this ?
>
> Regards
>
> Todor
>
>
> _______________________________________________
> gtk-app-devel-list mailing list
> gtk-app-devel-list@...
> http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: PyGTK site down

2009-11-10T15:43:10Z

...but is now up again.

_______________________________________________
pygtk mailing list pygtk@...
http://www.daa.com.au/mailman/listinfo/pygtk
Read the PyGTK FAQ: http://faq.pygtk.org/

From forum: Gtk+ - Python

gdk_pixbuf_composite leaves artifacts in the dest pixbuf, if,dest_x is > 0. Help please?

2009-11-10T15:39:48Z

Hello list,

I am trying to create a thumbnail from a digital image, which should not
exceed 128 px width or height. After loading the image, I apply the EXIF
orientation information, if available. In order to prettify the
resulting display, I would like to place the pixbuf horizontally
centered onto another transparent pixbuf with dimensions 128x128 px.
Here is (part of) the code I use (removed the error checking for
simplification):

gchar *imgSrcName;
GdkPixbuf *src, *interm, *result;
guint width, height, xOffset;
int i = 0;
...
src = gdk_pixbuf_load_from_file_at_scale( imgSrcName, 128, 128, TRUE,
NULL );
result = gdk_pixbuf_new( GDK_COLORSPACE_RGB, TRUE,
gdk_pixbuf_get_bits_per_sample( src ), 128, 128 );

interm = gdk_pixbuf_apply_embedded_orientation( src );
gdk_pixbuf_unref( src );
width = gdk_pixbuf_get_width( interm );
height = gdk_pixbuf_get_height( interm );
while ( i < 2 )
{
gchar *save;

gdk_pixbuf_fill( result, 0xffffff00 );
xOffset = i++ * ( 128 - width ) / 2;
gdk_pixbuf_composite( interm, result, xOffset, 0, width,
height, 0, 0, 1.0, 1.0, GDK_INTERP_NEAREST, 255 );

save = g_strdup_printf( "test%d.jpg", i );
gdk_pixbuf_save( result, save, "jpeg", NULL, "quality", "100",
NULL );
g_free( save );
}
gdk_pixbuf_unref( interm );
gdk_pixbuf_unref( result );

As you can see in the attached images, the result is OK only if the
xOffset on translation is 0. The centered version shows artifacts on the
right border of the image. And it does not make a speck of difference,
if the interpolation type is _BILINEAR or _HYPER ...

Could anyone please help in solving this ?

Regards

Todor

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: Converting clib FILE stream in GFileInputStream

2009-11-10T14:38:22Z

Am Montag, den 09.11.2009, 15:05 -0600 schrieb Shaun McCance:

> On Mon, 2009-11-09 at 21:44 +0100, Martin Kalbfuß wrote:
> > Hi again.
> >
> > Is it possible to convert a stadnard FILE stream into an
> > GFileInputStream and vice versa? The allegro library has functions to
> > read/write bitmaps in different formats to/from a FILE stream. I would
> > like to use this functions on GFileStreams.
>
> Do you really need a GFileInputStream, or just some sort of
> GInputStream? To get an input stream from a file descriptor,
> there's GUnixInputStream:
>
> http://library.gnome.org/devel/gio/stable/GUnixInputStream.html
>
> --
> Shaun
>
>

Thanks Shaun.

Wouldn't that be platform specific? I'm new to the GTK+ world so I'm
still confused about which file to use. Currently that's even not that
important, because the other library I use has it's own stream type like
GIO has and it hasn't the ability to convert it to a standard file
stream. So mixing the two APIs isn't possible. What a shame.

_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: Minutes of the GTK+ Team Meeting - 2009-11-10

2009-11-10T14:37:50Z

On Tue, 2009-11-10 at 17:19 -0500, Behdad Esfahbod wrote:

> On 11/10/2009 04:45 PM, Emmanuele Bassi wrote:
> > 4. text-buffer 3.0 request (jessevdk)
> > - split TextView: single TextBuffer driving two TextView widgets
> > - there are problems with selection and cursor handling
> > - move some things from the TextBuffer to the TextView, like the
> > new EntryBuffer in gtk+ 2.18 does
> > - worth supporting, targeting 3.0 - might cause deprecations
> > during the 2.19/2.20 cycle
>
> I'm not sure what this is about exactly. But something that I thought about
> working on vte (and specifically, thinking about breaking vte into model/view)
> is, if GtkTextBuffer and GtkTextView where proper interfaces, VteTerminal and
> VteTerminalView could implement them. In fact, VteTerminalView could be
> implemented only if GtkTextView was too slow for us. This could give us lots
> of neat stuff I imagine.

The maintenance on GtkTextView over the last 5+ years has been very
slight ... (150 open bugs at the moment)

I'd rather see someone pick that up rather starting on ambitious
rewrites.

- Owen

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

do the container types use g_malloc and co ?

2009-11-10T14:31:37Z

I wonder if it it's possible not to automatically abort the program if
there's not enough memory. Not that it's a bad behavior. But I'm mixing
to different apis and the other lib doesn't use the glib memory
functions. I would prefer to have a consistent behavior.

And what will show up when this happens?
If I would know I could exit by myself with the same results. When
there's not enough memory.

Thanks.

_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: GPtrArray has no insert?

2009-11-10T14:25:18Z

No one knows?

Especially the behavior on an out of range index would be interesting. I
would like to know if I have to implement a check myself.

Thanks

_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: Minutes of the GTK+ Team Meeting - 2009-11-10

2009-11-10T14:19:03Z

On 11/10/2009 04:45 PM, Emmanuele Bassi wrote:
> 4. text-buffer 3.0 request (jessevdk)
> - split TextView: single TextBuffer driving two TextView widgets
> - there are problems with selection and cursor handling
> - move some things from the TextBuffer to the TextView, like the
> new EntryBuffer in gtk+ 2.18 does
> - worth supporting, targeting 3.0 - might cause deprecations
> during the 2.19/2.20 cycle

I'm not sure what this is about exactly. But something that I thought about
working on vte (and specifically, thinking about breaking vte into model/view)
is, if GtkTextBuffer and GtkTextView where proper interfaces, VteTerminal and
VteTerminalView could implement them. In fact, VteTerminalView could be
implemented only if GtkTextView was too slow for us. This could give us lots
of neat stuff I imagine.

behdad
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Minutes of the GTK+ Team Meeting - 2009-10-27

2009-11-10T14:07:17Z

= minutes for the 2009-10-27 meeting =

1. fundamental types for gint16/guint16 (bug: 562498)
- initial request for D-Bus type system mapping
- also requested by Vala
- timj against the inclusion, and davidz and desrt agree that it's not
required for GDbus at the moment

2. use target milestone field in bugzilla
- mostly for QA, but requires help from developers
- facilitates queries for tracking new features/relevant bug fixes
- http://live.gnome.org/GTK%2B/BugzillaLinks - relevant queries for
outstanding patches

3. features set for gtk-next
- quick summary for 2.20: spinner widget and cell renderer
- tbf to brush up the tool palette before the end of november
- mclasen to push the tooltip drawing/positioning patches that made F12
- kris to review them
- bratsche asks for API to get the primary display
- mclasen agrees
- setter UI is still MIA
- missing accessors for 2.20

4. features set for glib-next
- kalikiana points out the diagnostic mode (runtime detection of
deprecated
properties and signals)
- still not well defined
- class private data
- pre-reviewed, timj to give final approval
- gvariant merge
- mclasen will review
- gdbus merge
- marked as 'further discussion needed'
- merge still planned for this cycle

next meeting: 2009-11-10

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: Building Gtk-OSX

2009-11-10T14:04:06Z

On Nov 10, 2009, at 11:50 AM, Jacob Juul Kolding wrote:

> On Nov 10, 2009, at 7:32 PM, John Ralls wrote:
>
>> Please rerun gtk-osx-build-install.sh to get the most recent
>> jhbuildrc. You'll have to build 32-bit to use ige-mac-integration
>> (it uses Carbon), but the latest jhbuildrc skips it for you if you
>> build for 64-bit. (For now, you can just abandon the module;
>> everything else is built.)
>
> But how do I build the framework or other apps without the ige stuff?
>

Just add

skip.append("ige-mac-integration")

to .jhbuildrc-custom and press on.

You're a bit on your own with 64-bit past gtk+, though. There may be
other libraries that choke; I can't test them all (and I haven't yet
even done PyGtk, though that's on my todo list). Note that modulesets-
stable and modulesets won't work in 64-bit because pango is held back
for Tiger compatibility.

You might find the results from adding

setup_sdk("10.6", "10.6", "i386")

to your jhbuildrc-custom INSTEAD and then starting over more
satisfying. That will build a fully-functional 32-bit build, which is
known to work for most everything.

Note well that the framework jhbuildrc files haven't been updated to
work with Snow Leopard, so you'll have to modify those to work
yourself if you're really intent on building a framework.

Regards,
John Ralls
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Ctrl-Click keyboard mouse event

2009-11-10T13:56:56Z

Hi list,
When a button press event is triggered, what is the proper way to know
if the user is pressing down the Ctrl key?

Tipically I would want to know this inside a button press event callback:

int handle_button_press (GtkWidget *widget,
GdkEventButton *event, void *data)
{
mouse button has been clicked: is Ctrl key down???
...
}

(I need this to implement multiple selections from my OpenGL code that
are consistent with multiple selections in GtkTreeView)

Thanks,
Carlos
_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Minutes of the GTK+ Team Meeting - 2009-11-10

2009-11-10T13:45:43Z

= minutes for the 2009-11-10 meeting =

1. bug 600141 (desrt)
- dbus-1 brings in pthread
- gio-2.0 doesn't
- dlopen()-ing a shared object linking to pthread from a library
that doesn't is a big no-no which kind of works in Linux but
breaks things like gdb
- big hammer: make gio-2.0 depend on gthread-2.0
- small hammer: make gio-2.0 pkg-config file include -pthread
ACTION: desrt should involve gtk-devel-list and gather feedback

2. libmodel (desrt)
- generic model API
- potentially intersecting with the container/iterator API
- discussion about GObject-based data structures library

3. gvariant merge (desrt)
- pending on more reviews on gtk-devel-list

4. text-buffer 3.0 request (jessevdk)
- split TextView: single TextBuffer driving two TextView widgets
- there are problems with selection and cursor handling
- move some things from the TextBuffer to the TextView, like the
new EntryBuffer in gtk+ 2.18 does
- worth supporting, targeting 3.0 - might cause deprecations
during the 2.19/2.20 cycle

5. bug 601234
- GDK_Uhorn is a non-BMP keysym that landed in the compose sequences table
- downcast from uint32 (0x10001af) to uint16 (0x01af)
- a uint32 table for non-BMP Unicode values should be added to avoid
wasting the work done in shrinking the current uint16 table

next meeting: 2009-11-24, 20:00 UTC

log: http://live.gnome.org/GTK+/Meetings

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

SOLVED: Re: Problem with gtk_tree_selection_select_path: background color does not change

2009-11-10T13:30:56Z

This is nicely working for me now, I should have been doing something
wrong before,
Carlos

> Hi list,
> I need to select multiple rows in a treeview,
> directly from my code, in MULTIPLE selection mode.
> This is a very simple treemodel, only one column
> with a string.
>
> Unfortunately, although gtk_tree_selection_select_path()
> seems to select the indicated rows, the background color
> of the rows do not change.
>
> For example, running this code:
> path = gtk_tree_path_new_from_string ("1");
> gtk_tree_selection_select_path (selection, path);
> path = gtk_tree_path_new_from_string ("3");
> gtk_tree_selection_select_path (selection, path);
>
> and then:
> gtk_tree_selection_selected_foreach (selection,
> static_foreach_list, data);
>
> shows that the second (1) and fourt (3) rows have indeed
> been selected, but the background color remains unchanged...
>
> So my question is: how can I force GTK to change the background color
> of the selected rows?
>
> Many thanks for your help, I searched the archives for
> this mailling list, I did see this question posted, but
> not a working answer... the same for the web...
>
> Carlos
>
>

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T11:50:27Z

On Nov 10, 2009, at 7:32 PM, John Ralls wrote:

On Nov 10, 2009, at 9:20 AM, Jacob Juul Kolding wrote:

That would be great!
I've been trying to build it on Snow Leopard, butI i'm stuck now with
jhbuild meta-gtk-osx-core failing to build ige-mac-integration:

Please rerun gtk-osx-build-install.sh to get the most recent jhbuildrc. You'll have to build 32-bit to use ige-mac-integration (it uses Carbon), but the latest jhbuildrc skips it for you if you build for 64-bit. (For now, you can just abandon the module; everything else is built.)

But how do I build the framework or other apps without the ige stuff?

Jacob Kolding

dacobi@...

Regards,
John Ralls

_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

smime.p7s (2K) Download Attachment

From forum: Gtk+ - Dev - General

Re: gobject ref and unref ?

2009-11-10T10:34:54Z

That was pretty neat thanks.

On 11/10/09, Tadej Borovšak <tadeboro@...> wrote:

> Hello.
>
>> Is there any documentation available for it ? as to when to ref and
>> unref it based upon the api's ?
>
> You don't need to know those things, since object's total reference
> count is of no importance to you. All you need to do is keep track of
> your own references on an object. Let me explain this based on your
> example.
>
> When you create new store for GtkTreeView (be it GtkListStore or
> GtkTreeStore), you get an object that has a reference count of 1 and
> this reference is owned by you and is the only thing that is keeping
> you newly created object "alive". Now you add this store to the
> GtkTreeView, which is from now "co-owner" of the store. I
> intentionally co-owner in last sentence, because you don't need to
> know what exactly GtkTreeView does to claim ownership (it can add 1, 2
> or 10 references if it needs to). If you don't need created store for
> anything else, you can now remove your own reference from it, which
> essentially makes GtkTreeView the only "owner" of the store.
>
> You can see that no explicit reference counting wizardry is needed to
> properly count references. Just make sure your own references are
> removed when you don't need an object to stick around.
>
> Tadej
>
>
> --
> Tadej Borovšak
> tadeboro.blogspot.com
> tadeboro@...
> tadej.borovsak@...
>

--
Regards,
~Sid~
I have never met a man so ignorant that i couldn't learn something from him
_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: gobject ref and unref ?

2009-11-10T10:34:54Z

That was pretty neat thanks.

On 11/10/09, Tadej Borovšak <tadeboro@...> wrote:

--
Regards,
~Sid~
I have never met a man so ignorant that i couldn't learn something from him
_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: Gtk-OSX Frameworks (was: Why are developers...)

2009-11-10T10:32:46Z

On Nov 10, 2009, at 9:20 AM, Jacob Juul Kolding wrote:

>
> That would be great!
> I've been trying to build it on Snow Leopard, butI i'm stuck now with
> jhbuild meta-gtk-osx-core failing to build ige-mac-integration:
>

Please rerun gtk-osx-build-install.sh to get the most recent
jhbuildrc. You'll have to build 32-bit to use ige-mac-integration (it
uses Carbon), but the latest jhbuildrc skips it for you if you build
for 64-bit. (For now, you can just abandon the module; everything else
is built.)

Regards,
John Ralls
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

Re: gobject ref and unref ?

2009-11-10T10:32:00Z

Please don't cross-post.

On Tue, Nov 10, 2009 at 11:14:24PM +0530, Siddu wrote:
> How does one get to know if an API is adding a reference of its own to
> an object

References represent ownership. Hence a reference is added if you
provide an object to be taken and used later. Most methods that have
other object arguments beside `self' or `this' are of this kind.

gtk_container_add(container, child);

Read:
container takes child (child becomes used/owned by container) so, it
adds a reference.

gtk_container_child_get(container, child, ...);

Read:
something happens with container and child but their mutual ownership
remains unchanged so, when the function returns the number of references
remain unchanged (even though it can change during the call).

The difference should be pretty clear if you undestand the purpose of
the method you use.

There are a few somehwat counterintuitive methods such as
gtk_tree_model_get() but they always mention the ownership rules in the
documentation.

> gtk_tree_view_set_model() is adding a reference count of its own
> to the store.

According to the above rules.

> /* Set treeview's model */
> gtk_tree_view_set_model( GTK_TREE_VIEW( treeview ),
> GTK_TREE_MODEL( store ) );
> g_object_unref( G_OBJECT( store ) );

You do the unref only if you want to get rid of your own model
reference, i.e. you are fine with hte model going poof the instant the
tree view decides it no longer needs the model.

To confuse things, widgets and other GtkObjects can exist in a state
not owned by anything (just initially after creation). See the
documentation of GInitiallyUnowned for details.

Yeti

_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: gobject ref and unref ?

2009-11-10T10:32:00Z

Please don't cross-post.

On Tue, Nov 10, 2009 at 11:14:24PM +0530, Siddu wrote:
> How does one get to know if an API is adding a reference of its own to
> an object

References represent ownership. Hence a reference is added if you
provide an object to be taken and used later. Most methods that have
other object arguments beside `self' or `this' are of this kind.

gtk_container_add(container, child);

Read:
container takes child (child becomes used/owned by container) so, it
adds a reference.

gtk_container_child_get(container, child, ...);

Read:
something happens with container and child but their mutual ownership
remains unchanged so, when the function returns the number of references
remain unchanged (even though it can change during the call).

The difference should be pretty clear if you undestand the purpose of
the method you use.

There are a few somehwat counterintuitive methods such as
gtk_tree_model_get() but they always mention the ownership rules in the
documentation.

> gtk_tree_view_set_model() is adding a reference count of its own
> to the store.

According to the above rules.

> /* Set treeview's model */
> gtk_tree_view_set_model( GTK_TREE_VIEW( treeview ),
> GTK_TREE_MODEL( store ) );
> g_object_unref( G_OBJECT( store ) );

You do the unref only if you want to get rid of your own model
reference, i.e. you are fine with hte model going poof the instant the
tree view decides it no longer needs the model.

To confuse things, widgets and other GtkObjects can exist in a state
not owned by anything (just initially after creation). See the
documentation of GInitiallyUnowned for details.

Yeti

_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: gobject ref and unref ?

2009-11-10T10:23:05Z

Hello.

> Is there any documentation available for it ? as to when to ref and
> unref it based upon the api's ?

You don't need to know those things, since object's total reference
count is of no importance to you. All you need to do is keep track of
your own references on an object. Let me explain this based on your
example.

When you create new store for GtkTreeView (be it GtkListStore or
GtkTreeStore), you get an object that has a reference count of 1 and
this reference is owned by you and is the only thing that is keeping
you newly created object "alive". Now you add this store to the
GtkTreeView, which is from now "co-owner" of the store. I
intentionally co-owner in last sentence, because you don't need to
know what exactly GtkTreeView does to claim ownership (it can add 1, 2
or 10 references if it needs to). If you don't need created store for
anything else, you can now remove your own reference from it, which
essentially makes GtkTreeView the only "owner" of the store.

You can see that no explicit reference counting wizardry is needed to
properly count references. Just make sure your own references are
removed when you don't need an object to stick around.

Tadej

--
Tadej Borovšak
tadeboro.blogspot.com
tadeboro@...
tadej.borovsak@...
_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev

Re: gobject ref and unref ?

2009-11-10T10:23:05Z

Hello.

> Is there any documentation available for it ? as to when to ref and
> unref it based upon the api's ?

You don't need to know those things, since object's total reference
count is of no importance to you. All you need to do is keep track of
your own references on an object. Let me explain this based on your
example.

When you create new store for GtkTreeView (be it GtkListStore or
GtkTreeStore), you get an object that has a reference count of 1 and
this reference is owned by you and is the only thing that is keeping
you newly created object "alive". Now you add this store to the
GtkTreeView, which is from now "co-owner" of the store. I
intentionally co-owner in last sentence, because you don't need to
know what exactly GtkTreeView does to claim ownership (it can add 1, 2
or 10 references if it needs to). If you don't need created store for
anything else, you can now remove your own reference from it, which
essentially makes GtkTreeView the only "owner" of the store.

You can see that no explicit reference counting wizardry is needed to
properly count references. Just make sure your own references are
removed when you don't need an object to stick around.

Tadej

--
Tadej Borovšak
tadeboro.blogspot.com
tadeboro@...
tadej.borovsak@...
_______________________________________________
gtk-list mailing list
gtk-list@...
http://mail.gnome.org/mailman/listinfo/gtk-list

From forum: Gtk+ - General

Re: Why are no developers completing/maintaining native Gtk+ for OS X?

2009-11-10T10:04:52Z

On Tue, 10 Nov 2009, John Ralls wrote:

> On Nov 10, 2009, at 7:50 AM, Allin Cottrell wrote:
> >
> > On Tue, 10 Nov 2009, Paul Davis wrote:
> >
> >> I do not believe that using a pre-built GTK OS X framework is
> >> desirable for users or developers. Include GTK as part of your
> >> app bundle. Its not hard to do and gives you complete control
> >> over which version of GTK is used by your app.
> >
> > I think this confuses two issues. I include a pre-built GTK as
> > part of my app bundle for OS X, just as I do for MS Windows.
> > It is difficult for me to build the GTK stack for OS X myself
> > because I work on Linux and have only occasional access to an OS X
> > machine.
> >
> > With the win32 binaries available from gtk.org, one can select the
> > pre-built package that works best (i.e. "complete control over GTK
> > version"). It would be great to be able to do that with OS X too,
> > though of course I understand that Kris Rietveld casn't be
> > expected to do all the work this would involve!
>
> Interesting point. ISTM that a framework is perhaps not the best
> solution in this case. A tarball of the gtk+ and dependencies
> installation folder would be easy to provide and would integrate well
> with autotools. Would that help you? If so, what would be a good
> prefix to build it into?

Thanks for the reply, and these are good questions. I admit the
analogy with building for MS Windows is not perfect.

* For a Windows build I use pre-built GTK "dev" files (for a
cross-build on Linux), and distribute pre-built runtime files
(DLLs etc.) with my app.

* On OS X I use a pre-built GTK framework for both development and
distribution. I currently use the OS X/X11 build of GTK 2.14.3
available from the R project. The dmg includes the required
headers and pkgconfig files, so in building a GTK app all one has
to do is give the right PKG_CONFIG_PATH. End-users don't need the
"dev" files, but they get them anyway.

It might be nice to have distinct pre-built "dev" and runtime GTK
packages for OS X, and in that case it would probably make sense
for the dev package to take the form of a tarball rather than a
framework dmg. I guess my choice would be to root it in /opt/gtk
or some such, but I don't think that matters a great deal.

Allin Cottrell
_______________________________________________
gtk-devel-list mailing list
gtk-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-devel-list

From forum: Gtk+ - Dev - General

gobject ref and unref ?

2009-11-10T09:44:24Z

Hi all,

How does one get to know if an API is adding a reference of its own to
an object

Passed as a Parameter and thereby incrementing the ref_count ?

Is there any documentation available for it ? as to when to ref and
unref it based upon the api's ?

For example in the below code

gtk_tree_view_set_model() is adding a reference count of its own
to the store.

I have looked at the documentation for this , and it doesnt say
anything about it

store = gtk_list_store_new( );
for( i = 0; i < 5; i++ )
{

gtk_list_store_append( store, &iter );
gtk_list_store_set( store, &iter, ........
-1 );
}

/* Set treeview's model */
gtk_tree_view_set_model( GTK_TREE_VIEW( treeview ),
GTK_TREE_MODEL( store ) );
g_object_unref( G_OBJECT( store ) );

Thanks in advance

--
Regards,
~Sid~
I have never met a man so ignorant that i couldn't learn something from him
_______________________________________________
gtk-app-devel-list mailing list
gtk-app-devel-list@...
http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list

From forum: Gtk+ - Apps Dev